Sustainable Strategy of Urban Green Space Landscape Planning based on Superpixel Segmentation Algorithm

Urban green spaces are vital for ecological services and social well-being, yet their optimization is challenged by land constraints and fragmentation. Enhancing connectivity and ecological benefits remains a critical focus in sustainable urban planning. This study proposes a sustainable strategy utilizing the SLIC superpixel segmentation algorithm for green space landscape planning. The methodology integrates remote sensing imagery and GIS for spatial data fusion, comparing SLIC’s performance against traditional K-means and Otsu algorithms. Experiments evaluated segmentation accuracy, connectivity optimization, and sustainability metrics. The SLIC algorithm demonstrated superior accuracy (91.2% user accuracy, 89.5% producer accuracy) and reduced fragmentation, yielding fewer connected regions. Ecological functionand social demand scores surpassed other methods, confirming its efficacy in balancing ecological and societal needs. The SLIC-based strategy significantly enhances green space connectivity and sustainability, offering a robust tool for urban planners. Limitations include reliance on simulated data; future work should incorporate real-world dynamics and socio-economic factors.