Real Time Parking Space Detection System Using A Pixel-Based Processing Technique

This study proposes a real-time parking space detection using a pixel-based image processing technique. The proposed algorithm detects free spaces in a busy parking place video stream. The performance of the algorithm is compared with CNN algorithm using YOLOv8 software. The detection rate of the developed algorithm is found to be 9% higher than the YOLOv8 classification model with 5 times less error rate. Also, the proposed system missed fewer objects and requires less memory space than the YOLOv8 model.