CN105342626B - Wearable and the fall detection method applied to Wearable - Google Patents
Wearable and the fall detection method applied to Wearable Download PDFInfo
- Publication number
- CN105342626B CN105342626B CN201510898194.6A CN201510898194A CN105342626B CN 105342626 B CN105342626 B CN 105342626B CN 201510898194 A CN201510898194 A CN 201510898194A CN 105342626 B CN105342626 B CN 105342626B
- Authority
- CN
- China
- Prior art keywords
- axis
- waveform
- amplitude
- extreme point
- value
- 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
Abstract
A kind of fall detection method an embodiment of the present invention provides Wearable and applied to Wearable, it senses the real time acceleration of three axial directions of Wearable by acceleration sensor, and chooses a wherein axis as differentiation axis according to three axis real time acceleration values.In addition, the extreme point waveform for meeting tumble condition is calculated on the differentiation axis of selection, then tumble event is detected by the variance of the variance of multiple data after in calculating before the extreme point waveform and the data point in latter section of plateau region of the waveform so that the detection of tumble event is more accurate and can efficiently avoid misidentifying.
Description
Technical field
The present invention relates to wearable device field, in particular to a kind of Wearable and applied to wearable
The fall detection method of equipment.
Background technology
At present, the Wearables such as intelligent glasses, Intelligent glove, Intelligent bracelet, smartwatch, intelligent dress ornament are widely
Approved by masses and obtained universal use.Wearable is applied to the elderly group to the hazard events such as fall or fall down
Be monitored and early warning be Wearable a kind of typical case.The existing fall detection method applied to Wearable
Including the fall detection method based on graph image, the fall detection method based on acoustics, the tumble based on wearable sensor
Detection method etc..However, current fall detection method is lacked there are accuracy of detection is low, misclassification rate and rate of failing to report are higher mostly
It falls into.Therefore, how accurately to falling or fall events are detected and are notified to, be that Wearable field faces at present one
Big subject.
Invention content
In view of this, the embodiment of the present invention is designed to provide a kind of Wearable and applied to Wearable
Fall detection method, to improve, misclassification rate low to accuracy of detection existing for fall detection and rate of failing to report are higher in the prior art
Problem.
A kind of Wearable provided in an embodiment of the present invention, including acceleration transducer and fall detection system.It is described
Fall detection system includes：
Acceleration acquisition module, for obtaining real-time three axis of the Wearable along three axial directions of a three-dimensional system of coordinate
Acceleration value；
Differentiate that axis chooses module, for being selected according to the 3-axis acceleration value of the multiple data points included in a time slip-window
It one of takes in three axis of three-dimensional system of coordinate as differentiating axis；
Tumble waveshape module, for calculating the data point extreme value in the corresponding time slip-window of the differentiation axis
Continuous three extreme points are formed an extreme point waveform by point, and judge whether fall in the time slip-window comprising satisfaction
The extreme point waveform of condition；Wherein, if the left side amplitude H of the extreme point waveform1Amplitude TH default more than first1, the extreme value
The right amplitude H of point waveform2Amplitude TH default more than second2, the left side amplitude H1With the right amplitude H2Average value be more than the
Three default amplitude TH3And the time interval in continuous three extreme points between first extreme point and the last one extreme point
T13Less than the first preset time threshold TH4, then judge waveform of the extreme point waveform to meet tumble condition；
Variance computing module, for when the time slip-window includes the extreme point waveform for meeting tumble condition, calculating
The waveform variance of multiple data points included after in before the extreme point waveform for meeting tumble condition, then judges the waveform side
Whether difference is more than one first default variance threshold values TH5；And
Plateau region computing module, for being more than the described first default variance threshold values TH in the waveform variance5When, it calculates
Meet the maximum value a of the data point of preset quantity after the extreme point waveform of the tumble conditionmaxWith minimum value amin, it is described pre-
If first data point in the data point of the variance var (a) of the data point of quantity and the preset quantity and the last one
Time interval T between data point1, and the maximum value a being calculated according to thismax, minimum value amin, variance var (a) is with timely
Between be spaced T1Judge whether to detect tumble event.
Preferably, in the Wearable that the embodiment provides, the fall detection system further includes tumble alarm mould
Block, for when detecting tumble event, sending out warning message to preset electronic device and alarming, the warning message packet
Include the current location information of Wearable.
Preferably, in the Wearable that the embodiment provides, the differentiation axis is chosen module and is selected in the following manner
Take the differentiation axis:
Maximum value ax is calculated in the time slip-window with multiple data points exported from x-axismaxAnd minimum value axmin,
Then x-axis amplitude is calculated, which is the maximum value axmaxWith minimum value axminDifference；
Maximum value ay is calculated in the time slip-window with multiple data points exported from y-axismaxAnd minimum value aymin,
Then y-axis amplitude is calculated, which is the maximum value aymaxWith minimum value ayminDifference；
Maximum value az is calculated in the time slip-window with multiple data points exported from z-axismaxAnd minimum value
azmin, z-axis amplitude is then calculated, which is the maximum value azmaxWith minimum value azminDifference；
Maximum amplitude is found out from the x-axis amplitude, y-axis amplitude and z-axis amplitude, the amplitude of the maximum is corresponded to
Axis determine and the differentiation axis；Wherein：
The x-axis, y-axis, z-axis represent three axis of the three-dimensional system of coordinate respectively.
Preferably, in the Wearable that the embodiment provides, before the extreme point waveform in after multiple numbers for including
All data points and the pole that multiple data points, the extreme point waveform include before strong point includes the extreme point waveform
Multiple data points after value point waveform；The calculation formula of the waveform variance is as follows：
Wherein, the average value of multiple data points included after during the x is represented before the extreme point waveform, described in n is represented
The quantity of multiple data points included after in before extreme point waveform, x1、x2…xnRepresent the corresponding acceleration value of each data point, s2
Represent the waveform variance.
Preferably, in the Wearable that the embodiment provides, plateau region computing module is in the maximum value amaxWith
Minimum value aminDifference be less than a preset difference value TH6, variance var (a) the variance threshold values TH default less than one second7And when described
Between be spaced T1Less than the second preset time threshold TH8When, judgement has detected that tumble event.
The fall detection method applied to above-mentioned Wearable that another embodiment of the present invention provides, including：
Acceleration obtaining step obtains the Wearable and accelerates along real-time three axis of three axial directions of a three-dimensional system of coordinate
Angle value；
Differentiate axis selecting step, three are chosen according to the 3-axis acceleration value of the multiple data points included in a time slip-window
As differentiation axis one of in three axis of dimension coordinate system；
Tumble waveshape step calculates the data point extreme point differentiated in the corresponding time slip-window of axis,
Continuous three extreme points are formed into an extreme point waveform, and judges whether to include in the time slip-window and meets tumble condition
Extreme point waveform；Wherein, if the left side amplitude H of the extreme point waveform1Amplitude TH default more than first1, the extreme point wave
The right amplitude H of shape2Amplitude TH default more than second2, the left side amplitude H1With the right amplitude H2Average value be more than third it is pre-
If amplitude TH3And the time interval T in continuous three extreme points between first extreme point and the last one extreme point13It is small
In the first preset time threshold TH4, then judge waveform of the extreme point waveform to meet tumble condition；
Variance calculates step, when the time slip-window includes the extreme point waveform for meeting tumble condition, described in calculating
The waveform variance of multiple data points included after in meeting before the extreme point waveform of tumble condition, then judges that the waveform variance is
It is no to be more than one first default variance threshold values TH5；And
Plateau region calculates step, is more than the described first default variance threshold values TH in the waveform variance5When, it calculates and meets
The maximum value a of the data point of preset quantity after the extreme point waveform of the tumble conditionmaxWith minimum value amin, the present count
First data point and the last one data in the data point of the variance var (a) of the data point of amount and the preset quantity
Time interval T between point1；According to the maximum value a being calculatedmax, minimum value amin, variance var (a) and between the time
Every T1Judge whether to detect tumble event.
Preferably, the fall detection method that another embodiment provides further includes：Tumble alarming step, falls detecting
The when of falling event, sends out warning message to preset electronic device and alarms, and the warning message includes working as Wearable
Front position information.
Preferably, in the fall detection method that another embodiment provides, the differentiation axis selecting step includes:
A maximum value ax is calculated in the time slip-window with multiple data points exported from x-axismaxAnd minimum value
axmin, x-axis amplitude is then calculated, which is the maximum value axmaxWith minimum value axminDifference；
Maximum value ay is calculated in the time slip-window with multiple data points exported from y-axismaxAnd minimum value aymin,
Then y-axis amplitude is calculated, which is the maximum value aymaxWith minimum value ayminDifference；
A maximum value az is calculated in the time slip-window with multiple data points exported from z-axismaxAnd minimum value
azmin, z-axis amplitude is then calculated, which is the maximum value azmaxWith minimum value azminDifference；
Maximum amplitude is found out from the x-axis amplitude, y-axis amplitude and z-axis amplitude, the amplitude of the maximum is corresponded to
Axis determine and the differentiation axis；Wherein：
The x-axis, y-axis, z-axis represent three axial directions of the three-dimensional system of coordinate respectively.
Preferably, in the fall detection method that another embodiment provides, before the extreme point waveform in after include
Multiple data points, the extreme point waveform include before multiple data points include the extreme point waveform all data points and
Multiple data points after the extreme point waveform；The calculation formula of the waveform variance is as follows：
Wherein, the average value of multiple data points included after during the x is represented before the extreme point waveform, described in n is represented
The quantity of multiple data points included after in before extreme point waveform, x1、x2…xnRepresent the corresponding acceleration value of each data point, s2
Represent the waveform variance.
In the fall detection method that another embodiment provides, if the maximum value amaxWith minimum value aminDifference be less than
One preset difference value TH6, variance var (a) the variance threshold values TH default less than one second7And the time interval T1Less than second
Preset time threshold TH8When, then judge to have detected that the tumble event.
Compared with prior art, Wearable of the invention and the fall detection method point applied to the Wearable
The other differentiation that three phases are passed through to the real-time 3-axis acceleration data that acceleration transducer senses.The differentiation of the three phases
Differentiation and extreme point including variance after in whether meeting to extreme point waveform before the differentiation of tumble condition, extreme point waveform
The data of stagnant zone (plateau region) are differentiated after waveform, so that the detection of tumble event is more accurate and effective
Ground avoids misrecognition.
For the above objects, features and advantages of the present invention is enable to be clearer and more comprehensible, preferred embodiment cited below particularly, and coordinate
Appended attached drawing, is described in detail below.
Description of the drawings
It in order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not construed as pair
The restriction of range, for those of ordinary skill in the art, without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 is the functional module framework schematic diagram of a kind of Wearable that present pre-ferred embodiments provide.
Fig. 2 is schematic diagram of the acceleration transducer shown in FIG. 1 in three axial directions of a three-dimensional system of coordinate.
Fig. 3 is the waveform of 3-axis acceleration sensed in acceleration transducer three-dimensional system of coordinate shown in Fig. 2
Schematic diagram.
Fig. 4 is the fall detection method applied to Wearable shown in FIG. 1 that present pre-ferred embodiments provide
Flow chart.
Main element symbol description
Specific embodiment
Below in conjunction with attached drawing in the embodiment of the present invention, the technical solution in the embodiment of the present invention is carried out clear, complete
Ground describes, it is clear that described embodiment is only part of the embodiment of the present invention, instead of all the embodiments.Usually exist
The component of the embodiment of the present invention described and illustrated in attached drawing can be configured to arrange and design with a variety of different herein.Cause
This, the detailed description of the embodiment of the present invention to providing in the accompanying drawings is not intended to limit claimed invention below
Range, but it is merely representative of the selected embodiment of the present invention.Based on the embodiment of the present invention, those skilled in the art are not doing
Go out all other embodiments obtained under the premise of creative work, shall fall within the protection scope of the present invention.
It should be noted that：Similar label and letter represents similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, does not then need to that it is further defined and explained in subsequent attached drawing.Meanwhile the present invention's
In description, term " first ", " second " etc. are only used for distinguishing description, and it is not intended that instruction or hint relative importance.
Embodiment
As shown in Figure 1, it is a kind of function module signal of preferably Wearable 100 provided in this embodiment of the invention
Figure.In the present embodiment, the Wearable 100 include fall detection system 10, acceleration transducer 20, storage unit 30 with
And processing unit 40.In the present embodiment, the Wearable 100 may be, but not limited to, intelligent glasses, Intelligent glove, intelligence
The wearable intelligent electronic devices with user such as energy bracelet, smartwatch, intelligent dress ornament.Preferably, it is described wearable to set
Refer to for 100 wearable in wrist belt-type wearable devices such as the Intelligent bracelet of user's wrist, smartwatch, Intelligent gloves.
The acceleration transducer 20 is sat for measuring the Wearable 100 in three-dimensional system of coordinate along the three-dimensional
The acceleration of three axial directions of mark system.Preferably, in the present embodiment, the acceleration transducer 20 is 3-axis acceleration sensor.
Such as shown in Fig. 2, it is assumed that three axial directions of the three-dimensional system of coordinate are respectively orthogonal X-axis, Y-axis and Z axis.It is described to add
Velocity sensor 20 can then sense the Wearable 100 respectively along the acceleration value of the X-axis, Y-axis and Z-direction.
In the present embodiment, the acceleration that acceleration transducer 20 senses in the X-direction is represented using ax, ay is used to represent and is accelerated
The acceleration that degree sensor 20 is sensed in the Y direction, acceleration transducer 20 is represented in the Z-direction sense using az
The acceleration measured.
The fall detection system 10 is used in the X-axis, Y-axis and Z axis be distinguished according to the acceleration transducer 20
The tumble or fall events that the user of the Wearable 100 may occur for the acceleration value sensed are detected, with
Conducive to the first time fallen in user by the tumble event notification to corresponding personnel, so as to avoid or reduce to tumble user's
Injury.Specifically, the fall detection system 10 includes acceleration acquisition module 101, differentiates that axis chooses module 102, tumble wave
Shape computing module 103, variance computing module 104, plateau region computing module 105 and tumble alarm module 106.This is preferably real
It applies in example, each function module included in the fall detection system 10 can be installed in the form of software or firmware (firmware)
It in the storage unit 30 or is solidificated in the operating system (OS) of the Wearable 100, by the processing unit 40
10 each function module of fall detection system is controlled to perform corresponding function.
In the present embodiment, the acceleration transducer 20, storage unit 30 and processing unit 40 between each other can be direct
Or it is electrically connected transmission and interaction to carry out data indirectly.The storage unit 30 may be, but not limited to, arbitrary access
Memory (Random Access Memory, RAM), read-only memory (Read Only Memory, ROM) may be programmed read-only
Memory (Programmable Read-Only Memory, PROM), erasable read-only memory (Erasable
Programmable Read-Only Memory, EPROM), electricallyerasable ROM (EEROM) (Electric Erasable
Programmable Read-Only Memory, EEPROM) etc..The processing unit 40 can be the Wearable 100
Central processing unit (Central Processing Unit, CPU) or other any processing lists for having data-handling capacity
Member, for example, Digital Signal Processing (Digital Signal Process, DSP) chip, integrated circuit (Integrated
Circuit) chip, microcontroller, field programmable gate array (Field Programmable Gate Array, FPGA) chip
And other programmable logic device, discrete gate or transistor logic, discrete hardware components etc..
Each function module included below to the fall detection system 10 is described in detail.
The acceleration acquisition module 101 is used to obtain the Wearable 100 from the acceleration transducer 20
Real-time 3-axis acceleration value, the 3-axis acceleration value respectively by ax, ay, az identify.
In the present embodiment, the 3-axis acceleration data that the acceleration transducer 20 exports are continuous Wave data.Example
As shown in figure 3, wherein w1, w2, w3 represent oscillogram be respectively the acceleration transducer 20 sense it is described wearable
Equipment 100 is respectively along the acceleration wave graphic data of three axis of x, y, z.Sliding-model control is carried out to the Wave data (such as to pass through
Fast Fourier variation carries out discretization to Wave data) it can obtain continuous discrete acceleration information.
The differentiation axis chooses module 102 and is used to be chosen in three axis of x, y, z according to the 3-axis acceleration value of the acquisition
One of them is as differentiation axis.Specifically, the theoretical foundation for choosing differentiation axis is to choose the acceleration information of three axis of x, y, z output
The most violent wherein axis of variation is as differentiation axis.It chooses after differentiating axis, subsequently then according to the corresponding acceleration of differentiation axis
Value is detected or judges to whether user occurs to fall.It chooses and differentiates that the method for axis is specific as follows.
First, maximum value ax is calculated from the time slip-window with multiple data points that x-axis exportsmaxAnd minimum value
axmin, x-axis amplitude is then calculated, which is the maximum value axmaxWith minimum value axminDifference.The calculating of X-axis amplitude
Formula is：Hx=axmax-axmin。
Secondly, maximum value ay is calculated from the time slip-window with multiple data points that y-axis exportsmaxAnd minimum value
aymin, y-axis amplitude is then calculated, which is the maximum value aymaxWith minimum value ayminDifference.The calculating of y-axis amplitude
Formula is：Hy=aymax-aymin。
Then, maximum value az is calculated from the time slip-window with multiple data points that z-axis exportsmaxIt is and minimum
Value azmin, z-axis amplitude is then calculated, which is the maximum value azmaxWith minimum value azminDifference.The meter of z-axis amplitude
Calculating formula is：Hz=azmax-azmin。
Finally, maximum amplitude arg is found out from the x-axis amplitude, y-axis amplitude and z-axis amplitudemax(hx, hy, hz),
The corresponding axis of amplitude of the maximum is determined and the differentiation axis.
In the present embodiment, the quantity of the data point included in the time slip-window is greater than or equal at 60 points.Preferably, institute
It states and includes 60 data points in time slip-window.In other embodiments, the quantity for the data point that the time slip-window includes
It is not limited thereto, for example, its number of data points included might be less that at 60 points.
The tumble waveshape module 103 is used in the corresponding time slip-window of the differentiation axis calculate to meet
The waveform of tumble condition.Specific computational methods are as follows.
First, extreme point is calculated in the corresponding time slip-window of the differentiation axis, then by continuous three pole
Value point composition extreme point waveform simultaneously calculates the left side amplitude of the extreme point waveform and the right amplitude, then according to the left side amplitude and
Whether amplitude features extreme point waveform in the right is the waveform for meeting tumble condition.Specifically, the extreme point includes maximum
(wave crest) and the corresponding data point of minimum (trough).The formula for finding extreme point is as follows：
Maximum：ai>ai-1&&ai>ai+1(1≤i≤59)
Minimum：ai<ai-1&&ai<ai+1(1≤i≤59)
In addition, the left side amplitude of the extreme point waveform and the calculation formula of the right amplitude are as follows：
H1=| A2-A1|, H2=| A3-A2|。
Wherein, A1,A2,A3The corresponding acceleration value of respectively continuous three extreme points, H1And H2The left side is represented respectively
Not care about one's appearance value and the right amplitude.
Finally, if meeting the following conditions, judge waveform of the extreme point waveform to meet tumble condition.
H1>TH1, H2>TH2,T13<TH4.Wherein, the T13It represents in continuous three extreme points
Time interval between first extreme point and the last one extreme point, i.e. A1With A3Time interval between point.
Specifically, above-mentioned condition is interpreted as：If the left side amplitude H1Amplitude TH default more than first1, described the right
Amplitude H2Amplitude TH default more than second2, the left side amplitude H1With the right amplitude H2Average value preset amplitude TH more than third3
And the time interval T in continuous three extreme points between first extreme point and the last one extreme point13It is pre- less than first
If time threshold TH4, then judge waveform of the extreme point waveform to meet tumble condition.
In the present embodiment, the first, second, third default amplitude TH1、TH2And TH3Add naturally more than or equal to 2
Velocity amplitude and less than or equal to 3 natural acceleration values.Preferably, the TH1、TH2And TH3Value for 2.4g, wherein, g
Represent a natural acceleration.In the present embodiment, the TH4Value range for 200ms to 300ms, preferably value is
250ms。
It should be noted that it is with the theoretical foundation that aforementioned four condition determines to meet tumble waveform：In the short period
More acute variation has occurred in the acceleration value that the differentiation axis is sensed in the interior deceleration sensor 20.If in addition,
The waveform for meeting tumble condition is not detected, then chooses module 102 back to differentiation axis and the time slip-window is moved one backward
The data of point obtain the new time slip-window for including another group of data, then choose again and repeat above-mentioned mistake after differentiating axis
Journey, until detecting qualified tumble waveform.
The variance computing module 104 is used for when the extreme point waveform is identified as meeting the waveform of tumble condition,
The waveform variance of multiple data points included after in calculating before the extreme point waveform, then judges whether the waveform variance is more than one
First default variance threshold values TH5。
Specifically, before the extreme point waveform in after multiple data points for including include the extreme point waveform before it is multiple
Multiple data points after all data points that data point, the extreme point waveform include and the extreme point waveform.Preferably,
In the present embodiment, before the extreme point waveform in after multiple data points for including include 10 data before the extreme point waveform
10 data points after all data points that point, the extreme point waveform include and the extreme point waveform.The waveform side
The calculation formula of difference is as follows：
Wherein, it is described represent before the extreme point waveform in after the average value of multiple data points that includes, n represents the pole
The quantity of multiple data points included after in before value point waveform, x1、x2…xnRepresent the corresponding acceleration value of each data point, s2Table
Show the waveform variance.It should be noted that the first default variance threshold values TH5It is according to lot of experiment validation and determining
The value of one suitable fall detection, the present embodiment are preferably 14000.
The variance computing module 104 by calculate meet before the extreme point waveform of tumble condition in after three phases number
According to variance, to falling, front and rear state carries out comprehensive descision, to enhance the accuracy rate of fall detection.If the in addition, waveform side
Difference is unsatisfactory for above-mentioned condition, then chooses module 102 back to differentiation axis and the time slip-window is moved to the data of any backward, weight
New selection repeats the above process after differentiating axis, until the waveform variance meets above-mentioned condition.
The plateau region computing module 105 is used to be more than the described first default variance threshold values TH in the waveform variance5
When, calculate the maximum value a of the data point of preset quantity after the extreme point waveformmaxWith minimum value amin, the preset quantity
The variance var (a) of data point and the data point of the preset quantity in first data point and the last one data point
Between time interval T1, and the maximum value a being calculated according to thismax, minimum value amin, variance var (a) and time interval
T1Judge whether to detect tumble event.
Specifically, if the maximum value amaxWith minimum value aminDifference be less than a preset difference value TH6, the variance var (a)
Variance threshold values TH default less than one second7And the time interval T1Less than the second preset time threshold TH8, then judge to have occurred
Tumble event.In the present embodiment, the data point of preset quantity is preferably 20 after the extreme point waveform after extreme point waveform
The data point of point.The preset difference value TH6Less than one natural acceleration value, the present embodiment preferred value are 0.38g.Described second
Default variance threshold values TH7For the value of a suitable fall detection determining according to lot of experiment validation, the present embodiment is preferably
500000.The second preset time threshold TH8Because being limited to a shorter time, the present embodiment preferred value is 600ms.
In the present embodiment, the plateau region computing module 105 passes through to one after qualified extreme point waveform
The data of plateau region are calculated, and to detect an effective tumble event, theoretical foundation is：The acceleration of acute variation
Whether appoint in a period of time after degree waveform and so keep a higher value, represent user just if a higher value is still maintained
More violent movement (as run) is being engaged in, so as to be judged as an invalid tumble event, so as to exclude user because acutely
Situations such as movement caused misrecognition.
In addition, if tumble event is not detected, returns to and differentiate axis selection module 102 by the time slip-window backward
The data of any are moved, chooses again and repeats the above process after differentiating axis, until detecting effective tumble event.
The tumble alarm module 106 is used for when detecting tumble event, is alarmed.Specifically, the mode of alarm
Can be control 100 warning message of Wearable, the warning message can be the audio by Wearable 100
The vibration that the alarm sound or vibration unit of output unit output generate.It is worn in addition, the tumble alarm module 106 also can control
It wears formula equipment 100 and sends out warning message to preset electronic device and alarm.The preset electronic device can be default
The corresponding communication equipment (such as mobile phone and computer) of communicating number (such as telephone number or IP address), the warning message can wrap
The current location information of Wearable 100 is included, so as to which the convenient user to Wearable 100 carries out Quick rescue.
Referring to Fig. 4, it is the fall detection applied to Wearable shown in FIG. 1 that present pre-ferred embodiments provide
The flow chart of method.It should be noted that fall detection method of the present invention and with Fig. 4 and as described below specific suitable
Sequence is limitation.It should be appreciated that in other embodiments, the sequence of fall detection method which part step of the present invention can
To be exchanged with each other according to actual needs or part steps therein can also be omitted or be deleted.It below will be to tool shown in Fig. 4
Body flow is described in detail.
Step S101 obtains real-time 3-axis acceleration of the Wearable 100 along three axial directions of a three-dimensional system of coordinate
Value.In the present embodiment, three axial directions of the three-dimensional system of coordinate are respectively by three axis of x, y, z.Step S101 can be examined by described fall
The acceleration acquisition module 101 of examining system 10 coordinates the acceleration transducer 20 to perform.Description has as described in step S101
Body can join the description to above-mentioned acceleration acquisition module 101.
Step S102 chooses three-dimensional coordinate according to the 3-axis acceleration value of the multiple data points included in a time slip-window
It is as differentiating axis one of in three axis.Specifically, step S102 can choose module 102 by the differentiation axis and perform,
It is specific to differentiate that axis choosing method join the above-mentioned detailed description to differentiating axis selection module 102.
Step S103 calculates extreme point, then by continuous three in the corresponding time slip-window of the differentiation axis
A extreme point forms an extreme point waveform.
Whether step S104 is judged in the time slip-window comprising the extreme point waveform for meeting tumble condition.If comprising
Meet the extreme point waveform of tumble condition, flow enters step S105, and otherwise, flow enters step S109.
Specifically, the step S103 and S104 can be performed by the tumble waveshape module 103.For judging
State in time slip-window whether can join comprising the method for extreme point waveform for meeting tumble condition it is above-mentioned to tumble waveshape mould
Block 103 elaborates.
Step S105, the waveform side of multiple data points included after calculating in meeting before extreme point waveform described in tumble condition
Difference, and judge whether the waveform variance is more than one first default variance threshold values TH5.If it is default that the waveform variance is more than described first
Variance threshold values TH5, flow enters step S106, and otherwise flow enters step S109.
Specifically, the step S105 can be performed by the variance computing module 104, and step S105 calculates waveform variance
Method can join the above-mentioned detailed description to variance computing module 104.
Step S106 calculates the maximum value of the data point of preset quantity after the extreme point waveform for meeting tumble condition
amaxWith minimum value amin, in the variance var (a) of data point of the preset quantity and the data point of the preset quantity
Time interval T between one data point and the last one data point1。
Step S107, according to the above-mentioned maximum value a being calculatedmax, minimum value amin, variance var (a) and time interval
T1Judge whether to detect tumble event.If detecting tumble event, flow enters step S108, and otherwise flow enters step
S109。
Specifically, the step S106 and step S107 can be performed by the plateau region computing module 105, judge whether to examine
The method for measuring tumble event can specifically join the detailed description to the plateau region computing module 105.
Step S108 alarms to the tumble event detected, terminates flow.Step S108 can be by the report
Alert module 106 performs, and specific alarm method can join the detailed description to the alarm module 106.
Step S109 moves a data point after time slip-window and obtains a new time slip-window, is then back to step
S102.The quantity of data point that the quantity for the data point that the new time slip-window includes is included with previous time slip-window
It is identical.That is, first data point of the previous time slip-window not included in new time slip-window.
In conclusion the embodiment of the present invention respectively leads to the real-time 3-axis acceleration data that acceleration transducer 20 senses
Cross the differentiation of three phases.The differentiation of the three phases is respectively whether to meet differentiation, the pole of tumble condition to extreme point waveform
After in before value point waveform after the differentiation of variance and extreme point waveform the data of stagnant zone (plateau region) differentiation, make
Tumble event detection it is more accurate and efficiently avoid misidentifying.
It is apparent to those skilled in the art that for convenience and simplicity of description, the method for foregoing description
The specific work process of step can refer to the specific descriptions of the corresponding function module in aforementioned fall detection system 10, herein
No longer repeat one by one.
In several embodiments provided herein, it should be understood that disclosed device and method can pass through it
Its mode is realized.The apparatus embodiments described above are merely exemplary, for example, the flow chart and block diagram in attached drawing are shown
The device of multiple embodiments according to the present invention, architectural framework in the cards, the work(of method and computer program product are shown
It can and operate.In this regard, each box in flow chart or block diagram can represent one of a module, program segment or code
Point, a part for the module, program segment or code includes one or more and is used to implement the executable of defined logic function
Instruction.It should also be noted that in some implementations as replacements, the function of being marked in box can also be to be different from attached drawing
The sequence marked occurs.For example, two continuous boxes can essentially perform substantially in parallel, they sometimes can also be by
Opposite sequence performs, this is depended on the functions involved.It is also noted that each box in block diagram and/or flow chart,
And the combination of the box in block diagram and/or flow chart, function as defined in performing or the dedicated of action can be used to be based on hardware
System realize or can be realized with the combination of specialized hardware and computer instruction.
In addition, each function module in each embodiment of the present invention can be integrated in a processing unit, it can also
Modules individualism, can also two or more modules be integrated in a module.
If the function is realized in the form of software function module and is independent product sale or in use, can be with
It is stored in a computer read/write memory medium.Based on such understanding, technical scheme of the present invention is substantially in other words
The part contribute to the prior art or the part of the technical solution can be embodied in the form of software product, the meter
Calculation machine software product is stored in a storage medium, is used including some instructions so that a computer equipment (can be
People's computer, server or network equipment etc.) perform all or part of the steps of the method according to each embodiment of the present invention.
And aforementioned storage medium includes：USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic disc or CD.It needs
Illustrate, herein, relational terms such as first and second and the like be used merely to by an entity or operation with
Another entity or operation distinguish, and without necessarily requiring or implying between these entities or operation, there are any this realities
The relationship or sequence on border.Moreover, term " comprising ", "comprising" or its any other variant are intended to the packet of nonexcludability
Contain so that process, method, article or equipment including a series of elements not only include those elements, but also including
It other elements that are not explicitly listed or further includes as elements inherent to such a process, method, article, or device.
In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including the element
Process, method, also there are other identical elements in article or equipment.
The foregoing is only a preferred embodiment of the present invention, is not intended to restrict the invention, for the skill of this field
For art personnel, the invention may be variously modified and varied.All within the spirits and principles of the present invention, that is made any repaiies
Change, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.It should be noted that：Similar label and letter exists
Similar terms are represented in following attached drawing, therefore, once being defined in a certain Xiang Yi attached drawing, are then not required in subsequent attached drawing
It is further defined and is explained.
The above description is merely a specific embodiment, but protection scope of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can readily occur in change or replacement, should all contain
Lid is within protection scope of the present invention.Therefore, protection scope of the present invention described should be subject to the protection scope in claims.
Claims (10)
- A kind of 1. Wearable, including acceleration transducer and fall detection system, which is characterized in that the fall detection system System includes：Acceleration acquisition module accelerates for obtaining the Wearable along real-time three axis of three axial directions of a three-dimensional system of coordinate Angle value；Differentiate that axis chooses module, for choosing three according to the 3-axis acceleration value of the multiple data points included in a time slip-window As differentiation axis one of in three axis of dimension coordinate system；Tumble waveshape module, for calculating the data point extreme point in the corresponding time slip-window of the differentiation axis, Continuous three extreme points are formed into an extreme point waveform, and judges whether to include in the time slip-window and meets tumble condition Extreme point waveform；Wherein, if the left side amplitude H of the extreme point waveform1Amplitude TH default more than first1, the extreme point wave The right amplitude H of shape2Amplitude TH default more than second2, the left side amplitude H1With the right amplitude H2Average value be more than third it is pre- If amplitude TH3And the time interval T in continuous three extreme points between first extreme point and the last one extreme point13It is small In the first preset time threshold TH4, then judge waveform of the extreme point waveform to meet tumble condition；Wherein, the extreme value The left side amplitude H of point waveform1With the right amplitude H2Calculation formula it is as follows：H1=| A2-A1|, H2=| A3-A2|；Wherein, A1, A2, A3The corresponding acceleration value of respectively continuous three extreme points；Variance computing module, for when the time slip-window includes the extreme point waveform for meeting tumble condition, described in calculating The waveform variance of multiple data points included after in meeting before the extreme point waveform of tumble condition, then judges that the waveform variance is It is no to be more than one first default variance threshold values TH5；Wherein, before the extreme point waveform in after multiple data points for including include it is described After multiple data points, the extreme point waveform include before extreme point waveform all data points and the extreme point waveform Multiple data points；AndPlateau region computing module, for when the waveform variance is more than the described first default variance threshold values TH5, calculating and meeting The maximum value a of the data point of preset quantity after the extreme point waveform of the tumble conditionmaxWith minimum value amin, the present count First data point and the last one data in the data point of the variance var (a) of the data point of amount and the preset quantity Time interval T between point1, and the maximum value a being calculated according to thismax, minimum value amin, variance var (a) and between the time Every T1Judge whether to detect tumble event.
- 2. Wearable according to claim 1, which is characterized in that the fall detection system further includes：Tumble alarm module, for when detecting tumble event, sending out warning message to preset electronic device and alarming, The warning message includes the current location information of Wearable.
- 3. Wearable according to claim 1, which is characterized in that the differentiation axis chooses module in the following manner Choose the differentiation axis：Maximum value ax is calculated in the time slip-window with multiple data points exported from x-axismaxAnd minimum value axmin, then X-axis amplitude is calculated, which is the maximum value axmaxWith minimum value axminDifference；Maximum value ay is calculated in the time slip-window with multiple data points exported from y-axismaxAnd minimum value aymin, then Y-axis amplitude is calculated, which is the maximum value aymaxWith minimum value ayminDifference；Maximum value az is calculated in the time slip-window with multiple data points exported from z-axismaxAnd minimum value azmin, so Z-axis amplitude is calculated afterwards, which is the maximum value azmaxWith minimum value azminDifference；Maximum amplitude is found out from the x-axis amplitude, y-axis amplitude and z-axis amplitude, by the corresponding axis of amplitude of the maximum It is determined as the differentiation axis；Wherein：The x-axis, y-axis, z-axis represent three axis of the three-dimensional system of coordinate respectively.
- 4. Wearable according to claim 1, which is characterized in that include after in before the extreme point waveform multiple Data point includes multiple data points, all data points for including of the extreme point waveform and the institute before the extreme point waveform State multiple data points after extreme point waveform；The calculation formula of the waveform variance is as follows：Wherein, the average value of multiple data points included after during the x is represented before the extreme point waveform, n represent the extreme value The quantity of multiple data points included after in before point waveform, x1、x2…xnRepresent the corresponding acceleration value of each data point, s2It represents The waveform variance.
- 5. Wearable according to claim 1, which is characterized in that as the maximum value amaxWith minimum value aminDifference Less than a preset difference value TH6, variance var (a) the variance threshold values TH default less than one second7And the time interval T1It is less than Second preset time threshold TH8When, the plateau region computing module then judges to have detected that tumble event.
- A kind of 6. fall detection method, applied to the Wearable including acceleration transducer, which is characterized in that the tumble Detection method includes：Acceleration obtaining step obtains real-time 3-axis acceleration of the Wearable along three axial directions of a three-dimensional system of coordinate Value；Differentiate axis selecting step, three-dimensional sit is chosen according to the 3-axis acceleration value of the multiple data points included in a time slip-window As differentiation axis one of in three axis of mark system；Tumble waveshape step calculates the data point extreme point differentiated in the corresponding time slip-window of axis, will even Whether three continuous extreme points form an extreme point waveform, and judge in the time slip-window comprising the pole for meeting tumble condition Value point waveform；Wherein, if the left side amplitude H of the extreme point waveform1Amplitude TH default more than first1, the extreme point waveform The right amplitude H2Amplitude TH default more than second2, the left side amplitude H1With the right amplitude H2Average value preset width more than third Value TH3And the time interval T in continuous three extreme points between first extreme point and the last one extreme point13Less than One preset time threshold TH4, then judge waveform of the extreme point waveform to meet tumble condition；Wherein, the extreme point wave The left side amplitude H of shape1With the right amplitude H2Calculation formula it is as follows：H1=| A2-A1|, H2=| A3-A2|；Wherein, A1, A2, A3The corresponding acceleration value of respectively continuous three extreme points；Variance calculates step, when the time slip-window includes the extreme point waveform for meeting tumble condition, calculates the satisfaction Then the waveform variance of multiple data points included after in before the extreme point waveform of tumble condition judges whether the waveform variance is big In one first default variance threshold values TH5；Wherein, before the extreme point waveform in after multiple data points for including include the extreme value It is multiple after multiple data points, the extreme point waveform include before point waveform all data points and the extreme point waveform Data point；AndPlateau region calculates step, is more than the described first default variance threshold values TH in the waveform variance5When, it calculates described in meeting The maximum value a of the data point of preset quantity after the extreme point waveform of tumble conditionmaxWith minimum value amin, the preset quantity First data point in the data point of the variance var (a) of data point and the preset quantity and the last one data point it Between time interval T1；According to the maximum value a being calculatedmax, minimum value amin, variance var (a) and time interval T1 Judge whether to detect tumble event.
- 7. fall detection method according to claim 6, which is characterized in that the fall detection method further includes：Tumble alarming step when detecting tumble event, sends out warning message to preset electronic device and alarms, described Warning message includes the current location information of Wearable.
- 8. fall detection method according to claim 6, which is characterized in that the differentiation axis selecting step includes：Maximum value ax is calculated in the time slip-window with multiple data points exported from x-axismaxAnd minimum value axmin, then X-axis amplitude is calculated, which is the maximum value axmaxWith minimum value axminDifference；Maximum value ay is calculated in the time slip-window with multiple data points exported from y-axismaxAnd minimum value aymin, then Y-axis amplitude is calculated, which is the maximum value aymaxWith minimum value ayminDifference；Maximum value az is calculated in the time slip-window with multiple data points exported from z-axismaxAnd minimum value azmin, so Z-axis amplitude is calculated afterwards, which is the maximum value azmaxWith minimum value azminDifference；Maximum amplitude is found out from the x-axis amplitude, y-axis amplitude and z-axis amplitude, by the corresponding axis of amplitude of the maximum It is determined as the differentiation axis；Wherein：The x-axis, y-axis, z-axis represent three axis of the three-dimensional system of coordinate respectively.
- 9. fall detection method according to claim 6, which is characterized in that include after in before the extreme point waveform more A data point include multiple data points before the extreme point waveform, all data points for including of the extreme point waveform and Multiple data points after the extreme point waveform；The calculation formula of the waveform variance is as follows：Wherein, the average value of multiple data points included after during the x is represented before the extreme point waveform, n represent the extreme value The quantity of multiple data points included after in before point waveform, x1、x2…xnRepresent the corresponding acceleration value of each data point, s2It represents The waveform variance.
- 10. fall detection method according to claim 6, which is characterized in that in the plateau region calculates step, when The maximum value amaxWith minimum value aminDifference be less than a preset difference value TH6, the variance var (a) is less than one second default side Poor threshold value TH7And the time interval T1Less than the second preset time threshold TH8When, then judge to have detected that tumble event.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510898194.6A CN105342626B (en) | 2015-12-08 | 2015-12-08 | Wearable and the fall detection method applied to Wearable |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510898194.6A CN105342626B (en) | 2015-12-08 | 2015-12-08 | Wearable and the fall detection method applied to Wearable |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105342626A CN105342626A (en) | 2016-02-24 |
| CN105342626B true CN105342626B (en) | 2018-06-29 |
Family
ID=55318807
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510898194.6A Active CN105342626B (en) | 2015-12-08 | 2015-12-08 | Wearable and the fall detection method applied to Wearable |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105342626B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10506990B2 (en) | 2016-09-09 | 2019-12-17 | Qualcomm Incorporated | Devices and methods for fall detection based on phase segmentation |
| CN106530611A (en) * | 2016-09-28 | 2017-03-22 | 北京奇虎科技有限公司 | Terminal, and method and apparatus of detecting fall of human body |
| CN106530607A (en) * | 2016-12-23 | 2017-03-22 | 天津永兴泰科技有限公司 | Intelligent glasses and alarm method |
| CN108175388B (en) * | 2017-12-01 | 2021-01-26 | 中国联合网络通信集团有限公司 | Behavior monitoring method and device based on wearable equipment |
| CN108564059B (en) * | 2018-04-26 | 2021-02-23 | 歌尔科技有限公司 | Wearable device, data processing method and device thereof, equipment and storage medium |
| CN112235464B (en) | 2019-06-28 | 2022-05-31 | 华为技术有限公司 | Falling detection-based help calling method and electronic equipment |
| CN110575176B (en) * | 2019-08-26 | 2024-03-26 | 南京理工大学 | Action segmentation method based on two-layer sliding window threshold |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101950464A (en) * | 2010-09-17 | 2011-01-19 | 中国科学院深圳先进技术研究院 | Method and system for fall monitoring and warning |
| CN102227189A (en) * | 2008-11-28 | 2011-10-26 | 皇家飞利浦电子股份有限公司 | Method and apparatus for fall detection |
| CN102626312A (en) * | 2012-04-01 | 2012-08-08 | 浙江大学城市学院 | Fall detection method and device and wearable device |
| CN102657533A (en) * | 2012-04-28 | 2012-09-12 | 浙江大学城市学院 | Falling detection method, falling detection device and wrist type device |
| US9005141B1 (en) * | 2007-10-12 | 2015-04-14 | Biosensics, L.L.C. | Ambulatory system for measuring and monitoring physical activity and risk of falling and for automatic fall detection |
| CN105046882A (en) * | 2015-07-23 | 2015-11-11 | 浙江机电职业技术学院 | Fall detection method and device |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9691253B2 (en) * | 2014-02-04 | 2017-06-27 | Covidien Lp | Preventing falls using posture and movement detection |
| WO2015143085A1 (en) * | 2014-03-18 | 2015-09-24 | Zhang Jack Ke | Techniques for wellness monitoring and emergency alert messaging |
-
2015
- 2015-12-08 CN CN201510898194.6A patent/CN105342626B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9005141B1 (en) * | 2007-10-12 | 2015-04-14 | Biosensics, L.L.C. | Ambulatory system for measuring and monitoring physical activity and risk of falling and for automatic fall detection |
| CN102227189A (en) * | 2008-11-28 | 2011-10-26 | 皇家飞利浦电子股份有限公司 | Method and apparatus for fall detection |
| CN101950464A (en) * | 2010-09-17 | 2011-01-19 | 中国科学院深圳先进技术研究院 | Method and system for fall monitoring and warning |
| CN102626312A (en) * | 2012-04-01 | 2012-08-08 | 浙江大学城市学院 | Fall detection method and device and wearable device |
| CN102657533A (en) * | 2012-04-28 | 2012-09-12 | 浙江大学城市学院 | Falling detection method, falling detection device and wrist type device |
| CN105046882A (en) * | 2015-07-23 | 2015-11-11 | 浙江机电职业技术学院 | Fall detection method and device |
Non-Patent Citations (1)
| Title |
|---|
| 基于三轴加速度传感器跌倒检测方法的研究;李飞龙;《万方数据企业知识服务平台》;20151203;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105342626A (en) | 2016-02-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105342626B (en) | Wearable and the fall detection method applied to Wearable | |
| CN104605827B (en) | The determination methods of a kind of human body strenuous exercise state and device | |
| EP3369375A1 (en) | Method and apparatus for identifying state of motion of human body | |
| CN104680046B (en) | A kind of User Activity recognition methods and device | |
| Timmer et al. | Pathological tremors: Deterministic chaos or nonlinear stochastic oscillators? | |
| JP5537553B2 (en) | Fall detection system and its operation method | |
| JP6083799B2 (en) | Mobile device location determination method, mobile device, mobile device location determination system, program, and information storage medium | |
| CN105559751B (en) | Monitor the method, apparatus and wearable device of micro- active state | |
| CN106540407B (en) | Treadmill method for control speed and device | |
| JP2011200390A (en) | Active meter, control program, and activity type specifying method | |
| JP2016097228A5 (en) | ||
| CN108814618A (en) | A kind of recognition methods of motion state, device and terminal device | |
| CN205103993U (en) | Intelligence human body guardianship device of tumbleing | |
| JPWO2008072678A1 (en) | Detection apparatus, system, program, and detection method | |
| CN109753777A (en) | A kind of personal identification method, device, storage medium and mobile terminal | |
| JP6689881B2 (en) | How to judge an ergonomic rating | |
| JP2022020722A5 (en) | ||
| TW201221929A (en) | Real-time detection system and the method thereof | |
| CN110222936A (en) | A kind of root of business scenario is because of localization method, system and electronic equipment | |
| CN105551191B (en) | A kind of fall detection method | |
| CN110226934A (en) | A kind of fall detection method, device and Wrist belt-type equipment | |
| CN105559789B (en) | Fall detection system and method | |
| CN115223316B (en) | Intelligent commodity anti-theft alarm detection method | |
| CN108777001A (en) | Surgical simulation method and device | |
| CN104281640B (en) | Characteristic quantity draw-out device, characteristic quantity abstracting method |
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 |