A dense spatial–spectral attention network for hyperspectral image band selection
- 31 January 2021
- journal article
- research article
- Published by Taylor & Francis in Remote Sensing Letters
- Vol. 12 (10) , 1025-1037
Abstract
Hyperspectral images (HSIs) are usually composed of hundreds of bands, which are highly correlated and redundant, leading to dimension disaster and high complexity of classification. In this paper, we propose an end-to-end dense spatial–spectral attention network (DSSAN) for hyperspectral image band selection to reduce the complexity of classification while ensuring the accuracy. In this network, an embeddable spatial–spectral attention module is designed, which can adaptively select the spectral bands from the raw input data. Moreover, this module is a plug-and-play complementary component and embedded in a dense convolutional network (DenseNet) for end-to-end training. The experimental results on two classic hyperspectral data sets demonstrate that the proposed method is superior to several mainstream band selection methods in classification accuracy and the selected band subset has lower redundancy, which can meet the application requirements.Keywords
This publication has 27 references indexed in Scilit:
- Advanced Spectral Classifiers for Hyperspectral Images: A reviewIEEE Geoscience and Remote Sensing Magazine, 2017
- Unsupervised Band Selection Based on Evolutionary Multiobjective Optimization for Hyperspectral ImagesIEEE Transactions on Geoscience and Remote Sensing, 2015
- Band Selection Using Improved Sparse Subspace Clustering for Hyperspectral Imagery ClassificationIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2015
- A New Sparsity-Based Band Selection Method for Target Detection of Hyperspectral ImageIEEE Geoscience and Remote Sensing Letters, 2014
- Hyperspectral Band Selection by Multitask Sparsity PursuitIEEE Transactions on Geoscience and Remote Sensing, 2014
- A Fast Volume-Gradient-Based Band Selection Method for Hyperspectral ImageIEEE Transactions on Geoscience and Remote Sensing, 2014
- Hyperspectral Band Selection Based on Trivariate Mutual Information and Clonal SelectionIEEE Transactions on Geoscience and Remote Sensing, 2013
- Hyperspectral Remote Sensing Data Analysis and Future ChallengesIEEE Geoscience and Remote Sensing Magazine, 2013
- Clustering-based hyperspectral band selection using sparse nonnegative matrix factorizationJournal of Zhejiang University SCIENCE C, 2011
- A joint band prioritization and band-decorrelation approach to band selection for hyperspectral image classificationIEEE Transactions on Geoscience and Remote Sensing, 1999