Application of pattern recognition on classifying texture images

Texture images differ from common gray images in pixels spatial distribution, and general gray image characteristic are unusable for classifying texture images. In order to classify the face and the inverse of leather which is typical texture image, achieve characteristic of face image and inverse image based on co-occurrence matrix and then design a linear classifier based on Fisher discrimination criterion to classify them. The experimental result indicates that the linear classifier is generally correct for classifying the face and the inverse of leather samples.