Method and system for identifying exercise intensity

A motion intensity, recognition system technology, applied in neural learning methods, character and pattern recognition, biological neural network models, etc., can solve the problems of inability to obtain robust deep learning features, poor generalization, rough indicators, etc. The robustness of the model and the robustness of the model, rich features, and the effect of reducing information loss

Active Publication Date: 2021-07-27
INST OF COMPUTING TECH CHINESE ACAD OF SCI +1
View PDF7 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] There are often problems such as rough indicators and poor generalization in detecting exercise intensity based on heart rate and stride frequency;
[0008] When faced with motion intensity det

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
  • Method and system for identifying exercise intensity
  • Method and system for identifying exercise intensity
  • Method and system for identifying exercise intensity

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below through specific embodiments in conjunction with the accompanying drawings. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

[0031] As mentioned in the background technology section, traditional machine learning methods can only train models based on existing data when facing motion intensity detection, and cannot obtain robust deep learning features and cannot meet the requirements of strong generalization capabilities. Aiming at this problem, the present invention constructs a motion intensity recognition system, which includes sequentially connected parallel convolution processing flow, bilinear pooling module and multi-task learning module, and the parallel convolution processing flow includes two paralle...

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 embodiment of the invention provides a method for training an exercise intensity recognition system. The method comprises the following steps of: A1, constructing an exercise intensity recognition system which comprises a data preprocessing module, a parallel convolution processing stream, a bilinear pooling module and a multi-task learning module which are connected in sequence, wherein the parallel convolution processing stream comprises two parallel convolution processing streams; A2, preprocessing motion data marked with labels through the data preprocessing module, and obtaining a training set composed of motion information and the corresponding labels, wherein the motion data are acquired by an inertial sensor, and the labels marked for the motion data comprise multiple different motion intensity labels and multiple different motion behavior labels; A3, training the exercise intensity recognition system by using the training set; according to the method, features extracted from a convolutional layer are optimized from the channel dimension and the space dimension, important features are concerned, unnecessary features are inhibited, and the recognition precision of the exercise intensity can be improved.

Description

technical field [0001] The present invention relates to the fields of pervasive computing and exercise health monitoring, specifically relates to the field of exercise intensity identification, and more specifically relates to a method and system for identifying exercise intensity. Background technique [0002] The "Physical Activity Guidelines for Chinese Adults" recommends that adults should spend more than 150 minutes of moderate-to-high-intensity physical activity per week to help their health. At the same time, children's exercise health includes many aspects such as exercise time, exercise intensity, exercise pattern and energy consumption level. As an important area of ​​research in children's physical health, studies have shown that increasing physical activity time in children with high-level exercise intensity has positive effects to varying degrees on the improvement of children's motor skills, obesity prevention interventions, psychosocial health, and cardiovascu...

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
IPC IPC(8): G06K9/00G06K9/46G06K9/62G16H50/30G06N3/04G06N3/08
CPCG16H50/30G06N3/084G06V40/20G06V10/40G06N3/045G06F18/253G06F18/24G06F18/214
Inventor 陈益强高晨龙张宇欣蒋鑫龙谷洋
Owner INST OF COMPUTING TECH CHINESE ACAD OF SCI
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
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap