Psychological Education and Emotional Model Establishment Analysis Based on Artificial Intelligence in the Intelligent Environment

Abstract

Emotion plays an important role in our daily life. It affects people s study and life in varying degrees. This study mainly discusses the psychological education and emotional model building based on artificial intelligence in intelligent environment. In this study, hidden Markov model (HMM) is used to recognize facial expression and describe the output probability of emotional state change. In the aspect of emotion feature extraction, acceleration sensor is used to judge the user s activity state, and optical sensor data and GPS data are used to collect environmental data. In order to reflect individual emotion and its intensity, emotion space method can be used to deal with the reflected emotion vector effectively. Because FACS system is too complex, this model simplifies it. The emotion reflected from emotion space corresponds to a series of AU parameters, which constitute the corresponding facial expression. The strength of these parameters is determined by the size of the emotion vector module. Finally, a sound processing module is added in front of the emotion parameter extraction module of the emotion model for better emotional interaction. In emotion recognition test, the accuracy rate of sensor data based on basic emotion model was 47.13%, 49.08% and 56.32%, respectively. The results show that the model attempts to achieve multi character expression by modifying the emotional space, and achieves the goal of multi modality of the model, which provides the possibility for personalized customization of emotional model in the future.