Fabric Defect Detection Using Salience Metric for Color Dissimilarity and Positional Aggregation

In this paper, inspired by an inherent characteristic of human visual system capable of recognizing salient regions from a complicated scene, we treat a defective region as a salient region in fabric images. A novel fabric defect detection method, which is based on saliency metric for color dissimilarity and positional aggregation, is proposed. In the method, the RGB color space of a given fabric image is first converted into the CIE L*a*b color space for feature representation. Then, the color dissimilarity and the positional distance between similar patches are jointly used to measure the defective values. To improve the contrast between the defective region and the non-defective region, a multi-scale analysis scheme performed on the pyramid images of the input fabric image is applied to calculate the defective values. Finally, the obtained defect map image is further enhanced by a certain threshold regarding to the obtained defective values. Thorough experimental results on several types of fabric images indicate the effectiveness of the newly proposed method.