SpaSSA: superpixelwise adaptive SSA for unsupervised spatial-spectral feature extraction in hyperspectral image.

Sun, Genyun; Fu, Hang; Ren, Jinchang; Zhang, Aizhu; Zabalza, Jaime; Jia, Xiuping; Zhao, Huimin

Abstract

Singular spectral analysis (SSA) has recently been successfully applied to feature extraction in hyperspectral image (HSI), including conventional (1-D) SSA in spectral domain and 2-D SSA in spatial domain. However, there are some drawbacks, such as sensitivity to the window size, high computational complexity under a large window, and failing to extract joint spectral-spatial features. To tackle these issues, in this article, we propose superpixelwise adaptive SSA (SpaSSA), that is superpixelwise adaptive SSA for exploiting local spatial information of HSI. The extraction of local (instead of global) features, particularly in HSI, can be more effective for characterizing the objects within an image. In SpaSSA, conventional SSA and 2-D SSA are combined and adaptively applied to each superpixel derived from an oversegmented HSI. According to the size of the derived superpixels, either SSA or 2-D singular spectrum analysis (2D-SSA) is adaptively applied for feature extraction, where the embedding window in 2D-SSA is also adaptive to the size of the superpixel. Experimental results on the three datasets have shown that the proposed SpaSSA outperforms both SSA and 2D-SSA in terms of classification accuracy and computational complexity. By combining SpaSSA with the principal component analysis (SpaSSA-PCA), the accuracy of land-cover analysis can be further improved, outperforming several state-of-the-art approaches.

Citation

SUN, G., FU, H., REN, J., ZHANG, A., ZABALZA, J., JIA, X. and ZHAO, H. 2022. SpaSSA: superpixelwise adaptive SSA for unsupervised spatial-spectral feature extraction in hyperspectral image. IEEE transactions on cybernetics [online], 52(7), pages 6158-6169. Available from: https://doi.org/10.1109/TCYB.2021.3104100

• Journal Article Type: Article
• Acceptance Date: Aug 7, 2021
• Online Publication Date: Sep 9, 2021
• Publication Date: Jul 31, 2022
• Deposit Date: Sep 16, 2021
• Publicly Available Date: Sep 16, 2021
• Journal: IEEE Transactions on Cybernetics
• Print ISSN: 2168-2267
• Electronic ISSN: 2168-2275
• Publisher: Institute of Electrical and Electronics Engineers (IEEE)
• Peer Reviewed: Peer Reviewed
• Volume: 52
• Issue: 7
• Pages: 6158-6169
• DOI: https://doi.org/10.1109/tcyb.2021.3104100
• Keywords: Feature extraction; Hyperspectral image (HSI); Land-cover analysis; Remote sensing; Singular spectrum analysis (SSA); Superpixelwise adaptive SSA (SpaSSA)
• Public URL: https://rgu-repository.worktribe.com/output/1457165