A fall recognition method and device, and user equipment
A recognition method and user technology, applied in neural learning methods, character and pattern recognition, applications, etc., can solve the problems of low fall detection accuracy and non-universality, and improve real-time performance, accuracy, and real-time performance Good, the effect of improving the matching degree
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0057] see figure 1 , figure 1 It is a schematic flowchart of a fall recognition method disclosed in an embodiment of the present invention. Such as figure 1 As shown, the fall recognition method may include the following steps:
[0058] 101. Obtain human body motion signals related to the falling action in real time. The device for collecting human motion signals is the motion capture system X-Sens, wherein the human motion signals include human physical motion acceleration, angular velocity, Euler angle and other signals.
[0059] 102. Input the human body motion signal related to the fall action into the strong classifier based on the BP_Adaboost algorithm model for analysis, and judge whether the user is about to fall according to the human body motion signal related to the fall action;
[0060] The human body motion signal data is collected in real time, and the collected human body motion signal data is input into the strong classifier in real time, and then the stro...
Embodiment 2
[0069] see figure 2 , figure 2 It is a schematic flowchart of another fall recognition method disclosed in the embodiment of the present invention. Such as figure 2 As shown, the fall recognition method may include the following steps:
[0070] 201. Establish a strong classifier based on the algorithm model of BP_Adaboost;
[0071] Establish a strong classifier based on the algorithm model of BP_Adaboost, and the object of the acquired human motion signal is the user. First of all, it is necessary to collect a certain amount of human body motion signal data. The collection device is the motion capture system X-Sens. First, the data is denoised, and then PCA is used to reduce the dimension of the data. Its goal is to use some kind of linear projection , to map the high-dimensional data collected from human motion signals to a low-dimensional space, and expect the variance of the data to be the largest in the projected dimension, so as to use fewer data dimensions while r...
Embodiment 3
[0095] see Figure 4 , Figure 4 It is a structural schematic diagram of a fall recognition device disclosed in an embodiment of the present invention. Such as Figure 4 As shown, the fall recognition device may include:
[0096] The acquisition module 401 is used to acquire human body motion signals related to falling movements in real time; specifically, the device for collecting human body motion signals is the motion capture system X-Sens, wherein the human body motion signals include human body physical motion acceleration, angular velocity, Euler angle Wait for the signal.
[0097] The analysis module 402 is used to input the human body motion signal related to the falling action into the strong classifier based on the BP_Adaboost algorithm model for analysis, and judge whether the user is about to fall according to the human body motion signal related to the falling action, that is, collect the human body motion signal data in real time , and input the collected hum...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More