Automatic Number Plate Recognition and Parking Management
Pratik Sharma, Shivam Gupta, Pritesh Singh, Kunal Shejul, Dinesh Reddy
Abstract
ANPR (Automatic number plate recognition) is a widely used technique globally. It is a real time recognition system for vehicle identification. Parking of vehicles plays an important role especially in country like India. Due to increase in number of vehicles, we need a vehicle identification system, as most of the car parks and educational institutions the procedure for continued vehicle registration for entry and parking for site employees, workforce/college student getting inside requires the security people to verify the info of the car via examining of the identification document of the driver. This procedure of manual work is monotonous and time ingesting and is susceptible to incorrect information noted down by the security guard, also the sharing and backup of this automobile data is hard since the information is difficult to reproduce. The proposed system will be having Automatic number plate detection capabilities which will help in detection and storing of vehicle license plate and then allotting it a free parking lot through our parking management system. This simple and not complex system has the ability to significantly reduce the time taken in registration and also create a hassle-free experience in parking. The ANPR system developed, applies OCR which uses the raspberry pi camera to detect and identify the license plate. Yolo image detection algorithm is used for fastest possible recognition. The performance comparison of Tesseract and CNN depicts the usage scenarios. A combination of Yolov4 and Contour detection also forms one of the main basis of this research.