基于数学形态学边缘检测的车牌字符分割算法  

Segmentation Algorithm of License Plate Characters Based on Mathematical Morphology

作　　者：熊哲源[1] 樊晓平[1] 黎燕[1] 

机构地区：[1]中南大学信息科学与工程学院,湖南长沙410075

出　　处：《计算机系统应用》2010年第9期155-158,共4页Computer Systems & Applications

摘　　要：汽车牌照字符分割是车牌识别过程中的关键步骤,直接影响到字符识别的效果。传统的方法对车牌图像质量要求较高,且抗干扰能力较差。提出一种基于Renyi熵和数学形态学边缘检测的车牌字符投影分割算法,首先用二维Renyi熵最大阈值法对车牌图像做二值化处理,然后用形态学腐蚀运算进行边缘检测,再去除车牌边框,最后通过投影分割提取车牌字符。仿真实验表明,基于Renyi熵最大阈值法和数学形态学边缘检测车牌图像预处理使得车牌字符边缘清晰,降低了噪声的干扰,有利于进行字符投影分割。该算法分割速度快,鲁棒性好,可获得比传统方法更好的分割效果。Vehicle-license-plate character segmentation is one of the most important steps in license plate recognition.Traditional methods always need high quality original images and thus, will inevitably affect the ultimate recognition rate.In this paper, a method for segmentation of license plate characters, based on Renyi＇s entropy and mathematical morphology edge detection, is introduced.The algorithm binaries the license plate by two-dimensional Renyi＇s entropy first and then detects the edges by morphology erosion.After getting rid of the outer bounding frame, the license plate character is segmented by projection.Computer simulation results show that the character is much clearer after preprocessing based on Renyi＇s entropy and mathematical morphology edge detection.This is favorable for the character segmentation based on projection.The algorithm is not only simple and fast, but also robust, and performs better than traditional methods.

关 键 词：字符分割 二维RENYI熵 数学形态学 边缘检测 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]