Development of a Metrological Database with Images of a Human Face in the Infrared Range to Evaluate the Effectiveness of Biometric Algorithms

Abstract

The work shows the promise of using biometric algorithms intended for remote determination of current human bio-parameters based on processing of the infrared image of his face. Problems that hinder the widespread use of remote biometric algorithms in practice are highlighted. The urgency of creating a metrological database containing video recordings of the human face image in the infrared range for the evaluation of the biometric algorithms efficiency is substantiated. Presented are the results of an analysis of existing databases containing images of a person’s face. Insufficient development of specialized databases containing infrared images of a person’s face was noted. The main shortcomings of such databases are highlighted. An approach to creating a metrological database based on the acquisition and storage of complex personal data is proposed. The main components of such data are video recordings of the face image in the visible and infrared ranges, time synchronized records of human bio-parameters, as well as information on the complexity of test tasks performed. The structure of the laboratory system for obtaining complex data on the person undergoing testing is considered. A typical data structure for a single test cycle is presented. The application results of the first version of the metrological database for experimental research of the effectiveness of algorithms for remote determination of human bio-parameters are presented.