UDC 004.93
PREDICTIVE PROCESSING OF VIDEO STREAM DATA IN CONTROLLED ACCESS SYSTEMS
V. L. Gorbunov, Doctor in technical sciences, full professor, National Research University «MIET», Moscow, Russia;
orcid.org/0009-0002-9892-6276, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
Ye Win Zaw, post graduate student, National Research University «MIET», Institute of Automatic Control and Monitoring Systems (SACMS); e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
I. V. Zhdanova, lecturer, Institute of Systems and Software Engineering and Information Technology, Na tional Research University «MIET», Moscow, Russia; e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
The increased demand for a reliable security system has led researchers to look for reliable technology that recognizes identities. Facial recognition systems can be a better solution due to their speed and conven ience compared to other biometric technologies. Depending on the applied tasks, it is necessary to analyze and select methods and algorithms that ensure the solution of the tasks set using limited resources. One of these tasks is a face recognition system for a limited number of people with the ability to rank weights for recognizable subjects. This task becomes relevant when using it for individual production facilities at enter prises or for monitoring residential premises with well-known guests and possible visitors. The purpose of the work is to build a system for identifying a limited number of people who have visited a controlled facility, provided that their access to this facility is certain.
Key words: : face recognition, signs, recognition accuracy, recognition speed, coordinates, key points, histograms.