发明名称 Method for face recognition
摘要 A method for face recognition is provided with collecting a match facial image; retrieving a reference image from image records of a database or an input image; selecting one or more facial features from each of the match facial image and the reference image; obtaining at least one match facial feature and a match deviation of the reference image corresponding to the facial features of the match facial image; creating a match geometric model and a reference geometric model; obtaining a model deviation by comparing the match geometric model and the reference geometric model; and employing a match deviation and a model deviation to obtain a recognition score based on a predetermined rule. The method involves a two-way face recognition by integrating facial features of block matching with geometric model comparison. It employs relationship of match deviation and model deviation.
申请公布号 US8953852(B2) 申请公布日期 2015.02.10
申请号 US201213527454 申请日期 2012.06.19
申请人 Chung Hua University 发明人 Huang Yea-Shuan;Peng Kuo-Ta
分类号 G06K9/00 主分类号 G06K9/00
代理机构 代理人 Shih Chun-Ming
主权项 1. A method for face recognition comprising the steps of: (a) collecting a match facial image and a reference image from an image taking device and an image record of a database, respectively; (b) normalizing the match facial image and the reference image to have a same predetermined image size and orientation; (c) selecting one or more facial feature points from each of the match facial image and the reference image; (d) obtaining corresponding matched facial feature points of the reference image and an average match deviation (AMD) for at least one facial feature point of the match facial image; (e) creating a match geometric model by using the at least one facial feature point of the match facial image, and a reference geometric model by using the corresponding matched facial feature points of the reference image; (f) obtaining a geometric model deviation (GD) by comparing the match geometric model and the reference geometric model; and (g) employing the average match deviation (AMD) and the geometric model deviation (GD) to obtain a recognition score (S) based on a predetermined rule, wherein one possibility of the predetermined rule is shown as below, and S=1/AMD×GD wherein the method is performed using a computer.
地址 Hsinchu TW