@InProceedings{8902983,
  author = {S. Kutluk and K. Kayabol and A. Akan},
  booktitle = {2019 27th European Signal Processing Conference (EUSIPCO)},
  title = {Spectral-Spatial Classification of Hyperspectral Images Using CNNs and Approximate Sparse Multinomial Logistic Regression},
  year = {2019},
  pages = {1-5},
  abstract = {We propose a technique for training convolutional neural networks (CNNs) in which the convolutional layers are trained using a gradient descent based method and the classification layer is trained using a second order method called approximate sparse multinomial logistic regression (ASMLR) which also provides a spatial smoothing procedure that increases the classification accuracy for hyperspectral images. ASMLR performs well on hyperspectral images, and CNNs are known to give good results in many applications such as image classification and object recognition. Thus, the proposed technique allows us to improve the performance of CNNs by training the whole network with an end-to-end framework. This approach takes advantage of convolutional layers for spectral feature extraction, and of the softmax classification layer for feature selection with sparsity constraints, and an intrinsic learning rate adjustment mechanism. In classification, we also use a spatial smoothing method. The proposed method was evaluated on two hyperspectral images for spectral-spatial land cover classification, and the results have shown that it outperforms the CNN and the ASMLR classifiers when they are used separately.},
  keywords = {convolutional neural nets;feature extraction;geophysical image processing;gradient methods;image classification;land cover;learning (artificial intelligence);object recognition;regression analysis;smoothing methods;classification accuracy;hyperspectral images;ASMLR;CNNs;image classification;convolutional layers;spectral feature extraction;softmax classification layer;spatial smoothing method;spectral-spatial land cover classification;approximate sparse multinomial logistic regression;training convolutional neural networks;gradient descent based method;second order method;spatial smoothing procedure;Training;Hyperspectral imaging;Feature extraction;Convolution;Convolutional neural networks;Smoothing methods;hyperspectral image classification;remote sensing;deep learning;convolutional neural networks;logistic regression},
  doi = {10.23919/EUSIPCO.2019.8902983},
  issn = {2076-1465},
  month = {Sep.},
  url = {https://www.eurasip.org/proceedings/eusipco/eusipco2019/proceedings/papers/1570533478.pdf},
}

{"_id":"5PJa5dsa3qsaD89CW","bibbaseid":"kutluk-kayabol-akan-spectralspatialclassificationofhyperspectralimagesusingcnnsandapproximatesparsemultinomiallogisticregression-2019","authorIDs":[],"author_short":["Kutluk, S.","Kayabol, K.","Akan, A."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["S."],"propositions":[],"lastnames":["Kutluk"],"suffixes":[]},{"firstnames":["K."],"propositions":[],"lastnames":["Kayabol"],"suffixes":[]},{"firstnames":["A."],"propositions":[],"lastnames":["Akan"],"suffixes":[]}],"booktitle":"2019 27th European Signal Processing Conference (EUSIPCO)","title":"Spectral-Spatial Classification of Hyperspectral Images Using CNNs and Approximate Sparse Multinomial Logistic Regression","year":"2019","pages":"1-5","abstract":"We propose a technique for training convolutional neural networks (CNNs) in which the convolutional layers are trained using a gradient descent based method and the classification layer is trained using a second order method called approximate sparse multinomial logistic regression (ASMLR) which also provides a spatial smoothing procedure that increases the classification accuracy for hyperspectral images. ASMLR performs well on hyperspectral images, and CNNs are known to give good results in many applications such as image classification and object recognition. Thus, the proposed technique allows us to improve the performance of CNNs by training the whole network with an end-to-end framework. This approach takes advantage of convolutional layers for spectral feature extraction, and of the softmax classification layer for feature selection with sparsity constraints, and an intrinsic learning rate adjustment mechanism. In classification, we also use a spatial smoothing method. The proposed method was evaluated on two hyperspectral images for spectral-spatial land cover classification, and the results have shown that it outperforms the CNN and the ASMLR classifiers when they are used separately.","keywords":"convolutional neural nets;feature extraction;geophysical image processing;gradient methods;image classification;land cover;learning (artificial intelligence);object recognition;regression analysis;smoothing methods;classification accuracy;hyperspectral images;ASMLR;CNNs;image classification;convolutional layers;spectral feature extraction;softmax classification layer;spatial smoothing method;spectral-spatial land cover classification;approximate sparse multinomial logistic regression;training convolutional neural networks;gradient descent based method;second order method;spatial smoothing procedure;Training;Hyperspectral imaging;Feature extraction;Convolution;Convolutional neural networks;Smoothing methods;hyperspectral image classification;remote sensing;deep learning;convolutional neural networks;logistic