![\"Loading..\"](\"//bibbase.org/img/ajax-loader.gif\")
\n\n\n generated by\n \n ![\"bibbase.org\"\n](\"//bibbase.org/img/logo.svg\"\n "\\"BibBase")
\n \n
\n \n\n \n \n \n\n
\n\n\n - \n\n
- \n \n Group by\n \n \n
- \n
- \n Year\n \n
- \n Author\n \n
- \n Type\n \n
- \n Keyword\n \n
- \n Downloads\n \n
\n\n\n - \n \n \n \n \n
- \n
- \n \n Expand/Collapse All\n \n \n
- \n \n Download BibTeX\n \n \n
- \n \n RSS Feed\n \n \n
\n\n \n\n\n \n
\n\n Excellent! Next you can\n create a new website with this list, or\n embed it in an existing web page by copying & pasting\n any of the following snippets.\n\n
\n\n \n JavaScript\n (easiest)\n
\n \n
\n\n PHP\n \n <script src=\"https://bibbase.org/service/mendeley/ab671682-19b4-3dc7-aa8e-f674831dab33/group/14dc25e5-4c22-3a10-8988-092d628a7743?jsonp=1&jsonp=1\"></script>\n \n \n
\n\n iFrame\n (not recommended)\n \n <?php\n $contents = file_get_contents(\"https://bibbase.org/service/mendeley/ab671682-19b4-3dc7-aa8e-f674831dab33/group/14dc25e5-4c22-3a10-8988-092d628a7743?jsonp=1\");\n print_r($contents);\n ?>\n \n \n
\n\n \n <iframe src=\"https://bibbase.org/service/mendeley/ab671682-19b4-3dc7-aa8e-f674831dab33/group/14dc25e5-4c22-3a10-8988-092d628a7743?jsonp=1\"></iframe>\n \n \n For more details see the documention.\n
\n\n\n This is a preview! To use this list on your own web site\n or create a new web site from it,\n create a free account. The file will be added\n and you will be able to edit it in the File Manager.\n We will show you instructions once you've created your account.\n
\n\n \n\n
\n\n To the site owner:
\n\nAction required! Mendeley is changing its\n API. In order to keep using Mendeley with BibBase past April\n 14th, you need to:\n
- \n
- renew the authorization for BibBase on Mendeley, and \n
- update the BibBase URL\n in your page the same way you did when you initially set up\n this page.\n \n
\n \n \n Fix it now\n
\n\n \n \n
\n\n\n \n\n \n \n \n \n\n\n
\n\n \n \n 2024\n \n \n (1)\n \n \n
\n \n \n \n
\n \n \n\n \n \n \n \n \n \n A Study of ConvNeXt Architectures for Enhanced Image Captioning.\n \n \n \n \n\n\n \n Ramos, L.; Casas, E.; Romero, C.; Rivas-Echeverría, F.; and Morocho-Cayamcela, M., E.\n\n\n \n\n\n\n IEEE Access,1-1. 2024.\n \n\n\n\n
\n\n\n\n \n \n![\"A](\"//bibbase.org/img/filetypes/pdf.svg\"\n\t "\\"Paper\\"\n\t")
Paper\n \n \n \n ![\"A](\"//bibbase.org/img/filetypes/link.svg\"\n\t "\\"Website\\"\n\t")
Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n\n@article{\n title = {A Study of ConvNeXt Architectures for Enhanced Image Captioning},\n type = {article},\n year = {2024},\n pages = {1-1},\n websites = {https://ieeexplore.ieee.org/document/10410861/},\n id = {1b039c3d-a3bd-34d4-8ded-75adee8e6e32},\n created = {2024-01-23T01:26:58.734Z},\n accessed = {2024-01-22},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T14:05:10.857Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n private_publication = {false},\n bibtype = {article},\n author = {Ramos, Leo and Casas, Edmundo and Romero, Cristian and Rivas-Echeverría, Francklin and Morocho-Cayamcela, Manuel Eugenio},\n doi = {10.1109/ACCESS.2024.3356551},\n journal = {IEEE Access}\n}\n\n
\n\n \n \n 2023\n \n \n (5)\n \n \n
\n \n \n \n
\n \n \n\n \n \n \n \n \n \n Mask R-CNN and YOLOv8 Comparison to Perform Tomato Maturity Recognition Task.\n \n \n \n \n\n\n \n Camacho, J., C.; and Morocho-Cayamcela, M., E.\n\n\n \n\n\n\n Communications in Computer and Information Science, 1885 CCIS: 382-396. 2023.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Mask R-CNN and YOLOv8 Comparison to Perform Tomato Maturity Recognition Task},\n type = {article},\n year = {2023},\n keywords = {Mask R-CNN,Object detection,YOLO,deep learning,maturity recognition,precision agriculture},\n pages = {382-396},\n volume = {1885 CCIS},\n websites = {https://link.springer.com/chapter/10.1007/978-3-031-45438-7_26},\n publisher = {Springer Science and Business Media Deutschland GmbH},\n id = {bca433c4-b3cf-3202-949d-4b04a5a2986a},\n created = {2023-12-12T18:34:50.275Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-02-23T19:51:56.571Z},\n read = {true},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n citation_key = {Camacho2023MaskTask},\n private_publication = {false},\n abstract = {This work explores the segmentation and detection of tomatoes in different maturity states for harvesting prediction by using the laboro tomato dataset to train a mask R-CNN and a YOLOv8 architecture. This work aims to test the mask R-CNN architecture and the proposed methodology efficiency on the benchmark paper [12]. The evaluation metric intersection over union (IoU) 0.5 showed an average precision of 67.2% with a recall of 78.9% over the laboro tomato dataset and an IoU average precision of 92.1% with a recall of 91.4% over the same dataset. The benchmark paper authors perform segmentation and classification in a separate process using color analysis algorithms and use the determination coefficient (R2 ) for how accurately the tomato was set into the three maturity classes. The results show that the state-of-the-art YOLOv8 has a R2 of 0.809, 0.897, and 0.968 in the ripe, half-ripe, and green categories, respectively. However, the Mask R-CNN results are acceptable, with 0.819, 0.809, and 0.893 in the ripe, half-ripe, and green categories, respectively. The YOLOv8 model performed better than the one used in the benchmark paper by detecting, segmenting, and classifying tomatoes. Moreover, the color-analysis technique used in the benchmark paper results inefficiently because the classification results showed no linear relation between the predictions and the real values.},\n bibtype = {article},\n author = {Camacho, Jean Carlo and Morocho-Cayamcela, Manuel Eugenio},\n doi = {10.1007/978-3-031-45438-7_26},\n journal = {Communications in Computer and Information Science}\n}\n\n This work explores the segmentation and detection of tomatoes in different maturity states for harvesting prediction by using the laboro tomato dataset to train a mask R-CNN and a YOLOv8 architecture. This work aims to test the mask R-CNN architecture and the proposed methodology efficiency on the benchmark paper [12]. The evaluation metric intersection over union (IoU) 0.5 showed an average precision of 67.2% with a recall of 78.9% over the laboro tomato dataset and an IoU average precision of 92.1% with a recall of 91.4% over the same dataset. The benchmark paper authors perform segmentation and classification in a separate process using color analysis algorithms and use the determination coefficient (R2 ) for how accurately the tomato was set into the three maturity classes. The results show that the state-of-the-art YOLOv8 has a R2 of 0.809, 0.897, and 0.968 in the ripe, half-ripe, and green categories, respectively. However, the Mask R-CNN results are acceptable, with 0.819, 0.809, and 0.893 in the ripe, half-ripe, and green categories, respectively. The YOLOv8 model performed better than the one used in the benchmark paper by detecting, segmenting, and classifying tomatoes. Moreover, the color-analysis technique used in the benchmark paper results inefficiently because the classification results showed no linear relation between the predictions and the real values.\n
\n\n\n\n \n\n \n \n \n \n \n \n Unraveling the Power of 4D Residual Blocks and Transfer Learning in Violence Detection.\n \n \n \n \n\n\n \n Bermeo, M.; Morocho-Cayamcela, M., E.; and Cuenca, E.\n\n\n \n\n\n\n Communications in Computer and Information Science, 1885 CCIS: 207-219. 2023.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Unraveling the Power of 4D Residual Blocks and Transfer Learning in Violence Detection},\n type = {article},\n year = {2023},\n keywords = {4D Residual Blocks,CNNs,MoViNet,RWF-2000,Video Classification,Violence Detection},\n pages = {207-219},\n volume = {1885 CCIS},\n websites = {https://link.springer.com/chapter/10.1007/978-3-031-45438-7_14},\n publisher = {Springer Science and Business Media Deutschland GmbH},\n id = {1fce7f8c-8494-3b7a-8a3a-902edff16d8c},\n created = {2023-12-12T18:59:31.956Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-02-23T19:52:03.680Z},\n read = {true},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n citation_key = {Bermeo2023UnravelingDetection},\n private_publication = {false},\n abstract = {In recent years, action recognition has seen significant advancements in using Convolutions Neural Networks (CNNs) models for video analysis. One of the essential fields in this area is violence detection, which determines whether or not violent scenes use videos from surveillance cameras. One popular approach to handle this is the Flow Gated Network, two separate networks that extract the features from the frames and the optical flow of the videos using convolutions. However, it cannot capture the spatio-temporal characteristics of the video, which are crucial for accurate action recognition. To address this limitation, researchers have proposed using 4D convolutions at the video level (V4D) and stream buffer in the case of MoViNets. These networks are designed to preserve the 3D spatio-temporal representation of the video while also incorporating residual connections, which allow for better feature propagation and improved performance. In this work, we propose using 4D residual blocks and MoViNets for violence detection on the data set RWF-2000 to achieve state-of-the-art results in action recognition. Furthermore, this approach compares the strengths of MoViNet and V4D, resulting in more robust and used models for violence detection getting 0.885 for 4D residual block and 0.895 for MoViNet A1 in terms of accuracy, beating the benchmark architecture.},\n bibtype = {article},\n author = {Bermeo, Mike and Morocho-Cayamcela, Manuel Eugenio and Cuenca, Erick},\n doi = {10.1007/978-3-031-45438-7_14},\n journal = {Communications in Computer and Information Science}\n}\n\n In recent years, action recognition has seen significant advancements in using Convolutions Neural Networks (CNNs) models for video analysis. One of the essential fields in this area is violence detection, which determines whether or not violent scenes use videos from surveillance cameras. One popular approach to handle this is the Flow Gated Network, two separate networks that extract the features from the frames and the optical flow of the videos using convolutions. However, it cannot capture the spatio-temporal characteristics of the video, which are crucial for accurate action recognition. To address this limitation, researchers have proposed using 4D convolutions at the video level (V4D) and stream buffer in the case of MoViNets. These networks are designed to preserve the 3D spatio-temporal representation of the video while also incorporating residual connections, which allow for better feature propagation and improved performance. In this work, we propose using 4D residual blocks and MoViNets for violence detection on the data set RWF-2000 to achieve state-of-the-art results in action recognition. Furthermore, this approach compares the strengths of MoViNet and V4D, resulting in more robust and used models for violence detection getting 0.885 for 4D residual block and 0.895 for MoViNet A1 in terms of accuracy, beating the benchmark architecture.\n
\n\n\n\n \n\n \n \n \n \n \n \n Hyperparameter Tuning in a Dual Channel U-Net for Medical Image Segmentation.\n \n \n \n \n\n\n \n Román, K.; Llumiquinga, J.; Chancay, S.; and Morocho-Cayamcela, M., E.\n\n\n \n\n\n\n Communications in Computer and Information Science, 1885 CCIS: 337-352. 2023.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Hyperparameter Tuning in a Dual Channel U-Net for Medical Image Segmentation},\n type = {article},\n year = {2023},\n keywords = {Convolutional Neural Network,DC-UNet,Hyperparameter Tuning,Medical Image Segmentation},\n pages = {337-352},\n volume = {1885 CCIS},\n websites = {https://link.springer.com/chapter/10.1007/978-3-031-45438-7_23},\n publisher = {Springer Science and Business Media Deutschland GmbH},\n id = {973bebd7-948e-3bf2-9ee1-69b8cacffe41},\n created = {2023-12-12T19:00:31.741Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-02-23T19:52:10.616Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n citation_key = {Roman2023HyperparameterSegmentation},\n private_publication = {false},\n abstract = {Deep learning has been receiving a lot of attention lately, specially in the computer vision research community. In particular, the image segmentation task has revolutionized the treatment of medical imagery for diagnose and illness prediction. U-Net is one of the most extensively employed convolutional neural network (CNN) architectures for medical image segmentation. Moreover, U-Net is recognized to excel in segmenting multimodal medical images in general and complicated circumstances. However, the traditional U-Net architecture still has significant limitations with medical images segmentation. To tackle this limitations, a novel CNN architecture called dual channel-UNet (DC-UNet) has emerged as a possible replacement for the U-Net model. However, the performance and accuracy of the model are not the greatest when inferring on endoscopy images. This paper contributes to the research field by proposing a hyperparameter tuning of DC-UNet on the CVC-ClinicDB endoscopy dataset. Among several factors, we determine the optimum hyperparameter setup for this dataset, considering different values of: (i) batch sizes, (ii) learning rates, (iii) gradient-based optimizers, and (iv) dropout layers. We have also included a contrast limited adaptive histogram equalization, as a preprocessing technique to analyze the effect of the different experiments on the model’s performance.},\n bibtype = {article},\n author = {Román, Krishna and Llumiquinga, José and Chancay, Stalyn and Morocho-Cayamcela, Manuel Eugenio},\n doi = {10.1007/978-3-031-45438-7_23},\n journal = {Communications in Computer and Information Science}\n}\n\n Deep learning has been receiving a lot of attention lately, specially in the computer vision research community. In particular, the image segmentation task has revolutionized the treatment of medical imagery for diagnose and illness prediction. U-Net is one of the most extensively employed convolutional neural network (CNN) architectures for medical image segmentation. Moreover, U-Net is recognized to excel in segmenting multimodal medical images in general and complicated circumstances. However, the traditional U-Net architecture still has significant limitations with medical images segmentation. To tackle this limitations, a novel CNN architecture called dual channel-UNet (DC-UNet) has emerged as a possible replacement for the U-Net model. However, the performance and accuracy of the model are not the greatest when inferring on endoscopy images. This paper contributes to the research field by proposing a hyperparameter tuning of DC-UNet on the CVC-ClinicDB endoscopy dataset. Among several factors, we determine the optimum hyperparameter setup for this dataset, considering different values of: (i) batch sizes, (ii) learning rates, (iii) gradient-based optimizers, and (iv) dropout layers. We have also included a contrast limited adaptive histogram equalization, as a preprocessing technique to analyze the effect of the different experiments on the model’s performance.\n
\n\n\n\n \n\n \n \n \n \n \n \n Deep Reinforcement Learning for Efficient Digital Pap Smear Analysis.\n \n \n \n \n\n\n \n Macancela, C.; Morocho-Cayamcela, M., E.; and Chang, O.\n\n\n \n\n\n\n Computation, 11(12): 252. 12 2023.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Deep Reinforcement Learning for Efficient Digital Pap Smear Analysis},\n type = {article},\n year = {2023},\n keywords = {Papanicolaou,cells classification,cervical cancer,convolutional neuronal network,deep reinforcement learning},\n pages = {252},\n volume = {11},\n websites = {https://www.mdpi.com/2079-3197/11/12/252/htm,https://www.mdpi.com/2079-3197/11/12/252},\n month = {12},\n publisher = {Multidisciplinary Digital Publishing Institute},\n day = {10},\n id = {52f5ee14-fe2b-3280-b600-ff5d1f0245cc},\n created = {2023-12-13T03:49:20.920Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-07T18:17:09.011Z},\n read = {true},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n citation_key = {Macancela2023DeepAnalysis},\n private_publication = {false},\n abstract = {In August 2020, the World Health Assembly launched a global initiative to eliminate cervical cancer by 2030, setting three primary targets. One key goal is to achieve a 70% screening coverage rate for cervical cancer, primarily relying on the precise analysis of Papanicolaou (Pap) or digital Pap smears. However, the responsibility of reviewing Pap smear samples to identify potentially cancerous cells primarily falls on pathologists—a task known to be exceptionally challenging and time-consuming. This paper proposes a solution to address the shortage of pathologists for cervical cancer screening. It leverages the OpenAI-GYM API to create a deep reinforcement learning environment utilizing liquid-based Pap smear images. By employing the Proximal Policy Optimization algorithm, autonomous agents navigate Pap smear images, identifying cells with the aid of rewards, penalties, and accumulated experiences. Furthermore, the use of a pre-trained convolutional neuronal network like Res-Net50 enhances the classification of detected cells based on their potential for malignancy. The ultimate goal of this study is to develop a highly efficient, automated Papanicolaou analysis system, ultimately reducing the need for human intervention in regions with limited pathologists.},\n bibtype = {article},\n author = {Macancela, Carlos and Morocho-Cayamcela, Manuel Eugenio and Chang, Oscar},\n doi = {10.3390/COMPUTATION11120252},\n journal = {Computation},\n number = {12}\n}\n\n In August 2020, the World Health Assembly launched a global initiative to eliminate cervical cancer by 2030, setting three primary targets. One key goal is to achieve a 70% screening coverage rate for cervical cancer, primarily relying on the precise analysis of Papanicolaou (Pap) or digital Pap smears. However, the responsibility of reviewing Pap smear samples to identify potentially cancerous cells primarily falls on pathologists—a task known to be exceptionally challenging and time-consuming. This paper proposes a solution to address the shortage of pathologists for cervical cancer screening. It leverages the OpenAI-GYM API to create a deep reinforcement learning environment utilizing liquid-based Pap smear images. By employing the Proximal Policy Optimization algorithm, autonomous agents navigate Pap smear images, identifying cells with the aid of rewards, penalties, and accumulated experiences. Furthermore, the use of a pre-trained convolutional neuronal network like Res-Net50 enhances the classification of detected cells based on their potential for malignancy. The ultimate goal of this study is to develop a highly efficient, automated Papanicolaou analysis system, ultimately reducing the need for human intervention in regions with limited pathologists.\n
