TY - GEN
T1 - Global-Local Channel Attention for Hyperspectral Image Classification
AU - Yan, Peilin
AU - Qin, Haolin
AU - Wang, Jihui
AU - Xu, Tingfa
AU - Song, Liqiang
AU - Li, Hui
AU - Li, Jianan
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - Hyperspectral image classification (HSIC) assigns a pixel-wise semantic label leveraging the rich information in the broad spectral band. However, most of the existent HSIC algorithm fail to take advantage of varying importance of different channels, which hinder the further improve of the performance. To this end, we originally propose a Global-Local Channel Attention (GLCA) module to assist the process of selecting useful channels while suppressing others. Working in a plug-and-play fashion, GLCA precisely re-calibrate channel-wise feature responses in a pixel-wise manner, flexible enough to be applied to any existing depth-based HSIC model with little additional computational cost. Rich experiments prove the effectiveness of our algorithm, and to the best of our knowledge, we establish new state-of-the-arts on multiple HSIC datasets.
AB - Hyperspectral image classification (HSIC) assigns a pixel-wise semantic label leveraging the rich information in the broad spectral band. However, most of the existent HSIC algorithm fail to take advantage of varying importance of different channels, which hinder the further improve of the performance. To this end, we originally propose a Global-Local Channel Attention (GLCA) module to assist the process of selecting useful channels while suppressing others. Working in a plug-and-play fashion, GLCA precisely re-calibrate channel-wise feature responses in a pixel-wise manner, flexible enough to be applied to any existing depth-based HSIC model with little additional computational cost. Rich experiments prove the effectiveness of our algorithm, and to the best of our knowledge, we establish new state-of-the-arts on multiple HSIC datasets.
KW - channel at-tention
KW - global-local feature
KW - hyperspectral image classification
KW - pixel-wise
UR - http://www.scopus.com/inward/record.url?scp=85127045229&partnerID=8YFLogxK
U2 - 10.1109/ICECET52533.2021.9698661
DO - 10.1109/ICECET52533.2021.9698661
M3 - Conference contribution
AN - SCOPUS:85127045229
T3 - International Conference on Electrical, Computer, and Energy Technologies, ICECET 2021
BT - International Conference on Electrical, Computer, and Energy Technologies, ICECET 2021
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2021 International Conference on Electrical, Computer, and Energy Technologies, ICECET 2021
Y2 - 9 December 2021 through 10 December 2021
ER -