A Literature Review on Quantifying Roadside Evolution Using Computer Vision and Deep Learning

This research uses computer vision and deep learning to automate the monitoring and analysis of land use changes in urban or rural areas.It replaces traditional manual methods with image segmentation models that classify satellite or aerial imagery into key land use categories such as residential areas, industrial zones, water bodies, and vegetation.By comparing these categories over time, the system tracks changes such as urban expansion, loss of green spaces, and variations in vegetation health.Using OpenCV, it calculates percentage changes in land use classes, providing a faster, cost-effective, and scalable approach.The findings offer valuable insights for urban planning, environmental management, and sustainable development, enabling data-driven decisions to address modern challenges.