CN104952210B - A kind of fatigue driving state detecting system and method based on decision making level data fusion - Google Patents
A kind of fatigue driving state detecting system and method based on decision making level data fusion Download PDFInfo
- Publication number
- CN104952210B CN104952210B CN201510249302.7A CN201510249302A CN104952210B CN 104952210 B CN104952210 B CN 104952210B CN 201510249302 A CN201510249302 A CN 201510249302A CN 104952210 B CN104952210 B CN 104952210B
- Authority
- CN
- China
- Prior art keywords
- pulse
- driver
- data
- driving state
- mrow
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 55
- 230000004927 fusion Effects 0.000 title claims abstract description 40
- 230000001133 acceleration Effects 0.000 claims abstract description 91
- 238000001514 detection method Methods 0.000 claims abstract description 52
- 230000033001 locomotion Effects 0.000 claims abstract description 31
- 238000012360 testing method Methods 0.000 claims abstract description 21
- 230000008859 change Effects 0.000 claims description 23
- 238000012549 training Methods 0.000 claims description 15
- 210000001367 artery Anatomy 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 11
- 230000009466 transformation Effects 0.000 claims description 11
- 230000005540 biological transmission Effects 0.000 claims description 7
- 238000009499 grossing Methods 0.000 claims description 7
- 210000003462 vein Anatomy 0.000 claims description 7
- 210000002321 radial artery Anatomy 0.000 claims description 6
- 210000000707 wrist Anatomy 0.000 claims description 6
- 238000013500 data storage Methods 0.000 claims description 5
- 235000013399 edible fruits Nutrition 0.000 claims description 2
- 238000012544 monitoring process Methods 0.000 claims description 2
- 238000005070 sampling Methods 0.000 claims description 2
- 238000003672 processing method Methods 0.000 claims 1
- 230000004044 response Effects 0.000 abstract description 2
- 206010016256 fatigue Diseases 0.000 description 67
- 230000006870 function Effects 0.000 description 17
- 210000004369 blood Anatomy 0.000 description 15
- 239000008280 blood Substances 0.000 description 15
- 230000000875 corresponding effect Effects 0.000 description 13
- 210000000709 aorta Anatomy 0.000 description 8
- 238000002474 experimental method Methods 0.000 description 6
- 230000007423 decrease Effects 0.000 description 5
- 230000000737 periodic effect Effects 0.000 description 5
- 230000000747 cardiac effect Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000001174 ascending effect Effects 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 3
- 230000017531 blood circulation Effects 0.000 description 3
- 230000008602 contraction Effects 0.000 description 3
- 210000005240 left ventricle Anatomy 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000010009 beating Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 210000004204 blood vessel Anatomy 0.000 description 2
- DDRJAANPRJIHGJ-UHFFFAOYSA-N creatinine Chemical compound CN1CC(=O)NC1=N DDRJAANPRJIHGJ-UHFFFAOYSA-N 0.000 description 2
- 238000005315 distribution function Methods 0.000 description 2
- 210000003017 ductus arteriosus Anatomy 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000009661 fatigue test Methods 0.000 description 2
- 210000003128 head Anatomy 0.000 description 2
- 230000010247 heart contraction Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000001361 intraarterial administration Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 230000036581 peripheral resistance Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 238000011084 recovery Methods 0.000 description 2
- 238000003786 synthesis reaction Methods 0.000 description 2
- 230000002861 ventricular Effects 0.000 description 2
- WQZGKKKJIJFFOK-GASJEMHNSA-N Glucose Natural products OC[C@H]1OC(O)[C@H](O)[C@@H](O)[C@@H]1O WQZGKKKJIJFFOK-GASJEMHNSA-N 0.000 description 1
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000036772 blood pressure Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 229940109239 creatinine Drugs 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000035080 detection of muscle activity involved in regulation of muscle adaptation Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000003205 diastolic effect Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- 210000000744 eyelid Anatomy 0.000 description 1
- 239000008103 glucose Substances 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000007914 intraventricular administration Methods 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000000149 penetrating effect Effects 0.000 description 1
- 230000035479 physiological effects, processes and functions Effects 0.000 description 1
- ZRHANBBTXQZFSP-UHFFFAOYSA-M potassium;4-amino-3,5,6-trichloropyridine-2-carboxylate Chemical compound [K+].NC1=C(Cl)C(Cl)=NC(C([O-])=O)=C1Cl ZRHANBBTXQZFSP-UHFFFAOYSA-M 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 230000033764 rhythmic process Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000009329 sexual behaviour Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000002618 waking effect Effects 0.000 description 1
Landscapes
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a kind of fatigue driving state detecting system and method based on decision making level data fusion, the system gathers recent movement acceleration first with acceleration transducer, recent movement state is judged based on acceleration dynamic threshold, driver tired driving state is tentatively judged based on the motionless theories of steering wheel 4s, also by the wiggly error amount of setting direction disk, strengthen the fault-tolerance of detection；Using the pulse time-domain value of pulse transducer collection driver, physiological driver's state is detected based on pulse frequency dynamic threshold；By carrying out decision level fusion, the testing result after being merged to two kinds of testing results.Designed in the fatigue driving state detection method based on decision making level data fusion and construct fatigue driving state detection prototype system.Compared with the existing methods, driver tired driving condition detection method disclosed by the invention based on decision making level data fusion has certain advantage in terms of Detection accuracy, and being reached in the response time of algorithm, time complexity and memory consumption etc. has preferable performance.
Description
Technical field
The present invention relates to driving condition detection method, more particularly to a kind of fatigue driving state detection method, belong to mobile
The interleaving techniques application field of calculating, sensing technology and data fusion.
Background technology
Fatigue driving (fatigue driving) refers generally to driver in startup procedure because fatigue occurs in body mechanism
Change and cause the not normal situation of its manipulation ability, seriously endanger traffic safety, it has also become what the whole world faced seriously asks
Topic.The report of National Highway Traffic safety management bureau is shown, because the traffic accident that driver tired driving induces account for handing over
The 20%-30% of logical total number of accident.For example, during August in 2012 26 days 2 31 divide about, driver Chen drives night coach, with
Heavy pot type semi-trailer train knocks into the back, and causes 36 people in motor bus to die instantly；According to vehicle-bone global positioning system (Global
Position System, GPS) record, the continuous driving time of motor bus driver Chen up to 4 hours 22 points, midway do not stop not
Breath, energy is not concentrated when fatigue driving causes to drive, and reaction and judgement decline, and cause the accident.
Efficiently detect driver tired driving state and carry out feedback in time can effectively to prevent similar traffic thing
Therefore generation.Blood by gathering driver can analyze blood glucose of the driver under fatigue driving state, plasma wrea and
Creatinine, this several category information of comprehensive analysis obtained one on driver whether fatigue, this method has higher inspection
Accuracy rate is surveyed, experimental result can be as the reference of other method, but real-time is bad, and needs the Medical Devices of specialty.
At present, research and technical staff have studied and have developed a series of achievement in research and product, are broadly divided into three classes：The first kind
It is the detection technique based on physiological signal, is based primarily upon the change etc. of brain wave, heart rate, pulse and skin voltage；Second class
It is to be based on driver's body physical state, is based primarily upon the change of inclined degree, eye, the change of face and gripping on head
Dynamics of steering wheel etc.；The third is to be based on travel condition of vehicle, is based primarily upon the characteristics of motion, the traveling of the vehicle speed of steering wheel
Running orbit of degree, the acceleration of vehicle and vehicle etc..
At present, researcher devises wearable electroencephalogram (Electroencephalograph, an EEG) inspection
Examining system, vigilant degree of the driver to itself driving behavior can be detected in real time, so as to reflect the fatigue conditions of driver；
Also researcher driver sleep it is bereft in the case of gather driver electrocardiogram (Electrocardiogram,
ECG) data, synthesis heart rate and frequency of wink two indices, the fatigue driving state of driver is analyzed；Also researcher passes through
The EEG data of driver is analyzed, change information of the driver in each frequency energy of fatigue stage is obtained, obtains driver's
In the tired variability of fatigue stage.Also researcher proposes a kind of mechanism of the driver tired driving state of detection in real time,
Using the EEG data, ECG data and electromyographic signal of driver come the fatigue driving state of comprehensive descision driver；Also study
PCA analysis experiment sample of the personnel based on core, selects suitable kernel function and relevant parameter to efficiently separate
Go out normal sample and tired sample, linear analysis is carried out to the ECG data of driver, obtains the experiment sample of driver's EEG data
This, analysis experiment sample is to belong to normally to still fall within tired sample, and then detects whether driver is in fatigue driving shape
State.Fatigue state detection method based on EEG and ECG has a preferable real-time and higher Detection accuracy, but hardware into
This is higher, and wears and be not easy.
