Unlock instant, AI-driven research and patent intelligence for your innovation.

An adaptive decision tree fall detection method and system

A detection method and decision tree technology, applied in the direction of neural learning methods, instruments, biological models, etc., can solve the problem that it is difficult to collect fall samples, etc., and achieve the effect of high accuracy and small amount of calculation

Active Publication Date: 2021-01-26
大同公元三九八智慧养老服务有限公司
View PDF13 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Another difficulty in fall detection is the collection of fall samples, it is difficult to collect real fall samples

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
  • An adaptive decision tree fall detection method and system
  • An adaptive decision tree fall detection method and system
  • An adaptive decision tree fall detection method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0056] (1) Data acquisition

[0057] The fall detection device uses MPU6050 as an attitude sensor to collect the wearer's triaxial acceleration and triaxial angular velocity data. The data is transmitted to the local computer through the WiFi module or to the network server through the SIM card module. The fall detection device is convenient to carry, has a buckle and can be hung on the belt of trousers, and can also be attached to a pocket or clothes.

[0058] Under normal circumstances, the duration of the falling action will not exceed 0.5 seconds. At the same time, considering that the wearer will remain basically stationary if an accidental fall occurs, the data of 1 second before and 4 seconds after the falling action are collected as samples. The data sampling rate is 100Hz, that is, 100 pieces of triaxial acceleration and triaxial angular velocity data are collected per second. The action of falling itself is a relatively dangerous action, which is likely to cause in...

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 discloses an adaptive decision tree fall detection method and system, belonging to the technical field of human behavior recognition and judgment. The method includes the following steps: Step 1: Obtain and screen the triaxial acceleration and triaxial angular velocity data of human fall and non-fall movements; Step 2: Calculate the combined acceleration and combined angular acceleration and divide the training set, test set and verification set into TSFRESH Calculate the features of the library, filter and delete useless features; step 3: use random forest to select preliminary important features; step 4: build a decision tree model for training and verification, and test the results; step 5: continue to obtain new samples, repeat the above step 2 Go to step 4, update the decision tree model. The present invention can obtain accurate fall judgment results through a decision tree algorithm with a small amount of calculation; and after a certain number of samples are collected, the decision tree model can be updated to further improve the judgment accuracy of the algorithm.

Description

technical field [0001] The invention belongs to the technical field of human behavior recognition and judgment, and in particular relates to an adaptive decision tree fall detection method and system. Background technique [0002] Aging is serious, and there are more and more cases of the elderly accidentally falling, especially when the elderly are alone, they cannot call the police after falling, no one knows, and the best time for rescue will be missed. It is imperative to develop a wearable device that can be carried around, which can automatically detect whether the elderly have fallen, and automatically alarm the device. [0003] At present, there are many similar devices, but because of the high false alarm rate of the device, which is difficult for users to accept, it is basically impossible to promote it. Since wearable devices are based on embedded computing modules with limited computing power, it is difficult to use complex machine learning or deep learning and ...

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): G06N3/00G06N3/08G06K9/62G08B21/04
CPCG06N3/006G06N3/08G08B21/043G08B21/0446G08B21/0492G06F18/24323G06F18/214
Inventor 刘超孔志强孔繁星向永红赵统旭
Owner 大同公元三九八智慧养老服务有限公司