基于MSER与SVM算法的车牌定位识别方法

针对实际车牌识别系统中车牌位置定位难、字符识别率低等问题,提出了一种基于MSER与SVM算法的车牌定位识别。该方法分为定位和识别两步,输入图像经过预处理,通过MSER与SVM算法直接提取出车牌的字符区域,然后将车牌字符图像裁剪送入识别阶段,识别阶段同样利用SVM算法对车牌字符进行识别。经验证,该车牌定位识别方法识别速度快、准确率高,能够适用于实际生活中较为复杂的交通环境。

A License Plate Location Recognition Method Based on MSER and SVM Algorithm

In order to solve the problems such as the difficulty in locating the position of the license plate and the low recognition rate of the characters in the actual license plate recognition system,an algorithm based on MSER and SVM was proposed.This method is divided into two steps:positioning and recognition.The input image is preprocessed,and the character area of license plate is directly extracted by MSER and SVM algorithm.Then the image clipping of license plate character is sent to the recognition stage.It is proved that the method is fast and accurate,and can be applied to the complicated traffic environment in real life.