\n\n\n\n \n\n \n \n \n \n \n Use of diffusion models for the prediction of the Septorhinoplasty surgeries results.\n \n \n \n\n\n \n Loor Duque, J., J.; Bravo Pita, R., Y.; Jiménez Narváez, A., D.; Guzmán Suarez, F., R.; and Morocho Cayamcela, M., E.\n\n\n \n\n\n\n In Applied Engineering and Innovative Technologies (AENIT2023), 2023. \n \n\n\n\n
\n\n\n\n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n\n\n\n \n\n\n\n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
\n\n\n\n@inproceedings{\n title = {Use of diffusion models for the prediction of the Septorhinoplasty surgeries results},\n type = {inproceedings},\n year = {2023},\n city = {Ibarra},\n id = {9d174a59-f1b3-3e80-a273-857d0103ae61},\n created = {2024-01-23T13:58:13.501Z},\n file_attached = {false},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T14:21:13.414Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n private_publication = {false},\n bibtype = {inproceedings},\n author = {Loor Duque, Jonathan Javier and Bravo Pita, Rosaura Yokasta and Jiménez Narváez, Ariana Deyaneira and Guzmán Suarez, Freddy Raúl and Morocho Cayamcela, Manuel Eugenio},\n booktitle = {Applied Engineering and Innovative Technologies (AENIT2023)}\n}\n\n
\n\n \n \n 2022\n \n \n (12)\n \n \n
\n \n \n \n
\n \n \n\n \n \n \n \n \n \n Parkinson's disease diagnosis through electroencephalographic signal processing and sub-optimal feature extraction.\n \n \n \n \n\n\n \n Pozo-Ruiz, S., A.; Morocho-Cayamcela, M., E.; Mayorca-Torres, D.; and Peluffo-Ordoñez, D., H.\n\n\n \n\n\n\n In Information Technology and Systems, pages 118-127, 3 2022. Springer, Cham\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n@inproceedings{\n title = {Parkinson's disease diagnosis through electroencephalographic signal processing and sub-optimal feature extraction},\n type = {inproceedings},\n year = {2022},\n pages = {118-127},\n websites = {https://link.springer.com/book/10.1007/978-3-030-96293-7},\n month = {3},\n publisher = {Springer, Cham},\n day = {2},\n city = {San Carlos, Costa Rica},\n id = {b428c1d0-11eb-3ced-bac4-bce44f0f7e9c},\n created = {2021-12-17T05:12:55.305Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T14:19:44.283Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n citation_key = {Pozo-Ruiz2022ParkinsonExtraction},\n private_publication = {false},\n abstract = {Parkinson's disease is the second most common neurological disorder after Alzheimer. Several limitations and challenges have arisen when aiming to diagnose this disease. In this regard, a computer-aided diagnosis system is enforced for the early detection of any abnormalities. Prominent research efforts have been developed based on speech and gait analysis; nonetheless, electroencephalographic (EEG)-signal-driven approaches have acquired some interest recently to diagnose an early Parkinson’s disease. According to recent studies, the angles and sharpness of brain waves may hold key hints to detect Parkinson's disease. In the present work, we conduct an exploratory study over digital signal processing, and machine learning techniques for characterizing and classifying Parkinson-diagnosed EEG signals; waveform shape, spectral, statistical and non-linear features are taken into account for the present study. Our results, without being definitive, propose a suitable set of processing techniques to increase the performance, estimation accuracy, and interpretation of this physiological phenomenon. At the end, we found that with the characterization we performed, k-NN is the classifier which performs better, obtaining a mean accuracy of 86% when differentiating Parkinson's disease patients and healthy control subjects.},\n bibtype = {inproceedings},\n author = {Pozo-Ruiz, Santiago Alexis and Morocho-Cayamcela, Manuel Eugenio and Mayorca-Torres, Dagoberto and Peluffo-Ordoñez, Diego Hernán},\n doi = {https://doi.org/10.1007/978-3-030-96293-7_12},\n booktitle = {Information Technology and Systems}\n}\n\n Parkinson's disease is the second most common neurological disorder after Alzheimer. Several limitations and challenges have arisen when aiming to diagnose this disease. In this regard, a computer-aided diagnosis system is enforced for the early detection of any abnormalities. Prominent research efforts have been developed based on speech and gait analysis; nonetheless, electroencephalographic (EEG)-signal-driven approaches have acquired some interest recently to diagnose an early Parkinson’s disease. According to recent studies, the angles and sharpness of brain waves may hold key hints to detect Parkinson's disease. In the present work, we conduct an exploratory study over digital signal processing, and machine learning techniques for characterizing and classifying Parkinson-diagnosed EEG signals; waveform shape, spectral, statistical and non-linear features are taken into account for the present study. Our results, without being definitive, propose a suitable set of processing techniques to increase the performance, estimation accuracy, and interpretation of this physiological phenomenon. At the end, we found that with the characterization we performed, k-NN is the classifier which performs better, obtaining a mean accuracy of 86% when differentiating Parkinson's disease patients and healthy control subjects.\n
\n\n\n\n \n\n \n \n \n \n \n \n End-to-end license plate recognition system for an efficient deployment in surveillance scenarios.\n \n \n \n \n\n\n \n Suntaxi Domínguez, D., H.; Quintanchala Sandoval, S., C.; and Morocho-Cayamcela, M., E.\n\n\n \n\n\n\n In Information Technology and Systems, pages 697-704, 3 2022. Springer, Cham\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n\n@inproceedings{\n title = {End-to-end license plate recognition system for an efficient deployment in surveillance scenarios},\n type = {inproceedings},\n year = {2022},\n pages = {697-704},\n websites = {https://link.springer.com/book/10.1007/978-3-030-96293-7},\n month = {3},\n publisher = {Springer, Cham},\n day = {2},\n city = {San Carlos, Costa Rica},\n id = {7b91c3ff-6848-327e-8bfa-2d36a7a56c82},\n created = {2021-12-17T05:12:56.033Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T14:20:01.538Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n citation_key = {Suntaxi-Domínguez2022End-to-endScenarios},\n private_publication = {false},\n bibtype = {inproceedings},\n author = {Suntaxi Domínguez, Diego Hernán and Quintanchala Sandoval, Samantha Carolina and Morocho-Cayamcela, Manuel Eugenio},\n doi = {https://doi.org/10.1007/978-3-030-96293-7_59},\n booktitle = {Information Technology and Systems}\n}\n\n \n\n \n \n \n \n \n \n Factores de Riesgo Asociados a la Mortalidad y Peso al Nacer de Pacientes Neonatos, Caso de Estudio: Hospital Pediátrico Baca Ortiz.\n \n \n \n \n\n\n \n Jaraiseh Abcarius, M.; Zambrano Bravo, B., A.; Morocho-Cayamcela, M., E.; and Tulcanaza-Prieto, A., B.\n\n\n \n\n\n\n Enfermería Investiga, 7(1): 17-23. 2022.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Factores de Riesgo Asociados a la Mortalidad y Peso al Nacer de Pacientes Neonatos, Caso de Estudio: Hospital Pediátrico Baca Ortiz},\n type = {article},\n year = {2022},\n pages = {17-23},\n volume = {7},\n websites = {https://revistas.uta.edu.ec/erevista/index.php/enfi/article/view/1473/1274},\n publisher = {Enero-Marzo},\n id = {a980928e-0bf0-3f9e-a248-1770677f2e82},\n created = {2022-01-19T02:00:11.564Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T14:20:23.913Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n private_publication = {false},\n abstract = {El peso al nacer es uno de los principales indicadores pronóstico de mortalidad neonatal, en el que influyen factores asociados con la madre, el neonato, y también con las características socioeconómicas del núcleo familiar. Los factores de riesgo implicancomorbilidades al momento del nacimiento, por lo que, la intervención adecuada y el oportuno acceso a los servicios de salud constituyen elementos primordiales para la reducción de la mortalidad neonatal. Objetivo:establecer los factores de riesgo asociados a la mortalidad y peso al nacer de pacientes neonatos, de la Unidad de Cuidados Intensivos Neonatales del Hospital Pediátrico Baca Ortiz. Métodos:Se realizó una investigación de diseño observacional, transversal de tipo descriptivo que con una muestrade 204 recién nacidos ingresados en laUnidad de Cuidados Intensivos Neonatales del Hospital Pediátrico Baca Ortizen la ciudad de Quito, Ecuador, durante el año 2019. Resultados: El peso al nacimiento tiene una asociación lineal negativa significativa con la mortalidad neonatal, siendo los neonatos de género masculino los más susceptibles a fallecer. También, existe mayor frecuencia de mortalidad neonatal en las madres que residen en el área urbana de la sierra ecuatoriana. Conclusiones: El peso al nacer es una variable de gran influencia en la salud y supervivencia infantil, debido a que los datos epidemiológicos muestran que un niño que nace con un peso por debajo de los límites normales tiene un mayor riesgo de fallecer, encomparación con los niños nacidos con un peso dentro del rango considerado normal.},\n bibtype = {article},\n author = {Jaraiseh Abcarius, Margaret and Zambrano Bravo, Berly Alejandra and Morocho-Cayamcela, Manuel Eugenio and Tulcanaza-Prieto, Ana Belén},\n journal = {Enfermería Investiga},\n number = {1}\n}\n\n El peso al nacer es uno de los principales indicadores pronóstico de mortalidad neonatal, en el que influyen factores asociados con la madre, el neonato, y también con las características socioeconómicas del núcleo familiar. Los factores de riesgo implicancomorbilidades al momento del nacimiento, por lo que, la intervención adecuada y el oportuno acceso a los servicios de salud constituyen elementos primordiales para la reducción de la mortalidad neonatal. Objetivo:establecer los factores de riesgo asociados a la mortalidad y peso al nacer de pacientes neonatos, de la Unidad de Cuidados Intensivos Neonatales del Hospital Pediátrico Baca Ortiz. Métodos:Se realizó una investigación de diseño observacional, transversal de tipo descriptivo que con una muestrade 204 recién nacidos ingresados en laUnidad de Cuidados Intensivos Neonatales del Hospital Pediátrico Baca Ortizen la ciudad de Quito, Ecuador, durante el año 2019. Resultados: El peso al nacimiento tiene una asociación lineal negativa significativa con la mortalidad neonatal, siendo los neonatos de género masculino los más susceptibles a fallecer. También, existe mayor frecuencia de mortalidad neonatal en las madres que residen en el área urbana de la sierra ecuatoriana. Conclusiones: El peso al nacer es una variable de gran influencia en la salud y supervivencia infantil, debido a que los datos epidemiológicos muestran que un niño que nace con un peso por debajo de los límites normales tiene un mayor riesgo de fallecer, encomparación con los niños nacidos con un peso dentro del rango considerado normal.\n
\n\n\n\n \n\n \n \n \n \n \n \n Deep Learning Approaches Based on Transformer Architectures for Image Captioning Tasks.\n \n \n \n \n\n\n \n Castro, R.; Pineda, I.; Lim, W.; and Morocho-Cayamcela, M., E.\n\n\n \n\n\n\n IEEE Access, 10: 33679-33694. 3 2022.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Deep Learning Approaches Based on Transformer Architectures for Image Captioning Tasks},\n type = {article},\n year = {2022},\n pages = {33679-33694},\n volume = {10},\n websites = {https://ieeexplore.ieee.org/document/9739703},\n month = {3},\n day = {22},\n id = {52fbeb2d-4dc6-3769-881b-e27460eb4edd},\n created = {2022-04-04T15:54:32.240Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T14:19:21.338Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n citation_key = {Castro2022DeepTasks},\n private_publication = {false},\n abstract = {This paper focuses on visual attention , a state-of-the-art approach for image captioning tasks within the computer vision research area. We study the impact that different hyperparemeter configurations on an encoder-decoder visual attention architecture in terms of efficiency. Results show that the correct selection of both the cost function and the gradient-based optimizer can significantly impact the captioning results. Our system considers the cross-entropy, Kullback-Leibler divergence, mean squared error, and negative log-likelihood loss functions; the adaptive momentum (Adam), AdamW, RMSprop, stochastic gradient descent, and Adadelta optimizers. Experimentation shows that a combination of cross-entropy with Adam is the best alternative returning a Top-5 accuracy value of 73.092 and a BLEU-4 value of 20.10. Furthermore, a comparative analysis of alternative convolutional architectures demonstrated their performance as an encoder. Our results show that ResNext-101 stands out with a Top-5 accuracy of 73.128 and a BLEU-4 of 19.80; positioning itself as the best option when looking for the optimum captioning quality. However, MobileNetV3 proved to be a much more compact alternative with 2,971,952 parameters and 0.23 Giga fixed-point Multiply-Accumulate operations per Second (GMACS). Consequently, MobileNetV3 offers a competitive output quality at the cost of lower computational performance, supported by values of 19.50 and 72.928 for the BLEU-4 and Top-5 accuracy, respectively. Finally, when testing vision transformer (ViT), and data-efficient image transformer (DeiT) models to replace the convolutional component of the architecture, DeiT achieved an improvement over ViT, obtaining a value of 34.44 in the BLEU-4 metric.},\n bibtype = {article},\n author = {Castro, Roberto and Pineda, Israel and Lim, Wansu and Morocho-Cayamcela, Manuel Eugenio},\n doi = {10.1109/ACCESS.2022.3161428},\n journal = {IEEE Access}\n}\n\n This paper focuses on visual attention , a state-of-the-art approach for image captioning tasks within the computer vision research area. We study the impact that different hyperparemeter configurations on an encoder-decoder visual attention architecture in terms of efficiency. Results show that the correct selection of both the cost function and the gradient-based optimizer can significantly impact the captioning results. Our system considers the cross-entropy, Kullback-Leibler divergence, mean squared error, and negative log-likelihood loss functions; the adaptive momentum (Adam), AdamW, RMSprop, stochastic gradient descent, and Adadelta optimizers. Experimentation shows that a combination of cross-entropy with Adam is the best alternative returning a Top-5 accuracy value of 73.092 and a BLEU-4 value of 20.10. Furthermore, a comparative analysis of alternative convolutional architectures demonstrated their performance as an encoder. Our results show that ResNext-101 stands out with a Top-5 accuracy of 73.128 and a BLEU-4 of 19.80; positioning itself as the best option when looking for the optimum captioning quality. However, MobileNetV3 proved to be a much more compact alternative with 2,971,952 parameters and 0.23 Giga fixed-point Multiply-Accumulate operations per Second (GMACS). Consequently, MobileNetV3 offers a competitive output quality at the cost of lower computational performance, supported by values of 19.50 and 72.928 for the BLEU-4 and Top-5 accuracy, respectively. Finally, when testing vision transformer (ViT), and data-efficient image transformer (DeiT) models to replace the convolutional component of the architecture, DeiT achieved an improvement over ViT, obtaining a value of 34.44 in the BLEU-4 metric.\n
\n\n\n\n \n\n \n \n \n \n \n \n Plant Disease Classification and Severity Estimation: A Comparative Study of Multitask Convolutional Neural Networks and First Order Optimizers.\n \n \n \n \n\n\n \n Lucero, V.; Noboa, S.; and Morocho-Cayamcela, M., E.\n\n\n \n\n\n\n Communications in Computer and Information Science, 1577 CCIS: 313-328. 2022.\n \n\n\n\n
\n\n\n\n \n \nWebsite\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Plant Disease Classification and Severity Estimation: A Comparative Study of Multitask Convolutional Neural Networks and First Order Optimizers},\n type = {article},\n year = {2022},\n keywords = {Agriculture technology,Artificial intelligence,Computer vision,Disease classification,Neural network,Optimizers,Severity estimation},\n pages = {313-328},\n volume = {1577 CCIS},\n websites = {https://link.springer.com/chapter/10.1007/978-3-031-04447-2_21},\n publisher = {Springer Science and Business Media Deutschland GmbH},\n id = {73f54ff6-e1a9-3f0a-b620-34f9c90b935e},\n created = {2023-12-18T21:23:51.901Z},\n accessed = {2023-12-18},\n file_attached = {false},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-31T04:32:23.967Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n private_publication = {false},\n abstract = {The detection of plant diseases has been a hot research topic lately, specially since deep learning models and state-of-the-art convolutional neural networks (CNNs) architectures came into play. For this reason, this paper aims to compare several multitask CNN architectures used for: (i) classifying the environmental stress of coffee leaves, and (ii) estimating the severity of diseases that affect coffee plantations. This study is performed in two stages. First, the best performing CNN architecture was obtained from the multitask tests, which was ResNet34. Second, we improved the performance of ResNet34 by training it with six different optimization functions, and three different initial learning rates. This comparison was based on the analysis of different performance metrics such as the classification accuracy, training loss, F1-score, and the required number of epochs to achieve convergence. Our results show that with an initial learning rate of 1 × 10 - 3, Adagrad and Adam are the best optimizers for disease classification and severity estimation, respectively. Likewise, stochastic gradient descent shows to be an acceptable optimizer when the momentum hyperparameter is tuned properly.},\n bibtype = {article},\n author = {Lucero, Valeria and Noboa, Sherald and Morocho-Cayamcela, Manuel Eugenio},\n doi = {10.1007/978-3-031-04447-2_21/},\n journal = {Communications in Computer and Information Science}\n}\n\n The detection of plant diseases has been a hot research topic lately, specially since deep learning models and state-of-the-art convolutional neural networks (CNNs) architectures came into play. For this reason, this paper aims to compare several multitask CNN architectures used for: (i) classifying the environmental stress of coffee leaves, and (ii) estimating the severity of diseases that affect coffee plantations. This study is performed in two stages. First, the best performing CNN architecture was obtained from the multitask tests, which was ResNet34. Second, we improved the performance of ResNet34 by training it with six different optimization functions, and three different initial learning rates. This comparison was based on the analysis of different performance metrics such as the classification accuracy, training loss, F1-score, and the required number of epochs to achieve convergence. Our results show that with an initial learning rate of 1 × 10 - 3, Adagrad and Adam are the best optimizers for disease classification and severity estimation, respectively. Likewise, stochastic gradient descent shows to be an acceptable optimizer when the momentum hyperparameter is tuned properly.\n
