Hyperspectral images classification based on dense convolutional networks with spectral-wise attention mechanism

Bei Fang, Ying Li, Haokui Zhang, Jonathan Cheung Wai Chan

Research output: Contribution to journalArticlepeer-review

163 Scopus citations

Abstract

Hyperspectral images (HSIs) data that is typically presented in 3-D format offers an opportunity for 3-D networks to extract spectral and spatial features simultaneously. In this paper, we propose a novel end-to-end 3-D dense convolutional network with spectral-wise attention mechanism (MSDN-SA) for HSI classification. The proposed MSDN-SA exploits 3-D dilated convolutions to simultaneously capture the spectral and spatial features at different scales, and densely connects all 3-D feature maps with each other. In addition, a spectral-wise attention mechanism is introduced to enhance the distinguishability of spectral features, which improves the classification performance of the trained models. Experimental results on three HSI datasets demonstrate that our MSDN-SA achieves competitive performance for HSI classification.

Original languageEnglish
Article number159
JournalRemote Sensing
Volume11
Issue number2
DOIs
StatePublished - 1 Jan 2019

Keywords

  • Attention mechanism
  • Dense connectivity
  • Spectral-spatial feature extraction
  • hyperspectral image classification

Fingerprint

Dive into the research topics of 'Hyperspectral images classification based on dense convolutional networks with spectral-wise attention mechanism'. Together they form a unique fingerprint.

Cite this