Vehicle License Plate Detection and their Count for Traffic Management

Abstract

Vehicle License Plate Detection and their count deals with providing an effective solution for traffic management and monitoring of vehicles. In this study, we construct a model that provides two functionalities, one is to effectively extract the license plate number from the given input image or live streaming data and the other one is to count the number of vehicles. There are many such Automatic Number Plate Recognition (ANPR) models prevailing but this particular model is built considering the Indian vehicles. For license plate detection, we used OpenCV which is a computer vision-based library, and for Character recognition and segmentation process we applied Optical Character Recognition (OCR) method. For the vehicle counting we applied a background subtraction algorithm that detects moving objects in a series of static camera frames. Our model has the simplest implementation and at the same time its highly accurate for Indian vehicles.