Color Clustering Analysis of Yarn-dyed Fabric in HSL Color Space

A novel method for realizing the color classifying in yarn-dyed fabric is proposed in this paper. The color image of yarn-dyed fabric was obtained by a flat scanner, and then it is converted from RGB color space to HSL color space. By analyzing the characters of hue component in HSL color space, the distance and cluster center of hue component in fuzzy clustering algorithms are redefined. During the iteration in FCM, membership degree and cluster center of H and (S, L) are calculated independently, and the membership degree is normalized in the process. A better clustering quality is shown in the experiment. The color yarn number is detected based on the validity for FCM clusters. Experimental comparisons on RGB color space and HSL color space show that the approach proposed in this article is more effective for color extracting and classifying in yarn-dyed fabric.