Extraction of Open Spaces and Identification of Suitable Rooftops for Urban Agriculture: Contribution of Geospatial Technologies for Sustainable Planning
Asmaa Moussaoui, Ouijdane Lahyan, Hajar Sbiki, Imane Sebari, Kenza Ait El Kadi
The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences
Urban Farm DB summary
This project aimed to extract vacant spaces and rooftops suitable for urban agriculture using geospatial technologies. For vacant space extraction, three classifiers were tested, with the Support Vector Machine (SVM) classifier achieving the highest accuracy at 75% precision. For rooftop extraction, object-based classification using SVM, the Footprint Building Extraction-USA deep learning model, and Mapflow's building model were tested, with the latter two achieving F-Factor values above 77%.