CN206563964U - A kind of old man's monitor system based on extreme learning machine - Google Patents
A kind of old man's monitor system based on extreme learning machine Download PDFInfo
- Publication number
- CN206563964U CN206563964U CN201720093180.1U CN201720093180U CN206563964U CN 206563964 U CN206563964 U CN 206563964U CN 201720093180 U CN201720093180 U CN 201720093180U CN 206563964 U CN206563964 U CN 206563964U
- Authority
- CN
- China
- Prior art keywords
- data
- mobile phone
- module
- old man
- learning machine
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Alarm Systems (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The utility model discloses a kind of old man's monitor system based on extreme learning machine, the system includes：Blue-teeth data acquisition module, mobile phone A PP modules, server module.Blue-teeth data acquisition module includes bluetooth module and sensor, passes through for gathering the activity data with old man, and by data in Bluetooth transmission to mobile phone APP modules.Mobile phone A PP modules are used for the gps data for collecting old man, and sensing data and gps data are transferred into server module by ICP/IP protocol.Server module includes server and GPRS module, for being handled in real time data using the extreme learning machine trained, and in time can be sent to warning information on guardian's mobile phone by way of short message in prediction occurring.The utility model effectively improves the speed of old man's abnormal conditions detection with extreme learning machine while old man's abnormal conditions Detection accuracy is improved, and monitor system is had quick, high accuracy for examination.
Description
Technical field
The utility model is related to medical treatment ＆ health and machine learning techniques field, and in particular to a kind of based on extreme learning machine
Old man's monitor system.
Background technology
The development with rapid changepl. never-ending changes and improvements of science and technology, drives constantly bringing forth new ideas for old man's monitoring technology.Former, if kinsfolk
In have the weaker elder of self-care ability, then generally require long-term someone side take care of, to prevent occur some can threaten its life
Order and fallen beyond the dangerous situation of safety, such as old man.And in scientific and technological flourishing today, can only monitoring technology had necessarily
The development of degree, user can remotely understand the active situation of elder in family by some equipment, such as occur abnormal conditions, user
It can in time learn and handle as early as possible.
Although many scholars propose old man's monitoring method at present, the research of current method still suffers from problems, such as
Existing old man's monitoring scenarios mostly just with 3-axis acceleration sensor and gyroscope as abnormal conditions judge according to
According to, because the species of gathered data is single, it is supplied to system to be differentiated without more useful informations, therefore system has necessarily
Rate of false alarm.In addition, current old man's monitoring algorithm is generally divided into threshold method and the class of intelligent algorithm two.The shortcoming of threshold method is
The threshold values of artificial setting early warning is needed, this threshold values is obtained by experience, it is difficult to determine optimal a numerical value and threshold values
Height have an effect on the accuracy rate of system detectio, the performance of this kind of algorithm is typically unstable.Now is used for what old man guarded
Although intelligent algorithm can preferably solve the problem of threshold method needs artificial setting early warning threshold values, the problem of bringing is intelligence
Energy algorithm, which needs to take much time, carries out the training of algorithm, typically also needs to spend substantial amounts of meter during algorithm performs
Calculation machine resource is calculated, and is unfavorable for the real-time working of system.
Utility model content
The purpose of this utility model is to be based on extreme learning machine there is provided one kind to solve drawbacks described above of the prior art
Old man's monitor system.
The purpose of this utility model can be reached by adopting the following technical scheme that：
A kind of old man's monitor system based on extreme learning machine, old man's monitor system includes：Blue-teeth data gathers mould
Block, mobile phone A PP modules, server module,
The blue-teeth data acquisition module includes bluetooth module and sensor, for collecting user's body Activities number
According to, and the mobile phone A PP modules are transmitted data to by the bluetooth module；The mobile phone A PP modules are used to gather GPS numbers
According to receive the data from the blue-teeth data acquisition module, and by data of the blue-teeth data acquisition module and described
Gps data is sent on the server module；The server module includes server and GPRS module, is come from for receiving
The data of the mobile phone A PP modules, and data are handled using the extreme learning machine trained, discriminate whether occur exception
Situation, and warning information is sent on guardian's mobile phone by the GPRS module.
Further, the sensor includes：Be worn on human body waist 3-axis acceleration sensor and three-axis gyroscope,
The heart rate sensor that is worn on human body wrist, the first pressure sensor that human body or so sole is worn on respectively and second pressure are passed
Sensor, the communication interface of each sensor is connected with the I/O interfaces of the bluetooth module, the power interface of each sensor and the indigo plant
The power supply of tooth module provides interface and is connected.
