Li-Fen Chen, Hong-Yuan Mark Liao, Chin-Chuan Han and Ja-Chen Lin
Face recognition, by definition, should be a recognition process in which recognition is based on the content of a face. The problem is: what is a "face"? Goudail et al. [1] and Swets and Weng [2] have recently proposed state-of-the-art statistics-based face recognition systems. However, they used "face" images that included hair, shoulders, face and background. Our intuition tells us that only a recognition process based on a "pure" face portion can be called face recognition. The mixture of irrelevant data may result in an incorrect set of decision boundaries. In this paper, we propose a statistics-based technique to quantitatively prove our assertion. For the purpose of evaluating how the different portions of a face image will influence the recognition results, two hypothesis testing models are proposed. We then implement the two above mentioned face recognition systems and use the proposed hypothesis testing models to evaluate the systems. Experimental results reflected that the influence of the "real" face portion is much less than that of the nonface portion. This outcome confirms quantitatively that a statistics-based face recognition system should base its recognition solely on the "pure" face portion.