基于字符纹理的车牌定位算法

License Plate Localization Algorithm on Character Texture Features

提出了一种基于小波变换和均值聚类的实时车牌定位算法,根据车牌字符的纹理特征,在小波变换域中结合均值聚类自适应动态阈值取得字符信息,然后通过连通区域分析,结合车牌的宏观特征对车牌进行定位,试验证明,该算法准确率高,抗干扰性强。

A real- time license plate location algorithm based on wavelet transform and K -means is put forward. By clustering analysis, the texture features of license plate characters can be acquired. Then by further operations, combining with multi -features of plate, the license plate region can be located. The experiment results show that the algorithm improves the ability of anti - interference and the accuracy of location.