Algorithm Analysis and System Implementation of Advanced Facial Information Extraction and Face Recognition under Wireless Networks

In recent years, there has been an emerging demand for utilizing biometrics especially face recognition in access control of secured areas and other applications like identity validation. In this paper, we develop a real-world face recognition system includes face detection and alignment, feature extraction under wireless networks. The proposed facial information extraction and face recognition system is capable of using real-time video from wireless networks. In order to reduce the influence of illumination, many preprocessing methods including Histogram equalization, Homomorphic filtering and MSR algorithm are applied. For enhancing the performance and accuracy with the variations of pose and illumination, Gabor feature is extracted to represent the facial images. Experiments on a real-world dataset of 100 images belong to 20 persons are tested and results prove the system has good performance in practical situation.