期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2014
卷号：2
期号：11
出版社：S&S Publications
摘要：A facial recognition system is a computer ap plication for auto matically identifying or verifying a person from a digital image or a video frame from a v ideo source. One of the ways to do this is by co mparing selected facial features from the image and a facial database. It is typically used in security systems and can be compared to other biometrics such as fingerprint or eye iris recognition systems. Face recognition is not perfect and struggles to perform under certain conditions. There are two key problems for any face recognition problem: the illumination problem and pose problem. The performance for face recognition system drops significantly when pose variations are present in input images. The task of face recognition becomes even more difficult when illumination variations are present. Face recognition applications commonly suffer from three main drawbacks: a reduced training set, information lying in high - dimensional subspaces, and the need to incorporate new people to recognize. We propose an accurate 3D facial feature localization approach based on 3D shape descriptions using Nearest Neighbor technique. An accurate face alignment using a combination of face alignment method PCA is implemented. The proposed method is sho wn in the experiment with more accuracy.
