Face Acknowledgment using Principle Component Analysis (PCA) of Eigenfaces

Abstract

Face acknowledgement is a biometric framework used to recognize or check an individual’s identity against a dataset of images. In recent times, there have been several different approaches utilized to try to achieve high accuracy rates. This paper presents a system that enables an individual’s identity to be determined based on a matching of their facial structure against a previously stored database. The matching compares the frontal view of the face with the two-dimensional images of the head already stored. In our system, the input image is sometimes enhanced using histogram equalization, before the matching takes place using the Euclidean distance between the face to be identified and those already stored. The developed acknowledgement system provides an accuracy of 97.5%.