基于灰度跳变与形态学的车牌定位法

车牌定位是车牌识别系统的关键步骤,定位的准确与否直接影响车牌识别的结果.提出了一种基于灰度跳变与形态学的定位方法.对于车牌候选区的二值图像,根据车牌的水平分布特点,从长宽之比、面积之比等多个方面进行综合分析,从而可以更加精确地对车牌进行识别.实验验证表明,基于该算法的车牌识别系统能够达到较高的车牌识别率.

License Plate Location Method Based on Gray-scale Transition and Morphology

The vehicle license plate location is key part of automatic vehicle plate recognition system. The accuracy of location will influence the plate recognition. This paper presents a positioning method combining gray transition with morphology. For binary image of the license plate candidate region, according to the characteristics of the horizontal distribution of license plate, this algorithm analyzes comprehensively from the length and width ratio, area ratio and other aspects, so that the license plate can be more accurately rec- ognized. The experimental results prove that the license plate recognition system based on this algorithm can achieve the higher license plate recognition rate.