Further, the bluetooth module uses the second generation New SmartRF development boards equipped with CC2540 Bluetooth chips,
Its expansion I/O interface is used to receive each sensing data, and its power supply, which provides interface, to be used to provide power supply to each sensor, main
Function is wanted to gather each sensing data and transmit data in mobile phone A PP modules.
Further, the mobile phone A PP modules pass through mobile phone bluetooth wireless interface and the blue-teeth data acquisition module
In bluetooth module connection, receive sensing data, in addition, the mobile phone A PP modules also pass through mobile Internet and the clothes
Device module of being engaged in connection, the server module is sent to by each sensing data and cellphone GPS data.
Further, the server module includes server and GPRS module, and wherein server passes through mobile Internet
It is connected with the mobile phone A PP modules, receives sensing data and gps data, carried out in real time with the extreme learning machine trained
Processing, GPRS module is sent to if occurring abnormal conditions by warning information.
Further, the GPRS module uses SIM800C development modules, and is connected with server by USB interface
Connect, for receiving the instruction with execute server.
Further, the warning information is sent on the mobile phone of guardian in the way of short message by GPRS module, short
The content of letter includes the unexpected content occurred and place.
The utility model has the following advantages and effect relative to prior art：
1) the utility model is used as positioning using foundation of the multiple sensors as judgement, and using the GPS navigation of mobile phone
Foundation.Improve the location information that abnormal conditions generation is additionally provided outside the accuracy rate that abnormal conditions differentiate.
2) the utility model effectively improves the training speed of algorithm using extreme learning machine as abnormal conditions distinguished number
Degree, the speed of service and the accuracy rate for detecting old man's abnormal conditions.
Brief description of the drawings
Fig. 1 is a kind of structured flowchart of old man's monitor system based on extreme learning machine disclosed in the utility model；
Fig. 2 is the structure composition figure of the blue-teeth data acquisition module disclosed in the utility model；
Fig. 3 is a kind of training flow chart of old man's monitoring method based on extreme learning machine disclosed in the utility model；
Fig. 4 is a kind of execution flow chart of old man's monitoring method based on extreme learning machine disclosed in the utility model.
Embodiment
It is new below in conjunction with this practicality to make the purpose, technical scheme and advantage of the utility model embodiment clearer
Accompanying drawing in type embodiment, the technical scheme in the utility model embodiment is clearly and completely described, it is clear that retouched
The embodiment stated is a part of embodiment of the utility model, rather than whole embodiments.Based on the implementation in the utility model
Example, the every other embodiment that those of ordinary skill in the art are obtained under the premise of creative work is not made is belonged to
The scope of the utility model protection.
Embodiment one
A kind of structured flowchart of old man's monitor system based on extreme learning machine in reference picture 1, as shown in Figure 1, this is old
People's monitor system includes blue-teeth data acquisition module, mobile phone A PP modules, server module.
Wherein, the engineering process of blue-teeth data acquisition module is as follows：A) collection is worn on each sensor with old man
Data, are sent on bluetooth module；B) sensing data is sent in mobile phone A PP modules by bluetooth module.
Wherein, the course of work of mobile phone A PP modules is as follows：A) sensing data from bluetooth module is received；B) gather
Gps data；C) sensing data and gps data are sent on server module in the lump.
Wherein, the course of work of server module is as follows：A) sensing data and GPS from mobile phone A PP modules are received
Data；B) extreme learning machine trained is called to handle data；C) warning information and dependent instruction are sent to GPRS
Module；D) warning information is sent on guardian's mobile phone.
The composition and working principle of the system modules is as follows：
(1) it is a kind of blue-teeth data acquisition module knot of old man's monitor system based on extreme learning machine shown in reference picture 2
Composition.Blue-teeth data acquisition module is by bluetooth module and some sensor groups into being worn on old man, for collecting old man
Body Activities data, and mobile phone A PP modules are transmitted data to by bluetooth module.Wherein bluetooth module is used and is furnished with
The second generation New SmartRF development boards of CC2540 Bluetooth chips, its power supply provides the power interface of interface and each sensor
It is connected, its expansion I/O interface is connected with the communication interface of each sensor, its bluetooth wireless interface and the indigo plant of mobile phone A PP modules
Tooth wave point is connected.Senior activity's data that bluetooth module is gathered by I/O interfaces each sensors, and utilize bluetooth
4.0 agreements are transmitted data in mobile phone A PP modules via bluetooth wireless interface.In addition included by blue-teeth data acquisition module
Sensor is as follows：
A) 3-axis acceleration sensor：During individual movement, different acceleration can be produced in three orthogonal directions, these add
The changing value of speed can be used to judge the change of body gesture, be to judge the foundation whether individual falls.The sensor is worn
Wear in old man's waist, data acquisition rate is 50Hz.
B) gyroscope：Present gyroscope can be accurately determined the corner of 3 orthogonal directions of moving object, pass through gyroscope
The change in human motion orientation can be obtained to judge to fall.The sensor is worn on old man's waist, and data acquisition rate is
50Hz。
C) heart rate sensor：It is red according to blood of human body, i.e., blood of human body can reflect the principle of red light absorption green glow,
Obtain heart rate data.Obtain human heart rate's change to judge to fall by gyroscope.The sensor is worn on old man's wrist, data
Acquisition rate is 25Hz.
D) pressure sensor：The change of pressure or the change for causing sensor internal resistance, therefore pressure sensor can be with
The change of accurately measure old man's step pressure, thus judges the active state of old man.The sensor have first sensor and
Second sensor, is worn on old man or so sole respectively, and data acquisition rate is 50Hz.
The communication interface of above-mentioned each sensor is connected with the I/O interfaces of the bluetooth module, solid arrow in such as Fig. 2, electricity
The power supply of source interface and bluetooth module provides interface and is connected, dotted arrow and removes the acquisition rate of heart rate sensor in such as Fig. 2 and is
Beyond 25Hz, other each sensor acquisition rates are 50Hz.
(2) mobile phone A PP modules are mounted to the APP on old man's smart mobile phone, for gathering gps data and receiving from indigo plant
The data of tooth data acquisition module, and the gps data of the data of blue-teeth data acquisition module and mobile phone is sent to server mould
On block.Its function realizes that step is as follows：
A) mobile phone A PP modules call the bluetooth wireless interface of smart mobile phone to be connected with blue-teeth data acquisition module, pass through
The agreement of bluetooth 4.0 is communicated, and receives the sensing data from blue-teeth data acquisition module.
B) mobile phone A PP modules call the GPS gain-of-function gps datas of smart mobile phone, the position for judging old man.
C) mobile phone A PP modules are connected via mobile Internet with server module, using ICP/IP protocol by sensor
Data and gps data are together sent on server module, for judging the activity situation of old man and determining old man position.
(3) server module is made up of server and GPRS module, for receiving the data from mobile phone A PP modules, and
Data are handled using the extreme learning machine trained, discriminate whether occur abnormal conditions, and warning information is passed through
GPRS module is sent on guardian's mobile phone.Its function realizes that step is as follows：
A) server in server module is connected by internet with mobile phone A PP modules, and is entered by ICP/IP protocol
Row communication, receive the sensing data related to geriatric state that is sended over from mobile phone A PP modules and with old man position phase
The gps data of pass.
B) after server is pre-processed to receiving sensing data, the extreme learning machine trained is called to data
Handled in real time, analyze the physiological status and activity situation of old man, judge whether old man occurs unexpected situation.
C) GPRS module in server module uses SIM800C development modules.The USB interface and server of GPRS module
USB interface be connected, for instruction and data transmission and GPRS module power supply supply.GPRS module is led to server
Cross AT instructions to be communicated, AT instructions are generally used for connection and communication between terminal device and PC applications.When server is sentenced
When abnormal conditions occur for disconnected old man, related AT will be instructed and warning information is sent to GPRS module by USB interface
On.
D) when GPRS module receives related AT instructions and warning information from server module, early warning will be believed
Breath is serviced to be sent in the form of short message on the mobile phone of guardian by GPRS, so as to reach the purpose of early warning.Wherein early warning is believed
The content of breath includes content and the location information that old man surprisingly occurs.
In summary, the utility model proposes a kind of old man's monitor system based on extreme learning machine and existing system phase
Than its feature mainly has：1) in addition to using 3-axis acceleration sensor and gyroscope, pressure sensor is also additionally employed
With foundation of the heart rate sensor as judgement, basis on location is used as using the GPS navigation of mobile phone.2) extreme learning machine conduct is used
Abnormal conditions detection algorithm, compared with threshold method, its advantage is without artificially setting the threshold values for being difficult to determine；Intelligently calculated with other
Method is compared, and its advantage is training speed is fast, the speed of service is fast, accuracy rate is high.
Embodiment two
Present embodiment discloses a kind of old man's monitoring method based on extreme learning machine, this method includes：Training stage and
The execution stage.The training stage includes：S1, each sensor of collection sample data；S2, sample data is pre-processed simultaneously
Construct sampling feature vectors；S3, with sampling feature vectors train extreme learning machine.The execution stage includes：S4, each biography of collection
The data of sensor；S5, data are pre-processed and characteristic vector is constructed；S6, with extreme learning machine to characteristic vector classify,
Output category result.
Wherein, sensor includes：It is worn on the 3-axis acceleration sensor and three-axis gyroscope of human body waist, is worn on people
The heart rate sensor of body wrist, the first pressure sensor and second pressure sensor for being worn on human body or so sole respectively.
Wherein, described pre-process is specially：The heart rate sensor is gathered using moving weighted average filtering algorithm
Heart rate data is pre-processed, and the acceleration gathered using the computational methods of Euclidean distance to the 3-axis acceleration sensor is passed
Sensor and the gyro data of three-axis gyroscope collection are pre-processed.
As shown in figure 3, be a kind of training flow chart of old man's monitoring method based on extreme learning machine, this method include with
Lower step：
S1, each sensor of collection sample data, sensor include the 3-axis acceleration sensor for being worn on human body waist
With three-axis gyroscope, be worn on the heart rate sensor of human body wrist, be worn on respectively human body or so sole first pressure sensing
Device and second pressure sensor；
S2, sample data is pre-processed using moving weighted average algorithm and Euclidean distance computational methods and sample is constructed
Eigen vector；
S3, with sampling feature vectors train extreme learning machine；
Step S1, the sample data for gathering each sensor, it is specific as follows：
Collected sensor include 3-axis acceleration sensor, gyroscope, heart rate sensor, first pressure sensor and
Second pressure sensor；
A) 3-axis acceleration sensor：For measuring the acceleration of wearer's body in three orthogonal directions, these surveys
Value can be used for the change for judging wearer's body posture, be to judge whether wearer occurs the foundation of the abnormal conditions such as tumble；
B) three-axis gyroscope：For measuring the angular speed of wearer's body in three orthogonal directions, these measured values are used
In the change for differentiating wearer's body motion orientation.
C) heart rate sensor：Heart rate sensor can reflect the principle of red light absorption green glow by human blood, obtain heart rate
Data.The heart rate data of measurement is used for judging whether wearer occurs abnormal conditions.
D) first pressure sensor and second pressure sensor：The pressure value of wearer's both feet is measured, its measured value is used for
Judge whether wearer occurs abnormal conditions.
Step S2, using moving weighted average algorithm and Euclidean distance computational methods sample data is carried out pre-processing and structure
Make sampling feature vectors；Detailed process is as follows：
S21, using moving weighted average algorithm sample heart rate data is pre-processed.Specific formula is as follows
Wherein, hbiFor pretreated i-th of sample heart rate data, o_hbiFor the heart rate sensor collection before pretreatment
I-th of the sample heart rate data arrived, n is total sample number, and setting o_hb0=o_hb-1=0.
S22, using Euclidean distance computational methods sample acceleration information and sample angular velocity data are pre-processed.Tool
Body formula is as follows：
Wherein acciAnd wiRespectively pretreated i-th of sample acceleration information and angular velocity data, acc_xi、
acc_yi、acc_ziIt is that i-th of sample on three orthogonal directions collecting of 3-axis acceleration sensor before pretreatment accelerates
Degrees of data, w_xi、w_yi、w_ziIt is i-th of sample angle on three orthogonal directions collecting of three-axis gyroscope before pretreatment
Speed data, n is total sample number.
The pressure data that S23, first pressure sensor and second pressure sensor are collected need not move through pretreatment and with
Other data after pretreated constitute sampling feature vectors.Sampling feature vectors form is as follows：
xi=[acci,wi,p1i,p2i,hbi]T, i=1 ..., n
Wherein xiIt is i-th of sampling feature vectors, hbi、acciAnd wiRespectively pretreated i-th of sample heart rate number
According to, sample acceleration information and sample angular velocity data, p1iAnd p2iIt is first pressure sensor and second pressure sensor respectively
I-th of the sample pressure data collected, n is total sample number.
