纺织学报 ›› 2012, Vol. 33 ›› Issue (8): 50-54.

• 纺织工程 • 上一篇    下一篇

基于AR模型的机织物线状疵点研究

 朱俊岭1, 汪军1,2, 张孝南3, 李立轻1, 陈霞1, 庞明军3   

    1. 东华大学纺织学院
    2. 东华大学纺织面料技术教育部重点实验室
    3. 中国纺织科学研究院江南分院
  • 收稿日期:2011-09-28 修回日期:2012-02-29 出版日期:2012-08-15 发布日期:2012-08-08
  • 通讯作者: 汪军 E-mail:junwang@dhu.edu.cn
  • 基金资助:

    纺织面料技术教育部重点实验室培育项目;绍兴科技攻关计划项目

Woven fabric linear defect research based on AR model

 ZHU  Jun-Ling1, WANG  Jun1,2, ZHANG  Xiao-Nan3, LI  Li-Qing1, CHEN  Xia1, PANG  Ming-Jun3   

    1. College of Textiles, Donghua University
    2. Key Laboratory of Textile Science & Technology, Ministry of Education, Donghua University
    3. Jiangnan Branch, China Textile Academy
  • Received:2011-09-28 Revised:2012-02-29 Online:2012-08-15 Published:2012-08-08
  • Contact: WANG Jun E-mail:junwang@dhu.edu.cn

摘要: 针对机织物线状疵点检测效果不佳的问题,本文采用AR模型的谱估计方法对此问题进行研究。首先将获取织物的图像按照一定的子窗口大小分割,并将子窗口内图像的灰度值按照纵向、横向两个方向分别使用方差的方式投影,得到方差序列;然后选择合适的AR模型阶数,依据Burg算法估计得到谱数据;最后通过求得带有疵点图像的谱估计与正常纹理图像得到的谱估计之间的相关系数检测出疵点及其位置。论文采用试验验证检验了该方法对机织物线状疵点检测的有效性。

关键词: AR模型 , Burg算法 , 机织物 , 疵点检测 , 线状疵点

Abstract: As detect the effect of woven fabric linear defects is poor, in this paper, the spectral estimation method of AR model was used to research that problem. Firstly, split fabric image obtained into certain size according to child windows. And also from both the vertical and horizontal directions, turn gray value of the child windows into variance sequence by the way of using the variance projection. Secondly, choose the appropriate AR model order and use Burg algorithm to estimate spectral data. Lastly, according to the correlation coefficient with the spectral estimation between a defect image and a normal one, defect and its location will be detected. The effectiveness of the method of the linear defects detection on woven fabric is proved by experiment.

Key words: AR model , Burg algorithm , woven fabric , defect detection , linear defect

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!