Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

A kind of fitness assistant method based on myoelectric signal

An EMG signal and muscle technology, applied in the field of pattern recognition and artificial intelligence, can solve problems such as inability to directly reflect the specific state of muscles, prone to misjudgment, and large amount of posture information

Active Publication Date: 2022-04-12
DALIAN UNIV OF TECH
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Current fitness assistance methods are mainly based on body posture, in Depari A, Ferrari P, Flammini A, Rinaldi S. Lightweight Machine Learning-Based Approach for Supervision of Fitness Workout[C] / / IEEE Conference Sensors Applications Symposium.Institute of Electrical and Electronics Engineers Inc , 2019. In this document, fitness assistance is provided through the posture of the x, y, and z axes, which can assist the fitness within a certain error tolerance range, but due to the large amount of posture information and the large amount of disturbed data, errors are prone to occur. Judgment, and the specific state of the muscles cannot be directly reflected with the help of posture information, and the reliability of the assistance is low

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A kind of fitness assistant method based on myoelectric signal
  • A kind of fitness assistant method based on myoelectric signal
  • A kind of fitness assistant method based on myoelectric signal

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0068] Embodiment 1, using the BP neural network to classify the classification results of the kg sitting single-arm dumbbell curl, 10kg sitting single-arm dumbbell curl, and 30kg arm strength machine overhand grip, as shown in Figure 9 shown. The three actions are labeled 1, 2, and 3, respectively. Do 20 sets of each of the three actions and input them into the trained BP neural network classification model to get the output value. Figure 9 a) is the ideal classification result, Figure 9 b) Classify the results for the model, from Figure 9 It can be seen that the classification effect of the label 1 action is the worst, but it can be completely separated from the other two actions.

[0069] Embodiment 2, using the arm strength machine to hold the simulated standard fitness action, such as figure 1 As shown in a), use the arm strength machine to simulate non-standard fitness movements, such as figure 1 b) as shown. Do 10 groups of actions for each of the two kinds of...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to a body-building assistance method based on an electromyography signal, which belongs to the technical field of pattern recognition and artificial intelligence, and relates to a body-building assistance method based on an electromyography signal. In this method, the electromyographic signals of the human body surface are collected through the electrode sheet, and the electromyographic signals of the human body are processed through various acquisition circuits, and the processed electromyographic signals are transmitted to the host computer, and the characteristic values ​​of the signals are extracted. The SVM support vector machine through machine learning is used as a classifier model to determine whether the action is standard. The linear kernel and the Gaussian kernel with different λ parameters are used to train the SVM model respectively, and the model with the best training effect is selected to determine the standard fitness action. A three-layer BP neural network is used as a classifier model to classify actions, and the extracted feature values ​​are used as input neurons of the neural network. The RELU function is used as the activation function to realize the non-linear mapping of the input information. After training, the classifier model is obtained to realize the action recognition, which has a high accuracy rate.

Description

technical field [0001] The invention belongs to the technical field of pattern recognition and artificial intelligence, and relates to a body-building assistance method based on electromyographic signals. Background technique [0002] Fitness can help prevent many diseases. In recent years, more and more people have realized the importance of fitness. The fitness industry has developed rapidly and penetrated into people of all ages. However, non-standard fitness movements will make the fitness effect poor, and in severe cases, it will cause muscle strain. Additionally, further research has shown that users are more motivated during training if the workout is supervised. Therefore, it is of great significance to study the fitness aid method. [0003] Current fitness assistance methods are mainly based on body posture, in Depari A, Ferrari P, Flammini A, Rinaldi S. Lightweight Machine Learning-Based Approach for Supervision of Fitness Workout[C] / / IEEE Conference Sensors Appl...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G16H20/30G06N20/10G06N3/04G06N3/08A61B5/397A61B5/00
CPCG16H20/30G06N20/10G06N3/084A61B5/7264A61B5/389G06N3/044
Inventor 张元良程绍珲蒋攀孙源贾海生杨贺宫迎娇
Owner DALIAN UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products