\n\n\n\n \n\n \n \n \n \n \n \n Identifying Defective Fruits and Vegetables with Hyper-spectral Images: A Brief Tutorial.\n \n \n \n \n\n\n \n Figueroa, S.; Morocho-Cayamcela, M., E.; and Pineda, I.\n\n\n \n\n\n\n Proceedings - 3rd International Conference on Information Systems and Software Technologies, ICI2ST 2022,21-27. 2022.\n \n\n\n\n
\n\n\n\n \n \nWebsite\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Identifying Defective Fruits and Vegetables with Hyper-spectral Images: A Brief Tutorial},\n type = {article},\n year = {2022},\n keywords = {Computational agriculture,hyper-spectral imaging,image binarization,image enhancement,image filtering},\n pages = {21-27},\n websites = {https://ieeexplore.ieee.org/document/10063753},\n publisher = {Institute of Electrical and Electronics Engineers Inc.},\n id = {01469e7f-9028-3a9f-95cd-f02c4b249dd6},\n created = {2024-01-03T18:07:31.265Z},\n file_attached = {false},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T01:23:35.645Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n citation_key = {Figueroa2022IdentifyingTutorial},\n private_publication = {false},\n abstract = {Horticulture is the branch of agriculture that addresses the crop of fruits and vegetables. This art of cultivating natural products has been a critical research area that helped to satisfy human's food consumption. The enhancement of quality control has improved the cost-effectiveness and increased the financial profit of the production. This study presents a processing pipeline architecture that assesses the quality of fruits and vegetables through image processing. The solution involves hyper-spectral imaging, image enhancement, image binarization, and median filtering. Our experiments demonstrate that our image processing pipeline proposal successfully identifies the defective regions in a publicly available dataset that contains images of apples and onions.},\n bibtype = {article},\n author = {Figueroa, Saul and Morocho-Cayamcela, Manuel Eugenio and Pineda, Israel},\n doi = {10.1109/ICI2ST57350.2022.00011},\n journal = {Proceedings - 3rd International Conference on Information Systems and Software Technologies, ICI2ST 2022}\n}\n\n Horticulture is the branch of agriculture that addresses the crop of fruits and vegetables. This art of cultivating natural products has been a critical research area that helped to satisfy human's food consumption. The enhancement of quality control has improved the cost-effectiveness and increased the financial profit of the production. This study presents a processing pipeline architecture that assesses the quality of fruits and vegetables through image processing. The solution involves hyper-spectral imaging, image enhancement, image binarization, and median filtering. Our experiments demonstrate that our image processing pipeline proposal successfully identifies the defective regions in a publicly available dataset that contains images of apples and onions.\n
\n\n\n\n \n\n \n \n \n \n \n An Efficient Deep Q-learning Strategy for Sequential Decision-making in Game-playing.\n \n \n \n\n\n \n Chang, O.; Morocho-Cayamcela, M., E.; Pineda, I.; and Cardenas, K.\n\n\n \n\n\n\n Proceedings - 3rd International Conference on Information Systems and Software Technologies, ICI2ST 2022,172-177. 2022.\n \n\n\n\n
\n\n\n\n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n
\n\n\n@article{\n title = {An Efficient Deep Q-learning Strategy for Sequential Decision-making in Game-playing},\n type = {article},\n year = {2022},\n keywords = {Artificial intelligence,deep Q-learning,game-playing,reinforcement learning,sequential decision-making},\n pages = {172-177},\n publisher = {Institute of Electrical and Electronics Engineers Inc.},\n id = {527c1f84-d309-3083-a89d-a68c7069bbc2},\n created = {2024-01-23T14:03:25.775Z},\n accessed = {2024-01-23},\n file_attached = {false},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T14:05:15.030Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n private_publication = {false},\n abstract = {This paper presents a deep reinforcement learning model that efficiently learns a sequential decision-making policy to play tic-tac-toe intelligently directly from a high-dimensional video. To produce a stable, sparse neural representation of the states of the tic-tac-toe board, a convolutional pre-trained neural network has been used, followed by a fully-connected sigmoidal network. The assemble behaves as a Q-matrix and produces the ultimate state-decision pairs that control a robotic arm placing physical tokens on the board. The hyperparameters in the whole network are tuned to produce a stable trainable array of elements. An internal clock composed of internal neurons is integrated to give the agent a sense of sequential timing. To solve the max(⊙) function, a novel algorithm is introduced to search for the Q-network values. The algorithm uses a dedicated, sigmoidal net initialized with random parameters. Under backpropagation it iteratively moves to a stable plateau that mimics the all-zeros condition of an initial Q-matrix. Next, the agent uses Bellman's reinforcement principles to learn an optimal policy with a noticeable look-ahead capability. Computer simulations driving a physical robot proved the convergence and effectiveness of the proposed methodology and demonstrated a marked ability in sequential decision-making, taking raw video frames as input.},\n bibtype = {article},\n author = {Chang, Oscar and Morocho-Cayamcela, Manuel Eugenio and Pineda, Israel and Cardenas, Kevin},\n doi = {10.1109/ICI2ST57350.2022.00032},\n journal = {Proceedings - 3rd International Conference on Information Systems and Software Technologies, ICI2ST 2022}\n}\n\n This paper presents a deep reinforcement learning model that efficiently learns a sequential decision-making policy to play tic-tac-toe intelligently directly from a high-dimensional video. To produce a stable, sparse neural representation of the states of the tic-tac-toe board, a convolutional pre-trained neural network has been used, followed by a fully-connected sigmoidal network. The assemble behaves as a Q-matrix and produces the ultimate state-decision pairs that control a robotic arm placing physical tokens on the board. The hyperparameters in the whole network are tuned to produce a stable trainable array of elements. An internal clock composed of internal neurons is integrated to give the agent a sense of sequential timing. To solve the max(⊙) function, a novel algorithm is introduced to search for the Q-network values. The algorithm uses a dedicated, sigmoidal net initialized with random parameters. Under backpropagation it iteratively moves to a stable plateau that mimics the all-zeros condition of an initial Q-matrix. Next, the agent uses Bellman's reinforcement principles to learn an optimal policy with a noticeable look-ahead capability. Computer simulations driving a physical robot proved the convergence and effectiveness of the proposed methodology and demonstrated a marked ability in sequential decision-making, taking raw video frames as input.\n
\n\n\n\n \n\n \n \n \n \n \n \n Implementation of a Lightweight CNN for American Sign Language Classification.\n \n \n \n \n\n\n \n Lomas, M., S.; Quelal, A.; and Morocho-Cayamcela, M., E.\n\n\n \n\n\n\n Communications in Computer and Information Science, 1647 CCIS: 197-207. 2022.\n \n\n\n\n
\n\n\n\n \n \nWebsite\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Implementation of a Lightweight CNN for American Sign Language Classification},\n type = {article},\n year = {2022},\n keywords = {American sign language,EfficientNet,Image classification convolutional neural network,Transfer learning},\n pages = {197-207},\n volume = {1647 CCIS},\n websites = {https://link.springer.com/chapter/10.1007/978-3-031-18347-8_16},\n publisher = {Springer Science and Business Media Deutschland GmbH},\n id = {effaf762-2931-31f3-a0b5-fd8aa6ff1d37},\n created = {2024-01-23T14:04:59.842Z},\n accessed = {2024-01-23},\n file_attached = {false},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T14:05:20.970Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n private_publication = {false},\n abstract = {The American sign language is the most popular and widely-accepted sign language for people with hearing difficulties. Computer vision techniques, such as skeleton recognition, depth recognition, 3D model recognition, or deep learning recognition, have helped to develop better systems for sign language classification and detection. Despite the promising results from baseline research efforts, overfitting problems have been detected when the training and testing accuracy are compared. In this work, we propose to exploit the scaling method on EfficientNet, which is a convolutional neural network architecture, in order to uniformly scale all the dimensions of depth, width, and resolution using a compound coefficient. Our results show that the overfitting problem can be solved by incorporating hyperparameter tunning and dropout as a regularization method. We also have the benefit of transfer learning to reduce the training time by reusing the weights of EfficientNet, pre-trained with the ImageNet dataset. Our results are compared with the benchmark paper, proving that our model generalizes better to unseen instances. In the first section of this study, we introduce American sign language meaning, hand gesture recognition, and its related works in the computer vision field. It allows us to mention transfer learning concepts and Efficient Nets architectures. In the second section, we establish the methodology by choosing the B0 model as the architecture selected to test with a Kaggle dataset. Adjusting hyperparameters, we enter in the third section, in the training and testing phase where overfitting problems were solved with high accuracy, finishing talking with contributions and future works.},\n bibtype = {article},\n author = {Lomas, Mateo Sebastián and Quelal, Andrés and Morocho-Cayamcela, Manuel Eugenio},\n doi = {10.1007/978-3-031-18347-8_16/COVER},\n journal = {Communications in Computer and Information Science}\n}\n\n The American sign language is the most popular and widely-accepted sign language for people with hearing difficulties. Computer vision techniques, such as skeleton recognition, depth recognition, 3D model recognition, or deep learning recognition, have helped to develop better systems for sign language classification and detection. Despite the promising results from baseline research efforts, overfitting problems have been detected when the training and testing accuracy are compared. In this work, we propose to exploit the scaling method on EfficientNet, which is a convolutional neural network architecture, in order to uniformly scale all the dimensions of depth, width, and resolution using a compound coefficient. Our results show that the overfitting problem can be solved by incorporating hyperparameter tunning and dropout as a regularization method. We also have the benefit of transfer learning to reduce the training time by reusing the weights of EfficientNet, pre-trained with the ImageNet dataset. Our results are compared with the benchmark paper, proving that our model generalizes better to unseen instances. In the first section of this study, we introduce American sign language meaning, hand gesture recognition, and its related works in the computer vision field. It allows us to mention transfer learning concepts and Efficient Nets architectures. In the second section, we establish the methodology by choosing the B0 model as the architecture selected to test with a Kaggle dataset. Adjusting hyperparameters, we enter in the third section, in the training and testing phase where overfitting problems were solved with high accuracy, finishing talking with contributions and future works.\n
\n\n\n\n \n\n \n \n \n \n \n \n Comparative Study of Image Degradation and Restoration Techniques.\n \n \n \n \n\n\n \n Pijal, W.; Pineda, I.; and Morocho-Cayamcela, M., E.\n\n\n \n\n\n\n Communications in Computer and Information Science, 1648 CCIS: 253-265. 2022.\n \n\n\n\n
\n\n\n\n \n \nWebsite\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Comparative Study of Image Degradation and Restoration Techniques},\n type = {article},\n year = {2022},\n keywords = {Computer vision,Image processing,Image restoration,Inverse filter,Wiener filter},\n pages = {253-265},\n volume = {1648 CCIS},\n websites = {https://link.springer.com/chapter/10.1007/978-3-031-18272-3_17},\n publisher = {Springer Science and Business Media Deutschland GmbH},\n id = {f7a5e21e-0887-3aeb-98f9-2131a7fcfd6d},\n created = {2024-01-23T14:06:38.273Z},\n accessed = {2024-01-23},\n file_attached = {false},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-02-23T19:21:30.328Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n private_publication = {false},\n abstract = {This paper implements airy disk smoothing, Poisson noise, Gaussian smoothing, Hanser’s phase term, and Zernike polynomial phase term as degradation techniques on images from the DIV2K dataset. These actions allows the generation of gray-scale degraded images to study the performance of the inverse filter, Wiener filter, and the Richardson-Lucy algorithm as image restoration techniques. Our experiments are conducted on two representative tasks: (i) intense image degradation, and (ii) image restoration from the degraded images. To measure the image degradation and the image approximation to the original image, this paper uses four similarity metrics: global dimensionless relative error of synthesis (ERGAS), mean squared error (MSE), spectral angle mapper (SAM), and visual information fidelity (VIFP). These similarity metrics determine which restoration technique can estimate the original image in more precisely, and enable the analysis of the required conditions for the estimation.},\n bibtype = {article},\n author = {Pijal, Washington and Pineda, Israel and Morocho-Cayamcela, Manuel Eugenio},\n doi = {10.1007/978-3-031-18272-3_17},\n journal = {Communications in Computer and Information Science}\n}\n\n This paper implements airy disk smoothing, Poisson noise, Gaussian smoothing, Hanser’s phase term, and Zernike polynomial phase term as degradation techniques on images from the DIV2K dataset. These actions allows the generation of gray-scale degraded images to study the performance of the inverse filter, Wiener filter, and the Richardson-Lucy algorithm as image restoration techniques. Our experiments are conducted on two representative tasks: (i) intense image degradation, and (ii) image restoration from the degraded images. To measure the image degradation and the image approximation to the original image, this paper uses four similarity metrics: global dimensionless relative error of synthesis (ERGAS), mean squared error (MSE), spectral angle mapper (SAM), and visual information fidelity (VIFP). These similarity metrics determine which restoration technique can estimate the original image in more precisely, and enable the analysis of the required conditions for the estimation.\n
\n\n\n\n \n\n \n \n \n \n \n \n A Computer Vision Model to Identify the Incorrect Use of Face Masks for COVID-19 Awareness.\n \n \n \n \n\n\n \n Crespo, F.; Crespo, A.; Sierra-Martínez, L., M.; Peluffo-Ordóñez, D., H.; and Morocho-Cayamcela, M., E.\n\n\n \n\n\n\n Applied Sciences 2022, Vol. 12, Page 6924, 12(14): 6924. 7 2022.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {A Computer Vision Model to Identify the Incorrect Use of Face Masks for COVID-19 Awareness},\n type = {article},\n year = {2022},\n keywords = {19,COVID,artificial intelligence,computer vision,deep learning,face mask recognition,image classification,object detection},\n pages = {6924},\n volume = {12},\n websites = {https://www.mdpi.com/2076-3417/12/14/6924/htm,https://www.mdpi.com/2076-3417/12/14/6924},\n month = {7},\n publisher = {Multidisciplinary Digital Publishing Institute},\n day = {8},\n id = {45f58bc4-172e-30cf-916b-1928df1512e3},\n created = {2024-01-23T14:16:05.270Z},\n accessed = {2024-01-23},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T14:16:42.665Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n private_publication = {false},\n abstract = {Face mask detection has become a great challenge in computer vision, demanding the coalition of technology with COVID-19 awareness. Researchers have proposed deep learning models to detect the use of face masks. However, the incorrect use of a face mask can be as harmful as not wearing any protection at all. In this paper, we propose a compound convolutional neural network (CNN) architecture based on two computer vision tasks: object localization to discover faces in images/videos, followed by an image classification CNN to categorize the faces and show if someone is using a face mask correctly, incorrectly, or not at all. The first CNN is built upon RetinaFace, a model to detect faces in images, whereas the second CNN uses a ResNet-18 architecture as a classification backbone. Our model enables an accurate identification of people who are not correctly following the COVID-19 healthcare recommendations on face mask use. To enable further global use of our technology, we have released both the dataset used to train the classification model and our proposed computer vision pipeline to the public, and optimized it for embedded systems deployment.},\n bibtype = {article},\n author = {Crespo, Fabricio and Crespo, Anthony and Sierra-Martínez, Luz Marina and Peluffo-Ordóñez, Diego Hernán and Morocho-Cayamcela, Manuel Eugenio},\n doi = {10.3390/APP12146924},\n journal = {Applied Sciences 2022, Vol. 12, Page 6924},\n number = {14}\n}\n\n Face mask detection has become a great challenge in computer vision, demanding the coalition of technology with COVID-19 awareness. Researchers have proposed deep learning models to detect the use of face masks. However, the incorrect use of a face mask can be as harmful as not wearing any protection at all. In this paper, we propose a compound convolutional neural network (CNN) architecture based on two computer vision tasks: object localization to discover faces in images/videos, followed by an image classification CNN to categorize the faces and show if someone is using a face mask correctly, incorrectly, or not at all. The first CNN is built upon RetinaFace, a model to detect faces in images, whereas the second CNN uses a ResNet-18 architecture as a classification backbone. Our model enables an accurate identification of people who are not correctly following the COVID-19 healthcare recommendations on face mask use. To enable further global use of our technology, we have released both the dataset used to train the classification model and our proposed computer vision pipeline to the public, and optimized it for embedded systems deployment.\n
