An Improved Adaptively Weighted Sub-pattern PCA Approach for Face Recognition

摘要

A face recognition algorithm based on adaptively weighted Sub-pattern PCA approach is presented in this paper. The proposed algorithm when compared with conventional PCA algorithm has an improved recognition rate for face images with large variations in lighting direction and facial expression. Unlike PCA based on a whole image pattern, the improved adaptively weighted Sub-pattern PCA (IAw-SpPCA) operates directly on its sub-patterns partitioned from an original whole pattern and separately extracts features from both same and different persons'face image, unlike mPCA that neglect different contributions made by different parts of the human face in face recognition, lAw-SpPCA can adaptively compute the contributions of each part and then endows them to a classification task in order to enhance the robustness to both expression and illumination variations. In the process of classification, we use face image set of both same and different person. Experiments on two standard face databases show that the proposed method is effective.