Fabric Defect Classification based on Deep Hashing Learning

Classifying categories of fabric defects can greatly help to identify the source of causing fabric defects in the textile manufacturing process. Most existing artificial intelligence based methods focus on identifying and locating defective regions and do not analyze the categories of the defects. On the other hand, as current fabric defect detection methods depend on handcrafted features, they can only handle fabric with specific patterns or textures. In this paper, we propose a novel model which can learn high-level representation from the automatic observations of the input images that can recognize the categories of the defects for various fabric patterns and textures, instead of only locating defects on specific patterns. Experimental results show that the proposed method is superior to the state-of-the-art deep hash methods in terms of fabric defect classification.