期刊名称：International Journal of Soft Computing & Engineering
电子版ISSN：2231-2307
出版年度：2015
卷号：5
期号：3
页码：66-72
出版社：International Journal of Soft Computing & Engineering
摘要：Humans have the ability to identify faces instantly with minimum effort and inspired by this, Face Recognition (FR) tries to imitate this ability by using numerous effective algorithms and has been extensively developed in the last decade. FR has received a lot of attention because of its wide range of its applications. Since Humans store and retrieve images instantly when needed, FR imitates this procedure by holding images in a database and trains them to recognize faces. Although many impactful algorithms have been developed, they are not entirely effective in unconstrained settings. Hence, we thoroughly compare the SIFT method and its two variations SIFT-PCA and SIFT-LDA to prove that the variations are better alternatives to regular SIFT.
