Integrated Urban Governance With Multimodal Remote Sensing and Vision-Language Evidence

Abstract

Integrated urban governance benefits from multimodal evidence that links spatial observation, economic structure, visual interpretation, and policy decision-making. This topic connects urban systems science with hyperspectral-LiDAR fusion, forensic vision-language models, instance-aware super-resolution, and financial risk assessment. Remote sensing and vision-language models support observation of urban growth, infrastructure change, environmental conditions, and spatial inequality. Token-efficient image understanding reduces the computational burden of large-scale urban monitoring, while super-resolution improves visual detail in complex built environments. Urban sub-center analysis provides a framework for understanding spatial economic concentration, and cross-scale dynamics help connect neighborhood-level phenomena with metropolitan governance. Financial fraud risk modeling contributes an additional risk-analysis perspective for urban governance systems that require transparent and explainable decision support.