基于彩色图像信息的车牌自动识别系统仿真研究  

Simulation Research on Automatic Recognition System Based on Color Image Information

作　　者：林国聪 王冬青[1] 薛斌强 刘金燕 LIN Guocong;WANG Dongqing;XUE Binqiang;LIU Jinyan(School of Automation,Qingdao University,Qingdao 266071,China)

机构地区：[1]青岛大学自动化学院

出　　处：《青岛大学学报（工程技术版）》2019年第4期24-28,共5页Journal of Qingdao University(Engineering & Technology Edition)

基　　金：国家自然科学基金项目(61603205)

摘　　要：针对汽车牌照自动识别技术在现代社会车辆管理中的应用,本文研究了一种基于Matlab软件的车牌识别系统。该系统包括图像预处理、车牌定位、字符分割及字符识别几部分。对采集好的车牌图像进行图像预处理,采用统计像素法对车牌图像进行定位,二值化已定位好的车牌图像,对处理过的车牌图像字符进行分割,并采用模板匹配法对分割好的车牌字符进行识别,同时选用100张不同场景、不同省份、不同品牌的汽车牌照,采用Matlab软件进行仿真实验。仿真结果表明,有79张汽车牌照识别正确,说明本文可以实现较为准确的车牌识别。该研究具有一定的实际应用参考价值。Aiming at the application of automatic license plate recognition technology in vehicle management in modern society,this paper studies a license plate recognition system based on Matlab software.The system includes image pre-processing,license plate location,character segmentation and character recognition.The image of the collected license plate image is preprocessed,the statistical pixel method is used to locate the license plate image,binarizing the already-positioned license plate image,segment the processed license plate image characters,and the template matching method is used to identify the segmented license plate characters.At the same time,100 different license plates of different scenes,different provinces and different brands are selected,and Matlab software is used for simulation experiments.The simulation results show that 79 vehicle license plates are correctly identified,which indicates that the present method can achieve more accurate license plate recognition.This research has certain practical application reference value.

关 键 词：车牌识别 车牌定位 字符切割 字符识别 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]