摘要
车辆牌照的准确定位是智能交通中车辆牌照识别技术的关键,提出一种基于小波变换的车牌质心定位方法,该方法可以很好地解决复杂背景与光照下的车牌定位。经过小波分析的车牌图像利用数学形态学进行车牌特征提取,对特征提取后的车牌图像采用连通区域质心的方法对车牌进行定位,最终得到车牌的准确区域。实验结果表明,该方法能够实现车牌的快速准确定位,是一种有效的车牌定位方法。
Vehicle license plate precise location takes an important part of vehicle license plate recognition in intelligent transpor tation system （ITS）. An approach of vehicle license plate centroid location based on wavelet transform is advanced to solve the vehicle license plate location problem of complex background and light influence. Firstly, vehicle license plate feature is extracted after wavelet analysis＇s plate image with mathematical morphology, then connect region＇s centroid method is adopted to locate the vehicle license plate with after characteristics extraction＇s vehicle license plate image, finally the vehicle license plate is gotten. The experiment indicates that the approach is a feasible vehicle license plate location method to locate the vehicle license plate precisely and quickly.
出处
《计算机工程与设计》
CSCD
北大核心
2012年第4期1491-1494,1503,共5页
Computer Engineering and Design
关键词
小波变换
小波去噪
数学形态学
边缘检测
车牌定位
wavelet transform
wavelet filtering
mathematical morphology
edge detection
vehicle license plate location
