Face Recognition Based on the Statistics Methods

摘要

In many face recognition methods, the ones based on the statistics theory are more commonly used and proved to be effective. This paper introduces two of these methods: an improved method named weighted modular Two-dimensional Principal Component Analysis (WM-2DPCA) and Bayesian classifier. And then combine the advantages of these two methods and apply them to face recognition. Experimental results show that the combined method can be used successfully in face recognition, and also illustrate the effectiveness of the combination.