An Unsupervised Segmentation Framework For Texture Image Queries

Texture image query is becoming an important and complex task. Texture is a phenomenon that is widespread; texture arises from a number of different sources. Firstly, views of large numbers of small objects are often best thought of as textures. Examples include grass, foliage, brush, pebbles and hair. Secondly, many surfaces are marked with orderly patterns that look like large numbers of small objects. Examples include: the spots of animals like leopards or cheetahs; the stripes of animals like tigers or zebras; the patterns on bark, wood and skin. There are three standard problems to do with texture: Texture segmentation is the problem of breaking an image into components within which the texture is constant. Texture synthesis seeks to construct large region of texture from small example images. Shape from texture involves recovering surface orientation or surface shape from image texture.In this paper, a novel unsupervised segmentation for texture image queries is presented. The proposed framework consists of an unsupervised segmentation method for texture images, and an multi-filter query strategy. By applying the unsupervised texture segmentation method on each texture image, a set of texture feature parameter for that texture image can be extracted automatically. Based up on these parameter, an effective multi-filter query strategy which allows the user to issue texture-based image queries is developed.Segmentation is an important part of the computer vision and image analysis, wherein regions of interest are identified and extracted for future processing. The definition of suitable similarity and homogeneity measure is a fundamental task...