A Novel DE-CNN-BiLSTM Multi-Fusion Model for EEG Emotion Recognition
As a long-standing research topic in the field of brain–computer interface, emotion recognition still suffers from low recognition accuracy. In this research, we present a novel model named DE-CNN-BiLSTM deeply integrating the complexity of EEG signals, the spatial structure of brain and temporal co...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-02-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/10/4/582 |
| _version_ | 1827654186324459520 |
|---|---|
| author | Fachang Cui Ruqing Wang Weiwei Ding Yao Chen Liya Huang |
| author_facet | Fachang Cui Ruqing Wang Weiwei Ding Yao Chen Liya Huang |
| author_sort | Fachang Cui |
| collection | DOAJ |
| description | As a long-standing research topic in the field of brain–computer interface, emotion recognition still suffers from low recognition accuracy. In this research, we present a novel model named DE-CNN-BiLSTM deeply integrating the complexity of EEG signals, the spatial structure of brain and temporal contexts of emotion formation. Firstly, we extract the complexity properties of the EEG signal by calculating Differential Entropy in different time slices of different frequency bands to obtain 4D feature tensors according to brain location. Subsequently, the 4D tensors are input into the Convolutional Neural Network to learn brain structure and output time sequences; after that Bidirectional Long-Short Term Memory is used to learn past and future information of the time sequences. Compared with the existing emotion recognition models, the new model can decode the EEG signal deeply and extract key emotional features to improve accuracy. The simulation results show the algorithm achieves an average accuracy of 94% for DEAP dataset and 94.82% for SEED dataset, confirming its high accuracy and strong robustness. |
| first_indexed | 2024-03-09T21:30:56Z |
| format | Article |
| id | doaj.art-e04aae6aaae142ceb608adecd1646126 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-09T21:30:56Z |
| publishDate | 2022-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-e04aae6aaae142ceb608adecd16461262023-11-23T20:56:59ZengMDPI AGMathematics2227-73902022-02-0110458210.3390/math10040582A Novel DE-CNN-BiLSTM Multi-Fusion Model for EEG Emotion RecognitionFachang Cui0Ruqing Wang1Weiwei Ding2Yao Chen3Liya Huang4College of Electronic and Optical Engineering, Nanjing University of Posts and Telecommunications, Nanjing 210023, ChinaCollege of Electronic and Optical Engineering, Nanjing University of Posts and Telecommunications, Nanjing 210023, ChinaCollege of Electronic and Optical Engineering, Nanjing University of Posts and Telecommunications, Nanjing 210023, ChinaCollege of Electronic and Optical Engineering, Nanjing University of Posts and Telecommunications, Nanjing 210023, ChinaCollege of Electronic and Optical Engineering, Nanjing University of Posts and Telecommunications, Nanjing 210023, ChinaAs a long-standing research topic in the field of brain–computer interface, emotion recognition still suffers from low recognition accuracy. In this research, we present a novel model named DE-CNN-BiLSTM deeply integrating the complexity of EEG signals, the spatial structure of brain and temporal contexts of emotion formation. Firstly, we extract the complexity properties of the EEG signal by calculating Differential Entropy in different time slices of different frequency bands to obtain 4D feature tensors according to brain location. Subsequently, the 4D tensors are input into the Convolutional Neural Network to learn brain structure and output time sequences; after that Bidirectional Long-Short Term Memory is used to learn past and future information of the time sequences. Compared with the existing emotion recognition models, the new model can decode the EEG signal deeply and extract key emotional features to improve accuracy. The simulation results show the algorithm achieves an average accuracy of 94% for DEAP dataset and 94.82% for SEED dataset, confirming its high accuracy and strong robustness.https://www.mdpi.com/2227-7390/10/4/582emotion recognitionDEtemporal and spatial featureDE-CNN-BiLSTM |
| spellingShingle | Fachang Cui Ruqing Wang Weiwei Ding Yao Chen Liya Huang A Novel DE-CNN-BiLSTM Multi-Fusion Model for EEG Emotion Recognition Mathematics emotion recognition DE temporal and spatial feature DE-CNN-BiLSTM |
| title | A Novel DE-CNN-BiLSTM Multi-Fusion Model for EEG Emotion Recognition |
| title_full | A Novel DE-CNN-BiLSTM Multi-Fusion Model for EEG Emotion Recognition |
| title_fullStr | A Novel DE-CNN-BiLSTM Multi-Fusion Model for EEG Emotion Recognition |
| title_full_unstemmed | A Novel DE-CNN-BiLSTM Multi-Fusion Model for EEG Emotion Recognition |
| title_short | A Novel DE-CNN-BiLSTM Multi-Fusion Model for EEG Emotion Recognition |
| title_sort | novel de cnn bilstm multi fusion model for eeg emotion recognition |
| topic | emotion recognition DE temporal and spatial feature DE-CNN-BiLSTM |
| url | https://www.mdpi.com/2227-7390/10/4/582 |
| work_keys_str_mv | AT fachangcui anoveldecnnbilstmmultifusionmodelforeegemotionrecognition AT ruqingwang anoveldecnnbilstmmultifusionmodelforeegemotionrecognition AT weiweiding anoveldecnnbilstmmultifusionmodelforeegemotionrecognition AT yaochen anoveldecnnbilstmmultifusionmodelforeegemotionrecognition AT liyahuang anoveldecnnbilstmmultifusionmodelforeegemotionrecognition AT fachangcui noveldecnnbilstmmultifusionmodelforeegemotionrecognition AT ruqingwang noveldecnnbilstmmultifusionmodelforeegemotionrecognition AT weiweiding noveldecnnbilstmmultifusionmodelforeegemotionrecognition AT yaochen noveldecnnbilstmmultifusionmodelforeegemotionrecognition AT liyahuang noveldecnnbilstmmultifusionmodelforeegemotionrecognition |