Face recognition using Kernel Fisher's Discriminant Analysis and nearest neighbor

Face recognition is a technology that is achieving more and more prominence today. This technology is now found in various applications such as automatic photo tagging and identification of criminal suspects. While the task of recognizing faces is easy for humans, the task of teaching a computer to do so is very challenging. This paper presents a face recognition system based on the Kernel Fishers Discriminant Analysis (KFDA) and Nearest Neighbor (NN) algorithms. We use the KFDA algorithm as a feature extractor and the NN algorithm as a classifier. Our current implementation of the system has achieved a recognition success rate of more than 83%.