Abstract

Facial recognition methods were first explored in security systems to identify and compare human faces and is far superior compared to biometric and iris recognition, this technique has been implemented in iris recognition, image detection, etc. Recently these methods have been explored in other fields of study and have become a commercial identification and marketing tool. This paper describes the different algorithms of facial recognition and compared their recognition accuracies. The face is detected through Haar Cascades algorithm which is saved into a database, after that, the study intended to compare facial recognition accuracy of the well-known algorithms Eigen faces with Principal Component Analysis (PCA), Support Vector Machine (SVM), K-Nearest Neighbor (KNN) and Convolutional Neural Network (CNN). The results showed out of the three algorithms we used CNN yielded the maximum accuracy.

Keywords: Eigenvalues Haar Cascades facial recognition Principal Component Analysis (PCA), Convolutional Neural Network (CNN), K-Nearest Neighbor (KNN), Support Vector Machine (SVM)