DCA: An Approach for Face Recognition through Component Analysis

Abstract

The developments in digital technologies is enhancing the communication between computers and Human. Face recognition is a vital technique that helps to develops user-friendly methods for Human Computer Interaction (HCI). In this paper we suggest a pattern recognition method using Difference Component Analysis (DCA) and present its application in face recognition. The DCA removes all the general features of images and compute the difference components of the images. The DCA is based on the principle that two similar images will have least difference components. Simulation of DCA is done by using Yale face database and AT&T face database. The experimental results indicate that, the presented approach is remarkably effective in recognizing faces under different conditions.