Resumen
Personal identification systems are widely used in everyday life. There are many methods for extracting faces in the original image, the most perspective of which is the usage of algorithms based on neural networks. The aim of the study was to design and test an intelligent system for face recognition based on this technology. In the course of a preliminary analysis of machine learning methods for solving the problem, it was found that it is most appropriate to use machine learning methods based on the analysis of facial micromotions with the subsequent construction of a map of points. Based on the data obtained, the development of a face identification system using computer vision technology was carried out, which was based on the method of creating complex architectures using various features with additional algorithms. A distinctive feature of the developed intelligent system is the ability to analyze several frames that confirm micromovements of the head or blinking. As a result of testing the resulting system using the algorithm of gradient boosting of regression trees, a map of 68 face points was obtained, on the basis of which human faces were identified with objects from the database.