Details-Enhanced Multiscale Network for Building Change Detection in Remote Sensing Images

Remote sensing image change detection (CD) aims to identify changes occurring in bi-temporal remote sensing image of the same area at different times. Due to the presence of targets with various scales in the change detection scenes, existing methods have designed multi-scale encoders to extract change information at different scales. However, these methods do not directly enhance or enrich high-frequency features, resulting in persistent blurry change boundaries. To address this issue, we propose a multi-scale enhanced change detection method based on detail supplementation, dubbed MEDS-CD. It designs a multi-scale edge-guided adaptive filter to extract high-frequency information and compensate for multi-scale features. Specifically, a multi-scale transformer encoder is employed to extract multi-scale features. To capture high-frequency details lost during the encoding process, a multi-scale edge-guided adaptive filter is introduced to extract high-frequency details. Furthermore, to fully exploit the restored high-frequency information for change extraction and edge enhancement, we propose a multi-scale convolution modulation module and a fuzzy weighting strategy. Finally, the above process is supervised by a hybrid loss function. The proposed method MEDS-CD is validated on three public datasets, LEVIR-CD, WHU-CD and SYSU-CD. The experimental results show that MEDS-CD achieves better segmentation performance.