\n\n\n\n \n\n \n \n \n \n \n \n An optimal location strategy for multiple drone base stations in massive MIMO.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; Lim, W.; and Maier, M.\n\n\n \n\n\n\n ICT Express, 8(2): 230-234. 6 2022.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {An optimal location strategy for multiple drone base stations in massive MIMO},\n type = {article},\n year = {2022},\n keywords = {Disaster scenario,Massive MIMO,Path planning,Unmanned aerial vehicle,User association},\n pages = {230-234},\n volume = {8},\n websites = {https://www.sciencedirect.com/science/article/pii/S2405959521000990},\n month = {6},\n publisher = {Elsevier},\n day = {1},\n id = {9180bede-8bae-3983-8482-927a3fe4df5a},\n created = {2024-01-23T14:17:42.977Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-31T04:27:18.243Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n private_publication = {false},\n abstract = {The concept of drone base stations (DBSs) has been applied to reduce the distance of the wireless link between a macro base station and its active users under diverse scenarios in military communications, smart industries, and high-density networks, and to provide service in topologies with damaged infrastructure. In this paper, we address the optimal positioning of multiple DBSs in a multiple-input multiple-output wireless network setting. We present a low-complexity machine learning-based algorithm to optimize the location of the DBSs by minimizing the collective wireless received signal strength experienced by the active terminals. The proposed algorithm reduces the propagation loss in the system and provides a lower bit error rate when compared with the Euclidean cost benchmark.},\n bibtype = {article},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu and Maier, Martin},\n doi = {10.1016/J.ICTE.2021.08.010},\n journal = {ICT Express},\n number = {2}\n}\n\n The concept of drone base stations (DBSs) has been applied to reduce the distance of the wireless link between a macro base station and its active users under diverse scenarios in military communications, smart industries, and high-density networks, and to provide service in topologies with damaged infrastructure. In this paper, we address the optimal positioning of multiple DBSs in a multiple-input multiple-output wireless network setting. We present a low-complexity machine learning-based algorithm to optimize the location of the DBSs by minimizing the collective wireless received signal strength experienced by the active terminals. The proposed algorithm reduces the propagation loss in the system and provides a lower bit error rate when compared with the Euclidean cost benchmark.\n
\n\n\n\n \n\n \n \n \n \n \n \n Predicting target data rates for dynamic spectrum allocation using Gaussian process regression.\n \n \n \n \n\n\n \n Njoku, J., N.; Morocho-Cayamcela, M., E.; Caliwag, A.; Xiao, P.; and Lim, W.\n\n\n \n\n\n\n ICT Express, 8(2): 207-212. 6 2022.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Predicting target data rates for dynamic spectrum allocation using Gaussian process regression},\n type = {article},\n year = {2022},\n keywords = {Gaussian process regression,Machine learning,Spectrum allocation,Target data rate},\n pages = {207-212},\n volume = {8},\n websites = {https://www.sciencedirect.com/science/article/pii/S2405959521001004},\n month = {6},\n publisher = {Elsevier},\n day = {1},\n id = {3d88edf7-7ff8-36df-b848-34b1baf00b24},\n created = {2024-01-23T14:18:40.040Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-31T04:28:56.008Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n private_publication = {false},\n abstract = {Issues in spectrum allocation between wireless network users have arisen due to the fast increase in the number of broadband services. Such issues include the failure to maximize the performance of all users by considering only a particular category of users. Specifically, a previously adopted selfish algorithm for spectrum allocation considers only the performance of the weakest user. To resolve this issue, we propose a new target data rate setting algorithm for dynamic spectrum allocation. In this algorithm, a Gaussian process regression model is trained to predict the target data rate. All users that perform below the defined target rate will have their frequency band allocations changed to one that guarantees a better performance. Through simulations, we show that the maximum data rate achieved by the weakest user in our algorithm is 121.7% higher than the selfish algorithm.},\n bibtype = {article},\n author = {Njoku, Judith Nkechinyere and Morocho-Cayamcela, Manuel Eugenio and Caliwag, Angela and Xiao, Pei and Lim, Wansu},\n doi = {10.1016/J.ICTE.2021.08.011},\n journal = {ICT Express},\n number = {2}\n}\n\n Issues in spectrum allocation between wireless network users have arisen due to the fast increase in the number of broadband services. Such issues include the failure to maximize the performance of all users by considering only a particular category of users. Specifically, a previously adopted selfish algorithm for spectrum allocation considers only the performance of the weakest user. To resolve this issue, we propose a new target data rate setting algorithm for dynamic spectrum allocation. In this algorithm, a Gaussian process regression model is trained to predict the target data rate. All users that perform below the defined target rate will have their frequency band allocations changed to one that guarantees a better performance. Through simulations, we show that the maximum data rate achieved by the weakest user in our algorithm is 121.7% higher than the selfish algorithm.\n
\n\n\n\n
\n\n \n \n 2021\n \n \n (8)\n \n \n
\n \n \n \n
\n \n \n\n \n \n \n \n \n \n Pattern recognition of soldier uniforms with dilated convolutions and a modified encoder-decoder neural network architecture.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; and Lim, W.\n\n\n \n\n\n\n Applied Artificial Intelligence, 35(6): 476-487. 2021.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Pattern recognition of soldier uniforms with dilated convolutions and a modified encoder-decoder neural network architecture},\n type = {article},\n year = {2021},\n pages = {476-487},\n volume = {35},\n websites = {https://www.tandfonline.com/doi/abs/10.1080/08839514.2021.1902124},\n publisher = {Bellwether Publishing, Ltd.},\n id = {1a900232-861e-38b8-bc2c-947419973b80},\n created = {2021-12-04T20:47:46.025Z},\n accessed = {2021-06-09},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T14:24:59.614Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n abstract = {In this paper, we study a deep learning (DL)-based multimodal technology for military, surveillance, and defense applications based on a pixel-by-pixel classification of soldier’s image dataset. We explore the acquisition of images from a remote tactical-robot to a ground station, where the detection and tracking of soldiers can help the operator to take actions or automate the tactical-robot in battlefield. The soldier detection is achieved by training a convolutional neural network to learn the patterns of the soldier’s uniforms. Our CNN learns from the initial dataset and from the actions taken by the operator, as opposed to the old-fashioned and hard-coded image processing algorithms. Our system attains an accuracy of over 81% in distinguishing the specific soldier uniform and the background. These experimental results prove our hypothesis that dilated convolutions can increase the segmentation performance when compared with patch-based, and fully connected networks.},\n bibtype = {article},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu},\n doi = {10.1080/08839514.2021.1902124},\n journal = {Applied Artificial Intelligence},\n number = {6}\n}\n\n In this paper, we study a deep learning (DL)-based multimodal technology for military, surveillance, and defense applications based on a pixel-by-pixel classification of soldier’s image dataset. We explore the acquisition of images from a remote tactical-robot to a ground station, where the detection and tracking of soldiers can help the operator to take actions or automate the tactical-robot in battlefield. The soldier detection is achieved by training a convolutional neural network to learn the patterns of the soldier’s uniforms. Our CNN learns from the initial dataset and from the actions taken by the operator, as opposed to the old-fashioned and hard-coded image processing algorithms. Our system attains an accuracy of over 81% in distinguishing the specific soldier uniform and the background. These experimental results prove our hypothesis that dilated convolutions can increase the segmentation performance when compared with patch-based, and fully connected networks.\n
\n\n\n\n \n\n \n \n \n \n \n \n CGDNet: Efficient Hybrid Deep Learning Model for Robust Automatic Modulation Recognition.\n \n \n \n \n\n\n \n Njoku, J., N.; Morocho-Cayamcela, M., E.; and Lim, W.\n\n\n \n\n\n\n IEEE Networking Letters, 3(2): 47-51. 2 2021.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n\n@article{\n title = {CGDNet: Efficient Hybrid Deep Learning Model for Robust Automatic Modulation Recognition},\n type = {article},\n year = {2021},\n pages = {47-51},\n volume = {3},\n websites = {https://ieeexplore.ieee.org/document/9349627},\n month = {2},\n publisher = {Institute of Electrical and Electronics Engineers (IEEE)},\n day = {9},\n id = {0ca64cc9-63d2-31cc-9fd8-61e6c380f6fb},\n created = {2021-12-04T20:47:46.787Z},\n accessed = {2021-06-09},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T14:25:32.681Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n bibtype = {article},\n author = {Njoku, Judith Nkechinyere and Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu},\n doi = {10.1109/lnet.2021.3057637},\n journal = {IEEE Networking Letters},\n number = {2}\n}\n\n \n\n \n \n \n \n \n \n An Assistive PIN Input Technology for the Visually Impaired.\n \n \n \n \n\n\n \n Jeon, I.; Morocho-Cayamcela, M., E.; Kim, M.; and Lim, W.\n\n\n \n\n\n\n The Journal of Korean Institute of Communications and Information Sciences, 46(5): 890-899. 5 2021.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n\n@article{\n title = {An Assistive PIN Input Technology for the Visually Impaired},\n type = {article},\n year = {2021},\n pages = {890-899},\n volume = {46},\n websites = {http://www.dbpia.co.kr/Journal/ArticleDetail/NODE10557257},\n month = {5},\n day = {31},\n id = {e6b3ceda-a51a-3cec-a3a7-41c7ad404114},\n created = {2021-12-04T20:47:47.286Z},\n accessed = {2021-06-09},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T14:25:17.931Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n bibtype = {article},\n author = {Jeon, Il-Soo and Morocho-Cayamcela, Manuel Eugenio and Kim, Myung-Sik and Lim, Wansu},\n doi = {10.7840/kics.2021.46.5.890},\n journal = {The Journal of Korean Institute of Communications and Information Sciences},\n number = {5}\n}\n\n \n\n \n \n \n \n \n \n Predicting target data rates for dynamic spectrum allocation using Gaussian process regression.\n \n \n \n \n\n\n \n Njoku, J., N.; Morocho-Cayamcela, M., E.; Caliwag, A.; Xiao, P.; and Lim, W.\n\n\n \n\n\n\n Proceedings of Symposium of the Korean Institute of Communications and Information Sciences,427-428. 2021.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Predicting target data rates for dynamic spectrum allocation using Gaussian process regression},\n type = {article},\n year = {2021},\n keywords = {Gaussian process regression,data rate,machine learning,selfish,spectrum allocation},\n pages = {427-428},\n websites = {https://www.dbpia.co.kr/Journal/articleDetail?nodeId=NODE10547590},\n id = {2907e342-24bd-3f34-8f5f-725c3e198e11},\n created = {2021-12-04T20:47:48.170Z},\n accessed = {2021-06-09},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T14:39:45.484Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n abstract = {Issues in spectrum allocation between wireless network users have arisen due to the fast increase in the number of broadband services. Such issues include the failure to maximize the performance of all users, by considering only a particular category of users. Specifically, a previously adopted selfish algorithm for spectrum allocation considers only the performance of the weakest user. To resolve this issue, we propose a new target data rate setting algorithm for dynamic spectrum allocation. In this algorithm, a Gaussian process regression model is trained to predict the target data rate. All users that perform below the defined target rate, will have their frequency band allocations changed to one that guarantees a better performance. Through simulations, we show that the maximum data rate achieved by the weakest user in our algorithm is 121.7% higher than the selfish algorithm.},\n bibtype = {article},\n author = {Njoku, Judith Nkechinyere and Morocho-Cayamcela, Manuel Eugenio and Caliwag, Angela and Xiao, Pei and Lim, Wansu},\n journal = {Proceedings of Symposium of the Korean Institute of Communications and Information Sciences}\n}\n\n Issues in spectrum allocation between wireless network users have arisen due to the fast increase in the number of broadband services. Such issues include the failure to maximize the performance of all users, by considering only a particular category of users. Specifically, a previously adopted selfish algorithm for spectrum allocation considers only the performance of the weakest user. To resolve this issue, we propose a new target data rate setting algorithm for dynamic spectrum allocation. In this algorithm, a Gaussian process regression model is trained to predict the target data rate. All users that perform below the defined target rate, will have their frequency band allocations changed to one that guarantees a better performance. Through simulations, we show that the maximum data rate achieved by the weakest user in our algorithm is 121.7% higher than the selfish algorithm.\n
\n\n\n\n \n\n \n \n \n \n \n \n BLER performance evaluation of an enhanced channel autoencoder.\n \n \n \n \n\n\n \n Njoku, J., N.; Morocho-Cayamcela, M., E.; and Lim, W.\n\n\n \n\n\n\n Computer Communications, 176: 173-181. 8 2021.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {BLER performance evaluation of an enhanced channel autoencoder},\n type = {article},\n year = {2021},\n keywords = {Autoencoder,Deep learning,End-to-end learning,Modulation,Neural network,Wireless communications},\n pages = {173-181},\n volume = {176},\n websites = {https://linkinghub.elsevier.com/retrieve/pii/S0140366421002188},\n month = {8},\n publisher = {Elsevier},\n day = {1},\n id = {caa4368d-d826-3833-a40f-3ff329cdd1fa},\n created = {2021-12-04T20:47:48.689Z},\n accessed = {2021-06-11},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T14:24:29.262Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n abstract = {The concept of using autoencoders (AEs) to represent wireless communication systems as an end-to-end reconstruction task that optimizes the transmitter and receiver components simultaneously in a single process has attracted the attention of wireless practitioners worldwide. This is attributable to the flexibility, and convenience of representing complex channel models. However, owing to the characteristics of deep neural networks (DNNs), as the AE learns the representation of the channel, overfitting limits its performance. In this paper, we propose RegAE, a regularized DNN architecture that overcomes the overfitting limitation in AEs and reduces their training complexity, which are characteristics of models with higher dimensions. We demonstrate that RegAE improves the block error rate (BLER) as compared with equivalent models from the literature. Thereby, it achieves a performance (1) better than that of a 4∕7 rate Hamming code with a 16 phase-shift keying (16PSK) modulation under an additive white Gaussian noise (AWGN) channel, (2) comparable to that of a 4∕7 rate maximum likelihood decoding (MLD) with a Eb∕N0 range from 1 dB to 5 dB, and (3) equivalent to that of an uncoded binary phase-shift keying (BPSK) modulation over a Eb∕N0 range from 0 dB to 10 dB.},\n bibtype = {article},\n author = {Njoku, Judith Nkechinyere and Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu},\n doi = {10.1016/j.comcom.2021.05.026},\n journal = {Computer Communications}\n}\n\n The concept of using autoencoders (AEs) to represent wireless communication systems as an end-to-end reconstruction task that optimizes the transmitter and receiver components simultaneously in a single process has attracted the attention of wireless practitioners worldwide. This is attributable to the flexibility, and convenience of representing complex channel models. However, owing to the characteristics of deep neural networks (DNNs), as the AE learns the representation of the channel, overfitting limits its performance. In this paper, we propose RegAE, a regularized DNN architecture that overcomes the overfitting limitation in AEs and reduces their training complexity, which are characteristics of models with higher dimensions. We demonstrate that RegAE improves the block error rate (BLER) as compared with equivalent models from the literature. Thereby, it achieves a performance (1) better than that of a 4∕7 rate Hamming code with a 16 phase-shift keying (16PSK) modulation under an additive white Gaussian noise (AWGN) channel, (2) comparable to that of a 4∕7 rate maximum likelihood decoding (MLD) with a Eb∕N0 range from 1 dB to 5 dB, and (3) equivalent to that of an uncoded binary phase-shift keying (BPSK) modulation over a Eb∕N0 range from 0 dB to 10 dB.\n
