车辆图像的车牌字符割方法设计与实现毕业论文.docVIP

LANZHOU UNIVERSITY OF TECHNOLOGY 毕业设计 题 目 车辆图像的车牌字符分割方法设计与实现 学生姓名 ****** 学 号 ******** 专业班级 通信工程* 指导教师 * 学 院 计算机与通信学院 答辩日期 *年6月16日 车辆图像的车牌字符分割方法设计与实现 Design and Implementation Method of Vehicle License Plate Character Image Segmentation * 摘要 随着社会的发展，车辆将越来越多，为了更好地管理车辆，智能交通系统将是发展的方向，而车牌自动识别系统是智能交通系统中一个重要的环节，它主要包括三个关键技术：车牌定位、车牌字符分割以及车牌字符识别。本文针对车牌自动识别系统中关键技术之一的车牌字符分割部分做了深入的研究。 首先，本文对车牌字符分割前的预处理进行了较深入研究，尤其是车牌图像二值化后的去噪处理，包括开闭运算、去除边缘、去除圆点等。其次，在字符分割算法中，给出两种分割算法，一种是固定边界法，另一种是垂直投影法，并对车牌字符进行了分割。然后，在分析了这两种方法的优缺点后，将这两种方法相结合进而得到改进的投影法，实验表明，就算是在除噪后仍然有少量噪声的车牌进行分割时，该方法也可以得到很好的字符分割效果。 关键词：字符分割；二值化；投影法Abstract There will be more and more cars with the development of society. Intelligent Transport System will be the direction of development. System of License Plate Automatic Recognition is an important part of the Intelligent Transport System. It is contain of three key technologies: the location of license plate, the division of license plate character and the recognition of license plate character. This paper studied at character recognition technology which is one key technology of the System of License Plate Automatic Recognition. At first, I study the license plate character segmentation image preprocessing, especially the license plate image binarization of denoising processing, including open arithmetic, close arithmetic, wipe of the edge, and wipe of the round spot etc.The second, there are two ways to divide the plate. one way is the fixed boundary, and the other is a vertical projection. Then I separate the license plate characters.finally ,I compared two kinds of methods, they can be incorporated into the new projection method.The experimental results show that?even after the noise reduction still contains a small amount of noise of the license plate segmentation,the method also can get very good effect. Keywords: The division of license plate character；Binarization；The method of sh