Basic Emotion Recognition in Facial Expressions using Deep CNN

One of the most common forms of non-verbal communication is facial expression. While there are many different types of facial expressions, psychologists have identified six basic expressions that are universally recognized: happiness, sadness, surprise, anger, fear, and disgust. Recognizing facial expressions is a complex task that relies on detecting significant movements in specific areas of the face. These innovative systems have a wide range of applications, including assisting children with autism, enhancing video games, and improving human-machine interaction. In this study, we propose developing a system that is able to identify and recognize individuals based on the emotional state of their facial expressions. The system uses a deep learning-based facial expression classifier that employs a convolutional neural network (CNN) algorithm. This paper presents a comparison of two Deep CNN architectures: a novel CNN architecture and the well-established VGG16 architecture. We chose to use VGG16 due to its success in image recognition. Both approaches were tested on the FER2013 database.