At present, also researcher judges the fatigue driving shape of driver based on driver's face behavioural characteristic to analyze
State.Such as：Comprehensive utilization driver's eyes closing time percentage, the degree that face opens and the inclined degree on head synthesis are sentenced
The fatigue driving state of disconnected driver；The position of human eye is positioned by comprehensively utilizing frame difference method, template matching method and Kalman's method
Put, and then position human eye and open closed state, based on eyes closed percentage of time (Percentage of Eyelid
Closure, PERCLOS) characteristic value detection driver fatigue driving state.Driver tired driving shape based on eye behavior
State detection method real-time is preferable, and Detection accuracy is higher, but this kind of method is confined to the good situation of driver's cabin light, nothing
Method is applied to the situation of driving at night, there is larger limitation.
The direction disc detector SAM of Electronic Safety Products companies development and production is that one kind can be examined
The device of the improper motion of vehicle steering is surveyed, when steering wheel is in the case of normal operation, sensor will not send alarm, but
In the case that pilot control steering wheel 4 seconds is motionless, sensor device will send alarm and reminding driver, until direction
Untill disk is returned under ordinary running condition by driver's control.A kind of corner based on steering wheel that also researcher proposes becomes
The driver tired driving condition detection method of change, angle displacement sensor and GPS module are embedded on steering wheel, to collection
Whether the angle change application model identification theoretical judgment driver arrived is in fatigue driving state, while is judged with GPS module
The transport condition of vehicle.Also researcher devises one kind and is based on steering wheel angle signal detection driver tired driving state
Method, the algorithm is by way of establishing the multiple linear regression model of steering wheel turn signal correlated variables and physiological signal
Realize, make use of it is preceding establish regression model to Sexual behavior mode method, can analyze whether driver locates by this regression model
In fatigue driving state.This kind of method suffers from preferable real-time, and expense is smaller, but Detection accuracy is relatively low.
In a word, current achievement in research is in the prevalence of Detection accuracy is not high, hardware cost is higher, equipment is worn not
Easily, by such environmental effects it is larger the defects of.And the present invention can solve the problems, such as above well.
The content of the invention
Present invention aims at a kind of fatigue driving state detecting system based on decision making level data fusion is proposed, this is
System efficiently can easily detect driver fatigue state, and the system gathers the motion of steering wheel by acceleration transducer first
Acceleration information, driver's pulse data is gathered by pulse transducer；Both data are pre-processed respectively, to pre- place
Two kinds of data after reason calculate dynamic threshold respectively, obtain whether being in the Preliminary detection knot of fatigue state on driver
Fruit；By carrying out decision level fusion to two kinds of testing results, the testing result after more accurately merging is obtained.
