基于数学形态学与数字图像处理的车牌识别系统-计算机技术专业论文.docx

南昌航 南昌航空大学硕士学位论文 目录 I I 摘 要 智能交通系统是现代交通管理发展热点和难点，而车牌识别是智能交通系统中 的关键技术之一。当今许多车牌识别系统已达到实用水平要求，但是一些车牌识 别系统在环境不理想的情况下，正确识别率有待提高。本文对图像预处理、车牌 定位、字符分割以及字符识别等车牌识别系统关键模块进行了研究开发，具体做 了以下工作： 首先，针对车牌图像在露天拍摄时，受到内外界干扰而产生的各类型噪声， 本文采用了一种新的去噪方法－自适应中值滤波法。并提出了一种新的边缘检测 法（Log-Prewitt、Log-Sobel），以此锐化图像，使得图像更加平滑。根据全局阈值 法，选取合适的阈值，进而对灰度图像进行二值化处理。 其次，给出了基于数学形态学与色彩特征的车牌定位分割方法。该方法能够准 确定位且分割出整个车牌区域。 再次，采用基于垂直投影与数学形态学中连通域法，提出特殊的车牌字符分割 方法。该方法能准确无误对车牌区域内每一字符进行分割。 最后，针对车牌的汉字、字母、数字等特征，设计了车牌识别的神经网络识别 器。该神经网络识别器采用标准 BP 网络算法与附加动量法和学习率相结合的综合 算法对车牌图像进行识别。实验结果表明，此法较为理想，而且识别正确率能够 达到 95%以上。 关键字：数字图像处理；BP 神经网络；边缘检测； 字符识别； 数学形态学； 字符分割；车牌定位；字符分割 II II Abstract Intelligent Transportation System, in which the License Plate Recognition is one of the key techniques, is the hot and difficult issue in the developtment of the modern traffic management. At present, many License Plate Recognition systems have reached the pratical level, while correct recognition rate of some License Plate Recognition system in an unideal circumstance needs to be improved. In this paper, some key modules of the License Plate Recognition system are studied, including image preprocessing, license plate location, character segmentation and character recognition and so on. The specific work is as follows: First of all, in allusion to the various types of noise caused by inside and outside interference when a license plate image is filmed in the open, this paper adoptes a new denoising method, adaptive median filtering method. It also puts forward a new edge method (Log - Prewitt, Log - Sobel), which is used to sharpen image to make the image more smooth. According to the global threshold method, we select the appropriate threshold, and then do binarization processing for gray image. Secondly, this paper combins with the mathematical morphology and color characters to locat and segment a license plate, which can locate and segment the license plate region accurately. Thirdly, this paper adopts a special license plate character segmentation me