A Novel Approach to Face Recognition using Image Segmentation Based on SPCA-KNN Method

In this paper we propose a novel method for face recognition using hybrid SPCA-KNN (SIFT-PCAKNN) approach. The proposed method consists of three parts. The first part is based on preprocessing face images using Graph Based algorithm and SIFT (Scale Invariant Feature Transform) descriptor. Graph Based topology is used for matching two face images. In the second part eigen values and eigen vectors are extracted from each input face images. The goal is to extract the important information from the face data, to represent it as a set of new orthogonal variables called principal components. In the final part a nearest neighbor classifier is designed for classifying the face images based on the SPCA-KNN algorithm. The algorithm has been tested on 100 different subjects (15 images for each class). The experimental result shows that the proposed method has a positive effect on overall face recognition performance and outperforms other examined methods.