Detecting and monitoring building changes is vital for urban development, safety, and environmental sustainability. It helps city planners track growth, enforce zoning laws, update property records, and identify illegal constructions. Additionally, it supports assessing the effects of urbanization on infrastructure and natural resources, as well as the impact of natural disasters on urban areas. Leveraging deep learning models with high-resolution aerial or drone imagery automates large-scale change detection and reveals subtle alterations that manual inspections might overlook. This approach improves accuracy, reduces labour, and enables data-driven urban planning.

This model detects building changes and generates a continuous change magnitude raster, providing the probability of change at each pixel rather than a simple binary output.