SSANet-BS: Spectral–Spatial Cross-Dimensional Attention Network for Hyperspectral Band Selection
Open Access
- 3 August 2024
- journal article
- research article
- Published by MDPI AG in Remote Sensing
- Vol. 16 (15) , 2848
Abstract
Band selection (BS) aims to reduce redundancy in hyperspectral imagery (HSI). Existing BS approaches typically model HSI only in a single dimension, either spectral or spatial, without exploring the interactions between different dimensions. To this end, we propose an unsupervised BS method based on a spectral–spatial cross-dimensional attention network, named SSANet-BS. This network is comprised of three stages: a band attention module (BAM) that employs an attention mechanism to adaptively identify and select highly significant bands; two parallel spectral–spatial attention modules (SSAMs), which fuse complex spectral–spatial structural information across dimensions in HSI; a multi-scale reconstruction network that learns spectral–spatial nonlinear dependencies in the SSAM-fusion image at various scales and guides the BAM weights to automatically converge to the target bands via backpropagation. The three-stage structure of SSANet-BS enables the BAM weights to fully represent the saliency of the bands, thereby valuable bands are obtained automatically. Experimental results on four real hyperspectral datasets demonstrate the effectiveness of SSANet-BS.Keywords
Funding Information
- Qingdao Natural Science Foundation (23-2-1-64-zyyd-jch, 2023M731843, QDBSH20230101012, 42301380, 2023KJ232)
- China Postdoctoral Science Foundation (23-2-1-64-zyyd-jch, 2023M731843, QDBSH20230101012, 42301380, 2023KJ232)
- Postdoctoral Applied Research Foundation of Qingdao (23-2-1-64-zyyd-jch, 2023M731843, QDBSH20230101012, 42301380, 2023KJ232)
- National Natural Science Foundation of China (23-2-1-64-zyyd-jch, 2023M731843, QDBSH20230101012, 42301380, 2023KJ232)
- Science and Technology Support Plan for Youth Innovation of Colleges and Universities of Shandong Province of China (23-2-1-64-zyyd-jch, 2023M731843, QDBSH20230101012, 42301380, 2023KJ232)
This publication has 49 references indexed in Scilit:
- A Metareinforcement-Learning-Based Hyperspectral Image Classification With a Small Sample SetIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2023
- Hyperspectral band selection using a decomposition based multiobjective wrapper approachInfrared Physics & Technology, 2023
- Three-dimensional singular spectrum analysis for precise land cover classification from UAV-borne hyperspectral benchmark datasetsISPRS Journal of Photogrammetry and Remote Sensing, 2023
- TAttMSRecNet:Triplet-attention and multiscale reconstruction network for band selection in hyperspectral imagesExpert Systems with Applications, 2022
- Graph regularized spatial–spectral subspace clustering for hyperspectral band selectionNeural Networks, 2022
- A band selection approach based on a modified gray wolf optimizer and weight updating of bands for hyperspectral imageApplied Soft Computing, 2021
- Nonlocal Band Attention Network for Hyperspectral Image Band SelectionIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021
- CNN based hyperspectral image classification using unsupervised band selection and structure-preserving spatial featuresInfrared Physics & Technology, 2020
- Hyperspectral Band Selection via Adaptive Subspace Partition StrategyIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2019
- A joint band prioritization and band-decorrelation approach to band selection for hyperspectral image classificationIEEE Transactions on Geoscience and Remote Sensing, 1999