The technical scheme adopted by the invention to solve the technical problem is that：A kind of fatigue based on decision making level data fusion is driven
Condition detecting system is sailed, the system includes acceleration information acquisition module, acceleration information transmission and pretreatment module, acceleration
Degrees of data dynamic threshold training module, algorithm application module, arteries and veins based on acceleration information detection driver tired driving state
Data acquisition module, the pulse data of fighting are stored with pretreatment module, pulse data dynamic threshold training module, based on pulse data
Detect algorithm application module, data fusion module of driver tired driving state etc.；The motion for comprehensively utilizing steering wheel accelerates
Spend collateral information and the pulse direct information of driver；The system is based on acceleration sensing using acceleration information acquisition module
Device collects recent movement acceleration information, using acceleration information transmission with pretreatment module to these initial data applications
The method of moving average is smoothed；Algorithm application module based on acceleration information detection driver tired driving state uses
The motionless theories of steering wheel 4s, the preliminary testing result for judging fatigue driving state；Meanwhile it is based on using pulse data acquisition module
Pulse transducer collects the pulse data in driver's driving procedure；Stored using pulse data and first stored with pretreatment module
The pulse data collected, the threshold method for being then based on wavelet transformation remove pulse signal noise, and it is mobile flat to reuse weighting
Equal method is smoothed to data；The pulse frequency analyzed using pulse data dynamic threshold training module and calculate driver is become
Change, the threshold value of corresponding normal driving state is established for different individuals；Driver tired driving shape is detected based on pulse data
The algorithm application module of state judges whether currently available driver's pulse frequency is normal, enters by the comparison with this normality threshold
And judge whether driver is in fatigue driving state；Data fusion module enters to both recognition result application evidence theories
Row decision level fusion, whether driver is obtained in tired by the fatigue driving state detection algorithm based on decision making level data fusion
The testing result of labor driving condition.
The function of acceleration information acquisition module is：Recent movement acceleration information is gathered using acceleration transducer.
Acceleration information transmits is with the function of pretreatment module：To being accelerated with acceleration transducer collection recent movement
The degree initial data application weighting method of moving average is smoothed.
The function of acceleration information dynamic threshold training module is：At by acceleration information transmission and pretreatment module
The data managed calculate average value again, dynamic threshold of the driver in current road segment are obtained, then by comparing acceleration information
Obtain the waving interval of steering wheel.
The function of the algorithm application module of driver tired driving state is detected based on acceleration information is：Using steering wheel
The motionless theories of 4s, the preliminary testing result for judging fatigue driving state.
The function of pulse data acquisition module is：Based on the pulse transducer collection being fixed at the wrist radial artery of human body
Pulse data in driver's driving procedure.
Pulse data stores is with the function of pretreatment module：The pulse data collected is stored, is then based on small echo change
The threshold method changed removes pulse signal noise, reuses the method for weighted moving average and data are smoothed.
The function of pulse data dynamic threshold training module is：Analyze and calculate the pulse frequency change of driver, for not
Same individual establishes the threshold value of corresponding normal driving state.
The function of the algorithm application module of driver tired driving state is detected based on pulse data is：By with normally driving
Whether the threshold value for sailing state relatively judges whether currently available driver's pulse frequency is normal, and then judge driver in tired
Labor driving condition.
The function of data fusion module is：To the algorithm application based on acceleration information detection driver tired driving state
Module and based on pulse data detection driver tired driving state algorithm application module recognition result application evidence theory
Decision level fusion is carried out, obtains whether driver is in by the fatigue driving state detection algorithm based on decision making level data fusion
The testing result of fatigue driving state.
The system driver of the present invention is during vehicle is driven, the acceleration and angle change of steering wheel for vehicle motion
Information is to analyze the important information of the driving condition of driver, can react the driving shape of driver indirectly to a certain extent
State.On the road of normally travel, if continuous more than the 4s of steering wheel is motionless, then driver is likely in fatigue driving
State, but if driver traveling on straight highway and around vehicle it is seldom, when at this moment the continuous 4s of steering wheel is motionless,
Can not judge whether driver is in fatigue driving state, thus the movable information of steering wheel be only capable of to a certain extent between it is reversed
Mirror the driving condition of driver.The pulse of people, which changes, can show the physiological status such as the fatigue of people, in driver's driving procedure
Pulse change, the especially change of pulse frequency, be the direct reflection to driver's driving condition.
The system of the present invention is the fatigue driving state detection method based on decision making level data fusion, comprehensively utilizes steering wheel
Acceleration of motion collateral information and driver pulse direct information, the detection that can effectively improve fatigue driving state is accurate
Rate：Recent movement acceleration information is collected using acceleration transducer, to these initial data application weighting rolling averages
Method is smoothed, based on the motionless theories of steering wheel 4s, the preliminary testing result for judging fatigue driving state；Meanwhile utilize
Pulse transducer collects the pulse data in driver's driving procedure, analyzes and calculates the pulse frequency change of driver, for
Different individuals establishes the threshold value of corresponding normal driving state, is judged by the comparison with this normality threshold currently available
Whether driver's pulse frequency is normal, and then may determine that whether driver is in fatigue driving state；To both recognition results
Decision level fusion is carried out using evidence theory, obtains more accurately whether being in the detection of fatigue driving state on driver
As a result.