Step S3, with sampling feature vectors train extreme learning machine.Specifically include：
S31, random generation hidden layer input weight and input biasing.Form is as follows：
ai∈R1×5, i=1 ..., L
bi∈ R, i=1 ..., L
Wherein aiFor the input weight of i-th of hidden layer node, aiIt is a row vector with 5 random real number elements；
biFor the input biasing of i-th of hidden layer node, biIt is a random real number；L is L in node in hidden layer, the utility model
=100.R represents all real numbers.
S32, hidden layer output matrix is calculated with sampling feature vectors.Formula is as follows：
Wherein, H is hidden layer output matrix, xiIt is i-th of sampling feature vectors, G (x) is the excitation letter of hidden layer node
Number, the present embodiment uses sigmoid functions, aiFor the input weight of i-th of hidden layer node, biFor i-th hidden layer node
Input biasing, n is total sample number, and L is L=100 in node in hidden layer, the present embodiment.
S33, calculating hidden layer output weight matrix.Formula is as follows：
WhereinIt is the intimate matrix of matrix H, HTIt is the transposed matrix of matrix H, T is the label matrix of sample data, and β is
Hidden layer exports weight matrix.
As shown in figure 4, be a kind of execution flow chart of old man's monitoring method based on extreme learning machine, this method include with
Lower step：
S4, each sensor of collection data；
S5, data are pre-processed and characteristic vector is constructed；
S6, with extreme learning machine to characteristic vector classify, output category result.
Step S4, each sensor of collection data
Collected sensor includes the 3-axis acceleration sensor and three axles for being worn on human body waist in actual applications
Gyroscope, the heart rate sensor for being worn on human body wrist, the first sensor for being worn on human body sole and second pressure sensor.
Step S5, data are pre-processed and characteristic vector is constructed.
Data are located in advance using the moving weighted average algorithm and Euclidean distance computational methods used in training flow
Reason, and constitutive characteristic vector.Form is as follows：
X '=[acc ', w ', p1 ', p2 ', hb ']T
Wherein x ' is characterized vector, and hb ', acc ' and w ' are respectively pretreated heart rate data, acceleration information and angle
Speed data, p1 ' and p2 ' are the pressure data that first pressure sensor and second pressure sensor are collected respectively.
Step S6, with extreme learning machine to characteristic vector classify, output category result.
S61, calculating hidden layer output, formula are as follows：
H '=[G (a1·x′+b1) … G(aL·x′+bL)]
Wherein H ' is hidden layer output matrix, and x ' is characteristic vector, and G (x) is the excitation function of hidden layer node, this practicality
New use sigmoid functions, aiFor the input weight of i-th of hidden layer node, biInput for i-th of hidden layer node is inclined
Put, L is L=100 in node in hidden layer, the utility model.
S62, calculating and output category result, formula are as follows：
Y=H ' * β
Wherein y is the classification results of extreme learning machine, and β is hidden layer output weight matrix, and H ' is hidden layer output matrix.
Above-described embodiment is the utility model preferably embodiment, but embodiment of the present utility model is not by above-mentioned
The limitation of embodiment, it is other it is any without departing from Spirit Essence of the present utility model with made under principle change, modify, replace
Generation, combination, simplification, should be equivalent substitute mode, are included within protection domain of the present utility model.
Claims (7)
1. a kind of old man's monitor system based on extreme learning machine, it is characterised in that old man's monitor system includes：Bluetooth number
According to acquisition module, mobile phone A PP modules, server module,
The blue-teeth data acquisition module includes bluetooth module and sensor, for collecting user's body Activities data,
And the mobile phone A PP modules are transmitted data to by the bluetooth module；The mobile phone A PP modules are used to gather gps data
With receive the data from the blue-teeth data acquisition module, and by the data and the GPS of the blue-teeth data acquisition module
Data are sent on the server module；The server module includes server and GPRS module, and institute is come from for receiving
The data of mobile phone A PP modules are stated, and data are handled using the extreme learning machine trained, discriminate whether occur abnormal feelings
Condition, and warning information is sent on guardian's mobile phone by the GPRS module.
2. a kind of old man's monitor system based on extreme learning machine according to claim 1, it is characterised in that the sensing
Device includes：It is worn on the 3-axis acceleration sensor and three-axis gyroscope of human body waist, is worn on the sensing heart rate of human body wrist
Device, the first pressure sensor and second pressure sensor for being worn on human body or so sole respectively, the communication interface of each sensor
It is connected with the I/O interfaces of the bluetooth module, the power supply of the power interface of each sensor and the bluetooth module provides interface phase
Even.
