Robust Parking Space Allocation System Using Open CV and Scikit-learn

Abstract

The proliferation of urbanisation has led to an increased demand for efficient parking management systems. In response, this project presents a Smart Parking Assistant system aimed at providing real-time space availability notifications to users through their smartphones. Leveraging computer vision techniques implemented via Open CV and machine learning algorithms from the Scikit-learn library, the system captures video feed from a webcam to detect and quantify the number of empty parking spaces in a designated area. Upon receiving a request from the user, the system processes the video feed to analyse the occupancy status of parking slots, utilising advanced image processing techniques to accurately identify empty spaces along with mask image of the parking lot. The model built using Scikit-learn efficiently categorises the available slots, enabling the system to relay the precise number of open spaces to the user's smartphone. Furthermore, the Smart Parking Assistant incorporates geo-spatial functionalities to enhance user experience. By integrating the Haversine distance formula, the system calculates the distances between the user's location and nearby parking areas. This information is then displayed to the user, allowing them to conveniently locate and navigate to the nearest available parking facility. The proposed system offers a comprehensive solution to address the challenges associated with parking management in urban environments. By harnessing the power of computer vision, machine learning, and geo-spatial technologies, it provides users with timely and accurate information regarding parking space availability, ultimately improving efficiency and convenience in urban parking scenarios.