Multi-type muscle fatigue detection method based on feature extraction and GRU deep learning model

Abstract

The invention discloses a muscle fatigue detection method based on feature extraction and a GRU deep learning model, and the method comprises the steps: 1, collecting surface electromyogram signals of a long-term back muscle group of a subject through a surface electromyogram sensor, carrying out the segmentation, abnormal value screening, filtering and denoising of sample data, and setting a classification label based on a fatigue limit; 2, extracting a cleaned data sample sliding window as a feature sequence of a shape [s, c], importing the feature sequence into a GRU neural network for training, and setting a sample sampling weight measure in the training process to solve the problem of unbalanced sample labels; and 3, adjusting the learning rate optimization model by using the verification set, selecting an optimal model by taking the accuracy rate of the verification set as a standard, and operating the final model on the test set, so that the accuracy rate of fatigue detection of each muscle region can reach 98% or above finally. According to the method, the limitation of a traditional single muscle detection method can be overcome, comprehensive fatigue detection is carried out on main muscle groups of the human body, and the detection accuracy is improved.

Description

technical field [0001] The invention relates to the technical field of physiological signal feature detection, in particular to a multi-type muscle fatigue detection method based on feature extraction and a GRU deep learning model. Background technique [0002] The muscular system is an important part of the human body, providing power for various movements of the human body. However, muscle fatigue will occur after long-term tension or repeated work, which will affect the normal movement of the human body and even cause damage to the muscles themselves. Therefore, the accurate detection of human muscle fatigue state is the basis of muscle fatigue relief and treatment, and has important kinematics and medical significance. [0003] Surface ElectroMyoGraphy (sEMG) signal is a weak current signal generated during muscle movement, and its change is related to factors such as the number of motor units participating in the activity, activity mode, and metabolic state, and can ac...

AI Technical Summary

Problems solved by technology

[0005] The present invention aims at the lack of universal applicability caused by single fatigue detection muscle in the prior art, and proposes a multi-type muscle fatigue detection method based on feature extraction and GRU deep learning model, in order to extract more effective feature combinations and combine GRU timing Deep network for multi-muscle fatigue detection, thereby improving the general applicability and accuracy of detection

Images