Face recognition 1

Abstract

Facial recognition software has been a hot topic for research due to its practicality in today’s society, be it in security applications such as identifying a suspect from an image source or video source, or in schools where face recognition technology can be used for attendance taking.

It has been observed that the accuracy and reliability of the face recognition system depends on many factors. Some of them include: the angle at which the face is facing the camera, the background noise accompanying the image source or video source, and lastly, the algorithm used for both face detection and recognition.

This paper aims to evaluate the effectiveness of face recognition systems using primarily the viola-jones object detection framework for face detection and Principal Component Analysis (PCA) for face recognition. This is done by evaluating a face sample, either from an image source or from a live video source against a reliable database of faces. Thus, the reliability of the face recognition system can then be measured. Last but not least, the technique of Principal Component Analysis is compared to other face recognition techniques, specifically the Fisher Linear Discriminating (FLD) approach and the Linear Discriminant Analysis. (LDA)