RETRACTED: An Image Processing based Fault Detection in Fabrics

Abstract

There are several defect that are occurring in fabrics they occur in the form of a hole, mark, improper stitch, oil stains, missed threads etc. The common thing about these defect are, they cannot be viewed by the naked eye. Hence we firmly say that they defects cannot be effectively identified by manual inspection. However, a much-automated method of inspection is essential. Hence, we are going for computer vision based defect detection. The primary requirement about computer vision is a full-fledged camera that can capture even minute defects. We use K-means clustering along with FCM segmentation in order to segment the flaws efficiently. The classifier used here is SVM classifier, with sensible classification rates of up to 98.9%.

Defective and nondefective classif ication of fabric images using shallow and deep networks

“…Jacintha et al 2 have presented a work on fault detection based on image processing in fabric materials. K‐means clustering is used along with FCM segmentation to segment the defects efficiently.…”