TAttMSRecNet:Triplet-attention and multiscale reconstruction network for band selection in hyperspectral images
作者：
Highlights：
• We propose an end-to-end network for band selection in hyperspectral images.
• It applies a triplet-attention with a multiscale reconstruction network.
• Captures the robust feature representations at a low computation overhead.
• Finds the most informative subset of spectral bands.
• Yields significant classification performance improvements.
摘要
•We propose an end-to-end network for band selection in hyperspectral images.•It applies a triplet-attention with a multiscale reconstruction network.•Captures the robust feature representations at a low computation overhead.•Finds the most informative subset of spectral bands.•Yields significant classification performance improvements.
论文关键词：Hyperspectral images (HSIs),Band selection (BS),Triplet attention,Multiscale reconstruction network
论文评审过程：Received 2 March 2022, Revised 29 August 2022, Accepted 5 September 2022, Available online 12 September 2022, Version of Record 18 September 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.118797
