DSA-Net: A novel deeply supervised attention-guided network for building change detection in high-resolution remote sensing images
Abstract: Highlights•A novel dual-branch end-to-end network is proposed for building change detection.•To improve detection accuracy, multi-level and multi-scale features are extracted.•Attention-guided skip-connection achieves better information fusion and retention.•Deep supervision module improves the network’s performance and robustness.•Experiments on two datasets demonstrate the superiority of the proposed network.
Loading