Answers are grounded in this paper and cite their sources.
This paper analysis the detection of defects in fabric based on Local Neighborhood Analysis. In the proposed algorithm full image is inspected by moving Local Neighborhood Window over it. For the homogeneity measure, Coefficient of variation is used. The value of coefficient of variation is maximum if there is defect in the fabric. Defected image is segmented by using thresholding. Morphological filtering on defect segmented image is used to enhance defective region. The computational efficiency increased by using integral image technique. It avoids complicated spectral decomposition. Image database include TILDA database and in-house database. Experiments are performed for defects likes slack end, loose weft, drop stitch, holes, broken end, missing plush loop etc.
No takes yet. Share an insight, caveat, or question.
Kure et al. (2017) studied this question.