Review of Vision-Based Smart Parking Detection and Space Management Systems
Main Article Content
Abstract
The rapid growth of urbanization and vehicle ownership has created significant challenges in parking management, leading to traffic congestion, fuel wastage, and inefficient utilization of parking spaces. Vision-based smart parking detection systems have emerged as an effective solution for automating parking space monitoring and improving parking efficiency. This review presents a comprehensive analysis of image processing and computer vision techniques used in smart parking management systems. Various approaches including grayscale conversion, image enhancement, thresholding, segmentation, edge detection, object recognition, and lane-wise parking analysis are examined for their role in identifying vacant and occupied parking slots. The study also reviews the application of cameras, surveillance systems, MATLAB-based processing, and machine learning algorithms for real-time parking detection. Vision-based systems provide advantages such as low installation cost, reduced hardware complexity, real-time monitoring, and improved parking guidance compared to traditional sensor-based approaches. Furthermore, the review highlights challenges associated with lighting variations, weather conditions, image blur, camera positioning, and occlusion problems that affect system accuracy. Recent developments in artificial intelligence, deep learning, and cloud-based smart parking architectures are also discussed as future directions for enhancing system reliability and scalability. Vision-based parking management systems contribute significantly toward intelligent transportation infrastructure and sustainable urban mobility solutions.