MF-SRCDNet: Multi-feature fusion super-resolution building change detection framework for multi-sensor high-resolution remote sensing imagery

Building change detection is essential for evaluating land use, land cover change, and sustainable development. However, owing to the mismatched resolutions from multi-sensors and the complexity of the features of high-resolution images, traditional methods of building change detection have problems with accuracy and appli-cability. In this study, we propose a deep-learning-based multi-feature fusion super-resolution building change detection framework (MF-SRCDNet), comprising super-resolution (SR), multi-feature fusion, and change detec-tion (CD) modules. The SR module introduces a Res-UNet network to generate unified SR images with rich se-mantic information. To enhance the performance of MF-SRCDNet for complex building detection, an effective right-angle edge vision feature was designed and fused with a CD module with an improved feature extractor. The proposed method achieved the highest F1 values of 0.881, 0.857, and 0.964 for the three datasets, respectively, compared with different modules. The results also show improved robustness in different bi-temporal image resolution scale-difference experiments. The method proposed in this study can be applied to a variety of complex scenarios for building CD tasks with strong model generalization.