Detecting three-dimensional changes in buildings plays a vital role in urban monitoring, sustainable development, and disaster management. This research presents an innovative method for detecting multi-class building changes. The proposed approach combines two advanced architectures—a Siamese transformer network and spatial-channel attention mechanisms—to enable the automated, precise detection of the structural changes. The primary innovation lies in the design of a dual-attention encoder block that simultaneously analyzes the local spatial relationships and channel dependencies to identify the changes.
A key challenge in change detection is the class imbalance (e.g., the unchanged buildings, the new constructions, and the demolished structures). To address this problem, methods such as the enhanced data augmentation and the overlapping patch extraction during preprocessing phase been employed. The proposed method was implemented on a stereo dataset from the GeoEye-1 satellite (0.5-meter resolution) and an aerial stereo dataset (0.08-meter resolution).
In the experiments, the proposed method achieved Kappa coefficients of 94% and 93% for the satellite and aerial datasets, respectively. This marks a significant improvement ( 3% increase in Kappa coefficient) compared to the state-of-the-art methods like ChangeFormer, which achieved 91% for both datasets. By enhancing feature extraction and performing robustly on diverse data, the model emerges as a powerful tool for urban environment monitoring, offering a scalable and reliable solution for urban planning and management.