Facial Feature Extraction with Weighted Modular Two-Dimensional PCA

Feature extraction is a key step in the process of face recognition. Principal component analysis (PCA), one of the methods to carry out feature extraction, is widely applied to the field of image recognition. Having studied traditional PCA and several extended measures, a method named weighted modular two-dimensional PCA is proposed in this paper. In this method, a two-dimensional face image is firstly divided into three parts. And then perform feature extraction respectively on these three parts. Finally endow different parts with unequal weights in classification. Experimental results illustrate the feasibility and effectiveness of the proposed algorithm.