Low cost, portable facial emotion recognition device for autistic children
This paper details the use of Raspberry Pi 4 to develop a facial emotion recognition device for autistic children. This device will be used to recognize seven basic emotions: Happiness, Anger, Sadness, Disgust, Fear, Surprise and Neutrality. The face detection algorithm used is the Viola-Jones face algorithm whilst the feature extraction algorithm used is a fusion of local binary patterns and histogram of oriented gradients. The machine learning algorithm used for classification is the Linear Support Vector Machine. This system produced an accuracy of 35.68% with the most accurately classified emotion being happy. This paper investigates the use of both local binary patterns and histogram of oriented gradients as feature extraction methods as well as both Support Vector Machine and Multilayer perceptron neural network as classification methods.
Capstone Project submitted to the Department of Engineering, Ashesi University in partial fulfillment of the requirements for the award of Bachelor of Science degree in Electrical and Electronic Engineering, May 2021