摘要
针对车牌定位难、定位速度慢,提出一种基于车牌背景和字符颜色特征的车牌定位算法。算法将车牌定位分为粗定位和细定位两部分。根据车牌背景和字符固定颜色搭配,利用新的颜色模型对车牌进行粗定位,在车牌图像中找出车牌区域的大概位置。对车牌进行细定位,设计竖向边缘检测算法——改进的Canny边缘检测算法,对粗定位的车牌区域进行边缘检测。利用形态学算法处理边缘检测后的区域,减少车牌区域的粘粘和断裂。通过投影定位法和车牌的几何特征完成车牌的定位。实验结果表明,该算法设计简单,定位速度快,且通用性较好。
Aiming at the difficulty and slow speed of license plate location, we proposed a license plate location algorithm based on background and character color feature. The algorithm divided license plate location into two parts： coarse positioning and fine positioning. According to the fixed color combination of license plate background and characters, a new color model was used to locate the license plate roughly, and the approximate location of the license plate area was found in the license plate image. We adopted the improved Canny edge detection algorithm to design a vertical edge detection method to detect the edge of the roughly positioned license plate area. The morphological algorithm was used to process the area after edge detection to reduce the stickiness and breakage of the license plate area. Location of license plate was completed by projection and geometric characteristics. The experimental results show that the algorithm is simple in design, fast in positioning and good in generality.
作者
杨鼎鼎
陈世强
刘静漪
Yang Dingding;Chen Shiqiang;Liu Jingyi(College of Science,Hubei University for Nationalities,Enshi 445000,Hubei,China;School of Information and Engineering,Hubei University for Nationalities,Enshi 445000,Hubei,China)
出处
《计算机应用与软件》
北大核心
2018年第12期216-221,共6页
Computer Applications and Software
基金
国家科技支撑计划课题(2015BAK27B03)
关键词
颜色特征
边缘检测
投影法
车牌定位
Color feature
Edge detection
Projection
License plate location
