Abstract

In order to accurately detect defects in patterned fabric images, a novel detection algorithm based on Gabor-HOG (GHOG) and low-rank decomposition is proposed in this paper. Defect-free pattern fabric images have the specified direction, while defects damage their regularity of direction. Therefore, a direction-aware descriptor is designed, denoted as GHOG, a combination of Gabor and HOG, which is extremely valuable for localizing the defect region. Upon devising a powerful directional descriptor, an efficient low-rank decomposition model is constructed to divide the matrix generated by the directional feature extracted from image blocks into a low-rank matrix (background information) and a sparse matrix (defect information). A nonconvex log det(.) as a smooth surrogate function for the rank instead of the nuclear norm is also exploited to improve the efficiency of the low-rank model. Moreover, the computational efficiency is further improved by utilizing the alternative direction method of multipliers (ADMM). Thereafter, the saliency map generated by the sparse matrix is segmented via the optimal threshold algorithm to locate the defect regions. Experimental results show that the proposed method can effectively detect patterned fabric defects and outperform the state-of-the-art methods.

Highlights

  • Fabric defect detection is the key step in quality control of textile products

  • THE PROPOSED ALGORITHM we describe the proposed fabric defect detection method, which includes feature extraction of GHOG descriptor, construction of low-rank decomposition model, optimization of the model and acquisition and segmentation of the saliency map

  • EXPERIMENTAL RESULTS we evaluate our proposed approach and compare with state-of-the-art methods in the datasets of 256by-256 patterned fabric images, which are from Industrial Automation Research Laboratory, Dept. of Electrical and Electronic Engineering, The University of Hong Kong, these images have three patterns: star, box- and dot-patterned fabric datasets for performance validation

Read more

Summary

INTRODUCTION

Fabric defect detection is the key step in quality control of textile products. Currently, it is mainly conducted visually by skilled workers. Directly using low-rank decomposition in the raw pixel space of images to detect defects in the complex patterned fabrics is not practical due to the low accuracy This promotes us to introduce new powerful descriptors to efficiently characterize the fabric texture, which make the defectfree or background regions lay in a low rank subspace, whilst the defective regions deviate from the subspace. HOG is generated by counting the gradient magnitude when the gradient orientation is consistent with the orientation of Gabor filtered maps; 2) Following the idea of [8], a spatial pooling strategy is utilized to enable small displacement of second order gradients in the neighborhood of a certain point; 3) A non-convex log det is exploited as a smooth surrogate function for the rank instead of the nuclear norm to improve the efficiency of the low-rank model; 4) Experimental results are presented to further demonstrate the efficiency of our proposed method.

RELATED WORK
OPTIMIZATION OF THE MODEL
SALIENCY MAP ACQUISITION AND SEGMENTATION
EXPERIMENTAL RESULTS
QUALITATIVE RESULTS
CONCLUSION

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.