The fatigue driving state detection model of the present invention includes pulse data acquisition module, pulse data storage and pretreatment
Module, pulse data dynamic threshold training module, the algorithm application mould based on pulse data detection driver tired driving state
Block, acceleration information acquisition module, acceleration information transmission with pretreatment module, acceleration information dynamic threshold training module,
Algorithm application module, data fusion module based on acceleration information detection driver tired driving state etc..
1st, data acquisition and pretreatment
(1) data acquisition
The advantages of recent movement acceleration is that changing features are obvious, can embody recent movement in real time
Change, and easily catch.The present invention gathers the acceleration of motion of steering wheel by acceleration transducer first, as judgement
One of foundation of driving condition of driver.
Collection for pulse data, carried out using pulse transducer.The wrist oar that pulse transducer is fixed on human body moves
At arteries and veins, because radial artery is the most strong place of human pulse.
(2) data prediction
Acceleration information and pulse data are all the data of time series, exist in the data collected by sensor and make an uproar
Sound.The data collected using sensor are in the prevalence of error, so the data for needing to collect sensor are carried out substantially
Data smoothing processing.The present invention is smoothed using the method for weighted moving average to data.Pulse signal have signal it is weak, frequency
The characteristics of rate is low and noise is strong, noise jamming may result in pulse signal distortion, can cause larger detection error, it is necessary to right
Pulse signal carries out denoising before carrying out feature extraction.Most of pulse signal of human body is distributed in low frequency region, and noise
Signal is generally uniformly distributed in high-frequency region, and the larger Wavelet Component of amplitude typically occurs in sign mutation region.The present invention
Threshold method based on wavelet transformation removes noise.
2nd, dynamic threshold is trained
(1) acceleration dynamic threshold
The method of present invention detection driver fatigue state is based on the motionless theories of steering wheel 4s, and carries out on this basis
Improve, the method for adding dynamic threshold, remove influence of change of the steering wheel with car body angle to testing result, more accurately
Judge the state of driver.
(2) pulse dynamic threshold
The present invention proposes a kind of method based on dynamic threshold detection driver tired driving state, according to Variation of Drivers ' Heart Rate
The degree that cycle declines judges the degree of fatigue of driver：A period of time that driver just starts to drive is usually relatively more clear-headed
, the pulse data of driver is detected during this period and calculates its corresponding heart rate periodic quantity, in this heart rate week
Time value has been crossed after this section of recovery time as the normal heart rate periodic quantity of driver and has continuously detected the pulse of driver and analyze
In the heart rate cycle, compared with the normal heart rate cycle, if declining degree has exceeded more than 10%, system judges that driver is in light
Fatigue state is spent, declines degree more than 20%, system judges that driver is in fatigue state.
3rd, the fatigue driving state detection algorithm based on decision making level data fusion
The present invention is tired by the driver fatigue testing result based on recent movement acceleration and the driver based on pulse
The carry out decision level fusion of labor testing result.
Present invention also offers one kind to be based on decision making level data fusion fatigue state recognition method, and this method includes following step
Suddenly：
Step 1 builds Basic probability assignment function, for two evidence bodies of acceleration and pulse, calculates both respectively
Probability distribution function f, while to ensure to be independent of each other between the two evidence bodies, independently of each other.
The rule of combination of step 2 application evidence theory obtains a new evidence body, and this new evidence body is by accelerating
Degree and pulse the two evidence bodies are combined into what is come, and the basic probability assignment that new evidence body shows shows pair closer to 1
The accuracy that proposition judges is higher.
Step 3 application decision rule, obtains the judgement result of decision on fatigue state and exports.
