Image Enhancement by Hypothesis based Combination of multiple Filtering with multiple algorithms

Abstract

In image enhancement process involves removal of noise and distortions like artifacts ,to improve the visual perspective for a viewer. Due to unique characteristics the document image having different contents with different regions which distorted differentially by noises and several artifacts. In many applications, single filtering approach will not give a good result at each pixel locations. So we provide a new approach a Combined Filtering based on a hypothesis for image quality improvement . It utilizes an approach to select a set of filters to improve the quality of distorted region of image with different regions. The HCF extracted the featured vector to predict the performance of filtering for estimating pixel intensity in original image. Maximum likelihood estimates of the model parameters are calculated by unsupervised clustering Expectation Maximization(EM) algorithm and FCM clustering ,which used to weighting for the filter output. In this way, the HCF serves as a framework for combining the outputs of a number of different user selected filters, each best suited for a different region of an image.To improve contrast quality Bi histogram equalization techniques applied for various images such as remote sensing images and general images which are obtained at the output of combined filtering The scheme consistently improves the quality of the decoded image over a variety of image content having