Research And Application Of Emotion Recognition Algorithm Based On EEG

With the constant exploration of researchers,the use of computers for emotion recognition research has been greatly developed.In the research of emotion recognition,identification contents and recognition algorithms are two important research hotspots.When the external physiological signal is used as the identification content,since the identification content in the actual application has a human control factor,the credibility of the emotion recognition result is low.Electroencephalography(EEG)has strong objectivity,so researchers chose to use different channels of EEG as identification content.Since human emotions are generated by the interaction of EEG from different channels,extracting correlation features between different channels is very important for emotion recognition algorithms.In order to obtain more accurate and objective emotional recognition results,this thesis has done the following work:(1)This thesis introduces the background and research significance of the subject,and studies and analyzes the research status in the field of emotion recognition from the aspects of recognition content and recognition algorithm,and introduces the process of emotion recognition and related technologies in the field of deep learning.(2)This thesis analyzes the selection method of stimulus materials in the current data collection process,chooses to use music video as the emotional induction material of the tester,designs the emotion collection process for data acquisition,and preprocesses the collected data.(3)When only the EEG of different channels are included as the identification content,this paper proposes an emotion recognition algorithm for EEG.The algorithm can extract the features between the EEG of different channels,and use the fully connected neural network to perform emotion recognition on the features.The simulation results show that the algorithm can achieve emotional recognition more accurately than the same type of algorithm.(4)When the recognition content includes EEG of different channels and various external physiological signals,this paper proposes an emotion recognition algorithm for physiological signals.The algorithm can perform feature extraction and feature fusion on two types of signals,and use the fully connected neural network to perform emotion recognition on features.Finally,through the public data set to verify,the algorithm can improve the accuracy of emotion recognition,and the algorithm has its practical application value through the collected data set verification.