Step 3 of the present invention uses the decision rule based on probability assignments, for determining based on probability assignments
Plan Rule Expression into：For any set M, ifMeet： If there are f (S1)-f(S2) ＞ θ1, f (M) ＜ θ2, f (S1) ＞ f (M), then S1It is pair
The result of decision of event, wherein θ1And θ2Represent the threshold value of setting.
Beneficial effect：
1st, fatigue driving state Detection accuracy of the invention is higher；Experiment shows the fatigue driving state detection of the present invention
Accuracy rate is then up to 91.67%.
2nd, system response time of the invention is short；Driver's indignation driving condition detecting system has 1s or so delay, main
On spending in pulse data collection and transmitting.
3rd, Space-time Complexity of the invention is low；Time complexity of the present invention is O (n)；Installed System Memory consumes, and Installed System Memory disappears
Consumption is between 50-70M.
Brief description of the drawings
Fig. 1 is the fatigue driving state detection model figure of the present invention.
Fig. 2 is one cycle pulse waveform figure of actual measurement.
Fig. 3 is the decision making level data fusion illustraton of model of the present invention.
Fig. 4 is flow chart of the method for the present invention.
Embodiment
The invention is further detailed with reference to Figure of description.
The motion of fatigue driving state detection method comprehensive utilization steering wheel of the present invention based on decision making level data fusion adds
Speed collateral information and the pulse direct information of driver, the Detection accuracy of fatigue driving state can be effectively improved：Utilize
Acceleration transducer collects recent movement acceleration information, and these initial data application weighting methods of moving average are put down
Sliding processing, based on the motionless theories of steering wheel 4s, the preliminary testing result for judging fatigue driving state；Meanwhile sensed using pulse
Device collects the pulse data in driver's driving procedure, analyzes and calculates the pulse frequency change of driver, for different
Body establishes the threshold value of corresponding normal driving state, and currently available driver's arteries and veins is judged by the comparison with this normality threshold
Whether rate is normal, and then may determine that whether driver is in fatigue driving state；To both recognition result application evidences
Theory carries out decision level fusion, obtains more accurately whether being in the testing result of fatigue driving state on driver.
As shown in figure 1, the present invention system include pulse data acquisition module, pulse data storage with pretreatment module,
Pulse data dynamic threshold training module, the algorithm application module based on pulse data detection driver tired driving state, add
Speed data acquisition module, acceleration information transmission with pretreatment module, acceleration information dynamic threshold training module, based on adding
Speed data detects the algorithm application module of driver tired driving state, data fusion module etc..
1st, data acquisition and pretreatment
(1) data acquisition
The advantages of recent movement acceleration is that changing features are obvious, can embody recent movement in real time
Change, and easily catch.The present invention gathers the acceleration of motion of steering wheel by acceleration transducer first, as judgement
One of foundation of driving condition of driver.
Pulse wave (Pulse Wave) is the pressure wave formed when blood flow is propagated from sustainer along arterial system, and blood
The contraction and diastole that cause sustainer with diastole are shunk in the propagation flowed in arterial system just because of the ventricular cycle of heart.
During each left ventricular contraction, penetrate blood and enter sustainer, expand aorta wall, and when LV Diastolic, aorta wall produces bullet
Property retraction.Pulse is originated in the beating of aortic root and propagated successively along ductus arteriosus wall to each artery of whole body.Pulse reacts
The cyclical upturn and downturn of endaortic pressure.With heart contraction and diastole, artery one opens the beating of a contracting.Under normal circumstances,
Pulse is consistent with heartbeat, and pulse is strong, and the rhythm and pace of moving things is uniform, and strong and weak consistent, interval is equal.The blood stream pumped out by heart is become owner of
Artery, causes the contraction and diastole of sustainer again, and blood flow is passed in the form of pressure wave from aortic root along arterial system
Broadcast, form pulse, heart, which often shrinks diastole, can once produce a cycle pulse wave.Oscillogram shown in Fig. 2 is to be
The cycle pulse wave intercepted in the cycle pulse waveform figure of system collection.
