2D face analysis: A comparative study

Face analysis is a critical and popular topic in the area of computer vision and image processing for its wide application potential and scientific challenge. 2D Face analysis has been considered as a major solution to deal with unsolved issues of reliable 2D face recognition. In this paper we study and compare two classical algorithms (PCA and LDA) of features extracted which provide theoretical and practical implementation details of the system. In the experimental results, we obtain a 95.7% rate of recognition performance using PCA and a 96.8% rate of recognition performance using LDA of the ORL database, which compares well with other state-of-the-art methods.