Automatic woven fabric classification based on support vector machine

In the textile industry, the classfication of woven fabric is usually manual. To improve work efficiency, this paper purposes a novel approach to extract image features for woven fabric´s recognition automatically. Firstly, the local binary pattern method and the gray level co-occurrence matrix are adopted to compute the fabric image features. Then, the principal component analysis is utilized to reduce the high dimensional feature data. Lastly, a support vector machine is used as a classifier to recognize the woven fabric type. The experiments show that these methods can automatically and accurately classify the plain weave fabrics, twill weave fabrics and satin weave fabrics.