A generalized segmentation approach for texture analysis and matching

In this paper we propose a texture segmentation and matching approach using the Local Binary Pattern (LBP) method. First, we examine the principle of segmentation and matching algorithm [1] that detects a unique given texture over a given image. Then we will describe our automatic segmentation algorithm which detects all textures in an image without any given information but the targeted image itself. Our method is applied on randomly generated images from a database of over 150 textures.