基于OpenCV的机动车牌照识别系统

发布时间：2018-11-15 14:25

【摘要】：车牌识别系统作为智能化交通管理系统的重要组成部分,已经越来越受到人们的重视,同时它也是计算机视觉、模式识别以及图像处理技术在智能交通管理领域中的研究热点。本文通过查阅大量文献,结合前人的工作基础与研究成果,搭建了一套车牌识别系统,该系统包括:首先使用车牌定位算法对车牌图片进行图像分割,提取出所有可能包含车牌的区域,接着使用支持向量机对所有分割出的图像区域进行分类,通过预测函数返回有效的车牌区域;然后将定位得到的车牌图像进行字符分割和字符识别,包括使用OCR算法检测车牌上的每个字符并将字符提取出来,然后使用训练好的人工神经网络对字符进行识别。本系统采用VS2010开发工具,结合C语言和OpenCV库函数对车牌识别系统进行编程设计,完成车牌图像的定位与识别,实验结果比较令人满意。
[Abstract]:As an important part of intelligent traffic management system, license plate recognition system has been paid more and more attention by people. It is also the research hotspot of computer vision, pattern recognition and image processing technology in the field of intelligent traffic management. In this paper, a license plate recognition system is built by consulting a lot of literature and combining the previous work and research results. The system includes: first, using license plate location algorithm to segment the license plate image. All the regions that may contain license plates are extracted, and then all the image regions are classified by support vector machine, and the effective license plate regions are returned by the prediction function. Then, the license plate image is segmented and recognized, including the OCR algorithm is used to detect each character on the license plate and the characters are extracted, and then the character recognition is carried out by using the trained artificial neural network. The system uses VS2010 development tools, combined with C language and OpenCV library functions to program the license plate recognition system, and completes the license plate image location and recognition. The experimental results are satisfactory.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U495;TP391.41

【相似文献】