Article 7311

Title of the article

LOOKING FOR A HUMAN FACE IMAGE BY A COMPOSITE PICTURE IN A LARGE DATABASE

Authors

Fedotov Nikolay Gavrilovich, Doctor of engineering sciences, professor, head of sub-department of economical
cybernetics, Penza State University, fedotov@pnzgu.ru
Petrenko Alexander Grigoryevich, Candidate of engineering sciences, associate professor, sub-department of economical cybernetics, Penza State University, alxndrpetrenko@mail.ru
Roy Aleksey Vladimirovich, Candidate of engineering sciences, senior lecturer, sub-department of economic
cybernetics, Moscow State Agroengineering University named after V. P. Goryachkin, ec@pnzgu.ru
Fionov Nikita Sergeevich, Postgraduate student, Penza State University, finik86@rambler.ru 

Index UDK

004.931:004.932 

Abstract

The authors present a new approach based on stochastic geometry and functional analysis to solve the problems of human face image retrieval from a database. A key element of this approach is the use of a new class of constructive features of recognition called triple features. An analysis of physiological principles of perception and recognition in conjunction with the theory of triple features determines a retrieval algorithm described in this paper. 

Key words

face recognition, image database, biometric search, trace transform, stochastic geometry, triple feature. 

Download PDF

 

Дата создания: 25.08.2014 15:44
Дата обновления: 26.08.2014 09:15