\n\n\n\n \n\n \n \n \n \n \n \n Addressing the Data Acquisition Paradigm in the Early Detection of Pediatric Foot Deformities.\n \n \n \n \n\n\n \n Rosero-Montalvo, P., D.; Fuentes-Hernández, E., A.; Morocho-Cayamcela, M., E.; Sierra-Martínez, L., M.; and Peluffo-Ordóñez, D., H.\n\n\n \n\n\n\n Sensors, 21(13): 4422. 6 2021.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Addressing the Data Acquisition Paradigm in the Early Detection of Pediatric Foot Deformities},\n type = {article},\n year = {2021},\n keywords = {children,data analysis,embedded systems,machine learning,plantar pressure},\n pages = {4422},\n volume = {21},\n websites = {https://www.mdpi.com/1424-8220/21/13/4422},\n month = {6},\n publisher = {Multidisciplinary Digital Publishing Institute},\n day = {28},\n id = {880cdc26-a3b0-3290-964f-8c5b0bf6593b},\n created = {2021-12-04T20:47:49.262Z},\n accessed = {2021-07-21},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T14:24:44.316Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n abstract = {<p>The analysis of plantar pressure through podometry has allowed analyzing and detecting different types of disorders and treatments in child patients. Early detection of an inadequate distribution of the patient’s weight can prevent serious injuries to the knees and lower spine. In this paper, an embedded system capable of detecting the presence of normal, flat, or arched footprints using resistive pressure sensors was proposed. For this purpose, both hardware- and software-related criteria were studied for an improved data acquisition through signal coupling and filtering processes. Subsequently, learning algorithms allowed us to estimate the type of footprint biomechanics in preschool and school children volunteers. As a result, the proposed algorithm achieved an overall classification accuracy of 97.2%. A flat feet share of 60% was encountered in a sample of 1000 preschool children. Similarly, flat feet were observed in 52% of a sample of 600 school children.</p>},\n bibtype = {article},\n author = {Rosero-Montalvo, Paul D. and Fuentes-Hernández, Edison A. and Morocho-Cayamcela, Manuel E. and Sierra-Martínez, Luz M. and Peluffo-Ordóñez, Diego H.},\n doi = {10.3390/s21134422},\n journal = {Sensors},\n number = {13}\n}\n\n
\n\n\nThe analysis of plantar pressure through podometry has allowed analyzing and detecting different types of disorders and treatments in child patients. Early detection of an inadequate distribution of the patient’s weight can prevent serious injuries to the knees and lower spine. In this paper, an embedded system capable of detecting the presence of normal, flat, or arched footprints using resistive pressure sensors was proposed. For this purpose, both hardware- and software-related criteria were studied for an improved data acquisition through signal coupling and filtering processes. Subsequently, learning algorithms allowed us to estimate the type of footprint biomechanics in preschool and school children volunteers. As a result, the proposed algorithm achieved an overall classification accuracy of 97.2%. A flat feet share of 60% was encountered in a sample of 1000 preschool children. Similarly, flat feet were observed in 52% of a sample of 600 school children.
\n\n \n\n \n \n \n \n \n \n Hyperparameter Tuning Over an Attention Model for Image Captioning.\n \n \n \n \n\n\n \n Castro, R.; Pineda, I.; and Morocho-Cayamcela, M., E.\n\n\n \n\n\n\n In Salgado Guerrero, J., P.; Chicaiza Espinosa, J.; Cerrada Lozada, M.; and Berrezueta-Guzman, S., editor(s), Information and Communication Technologies. TICEC 2021. Communications in Computer and Information Science, volume 1456, pages 172-183, 2021. Springer International Publishing\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n@inproceedings{\n title = {Hyperparameter Tuning Over an Attention Model for Image Captioning},\n type = {inproceedings},\n year = {2021},\n pages = {172-183},\n volume = {1456},\n websites = {https://link.springer.com/chapter/10.1007/978-3-030-89941-7_13},\n publisher = {Springer International Publishing},\n city = {Guayaquil},\n id = {f448a861-8787-3015-b75f-72a04fd20d88},\n created = {2021-12-04T20:47:50.553Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-02-23T19:03:03.608Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n abstract = {Considering the historical trajectory and evolution of image captioning as a research area, this paper focuses on visual attention as an approach to solve captioning tasks with computer vision. This article studies the efficiency of different hyperparameter configurations on a state-of-the-art visual attention architecture composed of a pre-trained residual neural network encoder, and a long short-term memory decoder. Results show that the selection of both the cost function and the gradient-based optimizer have a significant impact on the captioning results. Our system considers the cross-entropy, Kullback-Leibler divergence, mean squared error, and the negative log-likelihood loss functions, as well as the adaptive momentum, AdamW, RMSprop, stochastic gradient descent, and Adadelta optimizers. Based on the performance metrics, a combination of cross-entropy with Adam is identified as the best alternative returning a Top-5 accuracy value of 73.092, and a BLEU-4 value of 0.201. Setting the cross-entropy as an independent variable, the first two optimization alternatives prove the best performance with a BLEU-4 metric value of 0.201. In terms of the inference loss, Adam outperforms AdamW with 3.413 over 3.418 and a Top-5 accuracy of 73.092 over 72.989.},\n bibtype = {inproceedings},\n author = {Castro, Roberto and Pineda, Israel and Morocho-Cayamcela, Manuel Eugenio},\n editor = {Salgado Guerrero, Juan Pablo and Chicaiza Espinosa, Janneth and Cerrada Lozada, Mariela and Berrezueta-Guzman, Santiago},\n doi = {https://doi.org/10.1007/978-3-030-89941-7_13},\n booktitle = {Information and Communication Technologies. TICEC 2021. Communications in Computer and Information Science}\n}\n\n Considering the historical trajectory and evolution of image captioning as a research area, this paper focuses on visual attention as an approach to solve captioning tasks with computer vision. This article studies the efficiency of different hyperparameter configurations on a state-of-the-art visual attention architecture composed of a pre-trained residual neural network encoder, and a long short-term memory decoder. Results show that the selection of both the cost function and the gradient-based optimizer have a significant impact on the captioning results. Our system considers the cross-entropy, Kullback-Leibler divergence, mean squared error, and the negative log-likelihood loss functions, as well as the adaptive momentum, AdamW, RMSprop, stochastic gradient descent, and Adadelta optimizers. Based on the performance metrics, a combination of cross-entropy with Adam is identified as the best alternative returning a Top-5 accuracy value of 73.092, and a BLEU-4 value of 0.201. Setting the cross-entropy as an independent variable, the first two optimization alternatives prove the best performance with a BLEU-4 metric value of 0.201. In terms of the inference loss, Adam outperforms AdamW with 3.413 over 3.418 and a Top-5 accuracy of 73.092 over 72.989.\n
\n\n\n\n \n\n \n \n \n \n \n \n The Evolution and Takeoff of the Ecuadorian Economic Groups.\n \n \n \n \n\n\n \n Tulcanaza-Prieto, A., B.; and Morocho-Cayamcela, M., E.\n\n\n \n\n\n\n Economies, 9(4). 2021.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {The Evolution and Takeoff of the Ecuadorian Economic Groups},\n type = {article},\n year = {2021},\n volume = {9},\n websites = {https://www.mdpi.com/2227-7099/9/4/188},\n id = {dbacb0fd-6350-36d9-be38-f277ce05f4ce},\n created = {2021-12-04T20:47:51.137Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T14:23:10.408Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n source_type = {Article},\n private_publication = {false},\n abstract = {An economic group is a collection of parent and subsidiary corporations that operates as a single economic organism under the same legislature of control. The decisions taken by the economic groups in any country are among the most influential factors that impact its market and the country’s economic political scenario. This work studies the impact of the Ecuadorian economic groups from 2015 to 2019, where a historical peak of 300 economic groups was reached. However, the taxes representativeness of the Ecuadorian economic groups remained stable during the same period of analysis. We analyzed the financial and fiscal variables of the Ecuadorian ranking of firms, and detected the following of its economic groups: (i) They are still concentrating wealth despite the implementation of hard government policies to transparent the financial and economic information; (ii) They tend to compete in oligopolistic markets, given that their economic and financial decisions are interconnected with their family firms or consortium groups; (iii) They operate in a behavioral nature that follows a linear association between the total income, total assets, total equity, and total tax collection. We hope this work will serve as a future reference for researchers focused on the economic groups of Ecuador and Latin American countries.},\n bibtype = {article},\n author = {Tulcanaza-Prieto, Ana Belén and Morocho-Cayamcela, Manuel Eugenio},\n doi = {10.3390/economies9040188},\n journal = {Economies},\n number = {4}\n}\n\n An economic group is a collection of parent and subsidiary corporations that operates as a single economic organism under the same legislature of control. The decisions taken by the economic groups in any country are among the most influential factors that impact its market and the country’s economic political scenario. This work studies the impact of the Ecuadorian economic groups from 2015 to 2019, where a historical peak of 300 economic groups was reached. However, the taxes representativeness of the Ecuadorian economic groups remained stable during the same period of analysis. We analyzed the financial and fiscal variables of the Ecuadorian ranking of firms, and detected the following of its economic groups: (i) They are still concentrating wealth despite the implementation of hard government policies to transparent the financial and economic information; (ii) They tend to compete in oligopolistic markets, given that their economic and financial decisions are interconnected with their family firms or consortium groups; (iii) They operate in a behavioral nature that follows a linear association between the total income, total assets, total equity, and total tax collection. We hope this work will serve as a future reference for researchers focused on the economic groups of Ecuador and Latin American countries.\n
\n\n\n\n
\n\n \n \n 2020\n \n \n (12)\n \n \n
\n \n \n \n
\n \n \n\n \n \n \n \n \n \n Learning to Communicate with Autoencoders: Rethinking Wireless Systems with Deep Learning.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; Njoku, J., N.; Park, J.; and Lim, W.\n\n\n \n\n\n\n In 2020 International Conference on Artificial Intelligence in Information and Communication (ICAIIC), pages 308-311, 2020. \n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n\n@inproceedings{\n title = {Learning to Communicate with Autoencoders: Rethinking Wireless Systems with Deep Learning},\n type = {inproceedings},\n year = {2020},\n pages = {308-311},\n websites = {https://ieeexplore.ieee.org/document/9065246},\n city = {Fukuoka, Japan},\n id = {060d0c02-7288-3a61-a315-1d553cf6a39b},\n created = {2021-12-04T20:47:51.719Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:32:48.893Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n citation_key = {Morocho-Cayamcela2020},\n private_publication = {false},\n bibtype = {inproceedings},\n author = {Morocho-Cayamcela, Manuel Eugenio and Njoku, Judith Nkechinyere and Park, Jeonghun and Lim, Wansu},\n doi = {10.1109/ICAIIC48513.2020.9065246},\n booktitle = {2020 International Conference on Artificial Intelligence in Information and Communication (ICAIIC)}\n}\n\n \n\n \n \n \n \n \n \n Breaking Wireless Propagation Environmental Uncertainty with Deep Learning.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; Maier, M.; and Lim, W.\n\n\n \n\n\n\n IEEE Transactions on Wireless Communications, 19(8): 5075 - 5087. 2020.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Breaking Wireless Propagation Environmental Uncertainty with Deep Learning},\n type = {article},\n year = {2020},\n pages = {5075 - 5087},\n volume = {19},\n websites = {https://ieeexplore.ieee.org/document/9066907},\n id = {0a93a3eb-9307-3778-bc99-e4c6e7888869},\n created = {2021-12-04T20:47:52.254Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:32:08.504Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n abstract = {Wireless propagation loss modeling has gained significant attention due to its critical importance in forthcoming dynamic wireless technologies. Stochastic and map-based propagation models require more information (elevation extension, statistical scattering characteristics) than required by empirical models (i.e., operating frequency, distance between transceivers, and height of the antennas), but such information is not always available. Thus, empirical models are still widely used to evaluate coverage, link budget, and received signal strength. The drawback of empirical models is inaccuracy in highly dynamic transmitter and receiver environments. To reduce the error caused by the use of a single environment, we divide a geographical terrain to employ a specific propagation model in each segment of the wireless link. We enhance a deep learning (DL) encoder-decoder architecture to extract semantic information from satellite imagery to divide an environment into three classes. Our DL architecture achieved a segmentation accuracy of 89.41%, 86.47%, and 87.37% in urban, suburban, and rural classes, respectively. Simulation results indicate that estimating propagation loss with our multi-environment model reduced the root mean square deviation (RMSD) with respect to two publicly available wireless tracing datasets, CU-WART and Portland MetroFi, by 3.79dB and 4.09dB, respectively.},\n bibtype = {article},\n author = {Morocho-Cayamcela, Manuel Eugenio and Maier, Martin and Lim, Wansu},\n doi = {10.1109/TWC.2020.2986202},\n journal = {IEEE Transactions on Wireless Communications},\n number = {8}\n}\n\n Wireless propagation loss modeling has gained significant attention due to its critical importance in forthcoming dynamic wireless technologies. Stochastic and map-based propagation models require more information (elevation extension, statistical scattering characteristics) than required by empirical models (i.e., operating frequency, distance between transceivers, and height of the antennas), but such information is not always available. Thus, empirical models are still widely used to evaluate coverage, link budget, and received signal strength. The drawback of empirical models is inaccuracy in highly dynamic transmitter and receiver environments. To reduce the error caused by the use of a single environment, we divide a geographical terrain to employ a specific propagation model in each segment of the wireless link. We enhance a deep learning (DL) encoder-decoder architecture to extract semantic information from satellite imagery to divide an environment into three classes. Our DL architecture achieved a segmentation accuracy of 89.41%, 86.47%, and 87.37% in urban, suburban, and rural classes, respectively. Simulation results indicate that estimating propagation loss with our multi-environment model reduced the root mean square deviation (RMSD) with respect to two publicly available wireless tracing datasets, CU-WART and Portland MetroFi, by 3.79dB and 4.09dB, respectively.\n
