期刊名称：International Journal of Computer Science and Information Technologies
电子版ISSN：0975-9646
出版年度：2012
卷号：3
期号：5
页码：5141-5142
出版社：TechScience Publications
摘要：Facial recognition is considered to be the most vulnerable identification source for many of the criminal identification systems. In major security concerned areas, facial recognition is mostly used and the lack of security can be witnessed in issues like 26/11 Mumbai attacks, Washington attacks and other attacks. There are many algorithms that are developed for the effective analysis of the criminals based on facial recognition as the prime source of witness. However, the major drawback of any facial recognition algorithm is due to lightning conditions, occlusion and orientation issues and most of the available techniques are based on non-degenerative models, but generative models are much effective than nondegenerative models. Hence this paper highlights a model based approach for the effective identification of the face and thereby helping towards the criminal identification.
