A license plate recognition system based on machine vision

Abstract

License Plate Recognition based on machine vision has been widely used in ITS(Intelligent Transportation System) and Smart Parking System. In this paper, we developed a system which included the image acquisition, license plate location, character segmentation and character recognition. To improve the accuracy of the license plate location, we improved the effect of the binary image positioning method. In character segmentation, we solved several typical license plate characteristics. In order to improve the accuracy of segmentation, we have taken a variety of methods to deal with their respective characteristics. In character recognition, we use template matching method and considered whether to consider the background matching. In the system, we just adopted the image recognition to recognize the characters. With no significant exposure, the area of license plate is more than 1/5 of the total images, we can get a good recognition.