Geospatial mapping of urban utility networks integrating street view images and design rules

Urban utility networks (e.g., electric power networks [EPN] and water distribution networks [WDN]) are critical for urban management, yet large‐scale geospatial mapping of their topologies remains challenging. To this end, an intelligent method integrating street view images and design rules is proposed to generate geospatial mapping of EPN and WDN. First, utility markers are detected and located using street view images and an observation‐adaptive line‐of‐bearing model. Subsequently, EPN design rules are applied to classify localized markers into key EPN nodes, enabling automated mapping of multilevel EPN topologies. Finally, the framework is extended by incorporating WDN design rules, resulting in geospatial mapping for a hybrid loop‐tree WDN. A city‐scale application demonstrates that the proposed method achieves efficient and accurate mapping of EPN and WDN. Relying on street view images, the method constructs geospatial mappings of utility networks with community‐level granularity, supporting urban infrastructure management.