SAR Image Change Detection Based on Multiscale Capsule Network

TL;DR

This paper introduces a Multiscale Capsule Network (Ms-CapsNet) for SAR image change detection, effectively handling speckle noise and deformation sensitivity by exploiting spatial relationships and adaptive feature fusion.

Contribution

The paper proposes a novel Ms-CapsNet architecture combining capsule modules and adaptive fusion convolution to improve SAR change detection robustness and accuracy.

Findings

Outperforms four state-of-the-art methods on three SAR datasets

Effectively reduces speckle noise impact in change detection

Demonstrates superior spatial relationship modeling with capsule modules

Abstract

Traditional change detection methods based on convolutional neural networks (CNNs) face the challenges of speckle noise and deformation sensitivity for synthetic aperture radar images. To mitigate these issues, we proposed a Multiscale Capsule Network (Ms-CapsNet) to extract the discriminative information between the changed and unchanged pixels. On the one hand, the capsule module is employed to exploit the spatial relationship of features. Therefore, equivariant properties can be achieved by aggregating the features from different positions. On the other hand, an adaptive fusion convolution (AFC) module is designed for the proposed Ms-CapsNet. Higher semantic features can be captured for the primary capsules. Feature extracted by the AFC module significantly improves the robustness to speckle noise. The effectiveness of the proposed Ms-CapsNet is verified on three real SAR datasets. The comparison experiments with four state-of-the-art methods demonstrated the efficiency of the proposed method. Our codes are available at https://github.com/summitgao/SAR_CD_MS_CapsNet.