\n\n\n\n \n\n \n \n \n \n \n \n Machine Learning to Improve Multi-hop Searching and Extended Wireless Reachability in V2X.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; Lee, H.; and Lim, W.\n\n\n \n\n\n\n IEEE Communications Letters, 24(7): 1477-1481. 7 2020.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n\n@article{\n title = {Machine Learning to Improve Multi-hop Searching and Extended Wireless Reachability in V2X},\n type = {article},\n year = {2020},\n pages = {1477-1481},\n volume = {24},\n websites = {https://ieeexplore.ieee.org/document/9046001?source=authoralert},\n month = {7},\n id = {145b93cb-d692-32bd-9008-62d9eb4697c5},\n created = {2021-12-04T20:47:52.986Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:32:19.698Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n citation_key = {Morocho-Cayamcela2020},\n private_publication = {false},\n bibtype = {article},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lee, Haeyoung and Lim, Wansu},\n doi = {10.1109/LCOMM.2020.2982887},\n journal = {IEEE Communications Letters},\n number = {7}\n}\n\n \n\n \n \n \n \n \n \n Accelerating Wireless Channel Autoencoders for Short Coherence-time Communications.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; and Lim, W.\n\n\n \n\n\n\n Journal of Communications and Networks, 22(3): 215-222. 6 2020.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n\n@article{\n title = {Accelerating Wireless Channel Autoencoders for Short Coherence-time Communications},\n type = {article},\n year = {2020},\n pages = {215-222},\n volume = {22},\n websites = {https://ieeexplore.ieee.org/document/9143573},\n month = {6},\n id = {2cc55a26-4b0b-3bd9-a0ca-a44edc72a2e2},\n created = {2021-12-04T20:47:53.510Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:31:54.134Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n bibtype = {article},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu},\n doi = {10.1109/JCN.2020.000011},\n journal = {Journal of Communications and Networks},\n number = {3}\n}\n\n \n\n \n \n \n \n \n \n Lateral confinement of high-impedance surface-waves through reinforcement learning.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; and Lim, W.\n\n\n \n\n\n\n IET Electronics Letters, 56(23): 1262-1264. 2020.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n\n@article{\n title = {Lateral confinement of high-impedance surface-waves through reinforcement learning},\n type = {article},\n year = {2020},\n pages = {1262-1264},\n volume = {56},\n websites = {https://digital-library.theiet.org/content/journals/10.1049/el.2020.1977},\n id = {3d35c26b-21a3-3e8d-bd76-77688a3c1164},\n created = {2021-12-04T20:47:54.046Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:30:15.673Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n bibtype = {article},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu},\n doi = {10.1049/el.2020.1977},\n journal = {IET Electronics Letters},\n number = {23}\n}\n\n \n\n \n \n \n \n \n \n Impact of the human body in wireless propagation of medical implants for tumor detection.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; Kim, M.; and Lim, W.\n\n\n \n\n\n\n Journal of Internet Computing and Services, 21(2): 19-26. 4 2020.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n\n@article{\n title = {Impact of the human body in wireless propagation of medical implants for tumor detection},\n type = {article},\n year = {2020},\n pages = {19-26},\n volume = {21},\n websites = {https://www.koreascience.or.kr/article/JAKO202014761779391.page},\n month = {4},\n id = {7f5c80ee-44fd-3218-9a23-a34bc5933352},\n created = {2021-12-04T20:47:54.777Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:33:23.884Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n bibtype = {article},\n author = {Morocho-Cayamcela, Manuel Eugenio and Kim, Myung-Sik and Lim, Wansu},\n doi = {10.7472/jksii.2020.21.2.19},\n journal = {Journal of Internet Computing and Services},\n number = {2}\n}\n\n \n\n \n \n \n \n \n \n Expanding the Coverage of Multihop V2V with DCNNs and Q-learning.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; and Lim, W.\n\n\n \n\n\n\n The Journal of Korean Institute of Communications and Information Sciences, 4545(3): 622-627. 3 2020.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n\n@article{\n title = {Expanding the Coverage of Multihop V2V with DCNNs and Q-learning},\n type = {article},\n year = {2020},\n pages = {622-627},\n volume = {4545},\n websites = {https://www.researchgate.net/publication/340134056_Expanding_the_Coverage_of_Multihop_V2V_with_DCNNs_and_Q-Learning},\n month = {3},\n id = {9645cb4a-3f7b-3335-81e2-f637b05d5ce7},\n created = {2021-12-04T20:47:55.323Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:32:36.423Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n bibtype = {article},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu},\n doi = {https://doi.org/10.7840/kics.2020.45.3.622},\n journal = {The Journal of Korean Institute of Communications and Information Sciences},\n number = {3}\n}\n\n \n\n \n \n \n \n \n \n Automatic Radar Waveform Recognition using the Wigner-Ville distribution and AlexNet-SVM.\n \n \n \n \n\n\n \n Njoku, J., N.; Morocho-Cayamcela, M., E.; and Lim, W.\n\n\n \n\n\n\n In KICS Summer Conference 2020, pages 469-472, 2020. KICS\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n\n@inproceedings{\n title = {Automatic Radar Waveform Recognition using the Wigner-Ville distribution and AlexNet-SVM},\n type = {inproceedings},\n year = {2020},\n pages = {469-472},\n websites = {https://www.dbpia.co.kr/journal/articleDetail?nodeId=NODE10498610},\n publisher = {KICS},\n city = {Pyeongchang, Korea},\n id = {f64b5c59-e2b8-38d6-a0d7-ecd973f3692f},\n created = {2021-12-04T20:47:55.859Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:31:41.053Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n bibtype = {inproceedings},\n author = {Njoku, Judith Nkechinyere and Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu},\n booktitle = {KICS Summer Conference 2020}\n}\n\n \n\n \n \n \n \n \n \n Thermal Camera Face Detection and Alignment using MTCNN.\n \n \n \n \n\n\n \n Caliwag, E., M., F.; Morocho-Cayamcela, M., E.; and Caliwag, A.\n\n\n \n\n\n\n In KICS Summer Conference 2020, pages 321-323, 2020. KICS\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n\n@inproceedings{\n title = {Thermal Camera Face Detection and Alignment using MTCNN},\n type = {inproceedings},\n year = {2020},\n pages = {321-323},\n websites = {https://www.researchgate.net/publication/346446536_Thermal_Camera_Face_Detection_and_Alignment_using_MTCNN},\n publisher = {KICS},\n city = {Pyeongchang, Korea},\n id = {fcbb72c8-c69b-3580-b7d0-254147540650},\n created = {2021-12-04T20:47:56.366Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:30:35.267Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n bibtype = {inproceedings},\n author = {Caliwag, EJ Miguel Francisco and Morocho-Cayamcela, Manuel Eugenio and Caliwag, Angela},\n booktitle = {KICS Summer Conference 2020}\n}\n\n \n\n \n \n \n \n \n \n Optimizing the Energy Consumption of an Unmanned Aerial Base Station.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; and Lim, W.\n\n\n \n\n\n\n In KICS Summer Conference 2020, pages 467-468, 2020. KICS\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n\n@inproceedings{\n title = {Optimizing the Energy Consumption of an Unmanned Aerial Base Station},\n type = {inproceedings},\n year = {2020},\n pages = {467-468},\n websites = {https://www.dbpia.co.kr/journal/articleDetail?nodeId=NODE10498609},\n publisher = {KICS},\n city = {Pyeongchang, Korea},\n id = {045da416-d47b-3f60-a0e4-3e404f874261},\n created = {2021-12-04T20:47:56.931Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:30:20.934Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n bibtype = {inproceedings},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu},\n booktitle = {KICS Summer Conference 2020}\n}\n\n \n\n \n \n \n \n \n \n On the Improved Particle Filter by Mahalanobis Distance Consideration.\n \n \n \n \n\n\n \n Park, J.; Morocho-Cayamcela, M., E.; Lim, W.; and Yang, Y.\n\n\n \n\n\n\n In KICS Summer Conference 2020, pages 500-501, 2020. KICS\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n\n@inproceedings{\n title = {On the Improved Particle Filter by Mahalanobis Distance Consideration},\n type = {inproceedings},\n year = {2020},\n pages = {500-501},\n websites = {https://www.researchgate.net/publication/346446582_On_the_Improved_Particle_Filter_by_Mahalanobis_Distance_Consideration},\n publisher = {KICS},\n city = {Pyeongchang, Korea},\n id = {8acd43ef-37ee-3ad5-bd53-7f94aa01cfed},\n created = {2021-12-04T20:47:57.435Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:33:36.726Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n bibtype = {inproceedings},\n author = {Park, Jeong-Hun and Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu and Yang, Yeon-Mo},\n booktitle = {KICS Summer Conference 2020}\n}\n\n \n\n \n \n \n \n \n \n Increasing the Segmentation Accuracy of Aerial Images with Dilated Spatial Pyramid Pooling.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.\n\n\n \n\n\n\n Electronic Letters on Computer Vision and Image Analysis, 19(2): 17-21. 2020.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Increasing the Segmentation Accuracy of Aerial Images with Dilated Spatial Pyramid Pooling},\n type = {article},\n year = {2020},\n keywords = {computer vision,image analysis,image segmentation,pattern recognition,supervised learning,wireless communications},\n pages = {17-21},\n volume = {19},\n websites = {https://elcvia.cvc.uab.es/article/view/v19-n2-Morocho-cayamcela},\n id = {50c4428f-5939-36a3-a86c-1ef9b875946f},\n created = {2021-12-04T20:47:58.147Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:33:06.764Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n abstract = {This thesis addresses the environmental uncertainty in satellite images as a computer vision task using semantic image segmentation. We focus in the reduction of the error caused by the use of a single-environment models in wireless communications. We propose to use computer vision and image analysis to segment a geographical terrain in order to employ a specific propagation model in each segment of the link. Our computer vision architecture achieved a segmentation accuracy of 89.41%, 86.47%, and 87.37% in the urban, suburban, and rural classes, respectively. Results indicate that estimating propagation loss with our multi-environment model reduced the root mean square deviation (RMSD) with respect to two publicly available tracing datasets.},\n bibtype = {article},\n author = {Morocho-Cayamcela, Manuel Eugenio},\n doi = {https://doi.org/10.5565/rev/elcvia.1337},\n journal = {Electronic Letters on Computer Vision and Image Analysis},\n number = {2}\n}\n\n This thesis addresses the environmental uncertainty in satellite images as a computer vision task using semantic image segmentation. We focus in the reduction of the error caused by the use of a single-environment models in wireless communications. We propose to use computer vision and image analysis to segment a geographical terrain in order to employ a specific propagation model in each segment of the link. Our computer vision architecture achieved a segmentation accuracy of 89.41%, 86.47%, and 87.37% in the urban, suburban, and rural classes, respectively. Results indicate that estimating propagation loss with our multi-environment model reduced the root mean square deviation (RMSD) with respect to two publicly available tracing datasets.\n
\n\n\n\n
\n\n \n \n 2019\n \n \n (6)\n \n \n
\n \n \n \n
\n \n \n\n \n \n \n \n \n \n Fine-tuning a pre-trained Convolutional Neural Network Model to translate American Sign Language in Real-time.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; and Lim, W.\n\n\n \n\n\n\n In 2019 International Conference on Computing, Networking and Communications (ICNC), pages 100-104, 2019. \n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@inproceedings{\n title = {Fine-tuning a pre-trained Convolutional Neural Network Model to translate American Sign Language in Real-time},\n type = {inproceedings},\n year = {2019},\n keywords = {Artificial Intelligence,Convolutional Neural Network,Image Classification,Real-Time,Sign Language,Transfer Learning},\n pages = {100-104},\n websites = {https://ieeexplore.ieee.org/document/8685536},\n city = {Honolulu, HI, USA},\n id = {d055d546-6103-3858-88e8-6ad650d8f9d3},\n created = {2021-12-04T20:47:58.737Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:36:07.451Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n citation_key = {Morocho-Cayamcela2019a},\n private_publication = {false},\n abstract = {In this paper, we present a real-time American Sign Language (ASL) hand gesture recognizer based on an artificial intelligence execution, instead of the classical and outdated image processing modalities. Our approach uses a Convolutional Neural Network (CNN) to train a dataset of hundreds of instances from the ASL alphabet, extracting the features from each and every pixel and constructing an accurate translator based on predictions. This approach employs an atypical trade-off for a translator, where a superior precision and speed at the inference phase compensates for the computational expense at the early training. Furthermore, and to the best of our knowledge, the accuracy obtained by using the proposed deep learning technique, surpass the accuracy obtained using non-machine learning practices. The performance obtained by the proposed algorithm has also been compared with existing literature, showing that the suggested methodology outperformed the accuracy of its analogous counterparts.},\n bibtype = {inproceedings},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu},\n doi = {10.1109/ICCNC.2019.8685536},\n booktitle = {2019 International Conference on Computing, Networking and Communications (ICNC)}\n}\n\n In this paper, we present a real-time American Sign Language (ASL) hand gesture recognizer based on an artificial intelligence execution, instead of the classical and outdated image processing modalities. Our approach uses a Convolutional Neural Network (CNN) to train a dataset of hundreds of instances from the ASL alphabet, extracting the features from each and every pixel and constructing an accurate translator based on predictions. This approach employs an atypical trade-off for a translator, where a superior precision and speed at the inference phase compensates for the computational expense at the early training. Furthermore, and to the best of our knowledge, the accuracy obtained by using the proposed deep learning technique, surpass the accuracy obtained using non-machine learning practices. The performance obtained by the proposed algorithm has also been compared with existing literature, showing that the suggested methodology outperformed the accuracy of its analogous counterparts.\n
\n\n\n\n \n\n \n \n \n \n \n \n Research on discriminating the patterns of soldier uniforms using Deep Learning.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; and Lim, W.\n\n\n \n\n\n\n In 2019 Korean Institute of Communications and Information Sciences Conference (KICS), pages 493-494, 1 2019. KICS\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n\n@inproceedings{\n title = {Research on discriminating the patterns of soldier uniforms using Deep Learning},\n type = {inproceedings},\n year = {2019},\n pages = {493-494},\n websites = {https://www.researchgate.net/publication/335422006_Research_on_discriminating_the_patterns_of_soldier_uniforms_using_Deep_Learning},\n month = {1},\n publisher = {KICS},\n day = {23},\n city = {Pyeongchang, South Korea},\n id = {79d4a7c1-192a-3fe1-b5cb-b2849a886a71},\n created = {2021-12-04T20:47:59.244Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:36:23.271Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n bibtype = {inproceedings},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu},\n booktitle = {2019 Korean Institute of Communications and Information Sciences Conference (KICS)}\n}\n\n \n\n \n \n \n \n \n \n Finding the optimal path for V2V multi-hop connectivity with Q-learning and Convolutional Neural Networks.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; and Lim, W.\n\n\n \n\n\n\n In 2019 Korean Institute of Communications and Information Sciences Conference (KICS), pages 294-297, 6 2019. \n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n\n@inproceedings{\n title = {Finding the optimal path for V2V multi-hop connectivity with Q-learning and Convolutional Neural Networks},\n type = {inproceedings},\n year = {2019},\n pages = {294-297},\n websites = {https://www.researchgate.net/publication/335422295_Finding_the_optimal_path_for_V2V_multi-hop_connectivity_with_Q-learning_and_Convolutional_Neural_Networks},\n month = {6},\n day = {19},\n city = {Jeju, South Korea},\n id = {afb2d80e-b3aa-3a12-b36a-a37bc284d73b},\n created = {2021-12-04T20:47:59.801Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:35:45.907Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n citation_key = {Morocho-Cayamcela2019},\n private_publication = {false},\n bibtype = {inproceedings},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu},\n booktitle = {2019 Korean Institute of Communications and Information Sciences Conference (KICS)}\n}\n\n \n\n \n \n \n \n \n \n Machine Learning for 5G/B5G Mobile and Wireless Communications: Potential, Limitations, and Future Directions.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; Lee, H.; and Lim, W.\n\n\n \n\n\n\n IEEE Access, 7: 137184-137206. 2019.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Machine Learning for 5G/B5G Mobile and Wireless Communications: Potential, Limitations, and Future Directions},\n type = {article},\n year = {2019},\n keywords = {5G mobile communication,B5G,Machine learning,artificial intelligence,mobile communication,wireless communication},\n pages = {137184-137206},\n volume = {7},\n websites = {https://ieeexplore.ieee.org/document/8844682/},\n id = {7cafd101-9688-35cf-84a5-3bd3ff8ced78},\n created = {2021-12-04T20:48:00.431Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:35:30.734Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n citation_key = {Morocho-Cayamcela2019},\n private_publication = {false},\n abstract = {Driven by the demand to accommodate today’s growing mobile traffic, 5G is designed to be a key enabler and a leading infrastructure provider in the information and communication technology industry by supporting a variety of forthcoming services with diverse requirements. Considering the ever-increasing complexity of the network, and the emergence of novel use cases such as autonomous cars, industrial automation, virtual reality, e-health, and several intelligent applications, machine learning (ML) is expected to be essential to assist in making the 5G vision conceivable. This paper focuses on the potential solutions for 5G from an ML-perspective. First, we establish the fundamental concepts of supervised, unsupervised, and reinforcement learning, taking a look at what has been done so far in the adoption of ML in the context of mobile and wireless communication, organizing the literature in terms of the types of learning. We then discuss the promising approaches for how ML can contribute to supporting each target 5G network requirement, emphasizing its specific use cases and evaluating the impact and limitations they have on the operation of the network. Lastly, this paper investigates the potential features ofBeyond 5G (B5G), providing future research directions for how ML can contribute to realizing B5G. This article is intended to stimulate discussion on the role that ML can play to overcome the limitations for a wide deployment of autonomous 5G/B5G mobile and wireless communications.},\n bibtype = {article},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lee, Haeyoung and Lim, Wansu},\n doi = {10.1109/ACCESS.2019.2942390},\n journal = {IEEE Access}\n}\n\n Driven by the demand to accommodate today’s growing mobile traffic, 5G is designed to be a key enabler and a leading infrastructure provider in the information and communication technology industry by supporting a variety of forthcoming services with diverse requirements. Considering the ever-increasing complexity of the network, and the emergence of novel use cases such as autonomous cars, industrial automation, virtual reality, e-health, and several intelligent applications, machine learning (ML) is expected to be essential to assist in making the 5G vision conceivable. This paper focuses on the potential solutions for 5G from an ML-perspective. First, we establish the fundamental concepts of supervised, unsupervised, and reinforcement learning, taking a look at what has been done so far in the adoption of ML in the context of mobile and wireless communication, organizing the literature in terms of the types of learning. We then discuss the promising approaches for how ML can contribute to supporting each target 5G network requirement, emphasizing its specific use cases and evaluating the impact and limitations they have on the operation of the network. Lastly, this paper investigates the potential features ofBeyond 5G (B5G), providing future research directions for how ML can contribute to realizing B5G. This article is intended to stimulate discussion on the role that ML can play to overcome the limitations for a wide deployment of autonomous 5G/B5G mobile and wireless communications.\n
