文档介绍：汽车牌照自动识别算法研究
摘要:车牌识别系统是智能交通系统的核心组成部分,应用于各种交通管理中。本文对系统中汽车号码图像预处理、车牌定位、字符分割、字符识别进行了全面的论述及相关算法的研究,主要工作包括:
(1)车牌定位算法的研究及其校正;
(2)结合彩***像灰度化、车辆图像边缘检测以及图像的二值化对汽车牌照进行预处理,并做了相关实验进行论述;
(3)介绍了基于改进的扫描线算法确定字符串上下边界,利用全局分割和局部调整的技术实现了单个字符的分割,并对断裂部分进行了粘连处理;
(4)介绍了一种四灰度加权相关函数模板匹配集成方法来实现字符识别,使车牌字符识别的精度和实时性得到了改善。
实验结果表明,本课题研究的算法能比较准确地对车牌区域进行定位、分割车牌字符并进行字符识别,且性能良好。
关键词:车牌定位;字符分割;字符识别;边缘检测
Research of the Algorithm on Vehicle License Plate Automatic Recognition
Abstract: License plate recognition system is the core of intelligent transportation ponents used in a variety of traffic management. In this paper, image preprocessing system, license plate localization, character segmentation, character recognition to conduct prehensive research paper and related algorithms, the main activities include:
(1)This paper is about the research of license plate localization algorithms ,and correction;
(2)It is used gray scale and colored images, vehicle image edge detection and binary image to the pretreatment of vehicle license, and discuss the relevant experiments license plate location algorithm and its tilt correction;
(3)It is introduced , based on an improved scan line algorithm ,and to determine the string up and down the border, using the local adjustment of the global fragmentation and individual technology character segmentation and its adhesion to do break processing;
(4)It is introduced a correlation function of four gray-weighted template matching integrated approach to character recognition, and the license plate character recognition accuracy and timeliness has improved.
The results are shown that the algorithm of this research can be more accurate positioning of the license plate, license plate character segmentation and the character recognition, and good performance.
Key words: License plate location; Character segmentation; Character recognition; Edge detection
目录
1绪论 1
1
1
1
2
2车牌图像定位 3