Human Mood Identification (Data Recording System)

Abstract

Researchers across a variety of fields are becoming more interested in a human-computer interface system for automatic facial emotion identification. In our project, an Automatic Facial Expression Recognition System (AFERS) which records emotion for physiological analysis has been proposed. The suggested process consists of four steps: Identifying every face in the frame is part of the first stage. We focus on each face and adjust it in such a way that even if the face is turned in a different direction it is still able to identify the person. In the second stage, we will make use of a deep face algorithm to detect the emotion of the person. Unique features of the faces are selected that can be used to differentiate people and compare them with the faces of people stored in our database to recognize a person. The recognition rate for this method is around 90–95%. In the third stage, we have our Attendance System which will record the emotion of the person after every 1 sec which can be used to understand the person psychologically. At our final stage, we have developed GUI using Tkinter for user interaction.