摘要
为了提高智能交通系统的管理效率,给出一种基于芯片DM642的车牌识别系统。系统采用基于DM642的硬件平台,由车辆信息采集模块、DSP处理模块、监视器和PC系统4大模块组成。对采集图像进行灰度化、灰度增强和滤波降噪的预处理;利用基于字符竖向纹理特征的定位方法实现车牌定位;再经几何变换,完成字符分割;对分割图像进行字符分类并做归一化处理,通过模式匹配得出识别结果。实验选取3组不同特点的车牌图像,结果显示,所给系统的正确识别率可达93.5%~95.4%。
In order to improve the management efficiency of the intelligent traffic system,a vehicle plate recognition system based on chip DM642 is presented,Designed on DM642 hardware platform,the system is composed of 4 modules：vehicle information collection module,DSP processing module,monitor module and PC system.The collected images are preprocessed with grayscaling,gray enhancement and noise-reducing filter,and the vehicle plate is positioned by a locating method based on the vertical texture characteristics of the characters.After a geometric transformation,the characters are segmented,and then classified and normalized.Finally,by pattern matching,the vehicle plate can be recognized.On three groups of license plate image with different characteristics,a experiment is done,and the results show that,the recognition rate of the proposed system is about 93.5% ~ 95.4%.
出处
《西安邮电大学学报》
2016年第3期64-68,共5页
Journal of Xi’an University of Posts and Telecommunications
基金
国家自然科学基金资助项目(61179011)
福建省自然科学基金资助项目(2010J01327)
关键词
DM642
车牌定位
字符分割
字符识别
DM642
license plate location
character segmentation
character recognition
