Deep Convolutional Neural Networks for Hyperspectral Image Classification
Beijing University of Chemical Technology · University of Colorado Boulder · +1 more institution
Abstract
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…
Citation impact
- FWCI
- 94.24
- Percentile
- 100%
- References
- 33
Authors
5Topics & keywords
- Hyperspectral imaging
- Convolutional neural network
- Artificial intelligence
- Pattern recognition (psychology)
- Computer science
- Classifier (UML)
- Pooling
- Deep learning
- Reduced inequalities
Funding
- NNNational Natural Science Foundation of ChinaAwards: 61371165, NCET-11-0711, 61302164, YETP0500, YETP0501, 2011CB706900
- BUBeijing University of Chemical TechnologyAwards: YETP0500, 61371165, NCET-11-0711, 61302164, 2011CB706900, YETP0501
- PFProgram for New Century Excellent Talents in UniversityAwards: NCET-11-, 61302164, NCET-11-0711, 2011CB706900, 61371165, YETP0501, YETP0500
- BHBeijing Higher Education Young Elite Teacher ProjectAwards: YETP0500, 61302164, 61371165, YETP0501, 2011CB706900, NCET-11-0711
- NKNational Key Research and Development Program of ChinaAward: 2011CB706900