Research Of Fabric Pilling Objective Evaluation Based On Image Analysis

In the fabric pilling objective evaluation system, pilling image segmentation is a crucial step. Traditionally, the fabric pilling grades is judged according to the standard sample. But it is greatly subjective and will produce different results among different people. And it is difficult to describe the fabric precisely and quantificationally. Using image processing technique to evaluate the grades of the fabric can overcome the limitations of the traditional method and has important significance for promoting fabric performance digital testing.This paper researches the key technique of fabric pilling objective evaluation and includes pilling extraction and the segmentation of conglutination pilling grains. In the pilling extraction, the standard sample is deeply researched and mathematical morphological close-operation and Top-Hat transform are used to preprocess the image in order to reduce the influence of the holes between fabric voile and the kinks in the following step. Then, we realize fabric image binarization by using minimum bias-normal distribution and extract the image that only includes pillings. This method is simple and effective. In the research of the segmentation of conglutination pilling grains, a new method is proposed to segment the grains through analyzing the figure of pillings. It can locate the conglutination part and the segmentation points accurately, and realize the segmentation of pilling grains.