Design of License Plate Recognition System Based on Image Processing

Abstract

With the rapid development of road transportation, the method of relying on manpower for car management has fallen behind. The road traffic management of China has gradually developed towards the direction of Internet. The core of road traffic management is the license plate recognition system, and the license plate automatic recognition system is used in many occasions, such as bridge crossing charges, parking lot unmanaged, illegal vehicle automatic recording and so on. License plate recognition system makes traffic management more convenient and reduces human resources. However, in the process of license plate recognition, external factors are uncontrollable. Weather, sunlight and other external factors will affect the license plate recognition. The defects of the license plate itself, stains and so on will have a greater impact on the license plate recognition. License plate recognition system is based on the appearance and characteristics of license plate, through a series of algorithms to achieve the purpose of recognizing license plate characters. The research focus of this paper mainly includes blue area positioning algorithm, character segmentation algorithm, character recognition algorithm and template matching, and so on. Through the image processing and analysis of license plate photos, the recognition of license plate characters in the image is finally realized. By using the system designed in this paper to process and count the license plate information, it can be concluded that the recognition rate of the designed license plate recognition system can reach more than 92.9%, which meets the design requirement.