作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2013, Vol. 39 ›› Issue (2): 216-219. doi: 10.3969/j.issn.1000-3428.2013.02.044

• 人工智能及识别技术 • 上一篇    下一篇

基于主成分分析的表面缺陷自动检测算法

郭永彩,邓细凤,高 潮   

  1. (重庆大学光电工程学院光电技术及系统教育部重点实验室,重庆 400030)
  • 收稿日期:2012-03-07 修回日期:2012-04-04 出版日期:2013-02-15 发布日期:2013-02-13
  • 作者简介:郭永彩(1963-),女,教授、博士生导师,主研方向:机器视觉,数字信号处理,图像识别;邓细凤、硕士;高 潮,教授、博士生导师

Surface Defect Automatic Detection Algorithm Based on Principal Component Analysis

GUO Yong-cai, DENG Xi-feng, GAO Chao   

  1. (Key Laboratory of Optoelectronic Technology and System, Ministry of Education, College of Optoelectronics Engineering, Chongqing University, Chongqing 400030, China)
  • Received:2012-03-07 Revised:2012-04-04 Online:2013-02-15 Published:2013-02-13

摘要: 为检测产品表面的缺陷,提出一种基于主成分分析的自动检测算法。利用主成分分析法进行图像重构,以增强缺陷特征,对比原图像与重构图像,得到缺陷信息,通过统计过程控制二值化方法检测出缺陷。实验结果表明,该算法检测效果较好,运算速度较快,对于80张不同的表面图片,平均缺陷检测率达80%。

关键词: 缺陷检测, 主成分分析, 图像重构, 统计过程控制, 二值化, 机器视觉

Abstract: In order to detect the surface defects, this paper proposes an automatic detection algorithm based on Principal Component Analysis(PCA). The process uses PCA to make the image reconstruction to enhance the defects, and makes comparison with the original image and reconstruction image to get the defect information. The defects are detected after the statistical process control binarization. Experimental results show that the detection effect of this algorithm is good, operation speed is fast, for 80 different surface images, average defects detection rate is 80%.

Key words: defect detection, Principal Component Analysis(PCA), image reconstruction, statistical process control, binarization, machine vision

中图分类号: