Evaluation of fabric pilling based on Hough transform and Gabor filter

Abstract

Fabric pilling image acquisition process will inevitably cause noise. In order to effectively remove the noise from the image of pilled fabric obtained by circular locus method, this paper tried to use Otsu segmentation, edge detection algorithm and Hough transform to extract the fabric area, and aimed at the problems existing in the current fabric noise removal methods, such as fabric texture residual, pilling shadow sensitive, or the instability effect of texture noise removal, to use Gabor filter to filter fabric textures. The Gabor filter can make full use of not only the periodicity but also the directionality of fabric texture. Meanwhile heuristic global threshold segmentation algorithm was applied to keep pilling information, and the proportion of the pilling area was calculated for the evaluation of the grades of pilling, and the feasibility and accuracy of this method was verified by experiment.