Smart Urban Planning via Remote Sensing Data Fusion and GIS-Driven Computational Models

Accurate and rational urban planning is a major challenge in the 21st century. However, with the emergence of intelligent information systems (IIS), planning paradigms based on intelligent analysis are being continuously explored. Therefore, this study innovatively proposed a new urban planning framework that integrates remote sensing data analysis (change detection for data classification) and GIS functions. First, this study uses an improved multi-scale deep network and U-net framework to generate a new remote sensing image change detection algorithm. This change detection operation can automatically identify and quantify the changing trends in urban development. Subsequently, this study designed a remote sensing image classification method based on the improved YOLOv5s algorithm. The classification results can accurately analyze the specific categories of urban space, thereby conducting a specific study of the change trend. Finally, in the GIS part, this study used 3D GIS technology to deeply integrate remote sensing data with three-dimensional geographic information. The experiments were conducted on the Google Map dataset, and the results show that the proposed method achieves satisfactory results in terms of change detection and classification accuracy.