Face database compression by hotelling transform using a new method

Principal Component Analysis (PCA) is a method for compressing high dimensional databases [1]. If it used for image compression, it called Hotelling or KL transform. This method extracts q Eigen vectors and q Eigen values for database [2]. The quality of retrieved images various and some of them have very low quality. In this paper an optimized PCA method is introduced which can increase the quality of reconstructed images focusing on very noisy images. Keywords: Hotelling, Compression ratio, Eigen Face, Eigen value, Eigen vector, Principal Component Analysis;