Analyzing Face Recognition Using Pca and Comparison between Different Distance

iated with each of the eigenvector. This paper he position, PCA Algorithm Principal Component Analysis (PCA) dimensionality reduction technique in which multidimensional data is reduced by extracting the desired number of principal components. Without the need of intense effort PCA provides us with an outline that describes how to reduce a complex data set to a lower dimension to reveal certain unknown useful and simple information. This is the case when there is a strong correlation between observed variables. This strong correlation is present in human faces because every face has various features in common like two eyes, one nose, one mouth etc. at defined positions. By capturing the variation in a set of faces, independent of any judgment of features it is possible to extract the relevant information in a face image which is encoded as efficiently as po and compared with a database of encodings developed similarly for every face [1]. PCA approach transforms face images into a small set of relevant features images called Eigen faces which are ghostly images represented by eigenvectors that are the principle components of the faces and characterize the variation between them. Each individual face can be represented exactly in terms of a linear combination of the Eigen faces. Also not all but only few relevant eigenvectors with largest eigenvalues are sufficient to represent each face image. This method thus reduces the dimensionality of data space by projecting data from M-dimensional space to P dimensional space, where P<<M [5]. -9655 [683-686] on rate obtained