An Enhanced Deep Learning Model for Classifying Human Emotions Based on Facial Expressions

Human emotion detection (recognition or identification) is spread over an area of studies, and research on human emotions is continuously booming. One of the most important research fields is the human-computer relationship. Furthermore, in this interrelation, while humans are interacting with computers, the device (the computer) needs to start identifying root emotions. Identifying Human emotions is not everyone's cup of tea. Humans are extremely smart at covering their emotions. Therefore, recognizing emotion is challenging. Most human emotions are linked to facial expression. The goal of this paper is to classify individual emotions based on their facial expressions through image classification. Several deep learning algorithms such as Convolutional neural networks CNNs, Artificial neural networks ANN, Fully connected neural network FCNN are used for image classification. Keras library which is part of CNN, ANN and FCNN is used for powerful image classification. The results of the study depicts that CNN model is best in terms of accuracy to classify all types of facial expressions.