KGDNet: Knowledge-Guided Decoupled Modeling of Heterogeneous Change Mechanisms for Remote Sensing Change Detection (opens in new tab)

High-resolution remote sensing change detection (RSCD) methods typically rely on single-modal image features, making them highly susceptible to pseudo-changes induced by factors like illumination, shadows, and seasonal variations. Moreover, most approaches adopt a unified binary change paradigm, overlooking the heterogeneous mechanisms of change, such as appearance and disappearance. To address these limitations, this article presents KGDNet, a knowledge-guided tri-branch network designed for...