Deep Convolutional Neural Networks for Hyperspectral Image Classification

Journal of Sensors - Tập 2015 - Trang 1-12 - 2015
Wei Hu1, Yangyu Huang1, Li Wei1, Fan Zhang1, Heng-Chao Li2,3
1College of Information Science and Technology, Beijing University of Chemical Technology, Beijing 10029, China
2Department of Aerospace Engineering Sciences, University of Colorado, Boulder, CO 80309 USA
3Sichuan Provincial Key Laboratory of Information Coding and Transmission, Southwest Jiaotong University, Chengdu 610031, China

Tóm tắt

Recently, convolutional neural networks have demonstrated excellent performance on various visual tasks, including the classification of common two-dimensional images. In this paper, deep convolutional neural networks are employed to classify hyperspectral images directly in spectral domain. More specifically, the architecture of the proposed classifier contains five layers with weights which are the input layer, the convolutional layer, the max pooling layer, the full connection layer, and the output layer. These five layers are implemented on each spectral signature to discriminate against others. Experimental results based on several hyperspectral image data sets demonstrate that the proposed method can achieve better classification performance than some traditional methods, such as support vector machines and the conventional deep learning-based methods.

Từ khóa


Tài liệu tham khảo

10.1109/79.974718

10.1109/TGRS.2004.827257

10.1109/TGRS.2009.2016214

10.1109/tgrs.2011.2165957

10.1109/TGRS.2004.831865

10.1016/j.isprsjprs.2010.11.001

10.1109/TGRS.2012.2205263

10.1080/014311697218700

10.1109/36.752239

10.1109/TGRS.2009.2037898

10.1126/science.1127647

1988, Neural Networks, 1, 119, 10.1016/0893-6080(88)90014-7

10.1109/5.726791

10.1109/tpami.2012.231

10.1109/jstars.2014.2329330

10.1162/neco.2008.12-07-661

10.1113/jphysiol.1968.sp008455

2012, Efficient backprop, 9

2014, The Journal of Machine Learning Research, 15, 1929

Thông tin
Thông tin xuất bản

Journal of Sensors

Tập 2015

1-12

Thông tin tác giả