基于DM642的纹理检测与Adaboost分类器相结合的车牌定位  

Vehicle Plate Location by the Texture Recognition and Adaboost Classifier Based on DM642

作　　者：陈存弟 刘金清[1] 刘引[1] 蔡淑宽[1] 何世强[1] 周晓童 邓淑敏[1] 吴庆祥[1] 

机构地区：[1]福建师范大学光电与信息工程学院医学光电科学与技术教育部重点实验室,福州350007

出　　处：《计算机系统应用》2017年第7期56-64,共9页Computer Systems & Applications

基　　金：国家自然科学基金(61179011);福建教育厅项目(JAS151254);福建师大项目(I201502019)

摘　　要：智能交通系统(Intelligent Transportation System,简称ITS)是一种实时、准确、高效的综合交通运输管理系统,而车牌识别(License Plate Recognition LPR)技术是智能交通系统的关键技术之一.为了在嵌入式系统中实现车牌的实时检测,研究了基于TMS320DM642的车牌检测、定位与识别技术.本文提出一种基于纹理检测和Adaboost分类器相结合的车牌定位算法,针对该算法与DM642的结合进行了系统设计,并且基于EMCV与Opencv图像处理库进行了代码移植.该系统很好的避免了单独纹理检测的定位无方向性和单独Adaboost分类的定位不完整性的缺点,较好的提高了定位准确率,且定位后的车牌图像能够基于TCP协议发送到PC服务器,实现远程的车辆车牌的定位和监控.The Intelligent transportation system is a real-time, accurate and efficient integrated transportation management system, with the LPR（License Plate Recognition） as one of its key technologies. In order to realize the real-time detection of license plate in the embedded systems, the license plate detection, location and recognition technology based on TMS320DM642 are studied. This paper proposes a kind of license plate location algorithm by combining texture detection with the Adaboost classifier. The system design combining with DM642 is based on EMCV and Opencv image processing library to come true coding porting. The system avoids the non-directional under the condition of only texture detection and the non-integrity under the condition of only Adaboost Classification, it also improves the accuracy of the positioning. In addition, the license plate image which has been located can be sent to the PC server by the TCP protocol.Then the license plate location and monitoring of remote vehicles can be realized.

关 键 词：TMS320DM642 纹理检测 Adaboost分类 EMCV Opencv 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] U495[自动化与计算机技术—计算机科学与技术]