Principal Component Analysis Based Wavelet Transform


The principal component analysis (PCA) is a valuable statistical means,implemented in time domain that has found application in many fields such as facerecognition and image compression, and is a common technique for finding patterns indata of high dimension. This paper investigates the ability to implement PCA infrequency domain, by using the wavelet transform (WT), and evaluate its effectivenessbased on face recognition as a means to find patterns in data. The basic idea offrequency domain implementation of the PCA refers to the correlationimplementation using wavelet transform.The Min-max is invoked to increase wavelet based eigenface robustness tovariations in facial geometry and illumination. Two face images are contrast in termsof their correlation distance. A threshold is used to restrict the impostor face imagefrom being identified. Experimental results point up the effectiveness of a new methodin either using varying (noisy images, unknown images, face expressions, illumine,and scales ).