Automatic recognition of facial expressions using microsoft kinect with artificial neural network

摘要

Autism is a lifelong, developmental disability that affects how a person communicates with and relates to other people. It is defined by a certain set of behaviours such as, Facial Expression of Emotion. Autistic people often have difficulty recognising or understanding other people's emotions and feelings, or expressing their own. Such difficulties can make it very hard for autistic people to navigate the social world. Our study provides a novel clinical tool which facilitates for doctors the diagnosis of Autism Spectrum Disorders (ASD). We focus on automatic identification of six emotions among autistic children in real time: Happiness, Anger, Sadness, Surprise, Fear, and Disgust. In this paper, we present an emotion recognition system for children with autism, which consists of three modules: model-based face tracking, feature extraction, and emotion recognition using artificial neural network (ANN). The experimental results show that our system can achieve above 94.4% recognition rate.