Detection and Counting Vehicles in Parking Area using EasyOCR and DeepSORT

As the population has rapidly increased, the number of vehicles on the road has also grown, making intelligent transport detection systems increasingly necessary. Vehicle recognition and tracking are critical components in traffic surveillance systems, where managing traffic and ensuring safety are top priorities. However, with the growing number of vehicles, there is a need to increase the number of parking spaces and improve the organization of parking areas. While many studies have been conducted and various strategies used in this field, there is still room for further development and improvement. This study aims to implement a Vehicle Counting System (VCS) and a Vehicle Detection System (VDS) using deep learning and convolutional neural networks (CNN) to detect and count vehicles entering an open parking area. The proposed intelligent parking system is designed for effective parking area management. It employs EasyOCR for accurate license plate recognition, enabling comprehensive vehicle monitoring. The system combines this with DeepSORT learning to conduct real-time vehicle counting in comparison to available parking slots. The experimental results show that the proposed method can achieve more than 94% in the average detecting and counting accuracy.