首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于二值图像的车牌精确定位方法
引用本文:路小波,张光华.基于二值图像的车牌精确定位方法[J].东南大学学报(自然科学版),2005,35(6):972-974.
作者姓名:路小波  张光华
作者单位:东南大学智能运输系统研究中心,南京,210096
基金项目:江苏省自然科学基金,东南大学校科研和教改项目
摘    要:提出一种基于二值图像灰度变化特征进行车牌精确定位的方法.在车牌粗定位的基础上,对分割出的车牌图像进行二值化,对二值车牌图像进行逐行扫描,利用每一行像素的黑、白跳变规律确定车牌的上下边界;对二值车牌图像进行处理得到特征图像,通过对特征图像进行垂直投影确定车牌左右边界.测试结果表明,该方法精确定位的准确率达到99%.

关 键 词:车牌定位  二值化  纹理  图像识别痧
文章编号:1001-0505(2005)06-0972-03
收稿时间:04 22 2005 12:00AM
修稿时间:2005-04-22

Accurate localization of binarizing image based vehicle license plate
Lu Xiaobo,Zhang Guanghua.Accurate localization of binarizing image based vehicle license plate[J].Journal of Southeast University(Natural Science Edition),2005,35(6):972-974.
Authors:Lu Xiaobo  Zhang Guanghua
Institution:Research Center of Intelligent Transportation Systems, Southeast University, Nanjing 210096, China
Abstract:A method for vehicle license plate accurate localization based on binarized image is presented in this paper. Binarizing the segmented license plate image, the binarized image is scanned line by line on the basis of rough localization, the upper and lower borders are determined through the black-white variation of the pixels in every row. After processing the binarized license plate a characteristic image is obtained. The left and right borders are determined based on the vertical projection of the characteristic image. Testing results show that the presented method has an accurate localization rate of 99%.
Keywords:license plate localization  binarization  texture  image recognition
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号