Mapping Urban Villages in Beijing–Tianjin–Hebei Core Cities Through a Dual-Stage Fusion of Individual Building Footprints and Glimmer Imagery

Accelerated urbanization has led to the widespread emergence of informal urban settlements, particularly urban villages (UVs), posing significant challenges to social equity, public welfare, and sustainable urban development. Current research on UV identification often relies solely on high-resolution satellite imagery, neglecting the unique spatial patterns and socioeconomic characteristics of UVs, thus limiting the effective distinction between UVs and other confusing land-use types. In addition, the high acquisition cost of high-resolution imagery restricts its feasibility for large-scale UV mapping. To address these issues, this study proposes a dual-stage “candidate–classification” method integrating multisource remote sensing and geographic data. First, individual building footprints are analyzed using the density-based spatial clustering of applications with noise clustering algorithm to identify potential UV candidate areas based on spatial clustering characteristics. Subsequently, spatial structural and socioeconomic features are incorporated into a random forest classification model for the accurate UV identification. Experiments conducted in the core cities of the Beijing–Tianjin–Hebei region (Beijing, Tianjin, and Shijiazhuang) demonstrate excellent classification performance, achieving overall accuracy and F1-scores of 91.65% and 90.21%, respectively. Furthermore, gradually incorporating socioeconomic and building density features improves the average OA by 2.56% and 5.33%, respectively, confirming their significant contribution to UV identification accuracy. The proposed method effectively captures multidimensional UV characteristics, enabling the accurate identification of UVs and showing strong potential for nationwide application in UV mapping.