Traditional versus Neural Network Classification Methods for Facial Emotion Recognition

Abstract Facial emotion recognition (FER) is a topic that has gained interest over the years for its role in bridging gap between Human and Machine interactions. This study explores potential of real time FER modelling, to be integrated closed loop system, help treatment children suffering from Autism Spectrum Disorder (ASD). The aim this show differences implementing Traditional machine learning Deep approaches modelling. Two classification were taken, first approach was based on classic techniques using Histogram Oriented Gradients (HOG) feature extraction, with k-Nearest Neighbor Support Vector model as classifiers. second uses Transfer Learning popular “Alex Net” Neural Network architecture. performance accuracy randomly selected validation sets after training random Oulu-CASIA database. data analyzed shows traditional methods are effective deep neural net models good compromise accuracy, extracted features, computational speed costs.