JOURNAL OF TEXTILE RESEARCH ›› 2017, Vol. 38 ›› Issue (11): 162-167.doi: 10.13475/j.fzxb.20160901506

Previous Articles     Next Articles

Graphic contour extraction for printed fabric based on Ttxture smoothing

  

  • Received:2016-09-12 Revised:2017-08-15 Online:2017-11-15 Published:2017-11-15

Abstract:

In order to enrich printed products varieties, a method was put forward to effectively extract the contours of the printed fabric with high precision. Firstly, in order to smooth the image, by controlling the degree of smooth parameters and space scale, can eat off the fabric texture and structure of the image. Edge detection pattern with a Canny edge detection operator, Canny discriminant threshold segmentation of the default automatic threshold choice, can successfully segment the pattern on the fabric, after segmentation of image contour clear, continuous, edge and can segment the small structure in the images of the printed fabric. Experiments prove that in RGB color space of image smoothing and segmentation effect is superior to other color space, compared to other edge detection operator splitting pattern effect of printing fabrics, the results prove that Canny operator segmentation effect is best, segmentation results can be directly used in the production of printed fabric.

Key words: printed fabric, image smoothing, edge extraction, pattern segmentation, Canny operator

[1] . Determination on design cycle of printed fabrics based on distance matching function [J]. JOURNAL OF TEXTILE RESEARCH, 2015, 36(12): 98-0.
[2] PAN Ruru;GAO Weidong;QIAN Xinxin;ZHANG Xiaoting. Defect detection of printed fabrics using normalized cross correlation [J]. JOURNAL OF TEXTILE RESEARCH, 2010, 31(12): 134-138.
[3] ZHU Guiying;ZHANG Ruilin. Application of Canny operator on automatic detection of fabric waterproof performance [J]. JOURNAL OF TEXTILE RESEARCH, 2008, 29(3): 122-125.
[4] WANG Hua;SUN Li;ZHANG Shi. Analysis of ancient Egypt woven and dyed fabrics [J]. JOURNAL OF TEXTILE RESEARCH, 2007, 28(2): 123-125.
[5] On-line detection of the dyed and printed fabric defects by multi-features evidence learning and enhancement in spatiotemporal domain. On-line detection of the dyed and printed fabric defects by multi-features evidence learning and enhancement in spatiotemporal domain [J]. JOURNAL OF TEXTILE RESEARCH, 2006, 27(5): 1-5.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!