Deep Learning-Based Vehicle Parking Occupancy Detection

In urban locations, effective parking management is essential to reducing traffic and improving mobility in general. The increasing need for precise, real-time parking space monitoring has not been satisfied by conventional techniques of vehicle parking occupancy detection, such as human counting and sensor-based systems. This research proposes a deep learning-based approach for vehicle parking occupancy detection, leveraging convolutional neural networks (CNNs) and transfer learning techniques. In this study, we provide a system for tracking and analyzing the occupancy of the parking lots at the National Commercial Bank's (NCB) main building in real time. The technique combines image analysis with deep learning. In order to monitor each parking space independently, we specifically use YOLO (You Only Look Once) as a deep learning model for object detection and OpenCV for image analysis to determine the coordinates of each parking slot. This study aims to optimize the use of parking areas and to reduce the time wasted by daily drivers to find the right parking slot for their cars. Also, it helps to better manage the space in the parking areas. The inference model was used to evaluate the accuracy of the model on custom data collected from actual parking lot environments. This evaluation confirmed the model's effectiveness in handling real-world data and achieving high performance in parking space occupancy detection, enhancing the system's reliability for practical use. The proposed system achieves very good accuracy, outperforming traditional methods