CN105469546B - A kind of tumbling alarm system and method - Google Patents
A kind of tumbling alarm system and method Download PDFInfo
- Publication number
- CN105469546B CN105469546B CN201610023673.8A CN201610023673A CN105469546B CN 105469546 B CN105469546 B CN 105469546B CN 201610023673 A CN201610023673 A CN 201610023673A CN 105469546 B CN105469546 B CN 105469546B
- Authority
- CN
- China
- Prior art keywords
- mtd
- mrow
- sampling
- acceleration
- msub
- 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.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
- G08B21/0407—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis
- G08B21/043—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis detecting an emergency event, e.g. a fall
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
- G08B21/0438—Sensor means for detecting
- G08B21/0446—Sensor means for detecting worn on the body to detect changes of posture, e.g. a fall, inclination, acceleration, gait
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B25/00—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems
- G08B25/01—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems characterised by the transmission medium
- G08B25/016—Personal emergency signalling and security systems
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B25/00—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems
- G08B25/01—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems characterised by the transmission medium
- G08B25/08—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems characterised by the transmission medium using communication transmission lines
Landscapes
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Gerontology & Geriatric Medicine (AREA)
- Psychiatry (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Emergency Alarm Devices (AREA)
Abstract
The present invention relates to a kind of tumbling alarm system and method, system includes single-chip microcomputer, and single-chip microcomputer internal memory contains the programming module being capable of deciding whether as the state of falling down, the data input pin connection inertial data collecting unit of single-chip microcomputer, and position receiver unit;The data output end of single-chip microcomputer is connected with audio unit and radio communication unit, and radio communication unit is connected with designated terminal, by person under test fall down state and positional information is sent to designated terminal.Method comprises the following steps:Gather person under test's vector acceleration and angular velocity vector;The vector acceleration and angular velocity vector collected is filtered;Sampled acceleration vector is synthesized, and threshold decision is carried out to resultant acceleration, whether preliminary judgement may be the state of falling down;By sampled acceleration vector and the sensitization processing of sampling angular velocity vector;Set sampling window;To the sample magnitude Fuzzy processing in sampling window to determine whether to fall down;Signal will be fallen down transmit to designated terminal and fall down alarm to realize.
Description
Technical field
The present invention relates to a kind of tumbling alarm system and method, more particularly to a kind of fallen down using fuzzy self-adaption detection
Tumbling alarm system and method.
Background technology
Former generation progressively degenerates with the increase at age, body and mental function, and the incidence height of Falls in Old People, consequence are tight
Weight, if old man produces stupor after falling down, then situation is just more critical.This also turns into aged the first injury or dead
Die reason.Fallen in population of China and the 4th is come in accidental wound is killed because of cis-position, and then occupied in the elderly of over-65s
First place, and steeply risen with the death rate of the increase tumble at age, reach peak in the elderly of more than 85 years old.So design
A state that can accurately detect old man's real-time status and in time fall down old man informs that the device of household is very necessary.
In the prior art, falling down detection technique for inertia typically has two ways, and one kind is using neutral net mould
Formula knows method for distinguishing, and such method needs more hardware resource, and the requirement to hardware process speed is higher, and for difference
User establish neutral net, process is complicated, and reproducibility is poor;Another kind is by 3-axis acceleration vector modulus, and is set
The method that threshold decision is fallen down, such method is simple to operation, real-time, but judges excessively simple, False Rate height.
The content of the invention
In order to solve problem above, the invention provides a kind of tumbling alarm system and method, the system and method are simple
It is easy to operate, it is real-time, and judge precision height.
To achieve these goals, the present invention adopts the following technical scheme that:
A kind of tumbling alarm system, including single-chip microcomputer, single-chip microcomputer internal memory contain the volume being capable of deciding whether as the state of falling down
Journey module, the data input pin connection inertial data collecting unit of the single-chip microcomputer, and position receiver unit;The single-chip microcomputer
Data output end be connected with audio unit and radio communication unit, radio communication unit is connected with designated terminal, by person under test
Fall down state and positional information is sent to designated terminal.
The inertial data collecting unit includes the acceleration transducer of collection acceleration signal, and collection angular speed letter
Number gyroscope.
Low pass filter is provided between inertial data collecting unit and single-chip microcomputer, low pass filter gathers inertial data
The inertial data filtering gathered in unit, with exclusive PCR data.
Tumbling alarm system also includes the display button unit for operating warning system operation, and the unit includes display screen,
Multiple setting buttons are provided with below display screen, alarm in emergency situations is additionally provided between display screen and setting button
Button, the both sides of alarm key are respectively arranged with confirming button and negative button.
One kind falls down alarm method, comprises the following steps:
1) acceleration and angular speed on person under test three directions are gathered, and formed vector acceleration and angular speed to
Amount;
2) vector acceleration and angular velocity vector that collect are filtered, forms sampled acceleration vector and sampling angle
Velocity vector;
3) sampled acceleration vector is synthetically formed resultant acceleration, and threshold decision is carried out to resultant acceleration, tentatively
Determine whether and possible fall down state;By resultant acceleration compared with given threshold, if resultant acceleration is not less than setting threshold
Value, then judge that person under test is normal;If resultant acceleration is less than given threshold, judge that person under test may fall down;
4) by sampled acceleration vector and the sensitization processing of sampling angular velocity vector;
5) sampling window synthesis sampling matrix is set;
6) to the sampling matrix Fuzzy processing in sampling window to determine whether to fall down;
7) signal will be fallen down transmit to designated terminal and fall down alarm to realize.
The setting specific method of threshold value comprises the following steps in the step 3):
A. the multiple time sections of division in daily;
B. the amplitude of the resultant acceleration in each time section is recorded;
C. the quantity that resultant acceleration amplitude in each time section exceedes preset value is added up;
D. the section movement intensity in each time section is determined according to accumulated quantity;
E. given threshold is determined according to following formula:
TAcc=Texp+a·(E-Eexp), wherein TAccRepresent given threshold, unit m/s2;TexpIt is respectively to pass through with parameter a
The threshold preset parameter and default magnification ratio coefficient that overtesting obtains.E is section movement intensity, EexpFor section movement intensity
Parameter preset.
The step 4) is specially:It will be associated in sampled acceleration vector and sampling angular velocity vector for fall events
Spend low Data Synthesis reprocessing or reject;For the high data enhanced processing of the degree of association in fall events.
If judged result will trigger window judgement, window judges to receive at matrixing to fall down in threshold determination
Manage the sampling matrix of unit.
Setting sampling window synthesis sampling matrix detailed process is as follows in the step 5):After threshold determination is falls down, from
The sampling time point for obtaining next sampled acceleration and sampling angle number vector starts, and sets up a window at regular intervals
Judge sampling time point tj, m are set up altogether, t1 t2 Λ tm, judge that the collection of sampling time point is quick from data in each window
The n dimension datas of processing unit are helped to change, form sampling matrix S:
Wherein SijRepresent the element in sampling matrix S.The element representation of each row in matrix is in a corresponding time
Point tjN dimension datas (the S from data sensitive processing unit that place collects1j S2j Λ Snj)T。
The step 6), to determine whether to fall down, specifically includes following to the sampling matrix Fuzzy processing in sampling window
Step:
Using following formula to the sample magnitude Fuzzy processing in sampling window:
WhereinIt is each point S in the sampling matrix S determined through overtestingijDesired value, so not having m × n not
Same desired valueNw1、Nw2And Nw3For the fuzzy value in fuzzification function F.Tw1、Tw2And Tw3For
Default parameter in fuzzification function F.Point S in sampling matrixijThe result obtained after input function F is Nij.By above-mentioned mould
M × n numerical value additional combining after gelatinization processing in the fuzzy matrix that is formed judge to fall down parameter N, will judge to fall down parameter N and
It is default to fall down parameter NwinCompare, such as N > NwinWhen, it is judged to falling down, as N < NwinWhen be judged to not falling down.
Compared with prior art, the present invention have substantive distinguishing features prominent as follows and it is notable the advantages of:
(1) operating method of present system is simple, real-time;
(2) method that the inventive method is judged by adaptive threshold and self-adapting window judges, and to sensitive vector
Plus intensive treatment and set blurring self application regulation, make the present invention judgement precision it is higher, False Rate is smaller.
Brief description of the drawings
Fig. 1 is the schematic diagram of tumbling alarm system in the present invention.
Fig. 2 is the display interface schematic diagram of tumbling alarm system in the present invention.
Fig. 3 is the flow chart that alarm method is fallen down in the present invention.
Fig. 4 is the change schematic diagram of amplitude in a time section in the present invention.
Embodiment
The specific embodiment of the present invention is specifically described below in conjunction with the accompanying drawings.
As shown in figure 1, a kind of tumbling alarm system, including single-chip microcomputer, single-chip microcomputer internal memory, which contains, to be capable of deciding whether to fall
The programming module of state, the data input pin connection inertial data collecting unit of the single-chip microcomputer, and position receiver unit;
The data output end of the single-chip microcomputer is connected with audio unit and radio communication unit, radio communication unit and designated terminal phase
Even, by person under test fall down state and positional information is sent to designated terminal.
Above-mentioned inertial data collecting unit includes the acceleration transducer that can gather acceleration signal, and can acquisition angle speed
Spend the gyroscope of signal.
Low pass filter is provided between inertial data collecting unit and single-chip microcomputer, low pass filter can adopt inertial data
The inertial data filtering gathered in collection unit, low pass filter filter the noise signal of high frequency, eliminate interference data.
In addition, as shown in figure 1, this tumbling alarm system also includes the audio list being connected with single-chip microcomputer signal output end
Member, audio unit can send alarm signal, when person under test falls down, by the alarm of sound, cause the concern of people around, so as to seek
Ask help;Also can in audio unit typing sound, arrival take medicine the time when, remind user to take medicine the quantity etc. of species.
As shown in Fig. 2 in the present invention, tumbling alarm system also includes the display button of operable warning system operation
Unit, the unit include can show date, the time, the display screen 1 for setting of taking medicine, the lower section of display screen 1 is provided with multiple settings
Button 2, default numerical value in above-mentioned single-chip microcomputer can specifically be changed by different buttons, meanwhile, in display screen 1 and setting button
It is additionally provided with alarm key 5 in emergency situations between 2, the both sides of alarm key 5 are respectively arranged with confirming button 3 and no
Determine button 4, in specific judgement, as when system, which is sent, falls down alarm, person under test falls down state to be non-, then can press negative button
4, by negating button 4, error alarm can be carried out to default value in the tumbling alarm system, after error alarm, list can be started
Program restarting setting in piece machine, improves the accuracy for falling down judgement.
As shown in figure 3, the alarm method of falling down of the present invention comprises the following steps:
1) acceleration and angular speed on three directions of person under test are gathered, and forms vector acceleration VA(X, Y, Z) with
And angular velocity vector VG(X,Y,Z);
2) vector acceleration and angular velocity vector that collect are filtered to form sampled acceleration vector VAcc(X,Y,Z)
And sampling angular velocity vector VGyr(X,Y,Z);
3) sampled acceleration vector is synthetically formed resultant acceleration ΔAcc, and threshold decision is carried out to resultant acceleration,
Whether preliminary judgement may be the state of falling down;
4) by sampled acceleration vector and the sensitization processing of sampling angular velocity vector;
5) sampling window synthesis sampling matrix is set;
6) to the sample magnitude Fuzzy processing in sampling window, parameter is fallen down in synthesis judgement, and falls down parameter with default
Compare to determine whether to fall down;
7) signal will be fallen down transmit to designated terminal and fall down alarm to realize.
For the above method, the detailed description to each step is as follows:
X, Y, Z axis coordinate is set according to the position of person under test, then the acceleration and angular velocity vector measured is VA(X,
Y, Z) and VG(X,Y,Z);
Shown in Fig. 3, the vector acceleration and angular velocity vector collected is filtered to be formed sampled acceleration vector and
Sample angular velocity vector, VAcc(X, Y, Z) and VGyr(X,Y,Z)。
With continued reference to Fig. 3, sampled acceleration vector is synthetically formed resultant acceleration, and threshold value is carried out to resultant acceleration
Judge, whether preliminary judgement may be when falling down state, specifically, by resultant acceleration compared with given threshold, add if synthesizing
Speed is not less than given threshold, then judges that person under test is normal;If resultant acceleration is less than given threshold, judge that person under test may
Fall down.
The specific method of above-mentioned middle given threshold comprises the following steps:The multiple time sections of division in daily;Record is every
Amplitude in individual time section;Add up the quantity that amplitude in each time section exceedes preset value;Determined according to accumulated quantity every
Exercise intensity in individual time section;Given threshold is determined according to following formula:TAcc=Texp+a·(E-Eexp), wherein TAccTable
Show given threshold, unit m/s2;TexpWith the parameter a threshold preset parameters respectively obtained through overtesting and default amplification ratio
Example coefficient.E is section movement intensity, EexpFor the parameter preset of section movement intensity.As shown in figure 3, section movement Strength co-mputation
Threshold value T in automatic adjusument threshold decision after unit collection acceleration information is calculatedAccSize.
Then it is further elaborated with for above-mentioned threshold decision step as follows:
The number of axle of acceleration three is according to according to following formula in extraction vector of samples
Synthesized, to resultant acceleration ΔAccThreshold decision is carried out, works as ΔAcc<TAcc(wherein TaccFor given threshold) when, it is judged as
Do not fall down, then the data are given up automatically;Work as ΔAcc>TAccWhen be judged as falling down, by the judgement of threshold value, realize
Preliminary falls down judgement.It should be noted that because of threshold value TAccThere is different values the different periods in one day, these values
Size is (such as Fig. 3) determined according to the strength information E of following acceleration and angular speeds.A specific embodiment is lifted to illustrate,
Exemplified by 12 periods are divided into one day, that is, it is divided into 6:00-8:00,8:00-10:00,10:00-12:The time zone such as 00
Section, that is, correspond to 12 threshold value TAcc1ΛTAcc12.The setting of these threshold values can enter according to the service condition of user during use
The adaptive regulation of row.
Referring specifically to Fig. 4, be divided into some time section Q the time of 24 hours one day1ΛQn, in each time section
The strength information of interior record user's acceleration is E1ΛEn, strength information EiAcquisition be in corresponding time section QiIt is interior to shaking
The signal that width exceedes certain value carries out time cumulation, the length that passage time adds up, sets the intensity moved within the time period,
Formula is:Ei=TA1+TA2+Λ+TAn。
For example, 6:00 to 8:In 00 time section, i.e. Q in Fig. 4iIt is interior, strength information Ei.Acceleration amplitude
Time between A1 and A2 is TA1, time of the acceleration amplitude more than A2 is TA2.Motion so in whole time section
Intensity Ei=TA1+TA2,.Above-mentioned threshold value T is adaptively adjusted by the exercise intensity of above-mentioned determinationAccSize (such as in Fig. 3), i.e.,
TAcc=Texp+a·(E-Eexp).Wherein TexpFor threshold preset parameter, EexpFor section movement intensity parameter preset, a is put to be default
Large scale coefficient.
Threshold decision ensures that all situations about falling down all can be by judging, and the fuzzy judgement after being tentatively is sieved
Choosing, to improve the operating efficiency of detection.Therefore, threshold determination judges for first layer, but the accuracy rate that this layer judges is relatively low, because
As long as will judge that, based on this, the follow-up judgement that the present invention is set is more accurate by this layer for the possibility somewhat fallen down,
Operating efficiency is so not only increased, also improves the degree of accuracy of judgement.
With continued reference to Fig. 3, when sampled acceleration vector and the sensitization of sampling angular velocity vector are handled, will specially adopt
For the low Data Synthesis reprocessing of the fall events degree of association or reject in sample vector acceleration and sampling angular velocity vector;It is right
The high data of the degree of association are amplified processing in fall events.
Data selection gist described herein for fall events degree of association height is:Person under test produces during falling down
Angular speed and acceleration change and normal condition under the bigger data of the difference in change opposite sex be degree of association height, conversely, then
It is low for the degree of association.For example, people is in normal state, acceleration (i.e. z-axis acceleration) the change fluctuation ratio of vertical direction compared with
Greatly, the acceleration fluctuation of vertical direction is same bigger when falling down, then this data of the acceleration of vertical direction are with falling down thing
The degree of association of part is with regard to smaller.On the contrary, people is in normal state, the acceleration change in xy faces is smaller, falls down xy under state
Acceleration change in face is bigger, can substantially distinguish, then it is assumed that the data of x-axis and y-axis are high with the degree of association of fall events.
In addition, for the low Data Synthesis reprocessing of the fall events degree of association or reject, specific embodiment explanation is lifted such as
Under:As it appears from the above, the acceleration in z-axis direction, the normal fluctuation range under normal condition is that -2g arrives 2g, falls down the ripple under state
Dynamic scope is that -3g arrives 3g, and its maximum difference is 1g under two states, that is, shows that z-axis direction is low for the fall events degree of association, such as
This can be arranged using such a mode to data, i.e., by the acceleration of z-axis be multiplied by bigger numerical (numerical value can be selected arbitrarily,
Using 8-10 to be excellent), the larger numerical value change scope that can thus reach, so as to increase distinguishing for two states
Degree.
Data after sensitization is handled are changed into n dimensions from 6 dimensions.
If judged result will trigger window judgement to fall down in threshold determination.Window judges to receive first to carry out automoment
The sampling matrix of array processing unit.Sampling matrix obtain detailed process be:It is next from obtaining after threshold determination is falls down
The sampling time of sampled acceleration and sampling angle number vector point starts, when setting up a window judgement sampling at regular intervals
Between point tj, m are set up altogether, so the sampling time point that window judges is t1 t2 Λ tm, judge the sampling time in each window
N dimension data of the point collection from data sensitive processing unit, forms sampling matrix S:
Wherein SijRepresent the element in sampling matrix S.The element representation of each row in matrix is in a corresponding time
Point tjN dimension datas (the S from data sensitive processing unit that place collects1j S2j Λ Snj)T, when m window judges sampling
Between the time interval put be different.Window judge sampling time point position determination be according to human body under the state of falling down
According to variation characteristic.Because people's changing rule of different time points during falling down is different, then corresponding selection
Window judges that sampling time point is also different, and the inertial data (acceleration and angular speed) of human body changes in 0.1 second such as after falling down
Acutely, then in this time the window of setting judge sampling time point should more crypto set, in the 0.3-0.9 seconds of falling over of human body
The inertial data excursion of human body is smaller in period, and window judges that sampling time point can be with more sparse.If window judges
Sampling time point has m altogether, the n dimension datas from data sensitive processing unit that each time point collects, that is, is formed
State the sampling matrix S of m × n dimensions.
To the sampling matrix S Fuzzy processings in sampling window.Specifically include following steps:Using following formula to sampling
Matrix S Fuzzy processings:
WhereinIt is each point S in the sampling matrix S determined through overtestingijDesired value, so not having m × n not
Same desired valueNw1、Nw2And Nw3For the fuzzy value in fuzzification function F.Tw1、Tw2And Tw3For
Default parameter in fuzzification function F, to judgeSpan.Point S in sampling matrixijInput letter
The result obtained after number F is Nij, NijValue have four kinds of possibility, respectively Nw1、Nw2、Nw3Or 0.By sampling matrix S input functions F
After obtain be blurred matrix:
M × n numerical value additional combining in the fuzzy matrix that will be formed after above-mentioned Fuzzy processing judges to fall down parameter N, N
=N11+ΛNij+Λ+Nnm(i=1,2, Λ, n j=1,2, Λ, m).
It will judge that fall down parameter falls down parameter N with defaultwinCompare, as judged, fall down parameter falls down ginseng more than default judgement
During number, it is determined as the state of falling down;As judge to fall down parameter be less than it is default fall down parameter when, be judged to not falling down state, i.e., as N >
NwinWhen, it is judged to falling down, as N < NwinWhen be judged to not falling down.
T in above-mentioned function Fw1、Tw2And Tw3For variable, its value adjusts according to the erroneous judgement information self-adapting fallen down, that is, works as system
When being mistaken for falling down posture, user presses erroneous judgement button, and system is judged as erroneous judgement state.Adaptation module will adjust Tw1、Tw2With
Tw3Size, i.e., with accumulative, the T of erroneous judgement numberw1、Tw2And Tw3Value will more conform to the exercise habit of special user, make to sentence
It is disconnected stricter.
Claims (4)
1. one kind falls down alarm method, it is characterised in that comprises the following steps:
1) acceleration and angular speed on three directions of person under test are gathered, and forms vector acceleration and angular velocity vector;
2) vector acceleration and angular velocity vector that collect are filtered, forms sampled acceleration vector and sampling angular speed
Vector;
3) sampled acceleration vector is synthetically formed resultant acceleration, and threshold decision, preliminary judgement is carried out to resultant acceleration
Whether it is possible to fall down state;By resultant acceleration compared with given threshold, if resultant acceleration is not less than given threshold,
Judge that person under test is normal;If resultant acceleration is less than given threshold, judge that person under test may fall down;
4) by sampled acceleration vector and the sensitization processing of sampling angular velocity vector, it is specially:By sampled acceleration vector with
And for the low Data Synthesis reprocessing of the fall events degree of association or reject in sampling angular velocity vector;For being closed in fall events
Lian Dugao data enhanced processing;
5) sampling window synthesis sampling matrix is set;
6) to the sampling matrix Fuzzy processing in sampling window to determine whether to fall down;
7) signal will be fallen down transmit to designated terminal and fall down alarm to realize.
2. alarm method is fallen down according to claim 1, it is characterised in that the setting specific method of threshold value in the step 3)
Comprise the following steps:
A. the multiple time sections of division in daily;
B. the amplitude of the resultant acceleration in each time section is recorded;
C. the quantity that resultant acceleration amplitude in each time section exceedes preset value is added up;
D. the section movement intensity in each time section is determined according to accumulated quantity;
E. given threshold is determined according to following formula:
TAcc=Texp+a·(E-Eexp), wherein TAccRepresent given threshold, unit m/s2;TexpIt is respectively by examination with parameter a
Test obtained threshold preset parameter and default magnification ratio coefficient, E is section movement intensity, EexpFor the pre- of section movement intensity
Setting parameter.
3. alarm method is fallen down according to claim 1, it is characterised in that if judged result is falls down in threshold determination, just
Window judgement can be triggered, window judges to receive the sampling matrix from matrixing processing unit;Setting sampling in the step 5)
Window synthesis sampling matrix detailed process is as follows:Threshold determination is after falling down, from obtaining next sampled acceleration and sampling angle
The sampling time point of number of degrees vector starts, and sets up a window at regular intervals and judges sampling time point tj, m are set up altogether, t1
t2 … tm, judge n dimension data of the sampling time point collection from data sensitive processing unit in each window, form sampling
Matrix S:
<mrow>
<mi>S</mi>
<mo>=</mo>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<msub>
<mi>S</mi>
<mn>11</mn>
</msub>
</mtd>
<mtd>
<msub>
<mi>S</mi>
<mn>12</mn>
</msub>
</mtd>
<mtd>
<mn>...</mn>
</mtd>
<mtd>
<msub>
<mi>S</mi>
<mrow>
<mn>1</mn>
<mi>m</mi>
</mrow>
</msub>
</mtd>
</mtr>
<mtr>
<mtd>
<msub>
<mi>S</mi>
<mn>21</mn>
</msub>
</mtd>
<mtd>
<msub>
<mi>S</mi>
<mn>22</mn>
</msub>
</mtd>
<mtd>
<mn>...</mn>
</mtd>
<mtd>
<msub>
<mi>S</mi>
<mrow>
<mn>2</mn>
<mi>m</mi>
</mrow>
</msub>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
<mtd>
<mrow></mrow>
</mtd>
<mtd>
<mrow></mrow>
</mtd>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
<mtd>
<mrow></mrow>
</mtd>
<mtd>
<mrow></mrow>
</mtd>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
<mtd>
<mrow></mrow>
</mtd>
<mtd>
<mrow></mrow>
</mtd>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<msub>
<mi>S</mi>
<mrow>
<mi>n</mi>
<mn>1</mn>
</mrow>
</msub>
</mtd>
<mtd>
<msub>
<mi>S</mi>
<mrow>
<mi>n</mi>
<mn>2</mn>
</mrow>
</msub>
</mtd>
<mtd>
<mn>...</mn>
</mtd>
<mtd>
<msub>
<mi>S</mi>
<mrow>
<mi>n</mi>
<mi>m</mi>
</mrow>
</msub>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
Wherein SijRepresent the element in sampling matrix S, the element representation of each row in matrix is corresponding time point tj
N dimension datas (the S from data sensitive processing unit that place collects1j S2j … Snj)T。
4. alarm method is fallen down according to claim 1, it is characterised in that the step 6) is to the sampling square in sampling window
Battle array Fuzzy processing specifically includes following steps to determine whether to fall down:
Using following formula to the sample magnitude Fuzzy processing in sampling window:
WhereinIt is each point S in the sampling matrix S determined through overtestingijDesired value, so having m × n different phases
Prestige valueNw1、Nw2And Nw3For the fuzzy value in fuzzification function F, Tw1、Tw2And Tw3For blurring
Default parameter in function F, the point S in sampling matrixijThe result obtained after input function F is Nij, at above-mentioned blurring
M × n numerical value additional combining in the fuzzy matrix formed after reason judges to fall down parameter N, judgement is fallen down into parameter N and fallen with default
The N of falling parameterwinCompare, such as N > NwinWhen, it is judged to falling down, as N < NwinWhen be judged to not falling down.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610023673.8A CN105469546B (en) | 2016-01-14 | 2016-01-14 | A kind of tumbling alarm system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610023673.8A CN105469546B (en) | 2016-01-14 | 2016-01-14 | A kind of tumbling alarm system and method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105469546A CN105469546A (en) | 2016-04-06 |
CN105469546B true CN105469546B (en) | 2018-01-02 |
Family
ID=55607195
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610023673.8A Expired - Fee Related CN105469546B (en) | 2016-01-14 | 2016-01-14 | A kind of tumbling alarm system and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105469546B (en) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105930906A (en) * | 2016-04-15 | 2016-09-07 | 上海大学 | Trip detection method based on characteristic weighting and improved Bayesian algorithm |
CN106131792B (en) * | 2016-08-26 | 2019-11-01 | 北京小米移动软件有限公司 | User movement state monitoring method and device |
CN107067667A (en) * | 2017-02-27 | 2017-08-18 | 上海量明科技发展有限公司 | Trigger the method, system and shared bicycle for operation of requiring assistance |
GB2564167B (en) * | 2017-11-17 | 2019-05-22 | August Int Ltd | Improvements in or relating to fall detectors and fall detection |
CN108320456A (en) * | 2018-01-27 | 2018-07-24 | 西安交通大学 | It is a kind of fusion multisensor the elderly fall down prediction technique and system |
CN110738822B (en) * | 2019-10-24 | 2022-09-20 | 合肥盛东信息科技有限公司 | Tumble early warning system |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN201387660Y (en) * | 2009-04-28 | 2010-01-20 | 中国科学院合肥物质科学研究院 | Automatic detecting and alarming system for human body falling-over |
CN103393412A (en) * | 2013-08-15 | 2013-11-20 | 重庆邮电大学 | Intelligent home based old person caring device |
CN203535764U (en) * | 2013-10-28 | 2014-04-09 | 中山爱科数字科技股份有限公司 | Home elder monitoring device |
US8952818B1 (en) * | 2014-03-18 | 2015-02-10 | Jack Ke Zhang | Fall detection apparatus with floor and surface elevation learning capabilites |
CN104680525A (en) * | 2015-02-12 | 2015-06-03 | 南通大学 | Automatic human body fall-down detection method based on Kinect depth image |
-
2016
- 2016-01-14 CN CN201610023673.8A patent/CN105469546B/en not_active Expired - Fee Related
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN201387660Y (en) * | 2009-04-28 | 2010-01-20 | 中国科学院合肥物质科学研究院 | Automatic detecting and alarming system for human body falling-over |
CN103393412A (en) * | 2013-08-15 | 2013-11-20 | 重庆邮电大学 | Intelligent home based old person caring device |
CN203535764U (en) * | 2013-10-28 | 2014-04-09 | 中山爱科数字科技股份有限公司 | Home elder monitoring device |
US8952818B1 (en) * | 2014-03-18 | 2015-02-10 | Jack Ke Zhang | Fall detection apparatus with floor and surface elevation learning capabilites |
CN104680525A (en) * | 2015-02-12 | 2015-06-03 | 南通大学 | Automatic human body fall-down detection method based on Kinect depth image |
Non-Patent Citations (2)
Title |
---|
基于三轴加速度传感器的老年人摔倒检测系统;许继平等;《计算机仿真》;20141231;第31卷(第12期);第434-437、450页 * |
基于无线传感技术的人体姿态检测系统;季宇等;《通信技术》;20111130;第44卷(第11期);第46-48页 * |
Also Published As
Publication number | Publication date |
---|---|
CN105469546A (en) | 2016-04-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105469546B (en) | A kind of tumbling alarm system and method | |
CN105043410B (en) | A kind of pedometer | |
CN103399632B (en) | The method and mobile terminal of a kind of gesture control | |
CN105004349A (en) | Step calculation method | |
CN107358248B (en) | Method for improving falling detection system precision | |
CN103328968B (en) | The method for calibrating air borne sensor | |
CN104567912B (en) | Method for realizing pedometer on Android mobile phone | |
CN106446765B (en) | A kind of health status evaluation system based on multidimensional physiology big data deep learning | |
CN108992040A (en) | A kind of sleep quality state monitoring method and device | |
US10521720B2 (en) | Monitoring unit for the electronic monitoring of sensor signals within the framework of the monitoring of sepsis and a corresponding process | |
CN107121463A (en) | A kind of multi-channel electronic nose detecting system and its measuring method | |
CN103778916B (en) | The method and system of monitoring ambient sound | |
CN109603142A (en) | A kind of the identification method of counting and its device of dumbbell exercise | |
CN104836905A (en) | System adjusting method and apparatus based on user state | |
CN107105092A (en) | A kind of human body tumble recognition methods based on dynamic time warping | |
CN111772639A (en) | Motion pattern recognition method and device for wearable equipment | |
CN111657906B (en) | Heart rate calculating method, heart rate calculating device, calculating equipment and heart rate detecting device | |
WO2019196106A1 (en) | Method for calculating cadence of bicycle, wearable device, and storage medium | |
CN107704712A (en) | Mechanical failure diagnostic method and system based on vector spectrum feature extraction | |
CN106643785A (en) | Method for multi-source information self-adaption step number detection based on MEMS inertial measurement unit | |
CN108981744B (en) | Step frequency real-time calculation method based on machine learning and low-pass filtering | |
EP3330674A1 (en) | Method and apparatus for realizing step counting | |
CN112432640B (en) | Method, device, electronic equipment and medium for judging use state | |
CN113551687B (en) | Step counting method, step counting device, step counting equipment, computer storage medium and chip | |
Silva et al. | Non-invasive monitoring with Ballistocardiographic sensors for sleep management |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180102 Termination date: 20210114 |