\n\n\n\n \n\n \n \n \n \n \n \n Proposed cost function using wireless propagation for self-organizing networks.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; and Lim, W.\n\n\n \n\n\n\n In 2019 Korean Institute of Communications and Information Sciences Conference (KICS), pages 172-174, 11 2019. \n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n@inproceedings{\n title = {Proposed cost function using wireless propagation for self-organizing networks},\n type = {inproceedings},\n year = {2019},\n pages = {172-174},\n websites = {https://www.researchgate.net/publication/337335837_Proposed_cost_function_using_wireless_propagation_for_self-organizing_networks},\n month = {11},\n day = {16},\n city = {Seoul, South Korea},\n id = {de0717ca-1714-3825-8e65-d4d8a01d4055},\n created = {2021-12-04T20:48:00.954Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:34:47.891Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n private_publication = {false},\n abstract = {Deep learning has been helping communication networks to reconfigure and heal themselves dynamically. Self-organizing maps (SOM) have been used for this purpose in order to create self-organizing networks (SON) to meet the requirements of the actual fifth-generation (5G) network. In this paper, we create hexagonal and random topologies to simulate the performance of SON under certain conditions. Specifically we analyze the impact of the cost function in the neural network in charge of updating the weights of the SON. Results demonstrate that using the propagation loss as a cost function instead of the Euclidean distance increases the performance of the network. Furthermore, we optimize the training stage with stochastic gradient descent with momentum and prove that the convergence time can be reduced by 14%, dumping the gradient update oscillations on its way.},\n bibtype = {inproceedings},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu},\n booktitle = {2019 Korean Institute of Communications and Information Sciences Conference (KICS)}\n}\n\n Deep learning has been helping communication networks to reconfigure and heal themselves dynamically. Self-organizing maps (SOM) have been used for this purpose in order to create self-organizing networks (SON) to meet the requirements of the actual fifth-generation (5G) network. In this paper, we create hexagonal and random topologies to simulate the performance of SON under certain conditions. Specifically we analyze the impact of the cost function in the neural network in charge of updating the weights of the SON. Results demonstrate that using the propagation loss as a cost function instead of the Euclidean distance increases the performance of the network. Furthermore, we optimize the training stage with stochastic gradient descent with momentum and prove that the convergence time can be reduced by 14%, dumping the gradient update oscillations on its way.\n
\n\n\n\n \n\n \n \n \n \n \n \n Hunger marketing and Blockchain Technology: Applications in Wireless Spectrum Management.\n \n \n \n \n\n\n \n Njoku, J., N.; Morocho-Cayamcela, M., E.; and Lim, W.\n\n\n \n\n\n\n In 2019 Korean Institute of Communications and Information Sciences Conference (KICS), pages 168-171, 11 2019. \n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n\n\n\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n\n
\n\n\n@inproceedings{\n title = {Hunger marketing and Blockchain Technology: Applications in Wireless Spectrum Management},\n type = {inproceedings},\n year = {2019},\n pages = {168-171},\n websites = {https://www.researchgate.net/publication/337335916_Hunger_marketing_and_Blockchain_Technology_Applications_in_Wireless_Spectrum_Management},\n month = {11},\n day = {16},\n city = {Seoul, South Korea},\n id = {ed02f41c-2e44-389c-bb23-4bbb60d67402},\n created = {2021-12-04T20:48:01.705Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-02-18T23:46:52.828Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n abstract = {Due to the ever-increasing demand by bandwidth-hungry mobile applications and the prevalent growth in wireless communication, effective spectrum management continues to constitute an important issue. So many spectrum management techniques have been employed in different areas including broadband satellite systems, cognitive acoustic networks, railway cognitive radio networks, and smart grid network environments. Spectrum management mechanisms have evolved to meet the different requirements of increasing spectrum use efficiency. In this paper, we discuss two state of the art approaches for spectrum management: Hunger marketing and Blockchain technology. We summarize the pros and cons of these technologies and their application in spectrum management.},\n bibtype = {inproceedings},\n author = {Njoku, Judith Nkechinyere and Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu},\n booktitle = {2019 Korean Institute of Communications and Information Sciences Conference (KICS)}\n}\n\n Due to the ever-increasing demand by bandwidth-hungry mobile applications and the prevalent growth in wireless communication, effective spectrum management continues to constitute an important issue. So many spectrum management techniques have been employed in different areas including broadband satellite systems, cognitive acoustic networks, railway cognitive radio networks, and smart grid network environments. Spectrum management mechanisms have evolved to meet the different requirements of increasing spectrum use efficiency. In this paper, we discuss two state of the art approaches for spectrum management: Hunger marketing and Blockchain technology. We summarize the pros and cons of these technologies and their application in spectrum management.\n
\n\n\n\n
\n\n \n \n 2018\n \n \n (7)\n \n \n
\n \n \n \n
\n \n \n\n \n \n \n \n \n \n Artificial Intelligence in 5G Technology: A Survey.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; and Lim, W.\n\n\n \n\n\n\n In 2018 International Conference on Information and Communication Technology Convergence (ICTC), pages 860-865, 2018. \n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@inproceedings{\n title = {Artificial Intelligence in 5G Technology: A Survey},\n type = {inproceedings},\n year = {2018},\n keywords = {5G Networks,Artificial Intelligence,Convergence,Deep Learning,IT,Machine Learning,Next Generation Network},\n pages = {860-865},\n websites = {https://ieeexplore.ieee.org/document/8539642},\n city = {Jeju},\n id = {0d510e0d-5ae1-3526-8e54-c779fb654c8c},\n created = {2021-12-04T20:48:02.237Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:36:48.073Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n citation_key = {Morocho-Cayamcela2018},\n private_publication = {false},\n abstract = {A fully operative and efficient 5G network cannot be complete without the inclusion of artificial intelligence (AI) routines. Existing 4G networks with all-IP (Internet Protocol) broadband connectivity are based on a reactive conception, leading to a poorly efficiency of the spectrum. AI and its subcategories like machine learning and deep learning have been evolving as a discipline, to the point that nowadays this mechanism allows fifth-generation (5G) wireless networks to be predictive and proactive, which is essential in making the 5G vision conceivable. This paper is motivated by the vision of intelligent base stations making decisions by themselves, mobile devices creating dynamically-adaptable clusters based on learned data rather than pre-established and fixed rules, that will take us to a improve in the efficiency, latency, and reliability of the current and real-time network applications in general. An exploration of the potential of AI-based solution approaches in the context of 5G mobile and wireless communications technology is presented, evaluating the different challenges and open issues for future research.},\n bibtype = {inproceedings},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu},\n doi = {10.1109/ICTC.2018.8539642},\n booktitle = {2018 International Conference on Information and Communication Technology Convergence (ICTC)}\n}\n\n A fully operative and efficient 5G network cannot be complete without the inclusion of artificial intelligence (AI) routines. Existing 4G networks with all-IP (Internet Protocol) broadband connectivity are based on a reactive conception, leading to a poorly efficiency of the spectrum. AI and its subcategories like machine learning and deep learning have been evolving as a discipline, to the point that nowadays this mechanism allows fifth-generation (5G) wireless networks to be predictive and proactive, which is essential in making the 5G vision conceivable. This paper is motivated by the vision of intelligent base stations making decisions by themselves, mobile devices creating dynamically-adaptable clusters based on learned data rather than pre-established and fixed rules, that will take us to a improve in the efficiency, latency, and reliability of the current and real-time network applications in general. An exploration of the potential of AI-based solution approaches in the context of 5G mobile and wireless communications technology is presented, evaluating the different challenges and open issues for future research.\n
\n\n\n\n \n\n \n \n \n \n \n \n Using body-measurement indices and wrist-type photoplethysmography signals to categorize consumer electronic users' health state through a smartwatch application.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; Lim, W.; and Kwon, D.\n\n\n \n\n\n\n In 2018 International Conference on Electronics, Information, and Communication (ICEIC), pages 101-104, 2018. \n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@inproceedings{\n title = {Using body-measurement indices and wrist-type photoplethysmography signals to categorize consumer electronic users' health state through a smartwatch application},\n type = {inproceedings},\n year = {2018},\n keywords = {Internet of things,consumer electronics,health awareness,heart rate monitoring,photoplethysmography (PPG),signals,smartwatch application},\n pages = {101-104},\n websites = {http://ieeexplore.ieee.org/document/8330573/},\n city = {Honolulu, HI},\n id = {d9e99bcc-c622-393c-980e-654bc22cd831},\n created = {2021-12-04T20:48:02.802Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:38:11.341Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n abstract = {The development of wearable technologies has been boosted considerably. On one hand, due to Internet of Things market expansion, and on the other, owing to health awareness on users demanding the convergence of wearable technology with applications that track their activity during the day, providing feedback on how to improve their consuming experience employing human vital signs. The availability of sensors on most of the wrist smartwatches and fitness bands, make this convergence a precondition for consumer devices industry and stakeholders. This article presents an analytical exploitation of this valuable data, and uses the embedded heart rate sensor from an attainable smartwatch to meet the prior requirements, parting the users according to the level of physical activity in pursuance of make a suitable recommendation from a specific consumer device edible catalog, according to the number of calories recommended for a healthy state.},\n bibtype = {inproceedings},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu and Kwon, Donguk},\n doi = {10.23919/ELINFOCOM.2018.8330573},\n booktitle = {2018 International Conference on Electronics, Information, and Communication (ICEIC)}\n}\n\n The development of wearable technologies has been boosted considerably. On one hand, due to Internet of Things market expansion, and on the other, owing to health awareness on users demanding the convergence of wearable technology with applications that track their activity during the day, providing feedback on how to improve their consuming experience employing human vital signs. The availability of sensors on most of the wrist smartwatches and fitness bands, make this convergence a precondition for consumer devices industry and stakeholders. This article presents an analytical exploitation of this valuable data, and uses the embedded heart rate sensor from an attainable smartwatch to meet the prior requirements, parting the users according to the level of physical activity in pursuance of make a suitable recommendation from a specific consumer device edible catalog, according to the number of calories recommended for a healthy state.\n
\n\n\n\n \n\n \n \n \n \n \n \n An artificially structured step-index metasurface for 10GHz leaky waveguides and antennas.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; Angsanto, S., R.; Lim, W.; and Caliwag, A.\n\n\n \n\n\n\n In 2018 IEEE 4th World Forum on Internet of Things (WF-IoT), pages 568-573, 2018. \n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@inproceedings{\n title = {An artificially structured step-index metasurface for 10GHz leaky waveguides and antennas},\n type = {inproceedings},\n year = {2018},\n keywords = {Metasurface,antenna,high impedance surface,leaky-lave,surface waves},\n pages = {568-573},\n websites = {https://ieeexplore.ieee.org/document/8355195/},\n city = {Singapore},\n id = {3c6e2c5c-ae4b-3e79-bb25-9d4ae25e6596},\n created = {2021-12-04T20:48:03.332Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:37:59.174Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n abstract = {Metasurfaces can be engineered to guide surface waves in a homogeneous path, where sub-wavelength size printed patches are etched on a grounded high-frequency laminate. When the homogeneity of the patches is compromised or it is inappropriately excited, leakage takes place. This effect can be exploited to design leaky-wave antennas for a wide range of applications, starting for Internet of Things (IoT) to Smart Factories. Step-index waveguides and antennas are engineered to work by introducing perturbations on the pattern so radiation occur in a controlled manner. The aim of this paper is to propose certain candidate antenna designs. The engineering method to derive effective refractive index is comprehensively investigated to guide and radiate surface waves at a centre operating frequency of 10GHz as a validation of the theory proposed.},\n bibtype = {inproceedings},\n author = {Morocho-Cayamcela, Manuel Eugenio and Angsanto, Stephen Ryan and Lim, Wansu and Caliwag, Angela},\n doi = {10.1109/WF-IoT.2018.8355195},\n booktitle = {2018 IEEE 4th World Forum on Internet of Things (WF-IoT)}\n}\n\n Metasurfaces can be engineered to guide surface waves in a homogeneous path, where sub-wavelength size printed patches are etched on a grounded high-frequency laminate. When the homogeneity of the patches is compromised or it is inappropriately excited, leakage takes place. This effect can be exploited to design leaky-wave antennas for a wide range of applications, starting for Internet of Things (IoT) to Smart Factories. Step-index waveguides and antennas are engineered to work by introducing perturbations on the pattern so radiation occur in a controlled manner. The aim of this paper is to propose certain candidate antenna designs. The engineering method to derive effective refractive index is comprehensively investigated to guide and radiate surface waves at a centre operating frequency of 10GHz as a validation of the theory proposed.\n
\n\n\n\n \n\n \n \n \n \n \n \n Elasticity of the Total Production measured by the Investment in Information and Communication Technologies: Evidence from the Ecuadorian Manufacturing Companies.\n \n \n \n \n\n\n \n Tulcanaza-Prieto, A., B.; and Morocho-Cayamcela, M.\n\n\n \n\n\n\n X-Pedientes Económicos, 2(3): 6-27. 2018.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Elasticity of the Total Production measured by the Investment in Information and Communication Technologies: Evidence from the Ecuadorian Manufacturing Companies},\n type = {article},\n year = {2018},\n keywords = {Ecuador,ICT investment,Information and Communication Technologies (ICTs),Total Production,manufacturing companies,manufacturing sector},\n pages = {6-27},\n volume = {2},\n websites = {https://ojs.supercias.gob.ec/index.php/X-pedientes_Economicos/article/view/20},\n city = {Quito, Ecuador},\n id = {763c171c-d4d9-3fe6-a99d-defc1dc26842},\n created = {2021-12-04T20:48:03.848Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:40:26.414Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n abstract = {This article analyzes the use, access, and development of the Information and Communication Technologies (ICTs) in the Ecuadorian manufacturing companies during the period 2012-2014. The theoretical aspects around ICT, its importance, limitations, and challenges are summarized. Furthermore, this manuscript proves the lineal correlation between the total production in the manufacturing sector and the ICT investment, and develops a “log-log” regression model among both variables, which proves the elasticity of the total production with respect to the ICT investment. This document gives conclusions and recommendations for future studies in order to improve the ICT investment in the Ecuadorian manufacturing firms.},\n bibtype = {article},\n author = {Tulcanaza-Prieto, Ana Belén and Morocho-Cayamcela, Manuel},\n doi = {2602-831X},\n journal = {X-Pedientes Económicos},\n number = {3}\n}\n\n This article analyzes the use, access, and development of the Information and Communication Technologies (ICTs) in the Ecuadorian manufacturing companies during the period 2012-2014. The theoretical aspects around ICT, its importance, limitations, and challenges are summarized. Furthermore, this manuscript proves the lineal correlation between the total production in the manufacturing sector and the ICT investment, and develops a “log-log” regression model among both variables, which proves the elasticity of the total production with respect to the ICT investment. This document gives conclusions and recommendations for future studies in order to improve the ICT investment in the Ecuadorian manufacturing firms.\n
