International Journal of Scientific Engineering and Technology (ISSN : 2277-1581) www.ijset.com, Volume No.1, Issue No.2 pg:22-26 01 April 2012 22 Abstract: This paper presents the PCA (Principal Component Analysis) based algorithm for noise removal from noisy images, the concept is based on analysis which states that the noise components affects all parts of image uniformly, the noise can be from many independent components, in proposed algorithm the noisy vector is searched on the basis of global similarity of vectors of each segments from the images. The proposed algorithm simulated on MATLAB 7.5 and the results shows that the algorithm works well for de-noising with faster than the other wavelet based PCA methods. Keywords: PCA (Principal Component Analysis), image de-noising. 1. Introduction Image de-noising is a vivid research subject in signal processing because of its fundamental role in many applications. With the rapid development of modern digital imaging devices and their increasingly wide applications in our daily life, there are increasing requirements of new de-noising algorithms for higher image quality. In the past few years, image de-noising has been deeply impacted by a new approach,instead of processing each pixel individually; it has been shown to be preferable to de-noise the image block-wise. Taking advantage of the redundancy of small sub- images inside the image of interest, new robust methods have emerged that can properly handle constant, geometric and textured areas. There are several de-noising proposals are already available but limitation of processing power and resources availability like power, memory, etc. it is required to develop a new technique which will be capable of doing this work under limited resources. The proposed technique is a way for such systems here the complicated frequency and mixed domain transformation is avoided to reduce the complexity of the algorithm without compromising with the de- noising performance. The rest of the paper is arranged as that the 2 nd section a discussion on recently proposed works on same topic is done. The 3 rd section explains the basic of PCA after that 4 th section presents the proposed algorithm followed by simulation results and conclusion in 5 th and 6 th section. 2. Literature Review A comprehensive review of the literature on image restoration and de-noising is not necessary for this paper. Hence this section only gives a brief summary of the closest related work. One approach to image de-noising proposed by Lei Zhang, David Zhang et al., in their paper “Two-stage image de-noising by principal component analysis with local pixel grouping” [1], This paper presents an efficient image de-noising scheme by using principal component analysis (PCA) with local pixel grouping (LPG). For a better preservation of image local structures, a pixel and its nearest neighbours are modeled as a vector variable, whose training samples are selected from the local window by using block matching based LPG. Such an LPG procedure guarantees that only the sample blocks with similar contents are used in the local statistics calculation for PCA transform estimation, so that the image local features can be well preserved after coefficient shrinkage in the PCA domain to remove the noise. The LPG-PCA de- noising procedure is iterated one more time to further Image De-noising by Common Vector Elimination in PCA (Principal Component Analysis) Mr. Zahid Alam Associate Professor,Department of Electronics & Communication Lakshmi Narain College of Technology, Bhopal (M.P.) Email: zahidasif@yahoo.com