Research of recognition algorithms аnd identification when video recording in real time

Abstract

The study is devoted to the analysis of recognition and identification algorithms for real-time video recording. Modern methods of computer vision and machine learning used to highlight facial characteristics are considered. The stages of detection, feature extraction and authentication algorithms are discussed, with an emphasis on the application of machine learning. The prospects of application in the fields of security and video surveillance are outlined, as well as ethical and legal aspects related to the processing of personal data are highlighted. The study provides an overview of current trends and challenges in this area. An assessment was made of the influence of the system architecture on the efficiency of face recognition and identification processes from real-time video recordings, based on the results of which recommendations were developed for designing an effective video recording system in real-time conditions in the context of this study.