Domain adaptation spatial feature perception neural network for cross-subject EEG emotion recognition
Emotion recognition is a critical research topic within affective computing, with potential applications across various domains. Currently, EEG-based emotion recognition, utilizing deep learning frameworks, has been effectively applied and achieved commendable performance. However, existing deep lea...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-12-01
|
| Series: | Frontiers in Human Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnhum.2024.1471634/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846119269653282816 |
|---|---|
| author | Wei Lu Wei Lu Xiaobo Zhang Xiaobo Zhang Lingnan Xia Hua Ma Tien-Ping Tan |
| author_facet | Wei Lu Wei Lu Xiaobo Zhang Xiaobo Zhang Lingnan Xia Hua Ma Tien-Ping Tan |
| author_sort | Wei Lu |
| collection | DOAJ |
| description | Emotion recognition is a critical research topic within affective computing, with potential applications across various domains. Currently, EEG-based emotion recognition, utilizing deep learning frameworks, has been effectively applied and achieved commendable performance. However, existing deep learning-based models face challenges in capturing both the spatial activity features and spatial topology features of EEG signals simultaneously. To address this challenge, a domain-adaptation spatial-feature perception-network has been proposed for cross-subject EEG emotion recognition tasks, named DSP-EmotionNet. Firstly, a spatial activity topological feature extractor module has been designed to capture spatial activity features and spatial topology features of EEG signals, named SATFEM. Then, using SATFEM as the feature extractor, DSP-EmotionNet has been designed, significantly improving the accuracy of the model in cross-subject EEG emotion recognition tasks. The proposed model surpasses state-of-the-art methods in cross-subject EEG emotion recognition tasks, achieving an average recognition accuracy of 82.5% on the SEED dataset and 65.9% on the SEED-IV dataset. |
| format | Article |
| id | doaj-art-6751c7935a3c4a15bb0c042402c81a56 |
| institution | Kabale University |
| issn | 1662-5161 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Human Neuroscience |
| spelling | doaj-art-6751c7935a3c4a15bb0c042402c81a562024-12-17T06:23:06ZengFrontiers Media S.A.Frontiers in Human Neuroscience1662-51612024-12-011810.3389/fnhum.2024.14716341471634Domain adaptation spatial feature perception neural network for cross-subject EEG emotion recognitionWei Lu0Wei Lu1Xiaobo Zhang2Xiaobo Zhang3Lingnan Xia4Hua Ma5Tien-Ping Tan6Henan High-speed Railway Operation and Maintenance Engineering Research Center, Zhengzhou Railway Vocational and Technical College, Zhengzhou, ChinaSchool of Computer Sciences, Universiti Sains Malaysia, Penang, MalaysiaSchool of Computer Sciences, Universiti Sains Malaysia, Penang, MalaysiaJiangxi Vocational College of Finance and Economics, Jiujiang, ChinaHenan High-speed Railway Operation and Maintenance Engineering Research Center, Zhengzhou Railway Vocational and Technical College, Zhengzhou, ChinaHenan High-speed Railway Operation and Maintenance Engineering Research Center, Zhengzhou Railway Vocational and Technical College, Zhengzhou, ChinaSchool of Computer Sciences, Universiti Sains Malaysia, Penang, MalaysiaEmotion recognition is a critical research topic within affective computing, with potential applications across various domains. Currently, EEG-based emotion recognition, utilizing deep learning frameworks, has been effectively applied and achieved commendable performance. However, existing deep learning-based models face challenges in capturing both the spatial activity features and spatial topology features of EEG signals simultaneously. To address this challenge, a domain-adaptation spatial-feature perception-network has been proposed for cross-subject EEG emotion recognition tasks, named DSP-EmotionNet. Firstly, a spatial activity topological feature extractor module has been designed to capture spatial activity features and spatial topology features of EEG signals, named SATFEM. Then, using SATFEM as the feature extractor, DSP-EmotionNet has been designed, significantly improving the accuracy of the model in cross-subject EEG emotion recognition tasks. The proposed model surpasses state-of-the-art methods in cross-subject EEG emotion recognition tasks, achieving an average recognition accuracy of 82.5% on the SEED dataset and 65.9% on the SEED-IV dataset.https://www.frontiersin.org/articles/10.3389/fnhum.2024.1471634/fullaffective computingelectroencephalographyemotion recognitionconvolutional neural networkgraph attention networkdomain adaptation |
| spellingShingle | Wei Lu Wei Lu Xiaobo Zhang Xiaobo Zhang Lingnan Xia Hua Ma Tien-Ping Tan Domain adaptation spatial feature perception neural network for cross-subject EEG emotion recognition Frontiers in Human Neuroscience affective computing electroencephalography emotion recognition convolutional neural network graph attention network domain adaptation |
| title | Domain adaptation spatial feature perception neural network for cross-subject EEG emotion recognition |
| title_full | Domain adaptation spatial feature perception neural network for cross-subject EEG emotion recognition |
| title_fullStr | Domain adaptation spatial feature perception neural network for cross-subject EEG emotion recognition |
| title_full_unstemmed | Domain adaptation spatial feature perception neural network for cross-subject EEG emotion recognition |
| title_short | Domain adaptation spatial feature perception neural network for cross-subject EEG emotion recognition |
| title_sort | domain adaptation spatial feature perception neural network for cross subject eeg emotion recognition |
| topic | affective computing electroencephalography emotion recognition convolutional neural network graph attention network domain adaptation |
| url | https://www.frontiersin.org/articles/10.3389/fnhum.2024.1471634/full |
| work_keys_str_mv | AT weilu domainadaptationspatialfeatureperceptionneuralnetworkforcrosssubjecteegemotionrecognition AT weilu domainadaptationspatialfeatureperceptionneuralnetworkforcrosssubjecteegemotionrecognition AT xiaobozhang domainadaptationspatialfeatureperceptionneuralnetworkforcrosssubjecteegemotionrecognition AT xiaobozhang domainadaptationspatialfeatureperceptionneuralnetworkforcrosssubjecteegemotionrecognition AT lingnanxia domainadaptationspatialfeatureperceptionneuralnetworkforcrosssubjecteegemotionrecognition AT huama domainadaptationspatialfeatureperceptionneuralnetworkforcrosssubjecteegemotionrecognition AT tienpingtan domainadaptationspatialfeatureperceptionneuralnetworkforcrosssubjecteegemotionrecognition |