Recycling Material Classification using Convolutional Neural Networks. Liu*, K. & Liu, X. In Proceedings of the 21st IEEE International Conference on Machine Learning and Applications (ICMLA), 2022. IEEE (Full paper acceptance rate: <font color="red">32.4%</font>). ![Recycling Material Classification using Convolutional Neural Networks [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper ![Recycling Material Classification using Convolutional Neural Networks [pdf]](https://bibbase.org/img/filetypes/pdf.svg "pdf")
Paper abstract bibtex 35 downloads Using convolutional neural networks for classifying recyclable materials has shown promises for an effective and efficient way to classify recyclable trash. This work aims to demonstrate the most accurate CNN architecture for this task on our dataset combined from multiple sources, where in total 12,873 images of recyclable materials are collected over four classes: glass, metal, paper, and plastic. To this end, we use this dataset to train the CNN models, including a simple 8-layer CNN, AlexNet, VGGNet and InceptionNet are compared. Our empirical results show that VGGNet combined with transfer learning produces the best testing accuracy of 84.6%. Furthermore, we import this best model to a Raspberry Pi application and an Android application to demonstrate the potential for consumer and industrial usage.
@inproceedings{conf/icmla22/LiuL,
author = {Kaihua Liu* and Xudong Liu},
booktitle = {Proceedings of the 21st IEEE International Conference on Machine Learning and Applications (ICMLA)},
publisher = {IEEE (Full paper acceptance rate: <font color="red">32.4%</font>)},
url = {https://www.icmla-conference.org/icmla22/},
abstract = {Using convolutional neural networks for classifying recyclable materials has shown promises for an effective and efficient way to classify recyclable trash. This work aims to demonstrate the most accurate CNN architecture for this task on our dataset combined from multiple sources, where in total 12,873 images of recyclable materials are collected over four classes: glass, metal, paper, and plastic. To this end, we use this dataset to train the CNN models, including a simple 8-layer CNN, AlexNet, VGGNet and InceptionNet are compared. Our empirical results show that VGGNet combined with transfer learning produces the best testing accuracy of 84.6\%. Furthermore, we import this best model to a Raspberry Pi application and an Android application to demonstrate the potential for consumer and industrial usage.
},
url_Paper = {http://xudongliu.domains.unf.edu/resources/Recycle_icmla22.pdf},
title = {Recycling Material Classification using Convolutional Neural Networks},
year = 2022
}
Downloads: 35
{"_id":"8u3SZWC3tSJL4hh88","bibbaseid":"liu-liu-recyclingmaterialclassificationusingconvolutionalneuralnetworks-2022","author_short":["Liu*, K.","Liu, X."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["Kaihua"],"propositions":[],"lastnames":["Liu*"],"suffixes":[]},{"firstnames":["Xudong"],"propositions":[],"lastnames":["Liu"],"suffixes":[]}],"booktitle":"Proceedings of the 21st IEEE International Conference on Machine Learning and Applications (ICMLA)","publisher":"IEEE (Full paper acceptance rate: <font color=\"red\">32.4%</font>)","url":"https://www.icmla-conference.org/icmla22/","abstract":"Using convolutional neural networks for classifying recyclable materials has shown promises for an effective and efficient way to classify recyclable trash. This work aims to demonstrate the most accurate CNN architecture for this task on our dataset combined from multiple sources, where in total 12,873 images of recyclable materials are collected over four classes: glass, metal, paper, and plastic. To this end, we use this dataset to train the CNN models, including a simple 8-layer CNN, AlexNet, VGGNet and InceptionNet are compared. Our empirical results show that VGGNet combined with transfer learning produces the best testing accuracy of 84.6%. Furthermore, we import this best model to a Raspberry Pi application and an Android application to demonstrate the potential for consumer and industrial usage. ","url_paper":"http://xudongliu.domains.unf.edu/resources/Recycle_icmla22.pdf","title":"Recycling Material Classification using Convolutional Neural Networks","year":"2022","bibtex":"@inproceedings{conf/icmla22/LiuL,\n author = {Kaihua Liu* and Xudong Liu},\n booktitle = {Proceedings of the 21st IEEE International Conference on Machine Learning and Applications (ICMLA)},\n publisher = {IEEE (Full paper acceptance rate: <font color=\"red\">32.4%</font>)},\n url = {https://www.icmla-conference.org/icmla22/},\n abstract = {Using convolutional neural networks for classifying recyclable materials has shown promises for an effective and efficient way to classify recyclable trash. This work aims to demonstrate the most accurate CNN architecture for this task on our dataset combined from multiple sources, where in total 12,873 images of recyclable materials are collected over four classes: glass, metal, paper, and plastic. To this end, we use this dataset to train the CNN models, including a simple 8-layer CNN, AlexNet, VGGNet and InceptionNet are compared. Our empirical results show that VGGNet combined with transfer learning produces the best testing accuracy of 84.6\\%. Furthermore, we import this best model to a Raspberry Pi application and an Android application to demonstrate the potential for consumer and industrial usage.\n},\n url_Paper = {http://xudongliu.domains.unf.edu/resources/Recycle_icmla22.pdf},\n title = {Recycling Material Classification using Convolutional Neural Networks},\n year = 2022\n}\n\n\n","author_short":["Liu*, K.","Liu, X."],"key":"conf/icmla22/LiuL","id":"conf/icmla22/LiuL","bibbaseid":"liu-liu-recyclingmaterialclassificationusingconvolutionalneuralnetworks-2022","role":"author","urls":{"Paper":"https://www.icmla-conference.org/icmla22/"," paper":"http://xudongliu.domains.unf.edu/resources/Recycle_icmla22.pdf"},"metadata":{"authorlinks":{}},"downloads":35},"bibtype":"inproceedings","biburl":"http://xudongliu.domains.unf.edu/resources/mypubs.bib","dataSources":["t87zphR9BRLXbCGdz","HyNr96TFpCK8GJyjb","ZC4hkxYmM7m8bqanF","tckgwx8ugTmfQ6pQ4","nZ4KWueyniryTd5G3"],"keywords":[],"search_terms":["recycling","material","classification","using","convolutional","neural","networks","liu*","liu"],"title":"Recycling Material Classification using Convolutional Neural Networks","year":2022,"downloads":35}