Spatial–Spectral Split Attention Residual Network for Hyperspectral Image Classification
Open Access
- 1 December 2022
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 16 (19391404) , 419-430
Abstract
In the past few years, many convolutional neural networks (CNNs) have been applied to hyperspectral image (HSI) classification. However, many of them have the following drawbacks: they do not fully consider the abundant band spectral information and insufficiently extract the spatial information of HSI; all bands and neighboring pixels are treated equally, so CNNs may learn features from redundant or useless bands/pixels; and a significant amount of hidden semantic information is lost when a single-scale convolution kernel is used in CNNs. To alleviate these problems, we propose a spatial–spectral split attention residual networks (S $^{3}$ ARN) for HSI classification. In S $^{3}$ ARN, a split attention strategy is used to fuse the features extracted from multireceptive fields, in which both spectral and spatial split attention modules are composed of bottleneck residual blocks. Thanks to the bottleneck structure, the proposed method can effectively prevent overfitting, speeds up the model training, and reduces the network parameters. Moreover, the spectral and spatial attention residual branches aim to generate the attention masks, which can simultaneously emphasize useful bands and neighbor pixels and suppress useless ones. Experimental results on three benchmark datasets demonstrate the effectiveness of the proposed model for HSI classification.Keywords
Funding Information
- National Natural Science Foundation of China (61603159, 62162033, U21B2027)
- Yun nan Provincial Major Science and Technology Special Plan Projects (202002AD080001, 202103AA080015)
- Yun nan Foundation Research Projects (202101AT070438, 202101BE070001-056)
- Qing Lan Project of higher education of Jiangsu Province
This publication has 51 references indexed in Scilit:
- Spectral–Spatial Feature Extraction for Hyperspectral Image Classification: A Dimension Reduction and Deep Learning ApproachIEEE Transactions on Geoscience and Remote Sensing, 2016
- Hyperspectral Airborne “Viareggio 2013 Trial” Data Collection for Detection Algorithm AssessmentIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2016
- HD-CNN: Hierarchical Deep Convolutional Neural Networks for Large Scale Visual Recognition2015
- Multilayer manifold and sparsity constrainted nonnegative matrix factorization for hyperspectral unmixing2015
- Deep supervised learning for hyperspectral data classification through convolutional neural networks2015
- Convolutional neural networks at constrained time cost2015
- Going deeper with convolutions2015
- Deep Convolutional Neural Networks for Hyperspectral Image ClassificationJournal of Sensors, 2015
- Deep Learning-Based Classification of Hyperspectral DataIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014
- A model of saliency-based visual attention for rapid scene analysisIEEE Transactions on Pattern Analysis and Machine Intelligence, 1998