Use Principal Component Analysis Technique to Dimensionality Reduction to Multi Source

Abstract

This paper tackles with principal component analysis method (PCA ) to dimensionality reduction in the case of linear combinations to digital image processing and analysis. The PCA is statistical technique that shrinkages a multivariate data set consisting of inter-correlated variables into a data set consisting of variables that are uncorrelated linear combination, while ensuring the least possible loss of useful information. This method was applied to a group of satellite images of a certain area in the province of Basra, which represents the mouth of the Tigris and Euphrates rivers in the Shatt al-Arab in the province of Basra. In this research, when selected the best imaging band in terms of taking the highest eigen value, it is shown that the fourth image band is best when using the PCA method . The application principal component analysis, which depends on the eigen values, showed that the application of PCA gave high and accurate results in determining the best image among the six image beams. It was found that the fourth images which has the highest eigen value is the best and this indicates that it contains the most important Independent properties in images, which are used for analysis, such as isolating water areas, agricultural areas, soil type, etc.