Journal of South China University of Technology(Natural Science Edition) ›› 2012, Vol. 40 ›› Issue (3): 81-87.

• Electronics, Communication & Automation Technology • Previous Articles     Next Articles

Design of Machine Vision System to Detect Surface Stains of Non-Woven Fabric Products

Zhang Qian  Pei Hai-long  Shi Bu-hai  Zhao Yun-ji   

  1. Key Laboratory of Autonomous Systems and Networked Control of the Ministry of Education//School of Automation Science and Engineering,South China University of Technology,Guangzhou 510640,Guangdong,China
  • Received:2011-09-02 Revised:2011-12-08 Online:2012-03-25 Published:2012-02-01
  • Contact: 史步海(1963-) ,男,博士,副教授,主要从事大时滞系统建模与优化控制、图像处理算法研究. E-mail: bhshi@scut.edu.cn E-mail:zhangqian020@126.com
  • About author:张谦(1979-) ,男,博士生,主要从事嵌入式系统、机器视觉研究.
  • Supported by:

    国家自然科学基金重点项目( 60736024, 60574004, 61174053) ; 教育部科技创新工程重大项目( 708069)

Abstract:

In order to meet the requirements of high reliability,easy use,low cost,high accuracy and speed for the machine vision system to detect surface stains of non-woven fabric products,an embedded machine vision system,which integrates a human-computer interface to avoid the inconvenience induced by PC dependence,is designed. Then,the hardware structure of the system is introduced,and a dynamic threshold-based method is proposed to detect the stains in real time with high precision,low misjudgment rate and strong reliability. Practical application results indicate that the proposed system is of high performance,concretely,in a non-woven fabric production line with a linear velocity of 300 m/min,it can successfully detect the stain with 1 mm2 in square.

Key words: machine vision, image processing, non-woven fabric product, stain detection, CCD sensor

CLC Number: