Face Recognition Method Based on Adaptively Weighted Block-Two Dimensional Principal Component Analysis

Face recognition is one of the most important part of biometrical recognition. 2-Dimensional Principal Component Analysis (2DPCA) is a classic method in face recognition, which is proposed to reduce the computational cost of the standard Principal Component Analysis (PCA) algorithm, but the performance of 2-Dimensional Principal Component Analysis in reducing computational complexity and recognition rate is not satisfying. This paper mainly focuses on the feature extraction method of adaptively weighted Block 2-Dimensional Principal Component Analysis. The block methods divide a large picture into several smaller sub-blocks to get the local discrimination information and reduce the computational complexity. Then, a weighted Euclidean distance classifying algorithm is proposed to extract features of face images, and the Euclidean distance classifier is used for classifying. The experiments show that the Adaptively Weighted Block 2-Dimensional Principal Component Analysis method has better performance than standard 2-Dimensional Principal Component Analysis.