A fall detection method

Abstract

The invention discloses a fall detection method, comprising the following steps: S1, performing acceleration calculation on data collected by a triaxial acceleration sensor, determining a threshold range through a BP neural network, and comparing whether the acceleration exceeds to determine whether the current movement of the human body is extremely intense ; S2, multiply the acceleration data calculated by S1 by time and then further calculate the relative energy consumption of the human body by integrating the time, and compare whether it exceeds the set threshold to judge whether the current movement of the person is abnormally intense; S3, use the three-axis gyroscope to collect Data, calculate the tilt angle and roll angle of the human body through the Kalman filter algorithm, and determine whether the posture of the human body is abnormal by judging whether the two exceed the set threshold. The multi-level fall judgment of the present invention improves the reliability of output results, reduces the possibility of misjudgment and omission in the judgment algorithm as much as possible, and enables the system to have the ability of self-adaptation according to user characteristics.

Description

technical field [0001] The invention belongs to the technical field of health monitoring and relates to a fall detection method. Background technique [0002] Most of the existing elderly monitoring systems accumulate a large amount of user health data for a long time, and then mine and analyze these data. The cycle is too long and the cost is too high. This type of monitoring system often has complex structural design, slow data processing, and low efficiency, and cannot meet the needs of elderly caregivers to obtain emergency situations such as falls in a timely manner. As people pay more attention to health, in the care and health management of the elderly or patients, how to improve the efficiency of care, save the labor cost of care, and detect the abnormalities of the elderly or patients at the first time is the current Issues to be solved urgently in nursing and health management. In the event of a fall, timely and accurate fall detection and alarm can win valuable ...

AI Technical Summary

Problems solved by technology

[0002] Existing monitoring systems for the elderly are mostly based on long-term accumulation of a large amount of user health data, and mining and analysis of these data, the cycle is too long and the cost is too high

This type of monitoring system often has complex structural design, slow data processing, and low efficiency, which cannot meet the needs of elderly caregivers for timely access to emergency situations such as falls.

Images