Electroencephalogram emotion recognition method and system based on depth domain self-adaption

An emotion recognition and self-adaptive technology, applied in the field of EEG emotion recognition, can solve the problems of not meeting the recognition accuracy rate, not taking into account the channels of the individual differences of the EEG signals of the subjects, etc., so as to achieve accurate and accurate EEG emotion recognition results. The effect of reducing variance

Active Publication Date: 2022-02-18
SHANDONG UNIV
View PDF5 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the existing traditional deep learning research does not take into account the individual differences between the subjects and the relationship between the channels of EEG signals, resulting in the inability to meet the needs of higher recognition accuracy

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Electroencephalogram emotion recognition method and system based on depth domain self-adaption
  • Electroencephalogram emotion recognition method and system based on depth domain self-adaption
  • Electroencephalogram emotion recognition method and system based on depth domain self-adaption

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0043] The present embodiment 1 provides a kind of EEG emotion recognition system based on depth field adaptation, the system includes:

[0044] An acquisition module, configured to acquire the EEG signal to be identified;

[0045] The recognition module is used to process the acquired EEG signals by using a pre-trained recognition model to obtain an EEG emotion recognition result; wherein, the pre-trained recognition model is obtained by training a training set, and the training set includes multiple The EEG signals of different subjects and the labels of the emotional categories of the EEG signals;

[0046] Wherein, when the pre-trained recognition model processes the EEG signal, it extracts the differential entropy feature of the EEG signal, extracts the topological structure between the EEG signal channels based on the differential entropy feature, and extracts the feature based on the topological structure. Local information, based on the local information, the model rel...

Embodiment 2

[0102] combine figure 1 with Figure 4 As shown, in the present embodiment 2, a kind of EEG emotion recognition method based on cross-subject graph convolutional network is provided, comprising the following steps:

[0103] Step 1. Create a dataset

[0104] The public DEAP database was used, which was collected by experiments by Koelstra and others from Queen Mary University of London, University of Twente, University of Geneva, Switzerland, and the Swiss Federal College to study the diversity of human emotional states. channel data. The database is based on the physiological signals generated by the stimuli evoked by music video materials. It records 32 subjects, watching the physiological signals of 40 minutes of music videos (each music video is 1 minute) and the subject's Valence, Arousal, Dominance of the video. , Likening's psychological scale, and also included videos of the facial expressions of the top 22 participants. The database can study physiological signals ...

Embodiment 3

[0145]Embodiment 3 of the present invention provides a non-transitory computer-readable storage medium. The non-transitory computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by a processor, the above-mentioned depth-based A domain-adaptive EEG emotion recognition method, the method comprising:

[0146] Obtain the EEG signal to be identified;

[0147] The acquired EEG signal is processed by using a pre-trained recognition model to obtain the EEG emotion recognition result; wherein, the pre-trained recognition model is obtained by training a training set, and the training set includes brains of a plurality of different subjects. Labeling of electrical signals and emotional categories of EEG signals;

[0148] Wherein, when the pre-trained recognition model processes the EEG signal, the differential entropy feature of the EEG signal is extracted, the topological structure between the EEG signal channels is extracte...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention provides an electroencephalogram emotion recognition method and system based on depth domain self-adaption, and belongs to the technical field of electroencephalogram emotion recognition. The method comprises the steps: processing an obtained electroencephalogram signal through a pre-trained recognition model to obtain an electroencephalogram emotion recognition result; extracting differential entropy features of the electroencephalogram signals, extracting a topological structure between electroencephalogram signal channels based on the differential entropy features, extracting local information of the features based on the topological structure, obtaining a mode relation between the left brain and the right brain in emotional expression based on the local information, and conducting integrating; and performing emotion category judgment by using the integrated features. According to the method, a Coll loss function and graph convolution are combined with a traditional convolutional neural network, differential entropy local features are extracted, the relation between electroencephalogram signal channels is learned by using the graph convolution, information of a left brain and information of a right brain are extracted respectively, the extracted features are integrated, the difference between individuals is reduced by using adaptive loss, and the electroencephalogram emotion recognition result is more accurate.

Description

technical field [0001] The invention relates to the technical field of EEG emotion recognition, in particular to a method and system for EEG emotion recognition based on depth field self-adaptation. Background technique [0002] To a large extent, emotion reflects an individual's cognition and attitude towards things. It plays a role in information transmission and behavior regulation in people's daily life and work. Emotion is ubiquitous and has a non-negligible impact. In the human-computer interaction system, emotion recognition based on computer technology plays a very critical role. With the rapid development of artificial intelligence, the research based on emotion has also greatly increased. Correct emotion recognition can help monitor an individual's mental and emotional health. In medicine, it can help guide and judge patients with mental illness or expression disorders. [0003] Facial expressions, voice intonation, gestures, and physiological signals can all be u...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): A61B5/16A61B5/372A61B5/00
CPCA61B5/165A61B5/372A61B5/7267
Inventor 杨立才宋鑫旺闫丹丹
Owner SHANDONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap