Development Of An Automatic Parking Space Occupancy Detection With Vehicle License Palte Detection, Data Logging And Web-Based Monitoring

Parking space availability monitoring is a process of getting vacancy information in a given parking lot. Most parking spaces in the Philippines use manual management without giving drivers information about vacancies. This paper proposes a camera-based system running on a web-based monitoring center that utilizes Convolutional Neural Networks (CNN) to achieve the detection of license plates within and out of timestamp logs and the identification of the vacancy of parking spaces for a display to enter drivers. The license plate detection is achieved using a retrained tiny You Only Look Once (YOLO) model with centroid tracking for plate localization in conjunction with Tesseract Optical Character Recognition (OCR). The detected plate images are stored in separate directories according to date and filenames according to time. The parking occupancy network employs a simplified alexnet to classify parking spaces as either 'vacant" or "occupied". Testing for this system is done on an emulated parking lot. Where the live camera feed on the entrance, and exit is simulated by pre-recorded videos. The exit feed is a delayed version of the entrance feed. Testing shows that the license plate detection works on an average time of 0.023 seconds with 100% accuracy on still images. Parking Space occupancy identifier can classify the vacancy of 37 parking slots in an average time of o.06 seconds with 10.14% of false occupancy identification, 0.92% of false vacancy identification, and 88.94% of correct identification. Though tested in a simulated environment, the test results show that the system can be implemented and applied to actual parking lots.