regression","doi":"10.23919/EUSIPCO.2019.8902983","issn":"2076-1465","month":"Sep.","url":"https://www.eurasip.org/proceedings/eusipco/eusipco2019/proceedings/papers/1570533478.pdf","bibtex":"@InProceedings{8902983,\n author = {S. Kutluk and K. Kayabol and A. Akan},\n booktitle = {2019 27th European Signal Processing Conference (EUSIPCO)},\n title = {Spectral-Spatial Classification of Hyperspectral Images Using CNNs and Approximate Sparse Multinomial Logistic Regression},\n year = {2019},\n pages = {1-5},\n abstract = {We propose a technique for training convolutional neural networks (CNNs) in which the convolutional layers are trained using a gradient descent based method and the classification layer is trained using a second order method called approximate sparse multinomial logistic regression (ASMLR) which also provides a spatial smoothing procedure that increases the classification accuracy for hyperspectral images. ASMLR performs well on hyperspectral images, and CNNs are known to give good results in many applications such as image classification and object recognition. Thus, the proposed technique allows us to improve the performance of CNNs by training the whole network with an end-to-end framework. This approach takes advantage of convolutional layers for spectral feature extraction, and of the softmax classification layer for feature selection with sparsity constraints, and an intrinsic learning rate adjustment mechanism. In classification, we also use a spatial smoothing method. The proposed method was evaluated on two hyperspectral images for spectral-spatial land cover classification, and the results have shown that it outperforms the CNN and the ASMLR classifiers when they are used separately.},\n keywords = {convolutional neural nets;feature extraction;geophysical image processing;gradient methods;image classification;land cover;learning (artificial intelligence);object recognition;regression analysis;smoothing methods;classification accuracy;hyperspectral images;ASMLR;CNNs;image classification;convolutional layers;spectral feature extraction;softmax classification layer;spatial smoothing method;spectral-spatial land cover classification;approximate sparse multinomial logistic regression;training convolutional neural networks;gradient descent based method;second order method;spatial smoothing procedure;Training;Hyperspectral imaging;Feature extraction;Convolution;Convolutional neural networks;Smoothing methods;hyperspectral image classification;remote sensing;deep learning;convolutional neural networks;logistic regression},\n doi = {10.23919/EUSIPCO.2019.8902983},\n issn = {2076-1465},\n month = {Sep.},\n url = {https://www.eurasip.org/proceedings/eusipco/eusipco2019/proceedings/papers/1570533478.pdf},\n}\n\n","author_short":["Kutluk, S.","Kayabol, K.","Akan, A."],"key":"8902983","id":"8902983","bibbaseid":"kutluk-kayabol-akan-spectralspatialclassificationofhyperspectralimagesusingcnnsandapproximatesparsemultinomiallogisticregression-2019","role":"author","urls":{"Paper":"https://www.eurasip.org/proceedings/eusipco/eusipco2019/proceedings/papers/1570533478.pdf"},"keyword":["convolutional neural nets;feature extraction;geophysical image processing;gradient methods;image classification;land cover;learning (artificial intelligence);object recognition;regression analysis;smoothing methods;classification accuracy;hyperspectral images;ASMLR;CNNs;image classification;convolutional layers;spectral feature extraction;softmax classification layer;spatial smoothing method;spectral-spatial land cover classification;approximate sparse multinomial logistic regression;training convolutional neural networks;gradient descent based method;second order method;spatial smoothing procedure;Training;Hyperspectral imaging;Feature extraction;Convolution;Convolutional neural networks;Smoothing methods;hyperspectral image classification;remote sensing;deep learning;convolutional neural networks;logistic regression"],"metadata":{"authorlinks":{}},"downloads":0},"bibtype":"inproceedings","biburl":"https://raw.githubusercontent.com/Roznn/EUSIPCO/main/eusipco2019url.bib","creationDate":"2021-02-11T19:15:22.079Z","downloads":0,"keywords":["convolutional neural nets;feature extraction;geophysical image processing;gradient methods;image classification;land cover;learning (artificial intelligence);object recognition;regression analysis;smoothing methods;classification accuracy;hyperspectral images;asmlr;cnns;image classification;convolutional layers;spectral feature extraction;softmax classification layer;spatial smoothing method;spectral-spatial land cover classification;approximate sparse multinomial logistic regression;training convolutional neural networks;gradient descent based method;second order method;spatial smoothing procedure;training;hyperspectral imaging;feature extraction;convolution;convolutional neural networks;smoothing methods;hyperspectral image classification;remote sensing;deep learning;convolutional neural networks;logistic regression"],"search_terms":["spectral","spatial","classification","hyperspectral","images","using","cnns","approximate","sparse","multinomial","logistic","regression","kutluk","kayabol","akan"],"title":"Spectral-Spatial Classification of Hyperspectral Images Using CNNs and Approximate Sparse Multinomial Logistic Regression","year":2019,"dataSources":["NqWTiMfRR56v86wRs","r6oz3cMyC99QfiuHW"]}