Source:Textile Review

Alltextile industries aim to produce competitive fabrics. The competitionenhancement depends mainly on productivity and quality of the fabrics producedby each industry. In the textile sector, there have been an enlarge amount of lossesdue to faulty fabrics. In the least developed countries, most defects arisingin the production process of a textile material are still detected by humaninspection. The work of inspectors is very tedious and time consuming. Theyhave to detect small details that can be located in a wide area that is movingthrough their visual field. The identification rate is about 70%. In addition,the effectiveness of visual inspection decreases quickly with fatigue. Digitalimage processing techniques have been increasingly applied to textured samplesanalysis over the last ten years.

  • Wastage reduction through accurate and early stage detection of defects in fabrics is also an important aspect of quality improvement.
  • Summarize the comparison between human visual inspection and automated inspection.
  • Price of textile fabric is reduced by 45% to 65% due to defects.

 

Machine vision automated inspection system for textile defects has been in the research industry for a long time, Recognition of patterns independent of position, size, brightness and orientation in the visual field has been the goal of much recent work. However, there is stiII a lack of work in machine vision automated system for recognizing textile defects using a neural network pattern recognizer was developed. Fully connected three multilayer percetron network was used to identify different sizable objects. The input of this network is seven standardized invariant moment and the weights are trained using back propagation.

 

ReadFull Article

 

About the Author:

 

The author is Assistant Professor, Department of Fashion Technology, NIFT Bangalore

 

Originally published in Textile Review, September-2010