摘要
车牌预处理是汽车牌照自动识别系统中的一个重要的环节,预处理的好坏直接影响到后面的识别结果.在开发一套实用的自动车牌识别项目的过程中,分析研究了车牌预处理的各个环节,对其中关键算法———倾斜校正算法、二值化算法、字符分割算法做了详细的研究.在大量试验的基础上,提出了一种基于区域滤波的预处理算法和一种改进的车牌倾斜校正算法,并用于工程实际中,取得了良好的效果.
An improved vehicle license plate tilt correction method based on Sobel edge detector and a new plate pre-processing algorithm based on area filter is present. Experimental results show that both proposed algorithms are robust in dealing with uniform illuminated, contrast poor, degraded, dirt vehicle plates. With the proposed plate pre-processing method, 98% of 1000 sample vehicle license-plates images are correctly segmented; and recognition rate of the system can be improved notably.
出处
《武汉大学学报（工学版）》
CAS
CSCD
北大核心
2006年第3期131-134,共4页
Engineering Journal of Wuhan University
关键词
车牌
预处理
图像二值化
区域过滤
vehicle license plate
pre-processing
image binarization
area filter
