Hybrid CNNLBP using Facial Emotion Recognition based on Deep Learning Approach

Abstract

Listen

Translate article

Facial Emotion Recognition System is one of the emerging technologies and it has been a quite challenging task to recognize the human facial expressions using computer algorithms in emotion analysis rather than the humans. In human’s ideology, facial expressions like sorrowful, joyful, frightened, anger, disgust are assumed as a fundamental part. The main objective of Facial emotion Recognition System research is to focus on the traits of people with criminal associations facial expressions. As the traditional algorithms haven’t met the human needs, where Machine learning and deep learning algorithms have gained great success. There are many techniques applied to gain efficient results in identifying facial expressions for Human-machine interaction, monitoring security and treating patients in the medical field are some of the applications of facial expression recognition. So, it is necessary to build a model-based system to understand human emotions in different scenarios. The Convolutional Neural Network (CNN) algorithm is used to know the facial expressions of humans. On this framework, the CNN layers are upgraded, and the LBP is incorporated with it to merge multiple networking algorithms to develop the human emotion model. In the end, to check the validity of the new technique, CNNLBP is implemented. The experimental result show that accuracy of emotion recognition in between the training and testing phases, the CNNLBP model achieves an average accuracy of 98.3% compared to other traditional models.