车牌图像二值化方法研究分析论文.doc

题目 车牌图像二值化方法的研究 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，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 Study on license plate binarization method ABSTRACT With the development of our nation economy, there are more and more constructions in the expressway, city 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 important 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 adaptive 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 pixel 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. 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 propos