ParkIt: An ANPR based Automated Parking Registry Using YOLO And EasyOCR

In today’s busy environment, maintaining a congestion free and secure parking area is a difficult task to achieve. This project proposes a advanced solution to overcome this challenge by leveraging state-of-the-art artificial intelligence techniques. The AI-Powered Object Detection and OCR techniques aims to create a secure and trustworthy environment during parking, ensuring safe and reliable management of a parking space for academies or companies. The proposed system consists of high-resolution cameras strategically placed at entry points to get images of vehicles entering the premises. These images are processed using convolutional neural networks trained specifically for optical character recognition tasks, enabling accurate extraction of alphanumeric characters from the number plates. The system employs image preprocessing techniques such as grayscaling and thresholding to enhance the OCR accuracy. Once the number plates are detected and characters extracted, the system crossreferences the extracted plate numbers with an existing database or stores them as new entries. This association links each vehicle with relevant owner details, ensuring a comprehensive and up-to-date registry. The real-time data management is handled by a database, which tracks vehicles currently parked and those that have exited.