一种改进Radon变换的车牌图像识别  

An Improved License Plate Recognition Research and Realization of the Radon Transform

作　　者：幸坚炬 李军[1] 谢赞福[1] 

机构地区：[1]广东技术师范学院计算机科学学院,广东广州510665

出　　处：《广东技术师范学院学报》2016年第11期62-66,85,共6页Journal of Guangdong Polytechnic Normal University

基　　金：广东省科技计划工业高新技术领域攻关项目(No.2013B010401032)

摘　　要：随着物联网技术在交通运输领域的应用,以及人们对交通管理的要求越来越高,智能交通管理已成为了研究热点之一.而车牌识别技术(Vehicle License Plate Recognition,VLPR)是智能交通系统的重要环节,是视频图像识别技术在汽车牌照识别中的应用.车牌识别技术可以将运动中的汽车牌照从各种复杂环境中提取并识别出来,能够识别车辆牌号、颜色等信息.车牌识别技术在交通运输中的应用越来越明显,具有很大的研究价值.论文中利用图像处理技术对车牌图像进行处理,准确定位车牌区域,然后对由于运动等原因所造成的车牌倾斜进行Radon变换校正,实现了对车牌的字符进行分割和识别,倾斜车牌识别率为90%,经过校正后的识别率可达98%,达到了良好的效果.As the Internet of things technology applied in the field of transportation, and people more and more high to the requirement of traffic management, intelligent traffic management has become one of research hot spots. The License Plate Recognition technology （Vehicle License Plate Recognition, VLPR） is video image Recognition technology in the application of Vehicle License Plate Recognition, and a very important part of intelligent transportation system. Car license plate in the movement of license plate recognition technology can be extracted from the complex environment and identified, including the license plate extraction, image preprocessing, feature extraction, and license plate character recognition technology. Also, can identify the vehicle brand, color and other information. License plate recognition technology in the application of transportation is more and more obvious, and has broad application prospects. In the paper by using image processing technology of license plate image processing, it can accurate positioning license plate area, and then correct the movement by Radon transform to recognize the license plate. After correction the recognition rate can reach 98%.

关 键 词：模式识别 车牌识别 倾斜校正 图像处理 

分 类 号：P407[天文地球—大气科学及气象学]