Using Computer Vision and Machine Learning for Efficient Parking Management: A Case Study

Abstract

This paper addresses the challenges associated with urban mobility and introduces a low-complexity system for detecting parking lot occupancy using machine learning and computer vision techniques. Through a field experiment at a Czech university, images of parking areas were captured to create a dataset titled T10Lot, and classified to get parking spot occupancy using Raspberry Pi computer. Results indicate satisfactory accuracy despite challenges such as varying lighting conditions and weather.