基于神经网络和颜色特征的车牌字符分割方法  

A Kind of Segmentation Method of Vehicle License Plate Characters Based on Neural Networks and Color Feature

作　　者：王润民[1] 钱盛友[1] 宋平[1] 许慧燕[1] 

机构地区：[1]湖南师范大学物理与信息科学学院,长沙410081

出　　处：《计算机工程与应用》2006年第26期57-59,118,共4页Computer Engineering and Applications

基　　金：湖南省自然科学基金资助项目(编号:02JJY2059)

摘　　要：论文提出了一种基于神经网络和颜色特征的车牌字符分割方法。在颜色空间中利用树型判决结构,首先由亮度信息识别出车牌区域中的白色、黑色像素,然后利用网络对车牌区域中的蓝、红、黄色以及其他颜色进行识别。根据车牌的颜色特征,在判断出车牌的类型后,对车牌区域进行二值化处理。去除车牌边框和柳钉后,综合利用投影法和字符的连通性来分割车牌字符。实验结果表明,该方法是有效的。A kind of segmentation method of vehicle license plate characters based on neural networks and color feature is presented in this paper.ln the HSI color space,a judge structure of model tree is used.Firstly the white pixels and black pixels of the vehicle license plate are recognized by intensity information,then the blue pixels,red pixels, yellow pixels and the other color pixels of the vehicle license plate are recognized by a BP neural networks.The category of the vehicle license plate can be judged by its color feature,and the vehicle license plate can be binarized.After the vehicle license plate＇s frame and the nails have been moved from the figure,the vehicle license plate characters can be accurately segmented by using the projection method and the characters＇connexity.The experimental results demonstrate the efficiency of the proposed approach.

关 键 词：字符分割 神经网络 颜色特征 连通域 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]