Band Attention Convolutional Networks For Hyperspectral Image Classification

TL;DR

This paper introduces a band attention module (BAM) that enhances hyperspectral image classification by enabling end-to-end band selection and weighting within CNNs, improving performance on benchmark datasets.

Contribution

The novel BAM component allows integrated band selection and weighting in CNNs for hyperspectral images, addressing redundancy and noise issues effectively.

Findings

Outperforms classical and deep learning methods on benchmark datasets.

Enables end-to-end training with integrated band attention.

Improves classification accuracy by reducing band redundancy effects.

Abstract

Redundancy and noise exist in the bands of hyperspectral images (HSIs). Thus, it is a good property to be able to select suitable parts from hundreds of input bands for HSIs classification methods. In this letter, a band attention module (BAM) is proposed to implement the deep learning based HSIs classification with the capacity of band selection or weighting. The proposed BAM can be seen as a plug-and-play complementary component of the existing classification networks which fully considers the adverse effects caused by the redundancy of the bands when using convolutional neural networks (CNNs) for HSIs classification. Unlike most of deep learning methods used in HSIs, the band attention module which is customized according to the characteristics of hyperspectral images is embedded in the ordinary CNNs for better performance. At the same time, unlike classical band selection or weighting methods, the proposed method achieves the end-to-end training instead of the separated stages. Experiments are carried out on two HSI benchmark datasets. Compared to some classical and advanced deep learning methods, numerical simulations under different evaluation criteria show that the proposed method have good performance. Last but not least, some advanced CNNs are combined with the proposed BAM for better performance.