Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Detection method and system for sensing human body falling through Wi-Fi in bathroom scene

A detection method and bathroom technology, applied in sensors, diagnostic recording/measurement, medical science, etc., can solve the problems of complex model design, large computing power consumption, large amount of parameters, etc., to improve the recognition accuracy and generalization ability, the effect of reducing the size of the parameter

Pending Publication Date: 2021-10-15
ZHENGZHOU UNIV
View PDF20 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] Defects: In the current research on Wi-Fi sensing human fall detection and recognition based on cyclic neural network, most of the result model designs are too complicated, the parameter calculation is huge, the correct rate is not high, and an applicable scenario is not clearly proposed
[0010] Disadvantages: In the current research on Wi-Fi sensing human fall detection and recognition based on convolutional neural network, most of the achievement models are transformed from classic convolutional neural network or cyclic neural network, which has a huge amount of parameters and consumes a lot of training Computing power, and it is not clearly stated that it can be applied to the special scene of the bathroom

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
  • Detection method and system for sensing human body falling through Wi-Fi in bathroom scene
  • Detection method and system for sensing human body falling through Wi-Fi in bathroom scene
  • Detection method and system for sensing human body falling through Wi-Fi in bathroom scene

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0076] Such as Figure 1-3 As shown, the present embodiment provides a Wi-Fi sensing human body fall detection system (WiSFall) in a bathroom scene, including: a data processing module and a fall detection module, and the data processing module includes a sequentially connected data acquisition module, Data reconstruction module and data filtering module;

[0077] The system performs Wi-Fi sensing human fall detection method in the bathroom scene, including the following steps:

[0078] First of all, commercial Wi-Fi devices usually adopt MIMO (Multiple Input Multiple Output) technology, and the CSI data of each antenna pair includes multiple subcarrier information. Raw channel state information data H(f k ), expressed as formula (1);

[0079]

[0080] Among them, H(f k ) represents the CSI of the kth subcarrier, ||H(f k )|| and ∠H(f k ) represent the amplitude and phase, respectively. Since the phase information is susceptible to interference, this method only uses t...

Embodiment 2

[0103] The difference between this embodiment and Embodiment 1 is that this embodiment provides a specific deep neural network model.

[0104]The deep neural network model includes two ConvBN submodules for primary feature extraction and two CarConvBNMax modules for feature fusion;

[0105] The ConvBN sub-module includes three convolutional layers Conv2D and three batch normalization layers BatchNormalization. The ConvBN sub-module has designed a mechanism in which the convolutional layer Conv2D and the batch normalization layer BatchNormalization appear alternately, which can fully extract the preliminary Features and overall features; using batch normalization layer Batch Normalization can not only make the model converge quickly, but also disrupt the training data and improve the efficiency of data use;

[0106] The CarConvBNMax module includes a feature fusion layer Concatenate, a convolutional layer Conv2D, a batch normalization layer Batch Normalization, and a maximum po...

Embodiment 3

[0115] This embodiment provides a terminal, including a memory, a processor, and a computer program stored on the memory and capable of running on the processor, and the processor implements the first and second embodiments when executing the computer program. The steps of the Wi-Fi sensing human body fall detection method described in Example 2 in the bathroom scene obtain the fall detection result.

[0116] In particular, the terminal can be used as a care terminal. At present, our country has entered an aging society. Due to the limitation of medical resources, home care for the elderly has become very important, and its related research is also a hot spot at present. The terminal is used for home care, and the senseless intelligent monitoring of the elderly can be realized through the home Wi-Fi, especially the bathing situation of the elderly living alone. Detection can help in a timely and effective manner.

[0117] It should be understood that in this embodiment, the...

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 provides a detection method and system for sensing human body tumble through Wi-Fi in a bathroom scene, and the system comprises a data processing module and a tumble detection module. The data processing module is mainly used for the collection and preprocessing of sensing data, and comprises a data collection part, a data reconstruction part and a data filtering part. The method comprises the following steps: acquiring original channel state information data in a bathroom environment monitored by a single-transmitter-single-receiver Wi-Fi device, and extracting an amplitude frame; reconstructing the one-dimensional time sequence data into a two-dimensional matrix form, and filtering the one-dimensional time sequence data; enabling the tumble detection module to input the processed sensing data into a designed neural network model to extract features, and then calculating a monitoring result through a Softmax layer.

Description

technical field [0001] The invention relates to the field of behavior recognition, in particular to a Wi-Fi sensing human body fall detection method and system in a bathroom scene. Background technique [0002] In recent years, the elderly often fall due to the slippery ground when bathing, and the elderly living alone cannot seek help in time after falling, resulting in a greater degree of injury. [0003] At present, there are many human fall detection methods. From the perspective of signal collection, human fall detection can be divided into three types: vision-based fall detection, acoustic-based fall detection, and wearable sensor-based fall detection. Vision-based fall detection uses a camera to obtain images of human body movement, and uses image processing algorithms to extract image features of falls to determine whether a fall has occurred. This method has high accuracy, but this method easily exposes personal privacy and is costly. Acoustic-based fall detection ...

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 Applications(China)
IPC IPC(8): A61B5/11G08B21/04A61B5/00
CPCA61B5/1117A61B5/7264A61B5/7221A61B5/7203G08B21/043G08B21/0461
Inventor 段鹏松李婧馨李晨曹晨阳叶彪曹仰杰
Owner ZHENGZHOU UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products