Effective Independent Component Analysis Algorithm (EICA) for Blind Source Separation of Mixed Images for Biomedical Applications

Abstract

When dealing with medical applications like X-Rays, EEG, MRI and analyzing individual image component into individual images objects and thereby evaluating the exact image and process it the primary purpose is to convert analysis of images into information. The most basic problem encountered to this procedure is to present the database of mixed images or sample images patch during its analysis i.e. during image segmentation or de-noising for knowing the exact information provided through the mixed or individual sample of images. Independent component analysis (ICA) is new algorithmic approach in Image segmentation of investigate and is being applied for their mutually exclusive statistical PSNR in independent separation images. Independent component analysis is primarily a procedure class from the concept of BSS which is a theoretical concept given for the image and sound representation. In this paper we have developed an analytical approach for an effective distributive algorithmic approach for ICA-based blind source separation for separating out individual component from a mixed picture with maximum PSNR. In blind source separation (BSS) produces after simulation in MATLAB all original images from the observed mixtures. Independent Component Analysis (ICA) is built retrieve individual values called components from a non-regular combination called mixture which are more statistically independent from one another as possible (preferably non-Gaussian higher statistical calculations). The paper will be divided into 4 sections with ICA-BSS introduction, basic theoretical and mathematical model approach for image segmenting or image de-noising from mixed sample picture followed by simulation results followed by conclusion. The simulation of mixed images has been done in MATLAB.