3. a kind of old man's monitor system based on extreme learning machine according to claim 1, it is characterised in that the bluetooth
Module uses the second generation New SmartRF development boards equipped with CC2540 Bluetooth chips, and its expansion I/O interface is used to receive each
Sensing data, its power supply, which provides interface, to be used to provide power supply to each sensor, and major function is each sensing data of collection
And transmit data in mobile phone A PP modules.
4. a kind of old man's monitor system based on extreme learning machine according to claim 1, it is characterised in that the mobile phone
APP modules are connected by the bluetooth wireless interface of mobile phone with the bluetooth module in the blue-teeth data acquisition module, receive sensing
Device data, in addition, the mobile phone A PP modules are also connected by mobile Internet with the server module, by each sensor
Data and cellphone GPS data are sent to the server module.
5. a kind of old man's monitor system based on extreme learning machine according to claim 1, it is characterised in that the service
Device is connected by mobile Internet with the mobile phone A PP modules, sensing data and gps data is received, with the pole trained
Limit learning machine is handled in real time, and warning information is sent into GPRS module if occurring abnormal conditions.
6. a kind of old man's monitor system based on extreme learning machine according to claim 1, it is characterised in that the GPRS
Module uses SIM800C development modules, and is attached with server by USB interface, for receiving and execute server
Instruction.
7. a kind of old man's monitor system based on extreme learning machine according to claim 1, it is characterised in that the early warning
Information is sent on the mobile phone of guardian in the way of short message by GPRS module, and the content of short message includes the unexpected content occurred
And place.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201720093180.1U CN206563964U (en) | 2017-01-22 | 2017-01-22 | A kind of old man's monitor system based on extreme learning machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201720093180.1U CN206563964U (en) | 2017-01-22 | 2017-01-22 | A kind of old man's monitor system based on extreme learning machine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN206563964U true CN206563964U (en) | 2017-10-17 |
Family
ID=60029244
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201720093180.1U Active CN206563964U (en) | 2017-01-22 | 2017-01-22 | A kind of old man's monitor system based on extreme learning machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN206563964U (en) |
-
2017
- 2017-01-22 CN CN201720093180.1U patent/CN206563964U/en active Active
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105310696B (en) | A kind of fall detection model building method and corresponding fall detection method and device | |
| KR102252269B1 (en) | Swimming analysis system and method | |
| CN105125221B (en) | Detecting system and method are fallen down in cloud service in real time | |
| CN104888444B (en) | Wisdom gloves, the method and system of a kind of calorie of consumption and hand positions identification | |
| CN104436596B (en) | Device and motion support method are supported in motion | |
| CN105769205A (en) | Body information detection device and fall detection system | |
| CN106650300A (en) | Old person monitoring system and method based on extreme learning machine | |
| CN109816933B (en) | Intelligent monitoring system and monitoring method for preventing old people from falling down based on composite sensor | |
| CN103606248A (en) | Automatic detection method and system for human body falling-over | |
| JP2006068300A (en) | Body condition monitoring apparatus | |
| CN104224182B (en) | Method and device for monitoring human tumbling | |
| CN106923839A (en) | Exercise assist device, exercising support method and recording medium | |
| CN104297519B (en) | Human motion gesture recognition method and mobile terminal | |
| CN106725381A (en) | A kind of intelligent body-building motion bracelet | |
| CN105534500B (en) | The equilibrium function assessment device and method of a kind of integration of physiological parameter monitoring | |
| CN105212941B (en) | A kind of human body active state recognition methods and system | |
| CN106960544A (en) | A kind of fall detection system | |
| CN110464315A (en) | It is a kind of merge multisensor the elderly fall down prediction technique and device | |
| CN107657277A (en) | A kind of human body unusual checking based on big data and decision method and system | |
| CN107943271A (en) | Exercise data detection method, apparatus and system | |
| CN108827290A (en) | A kind of human motion state inverting device and method | |
| CN206563964U (en) | A kind of old man's monitor system based on extreme learning machine | |
| CN107569219A (en) | A kind of life sign monitor system | |
| Chen et al. | Improve the Accuracy of Fall Detection Based on Artificial Intelligence Algorithm. | |
| CN106472448A (en) | A kind of blue-tooth intelligence float for fishing of automatic data collection water quality environment state |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant |