Iot Based Stress Detection Using Cognitive Assistance For The Elderly

Facial expressions are one of the key features of human beings that can be used to speculate emotional state at a particular moment. The current work employs Convolution Neural Network to develop a facial recognition model that categorizes a facial expression into some different emotion categories Afraid, Anger, Disgust, Happiness, Neutral, Sad, and Surprise. Capturing facial expressions over a certain period of time can give an idea of what extent the elderly is feeling pain and can enable nurse/family members to decide their feelings and provide necessary assistance. For identification, the elderly’s photos are continuously taken with a smart camera and sent to the decision maker (laptop or desktop). Once the elderly are identified, he/she is monitored continuously for emotion recognition through facial expressions, and the detected emotions are stored. When an abnormal condition is detected, an alert message is sent to the caretaker/nurse. The system analyses visual cues, such as facial expressions and eye movements to detect signs of stress using computer vision and machine learning.