A Dual Attentive Generative Adversarial Network for Remote Sensing Image Change Detection

TL;DR

This paper introduces a dual attentive GAN framework for high-resolution remote sensing image change detection, effectively fusing multi-level features and improving detection accuracy without increasing model complexity.

Contribution

It proposes a novel dual attentive GAN architecture with multi-level feature fusion, multi-scale adaptive fusion, and context refinement modules for enhanced change detection.

Findings

Achieved 85.01% mean IoU on LEVIR dataset.

Outperformed existing methods in F1 score.

Enhanced spatial contiguity of change predictions.

Abstract

Remote sensing change detection between bi-temporal images receives growing concentration from researchers. However, comparing two bi-temporal images for detecting changes is challenging, as they demonstrate different appearances. In this paper, we propose a dual attentive generative adversarial network for achieving very high-resolution remote sensing image change detection tasks, which regards the detection model as a generator and attains the optimal weights of the detection model without increasing the parameters of the detection model through generative-adversarial strategy, boosting the spatial contiguity of predictions. Moreover, We design a multi-level feature extractor for effectively fusing multi-level features, which adopts the pre-trained model to extract multi-level features from bi-temporal images and introduces aggregate connections to fuse them. To strengthen the identification of multi-scale objects, we propose a multi-scale adaptive fusion module to adaptively fuse multi-scale features through various receptive fields and design a context refinement module to explore contextual dependencies. Moreover, the DAGAN framework utilizes the 4-layer convolution network as a discriminator to identify whether the synthetic image is fake or real. Extensive experiments represent that the DAGAN framework has better performance with 85.01% mean IoU and 91.48% mean F1 score than advanced methods on the LEVIR dataset.