基于Gabor滤波车牌区域定位算法研究及实现

Research and Implementation of Licence Plate Area Detecting Algorithm Based on Gabor Filter

车牌区域检测是车牌识别系统第一步。论文描述一种车牌区域检测的系统框架:首先是利用Gabor滤波对车牌图像进行滤波得到包含车牌区域的二值图,对二值图作数学形态学变换以增强(扩大)可能的连通车牌区域,然后用扫描策略标记出所有的连通区域,最后计算所有这些连通区域的颜色方差,判定出车牌区域。实验结果对上述步骤进行不同的组合会有不同的优缺点。经过2000张图像的实验,论文的这种优化组合可以很好地检测出车牌区域,车牌区域识别的正确率达94%,是一种可行的车牌区域检测算法。

Plate region detection is the first step in license plate recognition system.This paper describes a framework for license plate detction system：first is the use of Gabor filters to get the binary image containing the license plate area.For binary image as morphological transformations to enhance the（expanded）may connectivity plate area.Finally,calculat all these connected regions color variance,determines the license plate.The results have different advantages and disadvantages for different combinations of the above steps.After 2000 image of experiments,the optimal combination of this article may well detect license plate,license plate recognition accuracy rate of 94%,is a a viable plate region detection algorithm.