Mamba-FCS: Joint Spatio- Frequency Feature Fusion, Change-Guided Attention, and SeK Loss for Enhanced Semantic Change Detection in Remote Sensing

Researchers developed Mamba-FCS, an AI model that analyzes satellite images to detect and classify land-use changes, like deforestation or urban expansion. It combines spatial and frequency data for clarity, uses a specialized attention mechanism, and a new loss function to handle imbalanced data. The model achieved state-of-the-art accuracy, up to 96.25%, on benchmark datasets, outperforming previous methods based on CNNs and Transformers.