System of automatic analysis of methods for recognizing the emotional state of a person

Abstract

The introduction of methods and means of digital transformation of industry and the social sphere poses new challenges. One of these tasks is the management of active systems, in which simple registration and identification of the initiators of actions is not enough, but a deeper assessment of their state, including psychophysical and emotional, is required. The article is devoted to the analysis of methods and means of recognizing the emotional state of a person. Approaches to automated recognition of a person's emotional state based on primary, secondary and more complex features are analyzed. The main focus is on a comprehensive approach to recognition of a person's emotional state based on the analysis of visual and audio channels using neural networks and computer psyche algorithms. The target of the article is a formalization of methods and development of means to recognize a person's emotional state according to complex audiovisual criteria. Analyzed software for recognizing the emotional state of a person: FaceReaderNoldus, EmoDetect, FaceSecurity, Microsoft Oxford Project Emotion Recognition, eMotion Software, MMER_FEASy. Methods used to recognize the emotional state of a person by his face have been investigated, such as: the method of basic components, the Viola-Jones method, template comparison, the Hopfield neural network, the method based on the localization of key points on the face and the method based on texture information. Separately analyzed methods of recognizing the emotional state of a person from his speech. invention proposes a solution for multilevel recognition of emotional state of a person based on using algorithms of neural networks and computer psyche. Software for recognizing the emotional state of a person was analyzed: the developed approach can be used for various digital applications, ranging from the analysis of the psychophysiological and emotional state of a person - an operator (pilot, driver, etc.), to multimedia mobile applications for analyzing the emotional state of an interlocutor. The trend for remote work and self-employment opens up such areas of application of these applications as consulting psychologists, interviewing online, optimizing the human resources of companies and organizations. Prospects for using the latest versions of emotional analyzers are shown.