Perception, Detection, and Diagnosis of Appearance Defects in Fabrics

Abstract

High uniformity of appearance is essential for textiles. Unintended variations of color between and within fabrics pose serious problems: their avoidance makes great demands on manufacturing and quality control at all stages from the extrusion of synthetic fibers and yarns up to the final finishing of a fabric. Most fabric inspection for quality control is visual. Unfortunately in some cases, the appearance of a color defect is a very poor guide to the true nature of the defect, a fact not generally recognized by the industry. This paper describes studies of the responses of observers to well . characterized model color defects using standard psychophysical methods. The results from this new approach will help the industry to understand these problems and provide a basis for the design of sensor systems to be used in automated fabric in spection.