摘要
针对不同条件下车牌检测速度慢和检测精度低的问题,提出了一种改进自适应形态闭和开操作的车牌检测算法。该算法首先采用局部直方图对车牌图像均衡化处理,使用自适应形态闭操作对所有灰度化区域进行平滑处理,之后引入局部自适应阈值处理,能得到平滑图像和被分离的车牌,最后采用形态学开操作,将外部区域和车牌数连接部分分离。实验结果表明,所提方法的检测精度高于其他算法,同时,平均检测时间少于其他算法,适合不同条件下实时车牌检测。
According to the problem of slow license plate detection and low detection accuracy under different conditions,this paper proposed an improved adaptive morphological closed and open operation to improve the license plate detection algorithm.The algorithm first used the local histogram to equalize the license plate image,and used the adaptive morphological closed operation to smooth all grayscale regions,and then introduced the local adaptive threshold processing to get the smooth image and the separated license plate.Finally,it adopted the morphological open operation to separate the connection parts between the external region and the number of the license plates.The experimental results show that the detection accuracy of the proposed method is higher than other algorithms.At the same time,the average detection time is less than other algorithms.The proposed method is suitable for real-time license plate detection under different conditions.
作者
杨秀杰
李芳
Yang Xiujie;Li Fang(College of Computer,Chongqing College of Electronic Engineering,Chongqing 401331,China;School of Computer Science,Chongqing University,Chongqing 400044,China)
出处
《计算机应用研究》
CSCD
北大核心
2019年第11期3508-3510,共3页
Application Research of Computers
基金
重庆市教委课题(KJ1729408)
重庆市教委教改重点项目(162071)
关键词
车牌检测
局部直方图
自适应形态闭操作
局部自适应阈值
形态学开操作
license plate detection(LPD)
local histogram
adaptive morphological closed operation
local adaptive thres-hold
morphological opening operation
