Urban Change Detection

Urban change detection uses remote sensing data, primarily satellite imagery, to identify and map alterations in urban landscapes over time. Current research heavily employs deep learning, particularly convolutional neural networks and transformers, often within multi-task frameworks that integrate change detection with building segmentation for improved accuracy. This field is crucial for understanding urbanization's environmental and societal impacts, enabling better urban planning, resource management, and monitoring of sustainable development initiatives. The development of large, high-resolution datasets with fine-grained land use annotations is also a significant focus, improving model training and evaluation.