A method for recognizing and counting motion events based on a body-worn device

By collecting data through a six-axis inertial sensor and utilizing a physics-based decision model, the problems of independent algorithms for motion scenarios and adaptability to wearing positions in existing technologies have been solved. This enables unified identification and counting of various motion events, improving counting accuracy and analysis reliability.

CN122196665APending Publication Date: 2026-06-12SHENZHEN BEQI INFORMATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHENZHEN BEQI INFORMATION TECH CO LTD
Filing Date
2026-03-05
Publication Date
2026-06-12

AI Technical Summary

Technical Problem

Existing technologies require the development of independent algorithms for different sports scenarios, cannot adapt to different wearing positions, and are difficult to accurately identify various sports events such as tapping, rope skipping, and swinging from a physical perspective, resulting in insufficient counting accuracy and analysis reliability.

Method used

Motion data is collected in real time by a six-axis inertial sensor, a sliding window is divided, the event detection threshold is dynamically adjusted, multi-dimensional feature vectors are extracted, and input into a judgment model based on physical laws, including parallel judgment branches for impact, periodicity and attitude parameters, to achieve unified recognition and counting.

Benefits of technology

It achieves unified recognition and counting of motion events in multiple scenarios, improves the universality and robustness of the method, and can adapt to different wearing positions, accurately identify various motion events and output accurate counting signals or analysis results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122196665A_ABST
    Figure CN122196665A_ABST
Patent Text Reader

Abstract

The application provides a kind of motion event identification and counting method based on limb wearing device, comprising: real-time acquisition of user's motion data by six-axis inertial sensor;Motion data is divided into sliding window, threshold is dynamically adjusted based on statistical characteristics and candidate event window is extracted;Multi-dimensional feature vector including impact parameter, periodic parameter and posture parameter is extracted for candidate window;Feature vector is input into judgment model based on physical law, which contains parallel impact event judgment branch, periodic motion judgment branch and compound motion judgment branch, for identifying events such as knocking, skipping rope or hitting ball;When judged as valid motion event, output counting signal or motion analysis result.The application relates to the technical field of wearable device.
Need to check novelty before this filing date? Find Prior Art