《车牌图像值化方法的研究论文》由会员分享，可在线阅读，更多相关《车牌图像值化方法的研究论文（61页珍藏版）》请在金锄头文库上搜索。
1、I题目题目 车牌图像二值化方法的研究车牌图像二值化方法的研究I摘要随着我国经济的发展，高速公路、城市道路、安全管理停车场越来越多，对交通控制要求日益提高。智能交通系统已经成为当前交通管理发展的主要方向，汽车牌照自动识别技术是智能交通的一个重要环节。二值化是车牌识别系统图像预处理的关键一步，汽车牌照在复杂的光照环境下，经常会出现车牌光照不均匀的现象，给图像二值化带来困难。为此，本文研究一种解决方法，使用整体阈值和局部阈值相结合的自适应阈值方法。该算法的主要步骤如下，首先使用滤波器去掉车牌图像不均匀光照的影响，然后用差分算子检测图像的灰度变化较为剧烈的地方，在这些像素上进行二值化阈值的自适应选择，
2、其他像素则采取常规的方法决定二值化阈值，最后使用整体阈值和局部阈值相结合的自适应阈值方法进行二值化处理。实验表明，基于边界特征检测的算子的算法能很好的保留原图的边界特征，能有效克服不均匀光照的影响，二值化效果良好，车牌识别率得到显著提高。关键词：二值化 整体阈值 局部阈值 自适应 边界特征IIStudy on license plate binarization methodABSTRACTWith the development of our nation economy, there are more and more constructions in the expressway, ci
3、ty way and parking area, etc. The requisition on the traffic control, safety management improves day and day. The intelligent transportation system(ITS) has already become the main direction of present traffic administration , and License Plate Recongnition System as core of ITS, plays a very import
4、ant role. Binary is a key step of image preprocessing in License Plate Recognition system. It is a difficult task to binarize image under nonuniform illumination；and this problem is always met in license plate recognition system which must work under complex illumination.Thus, a new method of adapti
5、ve threshold selection is proposed which combines the local threshold and global threshold selection. The method as follow:first,we use filters to remove the nonuniform illumination of license plate image, then use difference operator detect the image intensity changes where are more severe, these p
6、ixel values on the threshold of the adaptive selection,Other pixels are taken to the conventional method to determine the threshold binarization. Finally, a new method of adaptive threshold selection is proposed which combines the local threshold and global threshold selection used for binarization.
7、 Our experiments show that the new algorithm keeps the original edge features well, this algorithm is robust in dealing with nonuniform illumination degraded license plates. With the proposed method, the recongnition rate of the system can be improve notedly.Keywords: Binarizationg; lobal threshold;
8、 local threshold; adaptive; edge detectionIII目录第一章 绪论.11.1 课题研究的背景及意义.1 1.2 二值化算法的发展状况.2 1.3 本文的工作和内容安排.3第二章 数字图像预处理.42.1 灰度处理.4 2.2 车牌定位.5 2.3 数学形态学运算.8 2.4 车牌校正.9 2.5 倾斜校正.13 2.6 本章小结.14第三章 图像二值化.153.1 引言.15 3.2 二值化的方法.15 3.3 全局阈值法.16 3.4 局部阈值法.18 3.5 基于边界特征的二值化算法.19 3.6 去除边框.24 3.7 实验结果.25 3.8 本章小结.27第四章 总结与展望.284.1 总结评价.28 4.2 展望.29参考文献.31附录.33外文原文.33 外文翻译.
