基于多颜色模型的车牌定位方法  

Multi-color models based car-plate location

作　　者：周泽华[1] 潘保昌[1] 郑胜林[1] 赵全友[1] 甘艳芬[1] 

机构地区：[1]广东工业大学信息学院数字图象技术研究所

出　　处：《微计算机信息》2007年第01S期283-285,共3页Control & Automation

基　　金：2006年国家自然科学基金项目(编号60572078)

摘　　要：提出了一种基于多颜色模型的车牌定位方法。首先把输入的RGB彩色图像转化到HSV和YIQ颜色空间,综合这两个颜色空间的信息进行颜色分割去除大量的背景干扰信息,得到了颜色为车牌照的一些区域。然后将颜色分割后的图像灰度化并分块,找出水平差分累加和最大的块确定车牌大致位置在原彩色图中实现车牌的粗定位。最后对粗定位图进行二次颜色分割得到车牌区域的信息利用投影法精确定位出车牌。实验结果表明该方法效果较好。This paper presents an car-plate location algorithm which based on multi-color models.Firstly,the input picture of RGB is transformed to HSV and YIQ color space,comprehensive information for the two color space division removed a lot of background interference color information,and some regions with the color of car-plate are obtained. Then the segmented picture is grayed and blocked. And in the way of seeking the maximum among difference accumulating values in the blocks, approximate region car-plate was found and approximate location is achieved in the original picture.At last,the approximate location picture is segmented for the second time and projection to get the accurate position of the plate.The experimental result shows that the proposed method is quite effective.

关 键 词：多颜色模型 颜色分割 分块 投影 车牌定位 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]