一种基于注意力机制和卷积网络的肌电信号实时分解方法
By proposing a real-time decomposition method for electromyography (EMG) signals based on attention mechanisms and convolutional networks, the time delay problem of real-time decomposition of high-density surface EMG signals is solved, achieving accurate decomposition within the EMG delay range and reducing computational costs.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- DALIAN MARITIME UNIVERSITY
- Filing Date
- 2024-12-04
- Publication Date
- 2026-07-17
AI Technical Summary
Existing methods for real-time decomposition of high-density surface electromyography signals suffer from time delays, failing to meet the requirements for real-time decomposition, and require complex preprocessing operations.
A real-time decomposition method for electromyography (EMG) signals based on attention mechanisms and convolutional networks is adopted. An offline dataset is constructed through a sliding window, and the initial decomposition is performed using convolutional kernel compensation algorithm and K-means clustering algorithm. A convolutional neural network with attention mechanism is constructed for training to directly decompose high-density surface EMG signals without preprocessing.
Real-time decomposition within the electromyographic delay range was achieved, improving decomposition accuracy and reducing computational costs. The decomposition delay time was 80ms, meeting the requirements for real-time decomposition.
Smart Images
Figure CN119587049B_ABST