Patterned fabric defect detection via convolutional matching pursuit dual-dictionary

Junfeng Jing; Xiaoting Fan; Pengfei Li

doi:10.1117/1.OE.55.5.053109

26 May 2016 Patterned fabric defect detection via convolutional matching pursuit dual-dictionary

Junfeng Jing, Xiaoting Fan, Pengfei Li

Author Affiliations +

Funded by: National Natural Science Foundation of China, Natural Science Foundation of China, Xi’an Polytechnic University, Xi’an Polytechnic University Scientific Research Foundation

Optical Engineering, Vol. 55, Issue 5, 053109 (May 2016). https://doi.org/10.1117/1.OE.55.5.053109

Abstract

Automatic patterned fabric defect detection is a promising technique for textile manufacturing due to its low cost and high efficiency. The applicability of most existing algorithms, however, is limited by their intensive computation. To overcome or alleviate the problem, this paper presents a convolutional matching pursuit (CMP) dual-dictionary algorithm for patterned fabric defect detection. A preprocessing with mean sampling is performed to eliminate the influence of background texture of fabric defects. Subsequently, a set of defect-free image blocks are selected as a sample set by sliding window. Dual-dictionary and sparse coefficiencies of the defect-free sample set are obtained via CMP and the K-singular value decomposition (K-SVD) based on a Gabor filter. Then we employ the defect-free and defective fabric image’s projections onto the dual-dictionary as features for defect detection. Finally, the test results are determined by comparing the distance between the features to be measured. Experimental results reveal that the proposed algorithm is effective for patterned fabric defect detection and an acceptable average detection rate reaches by 94.2%.

Citation Download Citation

Junfeng Jing, Xiaoting Fan, and Pengfei Li "Patterned fabric defect detection via convolutional matching pursuit dual-dictionary," Optical Engineering 55(5), 053109 (26 May 2016). https://doi.org/10.1117/1.OE.55.5.053109

Published: 26 May 2016

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

;

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE