一种基于HSI和YUV颜色模型的车牌定位方法

摘要

车牌定位是整个车牌识别模块实现的前提.针对车辆图像车牌位置定位不准等问题,采用HSI和YUV颜色模型,根据每个像素S值的分布,动态决定选用H值还是Y值,并经过水平垂直投影技术对车牌区域进行定位分割.实验表明,该方法提高了定位的精确性,克服了光照的影响和单一颜色空间的不足.%It is a premise for implementing the entire car license recognition module to locate the car license plate. Considering such problems as inaccuracy with car image and car license plate location, the thesis uses HIS and YUV color model, depending on the distribution of the S value of each pixel to dynamically decide whether to choose the H value or the Y value, then by means of horizontal and vertical mapping technology the car license plate area is located and divided. Experiments demonstrate that the method improves the accuracy of location and overcomes the influences of light as well as the shortcomings of the single color space.