Image Processing and Reconstructing Using Principal Components Analysis

Abstract

Principal components analysis (PCA) is a process of identifying image sequences in an
effective way to find differences and similarities of information. Generally, it is very hard to
find information from the high dimensional messages, where the graphical representation is
not available. In this regard, the PCA is an important statistical procedure for analyzing data
and find the reduce number of desired sequences. Essentially, the correlated sequences are
transformed into a set of uncorrelated ones which are arranged by reducing variability. The
PCA can reduce the computational complexity after reduction dimensions. This paper, we use
an original image of dimension 408x350. By applying PCA, we reduced the original
dimension to 408x57. Finally, we are remonstrated the original image by using selected
features vectors.