Face recognition via Lorentzian metric

Yerzhan Kerimbekov, Hasan Şakir Bılge

2017

Closed access Metadata (CSL-JSON)
DOI 10.1109/siu.2017.7960335
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

Nowadays the face recognition applications are extremely important in terms of information security. In this study, a new Lorentz Face Recognition (LFR) method based on special properties of Lorentz space was developed. The proposed method produces a similarity value for new test sample according to Lorentz distance. A similarity value is determined by nearest neighbor method as 1 or 0, namely, the true or false face image respectively. Moreover, we propose the Lorentz Feature Selection (LFS) based on Lorentz metric. The LFS is used for dimensionality reduction and to increase recognition accuracy of proposed method in multidimensional face data. The experimental results taken from face data sets show that the proposed LFR method is usable.