Texture Segmentation Algorithm Using Multivariate New Symmetric Mixture Model and Dct Coefficients

Abstract

Among the most challenging aspects of processing digital images is image classification, which handles the processing and analysis of the images. In higher level image processing applications like medical imaging, robotics, automation, and other key areas where the object study is thought to be of highest importance, the outputs obtained from the segmentation findings are treated as starting parameters. The segmentation process uses pre-requisite criteria to find objects of interest. The feature vector is typically taken to be this criterion. Separating the diverse set of images is the new goal of texture segmentation into regions of homogeneous pixels that have significance, based on this criterion. Texture is regarded as the most important of these feature vectors since it explains the spatial relationship between individual pixels in an image.