基于多颜色模型的车牌定位方法

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。

Multi-color models based car-plate location

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 inter- ference 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.