Fuzzy Based Image Enhancement and Segmentation Techniques

Abstract

The problem of segmenting gray scale, still images has been addressing in this newlinework. The definition of a general purpose segmentation technique has been revealed as a newlinecomplicated task. This complication is owing to the huge amount of different kind of newlinedata that a segmentation technique may have to handle. Although image segmentation is newlinebasic field of research, the most difficult problem faced is that the real objects have newlinecomplex shapes and boundaries. To tackle the difficult problem of image segmentation, newlineresearchers have proposed a variety of methods. The previous approaches to multistage newlinesegmentation at different scales were using structures at coarser scales but the problem newlineof structure of extraction is difficult task for getting optiomal threshold value. The newlineproposed work describes image segmentation at multiple scales by integrating with newlinedifferent structures. These techniques relying on boundary, textured and non-textured newlineinformation for image segmentation at multiple scales. This work argues that the issues newlineof scale selection and structure detection cannot be treated separately for segmentation. newlineSoft computing techniques are most suitable for addressing this kind of problems. Fuzzy newlineimage segmentation is a task that classifies pixels of an image using different labels so newlinethat the image partitioned into non-overlapped labeled regions. In this dissertation fuzzy newlineclustered based techniques are studied and developed Fuzzy Entropy technique, Rule newlinebased Type-II fuzzy logic, Edge detection based on gradient fuzzy logic, Generalized newlineFuzzy C-means and Fuzzy Entropy triangular model for super resolution images. The newlineexperiments have been done on well known image data bases and the results are newlineproduced in the form of tables and graphs for objective analysis and outputs of input newlineimages are placed for subjective analysis. In both objective and subjective analysis cases newlinethe proposed techniques have produced accurate results than traditional techniques. newlineKeywords: Image enhancement, Image Segmentation