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

基于机器视觉印品缺陷检测的滤波算法
引用本文:谢勇,王耀南,彭涛,龙永红.基于机器视觉印品缺陷检测的滤波算法[J].湖南大学学报(自然科学版),2005,32(4):53-57.
作者姓名:谢勇  王耀南  彭涛  龙永红
作者单位:1. 湖南大学,电气与信息工程学院,湖南,长沙,410082;株洲工学院,包装与印刷学院,湖南,株洲,412008
2. 湖南大学,电气与信息工程学院,湖南,长沙,410082
3. 株洲工学院,包装与印刷学院,湖南,株洲,412008
基金项目:国家自然科学基金资助项目(60375001)
摘    要:彩色印刷品是经过多色印刷后形成的,各色之间不可避免地存在一定程度的套印位置偏差等工艺缺陷.针对这个问题,提出了一种从印刷品差分图像中滤掉边缘噪声的算法.基于印刷工艺因素的影响及标准图像的自身结构特征,设计了滤波算法由图像乘、全局亮度门限和腐蚀/膨胀等部分组成.实验证明该算法能有效地降低边缘噪声。

关 键 词:计算机视觉  滤波算法  印品缺陷检测  工艺缺陷  边缘噪声
文章编号:1000-2472(2005)04-0053-05
收稿时间:11 10 2004 12:00AM
修稿时间:2004-11-10

A Filtering Algorithm Based on Machine Vision Inspecting for Printing Product Defects
Xie Yong;Wang YaoNa;Peng Tao;Long YongGong.A Filtering Algorithm Based on Machine Vision Inspecting for Printing Product Defects[J].Journal of Hunan University(Naturnal Science),2005,32(4):53-57.
Authors:Xie Yong;Wang YaoNa;Peng Tao;Long YongGong
Abstract:A color printing product is produced by means of multi-color printing. So it is unavoidable to bring about some technical defects between each color printing, such as overprint location deviation. For this problem, an algorithm has been developed to filter edge noise from the difference images of printing products.Based on the influence of technical factors and the structural features of the reference image, the filtering algorithm, consisting of an image multiplication, a global intensity threshold and an erosion/dilation, has been proved to be effective for reducing edge noise of sampling images.
Keywords:computer vision  filter algorithm  printing product defect
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《湖南大学学报(自然科学版)》浏览原始摘要信息
点击此处可从《湖南大学学报(自然科学版)》下载免费的PDF全文
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