A Review on GIS and Machine Learning Frameworks for Building Footprint Detection

This survey paper presents a comprehensive review of the latest techniques for extracting building footprints and identifying suitable solar sites, with a focus on the integration of Geographic Information Systems (GIS) and machine learning. As the demand for solar energy increases, the need for precise and efficient methods to locate potential solar installation sites becomes crucial. We conduct a structured analysis of recent developments in machine learning, particularly focusing on Convolutional Neural Networks (CNNs) and U-Net models for extracting building footprints from high-resolution remote sensing images. Additionally, we explore advanced techniques for assessing the suitability of solar sites by examining key factors such as solar radiation levels, roof orientation, and shading effects. Our review underscores the combined application of GIS and machine learning, highlighting optimization techniques that enhance decision-making in solar site selection. We address the technical challenges and solutions involved in merging these technologies, offering a detailed comparison of various approaches based on accuracy, efficiency, and scalability. The distinctive contribution of this work is its in-depth analysis of integrating machine learning with GIS for renewable energy applications, alongside a forward- looking view on future research directions. This paper serves as an essential resource for researchers, practitioners, and policymakers dedicated to advancing renewable energy generation and fostering urban sustainability through innovative technological advancements.