
A method for circumventing the biometric identification system based on facial recognition
Author(s) -
В. О. Антонов,
Nikolay Svistunov,
A. V. Garanzha,
I P Bychihin
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1069/1/012047
Subject(s) - biometrics , facial recognition system , computer science , identification (biology) , task (project management) , artificial intelligence , face (sociological concept) , three dimensional face recognition , pattern recognition (psychology) , computer vision , face detection , engineering , social science , botany , systems engineering , sociology , biology
The article deals with the task of identifying wanted people using two-dimensional face recognition. A potential method for circumventing the recognition system is described. “Potential” in this case means the absence of precedents for the use of the described method in governmental systems. An illustration of the algorithm for implementing a potential method for disrupting the functioning of the reference samples database of the facial recognition system is given. The results of testing the effectiveness of the presented method on the Microsoft Azure platform are presented. The results of the analysis of the modified image for obvious signs of montage are presented. As a result, the conclusion is presented that it is impossible to implement effective and operational protection of the reference samples database from the presented potential method of circumventing the facial recognition system.