Face Emotion Recognition to Detect Depression

Abstract

In the current competitive world, one of the most familiar and grave mental illness we encounter in humans is Depression also called as major depression or major depressive disorder. It makes you feel depressed and disinterested all the time, which has a bad impact on your thoughts and behaviour. Thus affecting not only the victim but also people associated with them, such as family, friends and society. If not treated properly, it can end up with adverse actions such as suicide or hurting others. Therefore, it is very essential to detect such people and provide them with the necessary required treatment. With advancements in Artificial Intelligence and Deep Learning, Facial Emotion Recognition has always been in focus. Through the facial expression, we can detect the emotions of a person. Convolutional Neural Network (CNN) method will be used to recognize facial expressions in real time from picture frames in video data. The video input will be taken through an external camera module fitted on the ESP32 chip. Image pre-processing will be done for better feature (expression, emotions) extraction. This processed images will be then sent into a deep learning model which predicts whether the facial expressions indicate depression. Depending on whether the Depression is detected or not, a message will be sent to that particular person and also to their well-wishers.