End-to-end Hyperspectral Image Change Detection Network Based on Band Selection

TL;DR

This paper introduces an end-to-end hyperspectral image change detection network with integrated band selection and band-specific attention, improving feature discrimination by focusing on critical bands for more accurate change detection.

Contribution

The proposed ECDBS network uniquely combines a deep learning band selection module with band-specific spatial attention blocks for joint optimization in change detection.

Findings

Outperforms state-of-the-art methods on three HSI-CD datasets.

Effectively retains critical bands to enhance change detection accuracy.

Demonstrates the benefit of band-specific feature extraction over uniform processing.

Abstract

For hyperspectral image change detection (HSI-CD), one key challenge is to reduce band redundancy, as only a few bands are crucial for change detection while other bands may be adverse to it. However, most existing HSI-CD methods directly extract change feature from full-dimensional HSIs, suffering from a degradation of feature discrimination. To address this issue, we propose an end-to-end hyperspectral image change detection network with band selection (ECDBS), which effectively retains the critical bands to promote change detection. The main ingredients of the network are a deep learning based band selection module and cascading band-specific spatial attention (BSA) blocks. The band selection module can be seamlessly integrated with subsequent CD models for joint optimization and end-to-end reasoning, rather than as a step separate from change detection. The BSA block extracts features from each band using a tailored strategy. Unlike the typically used feature extraction strategy that uniformly processes all bands, the BSA blocks considers the differences in feature distributions among widely spaced bands, thereupon extracting more sufficient change feature. Experimental evaluations conducted on three widely used HSI-CD datasets demonstrate the effectiveness and superiority of our proposed method over other state-of-the-art techniques.