Monitoring changes in urban green spaces is essential for sustainable urban planning and climate resilience. This study presents an integrated approach using multi-temporal satellite imagery, geographic information systems (GIS), and artificial intelligence (AI) to assess green space dynamics in three major parks in Hanoi—Thong Nhat, Thu Le, and Bach Thao—between 2016 and 2025. Sentinel-2 and PlanetScope data were processed to derive vegetation indices (NDVI, EVI), serving as inputs for Random Forest and deep learning models to classify land cover changes.

AI-enhanced analysis improved accuracy in detecting subtle vegetation changes and enabled fusion of multi-source imagery. The results reveal contrasting trends: Thong Nhat Park experienced gradual vegetation decline, Thu Le Park showed fragmented changes due to infrastructure expansion, while Bach Thao Park exhibited signs of ecological recovery.

Final outputs were deployed in an interactive WebGIS platform, offering high-resolution visualization and temporal querying of vegetation conditions.

This study demonstrates a scalable, AI-powered method for urban green space monitoring and supports data-driven environmental management, spatial planning, and increased public access to ecological information.