Human Stress Recognition System Using Deep Learning Techniques Based on Physiological Signals Data

Abstract

Computers are increasingly becoming an indispensable component of our daily existence. Therefore, it is imperative that we are able to engage in authentic conversations with them, akin to our connections with other human beings. Stressful conditions are crucial elements that provide instinctive engagement in human-computer interaction. Affective computing integrates human emotional emotions with computer systems, potentially enhancing communication between expressive humans and emotionally limited computers. Various modalities, including facial expressions and physiological markers, can be utilised to identify human stress levels. Physiological signals have distinct benefits compared to face expressions because they are more responsive to internal emotions and less affected by the social concealment of emotions. The development of small wearable physiological sensors has led to significant progress in stress identification, enabling the emergence of new applications in human-computer interaction (HCI) such as mental health care, intelligent tutoring, and transportation safety. A significant number of these wearable sensors are user-friendly and facilitate practical usage in real-world scenarios newline