Biometric System based on Facial Recognition

Abstract

This work relates to the implementation of an effective face detection and recognition method that can be used as a module in person identification depending on features. The algorithm used in the biometric system is devised into two steps. The first step, called facial detection, is based on Haar cascade technique - composed of sets of Haar-type features. The second step, called facial recognition, is based on principal component analyses (PCA) algorithm. The recognition process is accomplished by converting the current image into feature vector using PCA technique and then the face is recognized by comparing its characteristics with the already stored persons. The complete application was implemented in C# programming language and the results were achieved using the wireless IP camera of type SM6203.