摘要
车牌识别系统在高速路收费口与住宅小区车库管理中得到越来越多的应用,车牌定位是整个识别系统实现的前提。基于灰度图像的定位方法和基于彩色车牌图像的定位方法,实现效果均不大理想。在充分利用车牌先验信息的基础上,提出基于HSV色彩空间与数学形态学的车牌定位方法。先利用色彩信息对可能包含车牌目标的区域进行过滤,再利用数学形态学技术生成连通区域,判断并生成正确的车牌区域,最后,使用radon变换进行倾斜校正。
The license plate recognition system has been widely used in highway toll gate and garage management of residential quarters, and license plate location is the precondition of the realization of the whole recognition system. The effects of localization methods based on gray-scale image and color-plate image are not ideal. By making full use of the license plate of a priori information, a method of license plate location based on the HSV color space and mathematical morphology is proposed. First, get the possible objects by using color information as a filter, then generate connected regions by using mathematical morphology techniques and generate the final correct license plate region, and last use radon transform to tilt correction.
出处
《图学学报》
CSCD
北大核心
2013年第4期159-162,共4页
Journal of Graphics
