Electroencephalogram emotion recognition method based on attention mechanism and multi-feature fusion
By adopting an attention mechanism and multi-feature fusion method in EEG emotion recognition technology, the time domain, frequency domain and spatial characteristics of EEG signals are extracted, and the problem of insufficient robustness and generalization ability in the existing technology is solved, and more accurate emotion classification is achieved.
Patent Information
- Application Number
- CN202510155814.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-12
- Publication Date
- 2025-06-03
AI Technical Summary
The existing EEG emotion recognition technology has insufficient robustness and generalization ability in cross-individual experiments. Single-dimensional feature extraction cannot fully reflect emotional state, and it is difficult to capture the dynamic changes in time, frequency and spatial of EEG signals.
The EEG emotion recognition method based on attention mechanism and multi-feature fusion is adopted, and the time domain, frequency domain and airspace features are extracted and fusion through the TSS-ENet network model. Combined with the improved Attention module and the sparse multi-head attention module, the global context features of the time series are extracted, and spatial features are extracted through the dynamic graph convolution network.
It improves the emotional classification performance, enhances the importance of spectrum characteristics, captures the dynamic changes of EEG signals over time, and dynamically extracts the interaction relationships between brain regions, improving the accuracy of emotional classification results.