Texture classification of woven fabric based on a GLCM method and using multiclass support vector machine

This paper proposes the classification of three mean woven textile fabrics. The classifier is based on the texture analysis of woven fabric images for the recognition. In the pattern recognition phase, the co-occurrence matrix is applied to calculate the texture characteristics, such as the angular second moment, the correlation, the homogeneity and the contrast. We have varying the offset in distance and orientation, the best offset was retained for the classification. Taking advantage of the difference between the woven fabric textures of these parameters, a support vector machine is adopted as a classifier to categorize the type of woven fabric. Two types of multi-class classifiers were tested: the one against all and the one against one. The experimental results show that some of the studied parameters are more compatible with the SVM classifier than the others, for example the homogeneity leads to a recognition of 100% which makes it the most compatible parameter of the feature texture for the classification. We note that the one against all classifier gives better results than the other.