Spatial–Spectral Squeeze-and-Excitation Residual Network for Hyperspectral Image Classification

• Published in: Remote sensing (Basel, Switzerland) Vol. 11; no. 7; p. 884
• Main Authors: Wang, Li, Peng, Jiangtao, Sun, Weiwei
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.04.2019
• Subjects: Band spectra; Banded structure; Classification; convolutional neural networks; Deep learning; Earth science; Excitation spectra; hyperspectral images; Hyperspectral imaging; Image classification; International conferences; Machine learning; Modules; Neighborhoods; Neural networks; Pattern recognition; Pixels; Remote sensing; Spatial data; Spectral bands; spectral–spatial feature extraction; squeeze and excitation; Beijing China; United States > US; China
• DOI: 10.3390/rs11070884
• ISSN: 2072-4292

Jointly using spectral and spatial information has become a mainstream strategy in the field of hyperspectral image (HSI) processing, especially for classification. However, due to the existence of noisy or correlated spectral bands in the spectral domain and inhomogeneous pixels in the spatial neighborhood, HSI classification results are often degraded and unsatisfactory. Motivated by the attention mechanism, this paper proposes a spatial–spectral squeeze-and-excitation (SSSE) module to adaptively learn the weights for different spectral bands and for different neighboring pixels. The SSSE structure can suppress or motivate features at a certain position, which can effectively resist noise interference and improve the classification results. Furthermore, we embed several SSSE modules into a residual network architecture and generate an SSSE-based residual network (SSSERN) model for HSI classification. The proposed SSSERN method is compared with several existing deep learning networks on two benchmark hyperspectral data sets. Experimental results demonstrate the effectiveness of our proposed network.