Dual-Channel Densenet for Hyperspectral Image Classification. Yang, G., Gewali, U. B., Ientilucci, E., Gartley, M., & Monteiro, S. T. In IEEE International Geoscience and Remote Sensing Symposium, pages 2595-2598, July, 2018.
doi  abstract   bibtex   
Deep neural networks provide deep extracted features for image classification. As a high dimension data, hyperspectral image (HSI) feature extraction is unlike an RGB image whose feature representation could not be simply generated in the spatial domain. To take full advantage of HSI, a dual-channel convolutional neural network (CNN) is applied, 1D convolution for the spectral domain and 2D convolution for spatial domain. For pixel-wise classification of HSI, in our network model, one-dimensional customized DenseNet is for extracting the hierarchical spectral features and another customized DenseNet is applied to extract the hierarchical spatial-related feature. Furthermore, we experimentally tuned the several widen factors and dense-net growth rates to evaluate the impact of hyper-parameter. To compare our proposed method with HSI classification methods, we test other three DNNs based method in two real-world HSI dataset. The result demonstrated our approach outperformed the state-of-art method.
@INPROCEEDINGS{8517520, 
author={G. {Yang} and U. B. {Gewali} and E. {Ientilucci} and M. {Gartley} and S. T. {Monteiro}}, 
booktitle={IEEE International Geoscience and Remote Sensing Symposium}, 
title={Dual-Channel Densenet for Hyperspectral Image Classification}, 
year={2018}, 
pages={2595-2598}, 
abstract={Deep neural networks provide deep extracted features for image classification. As a high dimension data, hyperspectral image (HSI) feature extraction is unlike an RGB image whose feature representation could not be simply generated in the spatial domain. To take full advantage of HSI, a dual-channel convolutional neural network (CNN) is applied, 1D convolution for the spectral domain and 2D convolution for spatial domain. For pixel-wise classification of HSI, in our network model, one-dimensional customized DenseNet is for extracting the hierarchical spectral features and another customized DenseNet is applied to extract the hierarchical spatial-related feature. Furthermore, we experimentally tuned the several widen factors and dense-net growth rates to evaluate the impact of hyper-parameter. To compare our proposed method with HSI classification methods, we test other three DNNs based method in two real-world HSI dataset. The result demonstrated our approach outperformed the state-of-art method.}, 
keywords={convolution;feature extraction;feedforward neural nets;geophysical image processing;hyperspectral imaging;image classification;image representation;remote sensing;DNN;dual-channel Densenet;hierarchical spectral feature extraction;HSI pixel-wise classification;HSI feature extraction;feature representation;dense-net growth rates;hierarchical spatial-related feature;one-dimensional customized DenseNet;network model;spectral domain;dual-channel convolutional neural network;spatial domain;high dimension data;deep neural networks;hyperspectral image classification;Feature extraction;Hyperspectral imaging;Neural networks;Computer architecture;Training;Two dimensional displays;Hyperspectral image classification;Dual-channel DenseNet;spatial-spectral}, 
doi={10.1109/IGARSS.2018.8517520}, 
ISSN={2153-7003}, 
month={July}
}
Downloads: 0