Computer and Modernization ›› 2014, Vol. 0 ›› Issue (4): 16-19.

Previous Articles     Next Articles

Fabric Defect Detection Based on Periodicity and Local Directivity of Texture Edge

  

  1. College of Mechatronics Engineering, Xi’an Polytechnic University, Xi’an 710048, China
  • Received:2014-01-08 Online:2014-04-17 Published:2014-04-23

Abstract:  

According to the characteristic of the fabric texture image itself, a new method of fabric defect detection is presented, which based on texture periodicity and local directivity. Through eliminating the background texture information of similar texture with defect using its features of periodicity and directivity of normal texture edge, the defect information is highlighted, and then the defect information of no particular direction is detected quickly and efficiently. Through detection experiments of a large number of fabric image with defect, this method is proved to have good detection effect, when the fabric image has clear texture edge or good direction consistency.

Key words:

CLC Number: