Summary of the invention
According to an aspect of the present invention, provide a kind of intelligence and to fall monitor device, comprising: alarm modules, signal acquisition module and signal processing module, wherein,
Described signal acquisition module comprises 3-axis acceleration sensor, baroceptor and pressure transducer, for gathering human body attitude real time data, pressure altitude data and pressure data;
Described signal processing module comprises fall detection unit, human body attitude real time data, pressure altitude data and pressure data that described fall detection unit is set to according to described collection carry out analyzing and processing, to judge that human body is fallen state, export the first signal when judging that human body is fallen to alarm modules;
Described alarm modules generates according to described first signal and exports warning information.
Monitor device of the present invention carries out fall detection by gathering 3-axis acceleration, height and pressure data, and consider behavior characteristics and other environmental characteristic of human body, the accuracy of fall detection is higher.
In some embodiments, described signal acquisition module is also for harvester wearing information, described signal processing module also can comprise device and wear detecting unit, and described device is worn detecting unit and is set to carry out analyzing and processing according to device wearing information, exports wearing state mark; Described fall detection unit identifies according to the wearing state of described output, and human body is fallen state and export the first signal to described alarm modules; Wherein, described device wearing information comprises one of them or both above combinations of 3-axis acceleration signal, wearing site pressure signal, temperature signal and human biological signal.Thus, can detect device whether wear direction and/or wearing position correct, just carry out fall detection when device is worn correct, realize avoiding wearing the bad of the incorrect wrong report caused, further the accuracy rate of raising fall detection.
In some embodiments, described signal acquisition module also comprises temperature sensor and/or human-body biological electric transducer, one of them or both above combinations of the human biological signal of the pressure signal that described device wears the acceleration signal that detecting unit gathers according to described 3-axis acceleration sensor, described pressure transducer gathers, temperature signal that described temperature sensor gathers and described human biological electricity sensor acquisition, carry out analyzing and processing, export wearing state mark.Thus, can with or without situation about wearing, realize under avoiding device does not have wear condition, causing the bad of wrong report by checkout gear further, improve device and wear the accuracy rate with fall detection.
In some embodiments, described device also comprises locating module and wireless communication module, and described locating module gathers the positional information of described monitor device, and described alarm modules contains the warning information of positional information to remote terminal by wireless communication module transmission.Thus, when there is falling risk, guardian can be notified in time, and positional information be supplied to guardian, effectively helping timely to obtain.
In some embodiments, described signal processing module also comprise fall after state detection unit, be set to the data gathered according to described signal acquisition module, detect fall after body state.When judge human body remove fall state time, send secondary signal to described alarm modules, described alarm modules, according to described secondary signal, sends the information that departs from and fall to remote terminal by wireless communication module.Thus, can detect further user fall after disposition, notify guardian in time to depart to fall after state user, for guardian brings conveniently, effectively improve Consumer's Experience.Further, sound all clear in time, the consumption of device in alarm can be reduced, reduce the power consumption of device.
In some embodiments, described alarm modules comprises and carries out emergency voice broadcast according to the first signal enabling speaker and stop playing speaker distress signal according to secondary signal.Thus, can be implemented in while notifying emergency to guardian, people sends emergency towards periphery, to make user be succoured in time, is out of danger, and increases safety guarantee.
In some embodiments, this device also comprises human-computer interaction module, is placed through touch screen, button or phonetic incepting user input, inputs typing user basic information or start the process that described alarm modules carries out crying for help or removing emergency according to user.Thus, user can pass through touch screen typing user profile, carries out analyzing and processing to facilitate device according to user profile, also can realize when device occurs to detect wrong report, user is undertaken crying for help or remove wrong report emergency by button, effectively can improve Consumer's Experience, efficient and convenient.
According to another aspect of the present invention, additionally provide a kind of intelligence and to fall the processing method of monitor device, this monitor device comprises alarm modules, signal acquisition module and signal processing module, the treating method comprises:
Signal acquisition module Real-time Collection user behavior information data, exports signal processing module to, and described user behavior information data comprises 3-axis acceleration data, pressure altitude data and pressure data;
Signal processing module carries out analyzing and processing according to described user behavior information data, judging that human body is fallen state, when judging that human body is fallen, exporting the first signal to alarm modules;
Alarm modules generates according to described first signal and exports warning information.
Method of the present invention carries out fall detection by gathering 3-axis acceleration, height and pressure data, and consider behavior characteristics and other environmental characteristic of human body, the accuracy of fall detection is higher.
In some embodiments, described message processing module carries out analyzing and processing according to described user behavior information data, judges that the human body state of falling comprises:
The pressure altitude data that a, the acceleration information storing 3-axis acceleration sensor collection in certain hour interval, baroceptor gather and the pressure data that pressure transducer gathers;
B, according to store acceleration information judge the data whether occurring great change in described certain hour interval, according to the time point of great change data, calculate the pressure differential sum ∑ of health side and opposite side before and after the 3-axis acceleration change sum ACC_SUM in a period of time after the acceleration average AY_1 in the trunk direction in a period of time in a period of time before great change and after great change and AY_2, great change, human body pressure altitude difference H2-H1 before and after great change and great change | Pi1-Pi2|;
C, setting threshold values TH3, TH4, TH5, judge whether satisfy condition H2-H1>TH3, ACC_SUM<TH4 and ∑ | Pi1-Pi2|>TH5, and AY_1 is close to gravity acceleration g, AY_2 is close to 0, if met, then judge that human body is fallen.
Thus, the attitudes vibration of falling and lying low of human body is judged by 3-axis acceleration, by altitude information judgment means height change overhead, judge that the wearing site pressurized dynamics of a week changes by pressure data, in conjunction with user's attitude and behavior characteristics, can detect whether human body falls more exactly.
In some embodiments, described step b comprises:
According to the 3-axis acceleration data in the certain hour interval of described storage, to each 3-axis acceleration data gathered, calculate its 3-axis acceleration amplitude ACC, judge whether ACC is greater than the threshold values TH1 of setting, when being greater than setting threshold values, judge that namely the time point of this data acquisition is the time point occurring data great change;
According to the 3-axis acceleration data of described storage, calculate the acceleration information variable quantity ACC_CHG in great change surrounding time section, judge whether described acceleration information variable quantity ACC_CHG is greater than setting threshold values, if be greater than setting threshold values, then judge that namely great change surrounding time section is the moment of falling;
According to the 3-axis acceleration data of described storage, calculate health side and the opposite side pressure differential sum ∑ at wearable device position before and after the 3-axis acceleration change sum ACC_SUM in a period of time after the acceleration average AY_2 in the trunk direction in a period of time after the acceleration information average AY_1 in the trunk direction in a period of time before great change, great change, great change, the human body pressure altitude value H2-H1 before and after great change and great change | Pi1-Pi2|.
Thus, according to the big ups and downs of data, can judge the time that human body is fallen, the data obtained before and after falling compare, to determine whether human body falls.
In some embodiments, described method also can comprise: signal acquisition module real-time acquisition device wearing information data export signal processing module to, described signal processing module according to the process of device wearing information data analysis, judgment means wearing state, output device wearing state identify; Signal processing module read described wearing state mark judge, when wearing state be designated correctly wear time, carry out human body fall detection, and export the first signal to alarm modules.Thus, when detecting that device is correctly worn, just carrying out human body fall detection, when can avoid not wearing or wear incorrect or landing because of device, producing the problem of wrong report, improve the accuracy rate detected further.And, only when device is worn correct, just carry out the detection that the collection of user behavior information data and human body are fallen, unnecessary data processing operation can be reduced, raise the efficiency, save power consumption.
In some embodiments, described signal acquisition module also comprises temperature sensor and/or human-body biological electric transducer, and described device wearing information data comprise one of them or both above combinations of temperature data, human biological signal, wearing site pressure data and 3-axis acceleration data.
In some embodiments, described signal processing module carries out the above combinations of one of them or both of following A to D analyzing and processing, to generate and output device wearing state mark according to described device wearing information data:
The temperature signal T1 pressing close to human body side of A, reading temperature sensor and equipment are exposed to the temperature signal T2 of side in air, when judging that the temperature difference of T1 and T2 is greater than setting threshold values, arrange wearing state to be designated and correctly to wear, otherwise arrange wearing state be designated wear wrong;
B, read the human biological signal that human-body biological electric transducer exports, when judging human biological signal as high level, wearing state being set and being designated and correctly wearing, otherwise arrange wearing state be designated wear wrong;
C, obtain user waistline value L according to entry information, the number N=L/D of the sensor producing pressure is calculated according to the pressure transducer space D of matrix form distribution, read the pressure signal P 1 of N number of pressure transducer ... PN, when judging that the pressure signal of N number of pressure transducer all equals setting threshold values, arrange wearing state to be designated and correctly to wear, otherwise arrange wearing state be designated wear wrong;
D, the accekeration read on three direction of principal axis of 3-axis acceleration sensor, when the accekeration judging to represent trunk direction is g and other two accekerations are 0, arrange wearing state to be designated and correctly to wear, otherwise arrange wearing state be designated wear wrong;
Wherein, when carrying out the combinative analysis process more than both, when being judged as wearing wrong under either type, be namely set to wear wrong mark.
Thus, can realize judging whether human body has worn device by temperature and/or human biological signal, judge that whether the position of human body wearable device and direction be correct by pressure and acceleration, thus realize whether wearing device and wear whether correct detection, avoid reporting by mistake because of device problem.
In some embodiments, described device also comprises locating module and wireless communication module, described method also comprises: described locating module gathers the positional information of described monitor device, described alarm modules sends warning information to remote terminal by wireless communication module, and described warning information comprises the positional information and emergency content that obtain from described locating module.Thus, in time positional information and emergency content of falling can be sent to guardian, to be succoured timely and effectively.
In some embodiments, the method also can comprise:
When human body be in fall state time, described signal acquisition module continuous collecting 3-axis acceleration data, pressure altitude data and pressure data also store;
Signal processing module carries out analyzing and processing according to the 3-axis acceleration data of real-time update, pressure altitude data and pressure data, judge that human body removes the state of falling, when judging that human body releasing is fallen, secondary signal is sent to alarm modules, described alarm modules is according to secondary signal, and the information of being fallen by wireless communication module transmission disengaging is to remote terminal.
Thus, can be implemented in continuing to monitor to User Status after falling, so that after clearing, notify guardian in time, save time to guardian and reduce tutorial nervous anxiety, hommization more, Consumer's Experience is better.
In some embodiments, described signal processing module carries out analyzing and processing according to the 3-axis acceleration data of real-time update, pressure altitude data and pressure data, judges that human body is removed the state of falling and comprised:
Calculate the acceleration average AY_3 of the axle in the representative trunk direction after falling in a period of time;
Judge acceleration average AY_3, pressure altitude value H3 and force value (p13, P23, PN3) whether satisfy condition: | AY_3-g|<TH6, H3-H2>TH7 and P13=P23=...=PN3, wherein, TH6 and TH7 is the threshold values of setting, as satisfied condition, then judges that human body removes state of falling.
Thus, whether by the acceleration average in trunk direction close to g, whether device height becomes large gradually and whether wearing site pressure data becomes even, and judges whether human body stands up from the state of falling, thus when human body releasing is fallen, the very first time notifies guardian.
In some embodiments, the method also can comprise: receive and inputted by the signal of button, plays/suspends speaker alarm and send distress signals/releasing distress signals to remote terminal.Thus, when can be implemented in generation wrong report, carry out the active process of user, to reduce the adverse consequences reported by mistake and bring.
Detailed description of the invention
Below in conjunction with accompanying drawing, the present invention is further detailed explanation.
Fig. 1 show schematically show the surface structure of monitor device of falling according to the intelligent human-body of one embodiment of the present invention.As shown in Figure 1, this device comprises belt body 1 and bracelet 2, and belt body 1 is fixedly connected with bracelet 2 one end, and the other end can the fastening when being belt, connect and fastening mode with common belt.Bracelet 2 is provided with button 3, user can carry out man-machine interaction by pressing the button 3.The pressure transducer 5 of monitor device is uniformly distributed along belt body 1, and remaining functional module (other sensors, signal processing module, locating module, wireless communication module etc. as signal acquisition module) is all built on the integrated chip 4 in bracelet 2.In actual applications, user just can realize to user by wearing the belt shown in Fig. 1 daily monitoring.Because belt belongs to the daily dressing requirement of most people, easy to carry, without any subsidiary sense, not easily forget, very convenient.
Fig. 2 show schematically show the frame structure being built in each module in belt body of this device.As shown in Figure 2, this device comprises signal processing module 20, locating module 21, signal acquisition module 22, wireless communication module 23 and alarm modules 24.Locating module 21 adopts GPS or the locate mode such as the Big Dipper or mobile base station to realize, for providing the geographical location information of user.Wireless communication module 23 is GSM communication unit or bluetooth communication unit etc. can carry out communication to wirelessly chip or modules with mobile terminal device, for realizing the data interaction between remote terminal (as device ends such as mobile phone, computer, IPad).When alarm modules 24 is set to the first signal (as the distress signal) when receiving and fall corresponding, corresponding warning information (as distress signals) is sent to remote terminal by wireless communication module 23, need relief to notify that guardian user falls, corresponding distress signals can comprise the geographical location information of locating module 21 acquisition and content of specifically crying for help.Signal acquisition module 22, for Real-time Collection user behavior information data, is supplied to signal processing module 20 and carries out the analysis of human body fall detection.Signal acquisition module 22 realizes mainly through multiple sensors, include but not limited to 3-axis acceleration sensor, baroceptor and pressure transducer, 3-axis acceleration sensor is for gathering human body attitude data, baroceptor is for gathering pressure altitude data, and pressure transducer is for gathering pressure data.Signal processing module 20 is the microprocessors such as MCU.Wherein, signal processing module 20 comprises fall detection unit 202, fall detection unit 202 carries out fall detection for human body attitude data (comprising acceleration information AX, AY, AZ of three axis), pressure altitude data and the pressure data gathered according to signal acquisition module 22, when judging to fall, export distress signal (i.e. the first signal) to alarm modules 24, to start alarm modules 24, distress signals are sent to tutorial RTU by wireless communication module 23.Due to human body fall time, 3-axis acceleration can have of short duration " great change " instantaneously falling down, subsequently for some time can relatively " static ", pressure altitude data have certain difference in height before and after falling, pressure data can be different and certain according to stressing conditions in the side of human body ground proximity and the side deviating from ground difference, fall detection unit 202 just can carry out detection that whether human body fall and judgement according to these three kinds of data.
The 3-axis acceleration data broken line graph that the human body that Fig. 9 show schematically show a kind of typical case is fallen.Acceleration information broken line graph when first interval 90 is human body normal stand, second interval 91 is the acceleration information broken line graph of weightlessness, 3rd interval 92 is acceleration information broken line graph when falling, and the 4th interval 93 is the acceleration information broken line graph in a period of time after falling.As shown in Figure 9, when human body is fallen, when striking ground, 3-axis acceleration signal there will be the very violent one piece of data of fluctuation, as the 3rd interval 92 of signal generation acute variation in Fig. 9, be the moment of human body hitting ground, it is interval that the present invention is referred to as " great change ".Analyze the acceleration information in generation of falling rear a period of time, under normal circumstances, the interval of one section of geo-stationary is had after falling, (the 4th is interval) 93 between quiescent centre as shown in Figure 9, in the 4th interval 93, represent the acceleration of the Y-axis in trunk direction substantially close to 0 (because human body is from uprightly having become the attitude that lies low), it is interval that the present invention is referred to as " static ".
In use, signal acquisition module 22 Real-time Collection user behavior data (comprising human body attitude data, pressure altitude data and wearing site pressure data), and the human body attitude data (i.e. 3-axis acceleration data), pressure altitude data and the pressure data that are stored by FIFO (FirstInFirstOut, first in first out) form in a period of time (as 4 seconds).Fall detection unit 202 is according to the 3-axis acceleration data stored; analyze and whether occur the one piece of data (namely whether occurring " great change ") fluctuating very violent; be specially: the section setting threshold values TH1 of " great change " is occurring; by calculating the 3-axis acceleration data AX of each collection, the vectorial mould of AY, AZ obtains 3-axis acceleration amplitude ACC, namely has
judge whether the amplitude ACC of each 3-axis acceleration data gathered is greater than the threshold values TH1 of setting, when 3-axis acceleration amplitude is greater than the threshold values of setting, judges that the time point of this image data is the time of generation " great change ".According to the time point that " great change " occurs, read the 3-axis acceleration data AY that the representative trunk direction that (as 1 second) stores in " great change " front a period of time occurs, according to the accekeration AY of the Y-axis in the representative trunk direction gathered, calculate the average that " great change " front Y-axis occurs
(n is the numbering of the acceleration information gathered in " great change " front a period of time).Meanwhile, record occur the pressure altitude data H1 of " great change " front last collection and pressure data (P11, P21 ..., PN1).In order to confirm that whether human body falls during in great change further, acceleration information AX, AY, the AZ of (as in 0.1s after 0.1s-great change before great change) in the period before and after " great change " moment can be there is according to the image data record stored, acceleration information according to record calculates acceleration change amount ACC_CHG, and computing formula is:
(n is the numbering of the acceleration information gathered within this period moment).Setting threshold values TH2 is as being set to 2g, judge whether the acceleration change amount ACC_CHG of " great change " moment calculated meets ACC_AHG>TH2, meet illustrate this moment period human body fall, and after this falls moment human body by section quiescent time after entering great change, then record the 3-axis acceleration data AX of (in a period of time namely after great change after 0.1s) in " between quiescent centre ", AY, AZ, force value (the P12 of pressure altitude value H2 and wearing site, P22,, PN2).The average of the acceleration of the Y-axis in the representative trunk direction in this interval is calculated according to the data of record
with the 3-axis acceleration change sum in this interval
N is the numbering of the acceleration information gathered in quiescent centre.Setting device height threshold values TH3 overhead, resting state acceleration threshold values TH4 and pressure differential threshold values TH5, judge whether the difference in height of the pressure altitude value H2 in the pressure altitude value H1 before great change and quiescent centre meets the threshold values TH3 being greater than setting, namely whether H2-H1>TH3 is met, whether the acceleration change sum ACC_SUM in quiescent centre is less than the threshold values TH4 of setting, namely whether meets ACC_SUM<TH4.The judgment means wearing site pressure condition of a week simultaneously, the pressure differential sum that whether there is the force value of the pressure transducer of side and the force value of opposite side is greater than the threshold values TH5 of setting, i.e. ∑ | Pi1-Pi2|>TH5.And judge that whether the acceleration average AY_1 of the Y direction before great change is close to gravity acceleration g (illustrating that user is standing state), between quiescent centre, whether the acceleration average AY_2 of Y direction is close to 0 (user is lying status).Wherein, TH3 can be set to the height of waist to ankle as 80cm according to human body information data, if met, then illustrates before great change and between quiescent centre, and the height of human body is from uprightly becoming bending or lying status; TH4 is the tranquil period after falling, and this, acceleration change amount was very little in period, can be set to less value, as leveled off to 0.1g (g is acceleration of gravity).If met, then illustrate that human body remains static human height by a period of time after high step-down, and after generation of falling normally, before attitude is fallen in disengaging, all can occur this situation; The force value of side of landing when TH5 falls according to human body is arranged with away from the poor sum of the force value of side, ground, if satisfied, illustrates that human body is after height changes and enters resting state, has side to land.Thus, can judge to meet when three conditions simultaneously, be human body and there occurs and fall, Status Flag such as the FALL_DOWN_FLAG that then fallen by human body is set to TRUE, send distress signal (as character " 1 ") to alarm modules 24 simultaneously, thus start alarm, enter help-asking mode.The geographical location information that alarm modules 24 provides according to locating module 21, generates the distress signals comprising geographical location information and emergency content and is sent to tutorial remote terminal by wireless communication module 23, notify, to obtain relief.The fall detection mode that this embodiment provides, need the change of detected air pressure altitude information simultaneously, acceleration change and wearing site pressure data change in a week, can than behavior characteristics and the data more comprehensively considering user, the detection mode of relatively single acceleration or angle change, Detection accuracy of the present invention is higher more effective, so that user can send emergency request the very first time after falling, obtain relief.
Meanwhile, consider except the accuracy rate of algorithm aspect, device wear condition is also the key factor affecting Detection accuracy, invention also provides device in the solution not having to cause under wear condition or under wearing incorrect situation reporting by mistake.As shown in Figure 2, also comprise device in signal processing module 20 and wear detecting unit 201.Device is worn detecting unit 201 and is set to carry out analyzing and processing according to the device wearing information data (comprising 3-axis acceleration data and the wearing site pressure data of a week) of signal acquisition module collection, exports wearing state and controls signal to fall detection unit 202.The wearing state control signal that fall detection unit 202 exports carries out fall detection, when wearing correct at device, gathering user behavior data and carries out analysis detection, exporting distress signal when falling to alarm modules 24.
In use, after user's starting drive, signal acquisition module 21 gathers pressure data P1, P1 incessantly ..., PN.Device wears detecting unit 201 reduced pressure data P1, P1, PN, if human body is not worn or to wear degree of tightness undesirable, such as too loosen or wearing site inaccurate etc., each data of pressure transducer just have comparatively big difference, if correctly worn, the value of pressure data should meet P1=P2=substantially ...=PN=P, wherein P is elasticity force value when being correctly worn on waist.It should be noted that, pressure transducer is that matrix form is distributed on belt, and number is N=L/D, and wherein, L is the waistline information (essential information according to user's typing obtains) of user, and D is the spacing of the pressure transducer of matrix form distribution.If the pressure data that signal acquisition module 21 gathers meets P1=P2=...=PN=P, then can judge the elasticity that user wears and wearing position correctly, then wearing state be identified WARE_FLAG and be set to TRUE, otherwise be set to FALSE.
Preferably, after user's starting drive, signal acquisition module 21 constantly gathers 3-axis acceleration data AX, AY, AZ simultaneously.Whether device simultaneously can to wear orientation according to 3-axis acceleration data detection device correct if wearing detecting unit 201.Due to, under normal circumstances during person upright, correct wearing mode should only have an axle (i.e. the axle of trunk vertical direction) accekeration to be g (i.e. acceleration of gravity), and other two axle acceleration values are 0.Suppose that Y-axis represents trunk direction time human body is stood, then device wear detecting unit 201 judge gather AX, AY, AZ whether meet AY=g and AX=AZ=0, if met, then judge that user wears in the right direction, wearing state mark WARE_FLAG=TRUE is set, otherwise is set to FALSE.
Preferably, signal acquisition module 21 also comprises temperature sensor.Wherein, because temperature sensor itself has directivity (towards people side and toward the outer side), the temperature sensor of the present embodiment is set to two, the direction of one is set to towards the side of human body, for gathering human body temperature, another direction is set to towards the side of air, for gathering ambient temperature, after two temperature sensors set direction, be integrated in the enterprising trip temperature collection of integrated chip 4 shown in Fig. 1.After device starts, signal acquisition module 21 constantly harvester presses close to the temperature data T2 that the temperature data T1 of human body side and device are exposed to side in air, if human body does not have wearable device, then substantially should meet T1=T2 and (allow certain limit error, differential as T1 and T2 is bordering on the threshold values of setting as 0.5 °), if when wearable device, the temperature difference of both sides and T1 and T2 should have certain amplitude (as being greater than the threshold values 0.5 ° of setting).Whether device wears detecting unit 201 according to the temperature data gathered, and compares the temperature approach of T1 and T2, can wear by decision maker.If worn, then set wearing state and be designated TRUE, otherwise be set to FALSE.
Preferably, signal acquisition module 21 also can comprise human-body biological electric transducer.After device starts, whether signal acquisition module 21 constantly gathers the human biological signal that human-body biological electric transducer exports, be high level, judge whether human body has worn device according to human biological signal.If human biological signal is high level, then wearing state is set and is designated TRUE, otherwise be set to FALSE.
In actual applications, device wears detecting unit 201 can only according to the wherein detection carried out device and whether wear or whether correctly wear of above pressure data, 3-axis acceleration, temperature data and human biological signal, also the combination of wherein any more than two can be selected to detect simultaneously, the detection data selected are more, and the accuracy rate of detection is higher.Wherein, when selecting the combination of wherein any more than two to detect, as long as wherein the testing result of either type is wrong for wearing, wearing state all to be set and be designated FALSE.As, the combination simultaneously can carrying out four detects, comprise and first whether being worn by temperature data checkout gear, if human body and the external environment temperature difference very little time (as being all 37 degree), then whether worn by human biological signal's checkout gear, if worn, reduced pressure data judge that whether wearing position is correct, whether correct judge to wear direction again if correct according to 3-axis acceleration.If four is all correct, be then judged to be that device is worn correctly, wearing state be set and be designated TRUE, otherwise be set to FALSE, and proceed data acquisition.Fall detection unit 202 reads the value of wearing state mark, when for TRUE, gathers user behavior information data carry out fall detection by signal acquisition module 21.
Preferably, when carrying out wearing detection, after initialization or can also detect when wearing wrong, by the correct usual method of speech play, instructing user to wear.
As shown in Figure 2, this device can also comprise human-computer interaction module 25.Human-computer interaction module 25 can be touch screen, sound identification module or button, is set to receive user's input, carries out Data Enter, or carries out rescue alarm according to user instruction startup alarm modules 24 or remove rescue alarm.As passed through touch screen typing user basic information, or carry out a key warning by button, when also can meet because of problems affect testing results such as sample rate deficiency and algorithm discriminations, user can carry out falling warning in time, very efficient and convenient.
Preferably, in order to the demand of user more can be met to hommization, the present invention also can arrange the function automatically exiting and fall and report to the police further, to climb voluntarily rest a period of time after falling or other modes to be stood up etc. in situation to meet user, needed the demand of informing guardian in time and automatically exiting alarm mode.
As shown in Figure 2, signal processing module 20 also comprise fall after state detection unit 203, be set to after human body is fallen, signal acquisition module 21 continuous collecting 3-axis acceleration data (AX, AY, AZ), pressure altitude data (H) and wearing site pressure data (P1, P2 ..., PN), and store acceleration information, altitude information and the pressure data of (as in 4 seconds) in a period of time, analyze the Y-axis average representing trunk direction
pressure altitude value H3 with force value (P13, P23 ..., PN3) whether meet and transfer the condition of standing by falling to.Be specially, setting threshold values TH6 and TH7, judge whether to meet | AY_3-g|<TH6, H3-H2>TH7 and P13=P23=...=PN3, if all met, then be judged to be that user is stood voluntarily, setting status indicator of falling is FALSE, and send corresponding secondary signal (as removed distress signal) to alarm modules 24, alarm modules 24 departs from the information of state of falling to remote terminal, to remind guardian that user departs from the state of falling by wireless communication module 23 transmission.Wherein, TH6 is the threshold values representing difference between the accekeration g (i.e. acceleration of gravity) under the acceleration average AY_3 in trunk direction and standing state, can be set as close to 0, AY_3 more close to gravity acceleration g (namely | during AY_3-g|<TH6, TH6 is less), shows that human body is got over close to normal standing state; TH7 is difference in height threshold values, represent the degree of closeness of height when difference and standing state between the height of current state lower device and the height of timer of falling, 80cm can be set to, also can arrange according to the height information of user, H3-H2>TH7 represents that present apparatus height H 3 is higher than the device height H 2 when falling, and human body is described away from ground; And P13=P23=...=PN3 shows that the pressure of device wearing site has trended towards balance, does not namely have the point that pressure-bearing is larger, illustrate that user has not been be in side to land state.Thus, by judging whether to meet decision condition | AY_3-g|<TH6, H3-H2>TH7 and P13=P23=...=PN3, can judge whether user stands voluntarily, thus guardian can be notified in time when user departs from and falls pattern, effectively can promote Consumer's Experience, very convenient.Meanwhile, due at warning stage, device needs not stop to use wireless communication module, human-computer interaction module and alarm modules, and the power consumption of device can be higher, and automatically detect that user removes state backed off after random alarm of falling, and effectively can reduce the power consumption of device.
Alternatively, alarm modules 25 can also be speaker playing device, when starting help-asking mode, when sending distress signals by wireless communication module 23 to guardian, starting speaker simultaneously and playing voice help-asking signal, to be succoured in time; And when automatically exiting help-asking mode, sent the voice help-asking signal departing from fall status information and stopping broadcasting speaker to guardian by wireless communication module 23.
Intelligent human-body provided by the invention monitor device of falling can be worn on user's waist, uses as belt, very convenient.And device of the present invention carries out human body fall detection by 3-axis acceleration data, pressure altitude data and pressure data, and more meet the behavior characteristics of user, accuracy is higher.Simultaneously, inventive arrangement provides device and wear measuring ability, that can avoid because of wrong report when device is not worn or wears incorrect is bad, further increases the accuracy of fall detection, in time accurately fall distress signals and the positional information of user are sent to guardian.Device of the present invention also can provide the automatic detection after falling simultaneously, can continue to detect user behavior state after user falls, after user stands, in time the information removed of falling is sent to guardian, bring facility (as saved the tutorial time, reducing tutorial psychentonia pressure etc.) to guardian.Device of the present invention can also pass through the information interaction of the realizations such as touch screen, button, speech recognition and user, user-friendly, when emergency situation occurring or occurring wrong report, can be met the demand of user's emergency by button.
The intelligent human-body that Fig. 3 show schematically show an embodiment of the present invention is fallen the processing method (method of work) of monitor device.As shown in Figure 3, the method comprises:
Step S301: signal acquisition module gathers user behavior information data.
Signal acquisition module gathers user's human body attitude data (comprising 3-axis acceleration value AX, AY, AZ) by 3-axis acceleration sensor, by ambient pressure sensor acquisition user pressure altitude data (H), user's wearing site pressure data (P1 of a week is gathered by pressure transducer, P2, ..., PN).User's attitude data may be used for judging standing or lying status of user, and pressure altitude data may be used for judgment means height overhead, the pressure condition of land when the wearing site pressure data of a week may be used for judging that user lands side and the side that do not land.
Step S302: whether signal processing module falls according to the user behavior information data human body gathered.
Signal processing module is analyzed according to the user behavior information data collected, judging whether human body falls, when detecting that human body is fallen, carrying out step S303, if do not detect that human body is fallen, then proceed the data acquisition of step S301.Fig. 4 show schematically show the method flow of human body fall detection.As shown in Figure 4, the method comprises:
Step S401: Real-time Collection human body attitude behavioral data, pressure altitude data and pressure data.
Signal acquisition module carries out data acquisition in real time, main collection human body attitude real time data (i.e. 3-axis acceleration data AX, AY, AZ), ambient pressure height (H) data and wearing site pressure data (P1, P2 ..., PN).And adopt FIFO (FirstInFirstOut, the first in first out) data in pattern storage a period of time, as the acceleration information in 4 seconds, pressure altitude data and pressure data.
Step S402: judge whether " great change " that data occur.
Signal processing module is according to three axle expedited datas, analyze the data whether occurring " great change ", because when human body is fallen, when striking ground, 3-axis acceleration signal there will be the very violent one piece of data of fluctuation (specifically can describing see Fig. 9 above), at this setting threshold value TH1, according to 3-axis acceleration data AX, AY, the vectorial mould of AZ three calculates and obtains 3-axis acceleration amplitude ACC (computing formula is see describing) above, judge whether to there is ACC>TH1, if satisfied condition, then illustrate and now " great change " occurs, then carry out step S403, if do not satisfied condition, then proceed step S401.
Step S403: the height value (H1) before record change, Y-axis acceleration average (AY_1) and force value (P11, P21 ..., PN1).
According to the data of each sensor acquisition of signal acquisition module, record the Y-axis mean data AY_1 (computing formula is see describing) in the front representative trunk direction of generation " great change " above, pressure altitude data H1, pressure data (P11, P21 ..., PN1).
Step S404: the acceleration change value ACC_CHG calculating " great change " moment.
According to the data of each sensor acquisition of signal acquisition module, there is the acceleration information variable quantity ACC_CHG (computing formula is see describing) before and after " great change " moment in the period (as 0.1s after front 0.1s-) above in record.
Step S405: judge whether acceleration change value ACC_CHG is greater than setting threshold values TH2.
Because the moment of contacting to earth of normally falling is very of short duration; total variation when simultaneously contacting to earth can be very large; setting threshold value TH2; judge whether the ACC_CHG>TH2 that satisfies condition; if met, illustrate that human body has now contacted to earth and fall; after falling, human normal can enter between quiescent centre, then carry out step S406, otherwise carries out step S401 and continue to carry out data acquisition.
Step S406: the height value (H2) after record change, Y-axis acceleration average (AY_2), force value (P12, P22 ..., PN2) and the acceleration change summation ACC_SUM of " calmness " segment.
According to describing above to the broken line graph of Fig. 9, analyze the 3-axis acceleration in generation of falling rear a period of time, to detect between the quiescent centre after falling and (Y-axis representing trunk direction in this interval substantially close to 0, whether level off to by the acceleration of Y-axis in this interval and 0 to judge.), calculating 3-axis acceleration change sum ACC_SUM (computing formula is see describing) in this interval above, recording the human body pressure altitude value H2 of this time period simultaneously, and the force value (P12 at wearable device position, P22 ..., PN2).
Step S407: whether meet decision condition of falling.
Setting threshold values TH3, TH4, TH5 (value refers to and describes above), judge whether the difference in height of the pressure altitude value H2 in the pressure altitude value H1 before great change and quiescent centre meets the threshold values TH3 being greater than setting, namely whether H2-H1>TH3 is met, whether the acceleration change sum ACC_SUM (computing formula is see describing above) in quiescent centre is less than the threshold values TH4 of setting, namely whether ACC_SUM<TH4 is met, the judgment means wearing site pressure condition of a week simultaneously, the force value that whether there is the pressure transducer of side is greater than the threshold values TH5 of setting, i.e. ∑ | Pi1-Pi2|>TH5.And judge that whether the Y-axis acceleration average AY_1 before great change is close to gravity acceleration g, whether the Y-axis acceleration average between quiescent centre is close to 0, if meet H2-H1>TH3, ACC_SUM<TH4 and ∑ simultaneously | Pi1-Pi2|>TH5, and AY_1 is close to g, AY_2 is close to 0, then carry out step S408, otherwise carry out step S401.
Step S408: judge to fall, arranging status indicator of falling is TRUE, enters emergency state of falling.
If meet above condition simultaneously, then judge that human body there occurs fall (specifically see above and describe), then the status indicator FALL_DOWN_FLAG=TRUE that falls is set, sends the first signal as distress signal to alarm modules, to enter emergency state of falling simultaneously.
By above step, namely the direction of standing and the lying low change of human body is judged by 3-axis acceleration, by pressure altitude data judgment means height change overhead, judge that human body wearing site (the present invention is waist) the pressurized situation of a week changes by pressure data, thus detect whether human body falls, meet human body behavior characteristics, the accuracy rate of detection is higher.
Step S303: signal processing module sends distress signal to alarm modules, and obtain customer position information by locating module.
Signal processing module sends distress signal (as character " 1 ") to alarm modules, is obtained the geographical location information of user by locating module simultaneously.
Step S304: alarm modules carries out emergency response process.
After alarm modules receives distress signal, by wireless communication module, the geographical location information of user and emergency content are sent to the RTU of monitoring, notify guardian, to be succoured timely.
The intelligent human-body that Fig. 5 show schematically show another embodiment of the present invention is fallen the processing method of monitor device.As shown in Figure 5, this embodiment is from the different of Fig. 3 illustrated embodiment, and whether the present embodiment needs first checkout gear to wear correctly, if wear the detection correctly just carried out human body and whether fall.Specific as follows:
Step S501: signal acquisition module harvester wearing information data.
Signal acquisition module can be the combination of one of them or more than two of 3-axis acceleration sensor, pressure transducer, temperature sensor and human-body biological electric transducer, 3-axis acceleration data AX, AY, AZ can be gathered by 3-axis acceleration sensor, the wearing site pressure data P1 of a week is gathered by pressure transducer, P2, ..., PN, press close to the temperature data T1 of human body side by temperature sensor collection and be exposed to the temperature data T2 of air side, by human biological electricity sensor acquisition human biological signal.Signal acquisition module harvester wearing information data can be in above sensing data, also can be multiple combinations, and the scheme of the preferred three's combination of the present embodiment, is described in detail.This compound mode can improve the accuracy of detection.
Step S502: whether signal processing module is correctly worn according to the device wearing information data detection device gathered.
Signal processing module is according to the data gathered, and whether checkout gear is correctly worn.Fig. 6 show schematically show the detection method whether device is correctly worn, and as shown in Figure 6, the method comprises:
Step S601: opening device also initializes.
The power supply of user's opening device, waits for that device carries out the initialization of data automatically, giving initial value, being initialized as FALSE as wearing state identified WARE_FLAG by the state variable of device, status indicator FALL_DOWN_FLAG assignment of falling is FALSE etc.
Step S602: Real-time Collection temperature, pressure and acceleration information, and carry out voice guidance to user and wear.
Signal acquisition module Real-time Collection 3-axis acceleration data AX, AY, AZ, the wearing site pressure data P1 of a week, P2, ..., PN, press close to the temperature data T1 of human body side and be exposed to the temperature data T2 of air side, by speech play usual method, guidance being worn to user simultaneously.
Step S603: judge that human body side temperature T1 is compared with the temperature T2 of air side, whether T1-T2>TH.
If human body does not have wearable device, then T1=T2 in theory, has the error of about 0.5 °, if human body is worn in reality, then side is air themperature, side is that human body temperature will cause both sides to occur temperature difference, so just reaches the object whether checkout equipment is worn.Based on the fact that there is the temperature difference, whether setting threshold values TH, contrast T1 and T2 satisfied temperature difference are greater than the threshold values of setting, if be greater than, carry out step S604, otherwise continue the data acquisition carrying out step S602.
It should be noted that, owing to there is ambient temperature and the close situation of human body temperature, as preferred embodiment, human-body biological electric transducer can be increased in signal acquisition module, detect further, be specially and gather human biological signal, determine whether high level, if the output of human-body biological electric transducer is high level, then illustrate that human body has worn device, the pressure detecting of step S604 can be carried out, otherwise continue to carry out data acquisition.In actual applications, the replacement scheme whether human-body biological electric transducer also can be worn as temperature sensor judgment means, replace with human-body biological electric transducer by temperature sensor, carry out the judgement of human biological signal, the present invention does not limit compound mode.
Step S604: judge whether the value of each pressure transducer meets P1=P2=......=PN>0.
If human body is correctly worn, then the human body waist force value of a week meet equal and equal elasticity moderate time pressure value P, namely there is P1=P2=......=PN=P>0, judge whether to meet this condition and can judge device wearing site is correct and elasticity is suitable (specifically can see describing above).If met, then carry out step S605, otherwise continue the data acquisition carrying out step S602.
Step S605: under judging normal stand situation, whether 3-axis acceleration value meets AY=g and AX=AZ=0.
Under normal circumstances, when human body is stood, correct wearing mode should only have an axle acceleration value to be that other two axles of g (gravity and speed) should be 0, if Y-axis represents trunk direction time human body is stood, i.e. AY=g and AX=AZ=0, if met, illustrate that wearing of device is in the right direction, then carry out step S607, otherwise carry out step S606.
Step S606: user wears anisotropy by voice message.
Play voice message, reminding user wears anisotropy, and proceeds the data acquisition of step S602.
Step S607: judge to wear correctly, setting is worn correct status and is designated TRUE, enters fall detection state.
If the decision condition of satisfied temperature, pressure and 3-axis acceleration simultaneously, then illustrate that device is worn, and wearing site and direction are all correct, now will wear correct status mark WARE_FLAG and be set to TRUE, carry out the data acquisition of the fall detection of step S503 afterwards and judge whether human body falls, otherwise not carrying out fall detection.Thus, can avoid because of wrong report when device is not worn or wears incorrect, improve the accuracy rate of fall detection and emergency alarm.
Step S503: signal acquisition module gathers user behavior information data.
Step S504: speech play usual method.
Step S505: whether signal processing module falls according to the user behavior information data human body gathered.
Step S506: signal processing module sends distress signal to alarm modules, and obtain customer position information by locating module.
Step S507: alarm modules carries out emergency response process.
The realization of step S503 to step S507 can refer to above rapid S301 to step S304.By this embodiment, can realize under correct prerequisite is worn in judgement, then carry out fall detection, efficiency and the accuracy rate of detection can be improved.
The intelligent human-body that Fig. 7 show schematically show another embodiment of the present invention is fallen the processing method of monitor device.As shown in Figure 7, this embodiment is from the different of Fig. 5 illustrated embodiment, the present embodiment is detecting that human body is fallen and after entering help-asking mode, can continue to gather user behavior data, after falling, whether human body removes the detection of the state of falling, and detect human body remove fall state time, automatically exit help-asking mode of falling, for user and guardian provide convenience.Specifically comprise:
Step S701: signal acquisition module harvester wearing information data.
Step S702: whether signal processing module is correctly worn according to the device wearing information data detection device gathered.
Step S703: signal acquisition module gathers user behavior information data.
Step S704: speech play usual method.
Step S705: whether signal processing module falls according to the user behavior information data human body gathered.
Step S706: signal processing module sends distress signal to alarm modules, and obtain customer position information by locating module.
Step S707: alarm modules carries out emergency response process.
Step S708: signal acquisition module continuous collecting user behavior information data.
Step S709: whether signal processing module removes the state of falling according to the user behavior information data human body gathered.
Wherein, step S701 to step S707 with step S501 to step S507.Difference is, when after alarm of carrying out falling, need the data acquisition proceeding step S708, the data gathered comprise the pressure data of 3-axis acceleration data, pressure altitude data and wearing site, and need the releasing fall detection of the data gathered being carried out to step S709, so that after user is had a rest automatically or after depending on others' help and standing, can notify that guardian's user removes the state of falling in time, thus reduce tutorial nervous anxiety, save the tutorial time, to provide more intelligent better user's service.
Fig. 8 show schematically show whether signal processing module removes the state of falling method according to the user behavior information data human body gathered.As shown in Figure 8, the method comprises:
Step S801: judge whether status indicator of falling is TRUE.
Read the value of the status indicator FALL_DOWN_FLAG that falls, determine whether TRUE, fallen if TRUE then illustrates, then carry out step S802 and continue to carry out data acquisition, otherwise illustrate that user is not fallen, carry out step S803.
Step S802: Real-time Collection human body attitude behavioral data, pressure altitude data and pressure data.
Signal acquisition module continuous collecting 3-axis acceleration data (AX, AY, AZ), pressure-altitude sensor continuous collecting altitude information (H), pressure transducer Real-time Collection pressure data (P1, P2 ... and adopt FIFO form to store a period of time as the data gathered in 4 seconds PN).
Step S803: exit detection.
Step S804: record fall after height value (H3), Y-axis acceleration average (AY_3) and force value (P13, P23 ..., PN3).
Signal processing module, according to the 3-axis acceleration data stored, calculates the Y-axis acceleration average AY_3 (computing formula is see describing) in this time period above, and records the pressure altitude data H3 of Real-time Collection, with pressure data (P13, P23 ..., PN3).
Step S805: judge whether to meet the decision condition removing state of falling.
Setting threshold values TH6 and TH7, judge whether to satisfy condition | AY_3-g|<TH6, H3-H2>TH7, P13=P23=...=PN3, if met | AY_3-g|<TH6 illustrates that trunk is perpendicular to ground, namely stands.Meet H3-H2>TH7 and then illustrate that current device aspect ratio is high when falling.Meet P13=P23=...=PN3 then can judgment device current pressure state with fall before pressure state contrast and become even, be no longer the stressed situation that lands greatly in side.Wherein, the value of threshold values TH6, TH7 is specifically see describing above.If meet decision condition, then carry out step S806, otherwise carry out step S802 continuation image data.
Step S806: judge from mode release of falling, arranging status indicator of falling is FALSE, exports releasing and to fall distress signal.
If meet decision condition, then can judge that user is stood from the state of falling, now arranging status indicator of falling is FALSE, and sends releasing to alarm modules and to fall signal (as character " 0 "), to stop the alarm of crying for help.
Step S710: signal processing module sends releasing to alarm modules and to fall signal, alarm modules carries out the response process removing emergency.
Alarm modules to be fallen signal according to the releasing received, and send to have removed to guardian by wireless communication module and fall, stop the information of crying for help, meanwhile, alarm modules also can pass through mute speaker, and voice help-asking is carried out in stopping.
Preferably, in order to avoid when detecting mistake, report by mistake, the personal safety of better guarantee user, operational approach in monitor device of the present invention can also comprise: the signal input being received user by button, broadcasting/suspend speaker alarm and send emergency/releasing distress signals to remote terminal.As arranged a button on device, click if user is of short duration, then signal processing module receives the input of user, sends distress signal to alarm modules, and alarm modules is play speaker and comprised the distress signals of positional information to remote terminal to guardian's transmission.If with head of a household's button, then signal processing module receives user's input, sends remove distress signal to alarm modules, and alarm modules then stops the voice help-asking of broadcasting speaker and sends the information of having cleared to guardian.
By method of the present invention, achieve by 3-axis acceleration sensor, baroceptor and pressure transducer carry out human body behavioural information data collection and but by human body behavioural information data to human body fall state detection monitoring, accuracy rate is higher, meets the monitoring demand of falling to old man and patient.Simultaneously, method of the present invention additionally provides the detection that device is worn correctness and automatically exited the pattern of falling, the wrong report brought because device wears problem can be avoided, also the situation detecting user can be continued after judgement is fallen, reach timely notifier processes when standing, more intelligent convenience, the accuracy rate of detection is higher.
Above-described is only some embodiments of the present invention.For the person of ordinary skill of the art, without departing from the concept of the premise of the invention, can also make some distortion and improvement, these all belong to protection scope of the present invention.