An efficient Transformer with neighborhood contrastive tokenization for hyperspectral images classification

The success of vision Transformers (ViTs) relies heavily on the self-attention mechanism, which requires support from appropriate patch tokenization. However, hyperspectral image (HSI) often suffer from significant noise distortions and spectral uncertainty, which result in unstable attention patter...

Full description

Bibliographic Details

Published in:	International Journal of Applied Earth Observations and Geoinformation
Main Authors:	Miaomiao Liang, Xianhao Zhang, Xiangchun Yu, Lingjuan Yu, Zhe Meng, Xiaohong Zhang, Licheng Jiao
Format:	Article
Language:	English
Published:	Elsevier 2024-07-01
Subjects:	Hyperspectral image (HSI) Transformer Contrastive tokenization Self-supervised learning (SSL)
Online Access:	http://www.sciencedirect.com/science/article/pii/S1569843224003339