Boosting the Generalization Ability for Hyperspectral Image Classification using Spectral-spatial Axial Aggregation Transformer

TL;DR

This paper introduces SaaFormer, a spectral-spatial axial aggregation transformer that enhances hyperspectral image classification generalization across dataset partitions by emphasizing spectral features and spatial-spectral integration.

Contribution

The paper proposes SaaFormer, a novel transformer model with multi-level spectral extraction and axial aggregation attention to improve generalization in hyperspectral image classification.

Findings

SaaFormer performs well on random dataset partitions.

SaaFormer significantly outperforms other methods on non-overlapping partitions.

The model shows strong background classification capabilities.

Abstract

In the hyperspectral image classification (HSIC) task, the most commonly used model validation paradigm is partitioning the training-test dataset through pixel-wise random sampling. By training on a small amount of data, the deep learning model can achieve almost perfect accuracy. However, in our experiments, we found that the high accuracy was reached because the training and test datasets share a lot of information. On non-overlapping dataset partitions, well-performing models suffer significant performance degradation. To this end, we propose a spectral-spatial axial aggregation transformer model, namely SaaFormer, that preserves generalization across dataset partitions. SaaFormer applies a multi-level spectral extraction structure to segment the spectrum into multiple spectrum clips, such that the wavelength continuity of the spectrum across the channel are preserved. For each spectrum clip, the axial aggregation attention mechanism, which integrates spatial features along multiple spectral axes is applied to mine the spectral characteristic. The multi-level spectral extraction and the axial aggregation attention emphasize spectral characteristic to improve the model generalization. The experimental results on five publicly available datasets demonstrate that our model exhibits comparable performance on the random partition, while significantly outperforming other methods on non-overlapping partitions. Moreover, SaaFormer shows excellent performance on background classification.