Fuzzy logic and co occurring pattern Based texture descriptors For image analysis

Abstract

Image analysis plays a prominent role in image processing and newlinecomputer vision applications The primary goal of image analysis is to extract newlinethe attributes from the image to represent it in a more suitable form for object newlinerecognition and image understanding tasks Edges regions texture color and newlineshape are some of the useful features extracted from the image Texture is newlinefound to be one of the potential features to characterize images Texture newlineanalysis includes texture feature extraction texture classification texture newlinesegmentation etc This thesis aims to propose methods for describing texture newlinecontents of gray scale and color images which aid in the performance newlineenhancement of image analysis newlineDescribing the texture in the images is an important step in image newlineAnalysis The global texture description in terms of local texture description is newlineone among the number of techniques reported in the literature In this newlineapproach texture information contained in the small regions of the image is newlinerepresented by local description procedure and the texture content of the newlineentire image is globally represented by the collection of these local newlinedescriptors Due to the vagueness and diversity of properties of textures newlineuncertainty arises in describing the texture Fuzzy logic is an efficient tool for newlinehandling uncertainties newline newline