Semantic Change Detection in HR Remote Sensing Images With Joint Learning and Binary Change Enhancement Strategy

Semantic change detection (SCD) in high-resolution (HR) remote sensing images faces two issues: 1) isolated network for binary change detection (BCD) within multitask architecture result in suboptimal SCD performance and 2) false alarms or missed detections caused by illumination differences or seasonal transform. To address these issues, this study proposes a bi-temporal binary change enhancement network (Bi-BCENet). Specifically, we introduce a binary change enhancement (BCE) strategy based on multinetwork joint learning to achieve superior SCD via improving change areas prediction. Within the network’s high-level forward process, we develop a cross-attention fusion module (CAFM) to enhance global similarity modeling via cross-network matching fusion, and employ a cosine similarity-based auxiliary loss to optimize non-change’s semantic consistency. The experiments on SECOND and CINA-FX datasets demonstrate that Bi-BCENet outperforms representative SCD networks, achieving 62.08% and 84.95% in FSCD, where 66.88% and and 83.10% in mIoUSCD, respectively. And the ablation analysis validates Bi-BCENet’s effectiveness in reducing false alarms and missed detections in SCD results. Moreover, for specific SCD of cropland, Bi-BCENet shows its strong potential in single-to-multi-SCD.