\n\n\n\n \n\n \n \n \n \n \n \n Application of Image Classification using Machine Learning Technique on Smart Device.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; and Lim, W.\n\n\n \n\n\n\n The Journal of Korean Institute of Next Generation Computing, 14(1): 16-26. 2 2018.\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@article{\n title = {Application of Image Classification using Machine Learning Technique on Smart Device},\n type = {article},\n year = {2018},\n keywords = {Artificial intelligence,Image classification,Internet of things,Smartwatch application,사물인터넷,스마트워치 어플리케이션,이미지 분류,인공지능},\n pages = {16-26},\n volume = {14},\n websites = {https://www.kci.go.kr/kciportal/ci/sereArticleSearch/ciSereArtiView.kci?sereArticleSearchBean.artiId=ART002322791},\n month = {2},\n publisher = {KINGComputing},\n id = {c37ef1d8-afc2-3d98-8bb8-7d60855bc3e5},\n created = {2021-12-04T20:48:04.352Z},\n accessed = {2019-01-08},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:37:46.942Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n abstract = {This paper combines two of the most recent research topics for consumer electronic devices: Internet of Things (IoT) and Artificial Intelligence (AI), applied to solve an image classification problem for an electrical appliance’s recipe database. The first part of this article presents the development of an Android mobile application containing all the options from its recipe book manual. The problem addresses the inconvenience to manually search over the catalog categories to find the recipe that matches the actual ingredient available for the users in their home. The proposed solution establishes to recognize the ingredient with the mobile device camera using transfer learning over a pre-trained Convolutional Neural Network (CNN) to distinguish between the recipes that uses the ingredient, and exclude the recipe sets that are unrelated to the ingredient acquired with the camera. In the second part of the article, the availability of sensors on most of the wrist smartwatches and fitness bands is exploited to categorize the users according to the level of physical activity, in pursuance of making a healthy recommendation according to the number of calories in each of the recipes.},\n bibtype = {article},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu},\n doi = {1975-681X},\n journal = {The Journal of Korean Institute of Next Generation Computing},\n number = {1}\n}\n\n This paper combines two of the most recent research topics for consumer electronic devices: Internet of Things (IoT) and Artificial Intelligence (AI), applied to solve an image classification problem for an electrical appliance’s recipe database. The first part of this article presents the development of an Android mobile application containing all the options from its recipe book manual. The problem addresses the inconvenience to manually search over the catalog categories to find the recipe that matches the actual ingredient available for the users in their home. The proposed solution establishes to recognize the ingredient with the mobile device camera using transfer learning over a pre-trained Convolutional Neural Network (CNN) to distinguish between the recipes that uses the ingredient, and exclude the recipe sets that are unrelated to the ingredient acquired with the camera. In the second part of the article, the availability of sensors on most of the wrist smartwatches and fitness bands is exploited to categorize the users according to the level of physical activity, in pursuance of making a healthy recommendation according to the number of calories in each of the recipes.\n
\n\n\n\n \n\n \n \n \n \n \n \n Performance Evaluation of Sound Localizing using Response Power and Cross Correlation.\n \n \n \n \n\n\n \n Fadhil Zuandi, M.; Pratiwi Maharani, M.; Morocho-Cayamcela, M., E.; and Lim, W.\n\n\n \n\n\n\n In 2018 Korean Institute of Communications and Information Sciences Conference (KICS), pages 320, 1 2018. \n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@inproceedings{\n title = {Performance Evaluation of Sound Localizing using Response Power and Cross Correlation},\n type = {inproceedings},\n year = {2018},\n keywords = {Cross Correlation,Response Power,Sound Localization},\n pages = {320},\n websites = {https://www.eiric.or.kr/literature/ser_view.php?SnxGubun=INME&mode=total&searchCate=literature&more=Y&research=Y&re_q1=&gu=INME011A8&cmd=qryview&SnxIndxNum=210973&rownum=10&totalCnt=15&q1_t=V2Fuc3UgTGlt&listUrl=L3NlYXJjaC9yZXN1bHQucGhwP1NueEd1YnVuPUlOTUUm},\n month = {1},\n day = {17},\n city = {Jeongseon-gun, South Korea},\n id = {de5e8062-814c-3ddc-8659-05f06599933d},\n created = {2021-12-04T20:48:04.851Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:40:39.859Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n abstract = {Acoustic-based source localization is being widely developed in target localization due to its advantages over visual-based localization. In this paper, a comparison of CC-PHAT and RP-PHAT methods for acoustic source localization is introduced in order to determine the accuracy and response in speed for acoustic source localization applications. The results obtained from several experiments shows the performance comparison in terms of accuracies and computational times between the two approaches. Response power method is suitable for wider environment where large data can be obtained and beamforming can be performed. Cross Correlation is suitable for narrow environment where the distance between the source and the sound localizer is small thus reducing time difference of arrival and increasing accuracy.},\n bibtype = {inproceedings},\n author = {Fadhil Zuandi, Muhammad and Pratiwi Maharani, Mareska and Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu},\n doi = {NODE07368691},\n booktitle = {2018 Korean Institute of Communications and Information Sciences Conference (KICS)}\n}\n\n Acoustic-based source localization is being widely developed in target localization due to its advantages over visual-based localization. In this paper, a comparison of CC-PHAT and RP-PHAT methods for acoustic source localization is introduced in order to determine the accuracy and response in speed for acoustic source localization applications. The results obtained from several experiments shows the performance comparison in terms of accuracies and computational times between the two approaches. Response power method is suitable for wider environment where large data can be obtained and beamforming can be performed. Cross Correlation is suitable for narrow environment where the distance between the source and the sound localizer is small thus reducing time difference of arrival and increasing accuracy.\n
\n\n\n\n \n\n \n \n \n \n \n \n Learning the patterns of soldier uniforms from the weights of a Semantic Segmentation Network.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; and Lim, W.\n\n\n \n\n\n\n In 2018 International Workshop on Industrial IT Convergence 2018 (WIITC), 11 2018. \n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@inproceedings{\n title = {Learning the patterns of soldier uniforms from the weights of a Semantic Segmentation Network},\n type = {inproceedings},\n year = {2018},\n keywords = {artificial intelligence,deep learning,image,military defence,pattern recognition,processing},\n websites = {https://www.researchgate.net/publication/335422521_Learning_the_patterns_of_soldier_uniforms_from_the_weights_of_a_Semantic_Segmentation_Network},\n month = {11},\n day = {2},\n city = {Gumi, South Korea},\n id = {4a2db1f0-ec59-32a5-a046-b1566c89ccaa},\n created = {2021-12-04T20:48:05.830Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:40:11.141Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n private_publication = {false},\n abstract = {In this paper, we describe an artificial intelligence (AI)-based application for military and defence purposes, based on the detection of the patterns of soldier’s uniforms. Our approach uses two convolutional neural networks (CNN) to generate a segmentation network (SegNet) capable of being trained to perform semantic segmentation of any image pixel-by-pixel. Our approach learns from the data, as opposed to the traditional techniques based on image processing techniques that relies on fixed rules and needs to be programmed differently every time that the target suffers alterations.},\n bibtype = {inproceedings},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu},\n booktitle = {2018 International Workshop on Industrial IT Convergence 2018 (WIITC)}\n}\n\n In this paper, we describe an artificial intelligence (AI)-based application for military and defence purposes, based on the detection of the patterns of soldier’s uniforms. Our approach uses two convolutional neural networks (CNN) to generate a segmentation network (SegNet) capable of being trained to perform semantic segmentation of any image pixel-by-pixel. Our approach learns from the data, as opposed to the traditional techniques based on image processing techniques that relies on fixed rules and needs to be programmed differently every time that the target suffers alterations.\n
\n\n\n\n
\n\n \n \n 2017\n \n \n (2)\n \n \n
\n \n \n \n
\n \n \n\n \n \n \n \n \n \n A Transfer Learning approach for Image Classification on a Mobile Device.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; Lim, W.; and Kwon, D.\n\n\n \n\n\n\n In 2017 International Conference on Next Generation Computing (ICNGC), pages 180-182, 12 2017. Korean Institute of Next Generation Computing\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@inproceedings{\n title = {A Transfer Learning approach for Image Classification on a Mobile Device},\n type = {inproceedings},\n year = {2017},\n keywords = {convolutional neural network,image classification,internet of things,mobile application,transfer learning},\n pages = {180-182},\n websites = {http://www.kingpc.or.kr/wp/,http://www.icngc.org/img/file/program Schedule_2017b.pdf},\n month = {12},\n publisher = {Korean Institute of Next Generation Computing},\n day = {21},\n city = {Kaohsiung, Taiwan},\n id = {523960b7-a8f0-3429-9cd6-2fd0a788f5ad},\n created = {2021-12-04T20:48:06.388Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:42:40.112Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n abstract = {This paper combines two of the most recent research topics for consumer electronic devices: Internet of Things (IoT) and Artificial Intelligence (AI), applied in a commercial Whole Slow Juicer. The article presents the development of the Android mobile application containing all the options from its recipe book manual. The problem addresses its inconvenience to manually search over the catalog categories to find the recipe that matches the actual ingredient available for the users in their home. The proposed solution establishes to recognize the ingredient with the mobile device camera using transfer learning over a pre-trained Convolutional Neural Network (CNN) to distinguish between the recipes that uses the ingredient, and exclude the recipe sets that are unrelated to the ingredient acquired with the camera.},\n bibtype = {inproceedings},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu and Kwon, Donguk},\n booktitle = {2017 International Conference on Next Generation Computing (ICNGC)}\n}\n\n This paper combines two of the most recent research topics for consumer electronic devices: Internet of Things (IoT) and Artificial Intelligence (AI), applied in a commercial Whole Slow Juicer. The article presents the development of the Android mobile application containing all the options from its recipe book manual. The problem addresses its inconvenience to manually search over the catalog categories to find the recipe that matches the actual ingredient available for the users in their home. The proposed solution establishes to recognize the ingredient with the mobile device camera using transfer learning over a pre-trained Convolutional Neural Network (CNN) to distinguish between the recipes that uses the ingredient, and exclude the recipe sets that are unrelated to the ingredient acquired with the camera.\n
\n\n\n\n \n\n \n \n \n \n \n \n Indoor Propagation Loss and Fading Estimation of a Short-Range Wireless Implant based on ITU-R Prediction Models at 915MHz for Tumor Detection.\n \n \n \n \n\n\n \n Morocho-Cayamcela, M., E.; Lim, W.; Angsanto, S., R.; and Kwon, J.\n\n\n \n\n\n\n In 2017 Asia Pacific International Conference on Information Science and Technology (APIC-IST), pages 299-301, 6 2017. Korean Society for Internet Information\n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@inproceedings{\n title = {Indoor Propagation Loss and Fading Estimation of a Short-Range Wireless Implant based on ITU-R Prediction Models at 915MHz for Tumor Detection},\n type = {inproceedings},\n year = {2017},\n keywords = {channel modeling,fading,human tissue,indoor propagation,itu models,tumor detection},\n pages = {299-301},\n websites = {http://apicist.org/2017/APIC-IST_2017_program_download2.pdf},\n month = {6},\n publisher = {Korean Society for Internet Information},\n day = {25},\n city = {Chiang Mai, Thailand},\n id = {222ca4a5-a3c3-33cc-9407-134a5babc296},\n created = {2021-12-04T20:48:06.913Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2024-01-23T15:45:04.139Z},\n read = {false},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n private_publication = {false},\n abstract = {This paper proposes a methodological design practice for the estimations of fading and propagation losses for an implant-to-bedside short-range wireless link. The breakdown is based on the ITU-R propagation data and prediction models for the planning of indoor radiocommunication systems and radio local area networks in the frequency range of 300 MHz to 100 GHz, where primary estimations are based on a frequency of 915 MHz. The study also confirms the return loss variations for a spiral antenna implanted in humans when a tumor is detected near the implant. An explanation of the feasibility of the system implementation and technical recommendations to avoid interference are also proposed.},\n bibtype = {inproceedings},\n author = {Morocho-Cayamcela, Manuel Eugenio and Lim, Wansu and Angsanto, Stephen Ryan and Kwon, Jonghun},\n booktitle = {2017 Asia Pacific International Conference on Information Science and Technology (APIC-IST)}\n}\n\n This paper proposes a methodological design practice for the estimations of fading and propagation losses for an implant-to-bedside short-range wireless link. The breakdown is based on the ITU-R propagation data and prediction models for the planning of indoor radiocommunication systems and radio local area networks in the frequency range of 300 MHz to 100 GHz, where primary estimations are based on a frequency of 915 MHz. The study also confirms the return loss variations for a spiral antenna implanted in humans when a tumor is detected near the implant. An explanation of the feasibility of the system implementation and technical recommendations to avoid interference are also proposed.\n
\n\n\n\n
\n\n\n\n\n \n \n 2014\n \n \n (1)\n \n \n
\n \n \n \n
\n \n \n\n \n \n \n \n \n \n Integration of a triple-play platform service to the GPON infrastructure of the National Telecommunications Corporation of Ecuador.\n \n \n \n \n\n\n \n Coronel Gonzalez, E., J.; and Morocho-Cayamcela, M., E.\n\n\n \n\n\n\n In 2014 IEEE ANDESCON, 2014. \n \n\n\n\n
\n\n\n\n \n \nPaper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n\n\n\n\n\n \n\n\n\n \n \n
Paper\n \n \n \n Website\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n\n
\n\n\n@inproceedings{\n title = {Integration of a triple-play platform service to the GPON infrastructure of the National Telecommunications Corporation of Ecuador},\n type = {inproceedings},\n year = {2014},\n keywords = {Optical communication equipment,optical fiber networks,optical reflectometry measurements,telecommunications network topologies,time-domain reflectometer testing,triple-play service},\n websites = {http://ieeexplore.ieee.org/document/7098573/},\n city = {Cochabamba},\n id = {badf9000-83b0-32ba-81aa-8e7e52dcc946},\n created = {2021-12-04T20:48:07.717Z},\n file_attached = {true},\n profile_id = {ab671682-19b4-3dc7-aa8e-f674831dab33},\n group_id = {14dc25e5-4c22-3a10-8988-092d628a7743},\n last_modified = {2023-12-13T03:04:44.013Z},\n read = {true},\n starred = {true},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n citation_key = {Coronel-Gonzalez2014IntegrationEcuador},\n private_publication = {false},\n abstract = {In this paper we analyze the integration of HSI, VoIP and IPTV services into the optical network owned by the National Telecommunications Corporation of Ecuador; across the study of FTTx network topologies, convergence of technologies and access to the company services from the nodes. We overhaul the implementation process of the contract for installation and equipment supply in furtherance to submit the packaging distribution described. This article also evaluates the optical attenuation and reflection measurements in a common deployment of a fiber network scenario in pursuance to confirm the technical feasibility of using a MA5600T OLT for commissioning and a HG8245 ONT for content access. We additionally underscore a participatory research to study consumer demand and requirements in order to obtain a community-based perception of their assumption for the services received by different local suppliers, toward develop a fiber optic network with all the added values it entails.},\n bibtype = {inproceedings},\n author = {Coronel Gonzalez, Edwin Jonathan and Morocho-Cayamcela, Manuel Eugenio},\n doi = {10.1109/ANDESCON.2014.7098573},\n booktitle = {2014 IEEE ANDESCON}\n}\n\n In this paper we analyze the integration of HSI, VoIP and IPTV services into the optical network owned by the National Telecommunications Corporation of Ecuador; across the study of FTTx network topologies, convergence of technologies and access to the company services from the nodes. We overhaul the implementation process of the contract for installation and equipment supply in furtherance to submit the packaging distribution described. This article also evaluates the optical attenuation and reflection measurements in a common deployment of a fiber network scenario in pursuance to confirm the technical feasibility of using a MA5600T OLT for commissioning and a HG8245 ONT for content access. We additionally underscore a participatory research to study consumer demand and requirements in order to obtain a community-based perception of their assumption for the services received by different local suppliers, toward develop a fiber optic network with all the added values it entails.\n
\n\n\n