In fig. 2, abscissa T represents time, ordinate P representative pressure values, and the waveform within the time [0, t] is an allusion quotation
The pulse cycle waveform of type, t represent a cycle of pulse wave.H1 represents main wave amplitude, is main crest top to pulse wave figure
The height of baseline, h2 represent wave amplitude before dicrotic pulse, be dicrotic pulse prewave bind reach pulse wave figure baseline height, h3 represent drop in
Gorge amplitude, it is height of the dicrotic notch the lowest point to pulse wave figure baseline, h4 represents dicrotic pulse wave amplitude, for dicrotic pulse crest top to dicrotic notch
Height between the baseline parallel lines that the lowest point is made, t1 represent duration of the pulse wave figure starting point to main wave crest point, and t1 is corresponding left
The phase of maximum ejection of ventricle, t2 represent pulse wave figure starting point and correspond to the systole phase of left ventricle to the duration between dicrotic notch, t2,
T2-t represents dicrotic notch to the duration between pulse wave figure terminating point this period, and the diastole of corresponding left ventricle, 0-t represents arteries and veins
Fight ripple figure starting point to the duration between terminating point, t corresponds to a cardiac cycle of left ventricle, corresponding to pulse, and
The cycle of one pulse.
Physiologic meaning corresponding to each characteristic point is as follows in Fig. 2：
P1 points：Sustainer opening point is represented, that is, begin exit point.It is the minimum point of whole pulse waveform figure, indicates that heart is fast
The end of term endovascular pressure and volume are shunk in the beginning of rapid fire blood phase, main reflection.
P2 points：Aortic pressure peak.It is main ripple herein, is that a rising on baseline to the main crest top of oscillogram is bent
The maximum of line, summit reflection Intraarterial pressure and volume, forms the ascending branch of main ripple, reflects the fast rapid fire blood of ventricle, angiosthenia
Rapid to rise, tube wall is expanded suddenly.Its rate of climb mainly with cardiac output, Ve speed, Artery resistance and tube wall bullet
Property is relevant, can be represented with ascending branch slope.If cardiac output is more, blood speed is penetrated, aorta elasticity reduces, then tiltedly
Rate is larger, and wave amplitude is higher；If cardiac output is less, penetrate that blood speed is slower, and aorta elasticity is larger, then slope reduces, wave amplitude
It is relatively low.
P4 points：The left heart penetrates blood halt, is herein tidal wave, is also dicrotic wave prewave.Positioned at the decent of oscillogram, typically
Delay after main ripple, less than main ripple, position is higher than dicrotic wave.It is to stop penetrating blood, artery in later stage reduced ejection period ventricle
Expansion, drop in blood pressure, the reverse flow of intra-arterial blood and the back wave that is formed, mainly with peripheral resistance, blood vessel elasticity and descending branch
The paces of change such as decrease speed are relevant.
P5 points：Dicrotic wave trough, it is the main ripple decent incisura ripple downward with the waveform of dicrotic wave ascending branch composition.It leads
Sustainer static pressure emptying time is reacted, is the separation of heart contraction and diastole, easily by peripheral resistance and descending branch decrease speed
Influence.
P6 points：Aorta elasticity retraction ripple, i.e. dicrotic wave.It is to be located at a prominent small echo after dicrotic wave trough,
Its formation is after ventricle reduced ejection period, and ventricle starts diastole, and intraventricular pressure is dropped rapidly to significantly lower than aortic pressure, main
Endarterial blood starts to backflow to ventricle direction.Because of the impact for the blood that backflows, aorta petal is closed suddenly, and the blood to backflow is hit
Hit on the aorta petal closed suddenly and be shot back, aortic pressure is slightly increased again, ductus arteriosus wall also slightly expands therewith
.Therefore, a upward small echo, i.e. dicrotic wave are formed in the stage casing of decent.It can react aorta petal function status,
Blood vessel elasticity and blood flow flow regime.
Collection for pulse data, carried out using pulse transducer.The wrist oar that pulse transducer is fixed on human body moves
At arteries and veins, because radial artery is the most strong place of human pulse, the convenient pulse data for accurately obtaining human body, and wear conveniently.
(2) data prediction
Acceleration information and pulse data are all the data of time series, exist in the data collected by sensor and make an uproar
Sound.The data collected using sensor are in the prevalence of error, so the data for needing to collect sensor are carried out substantially
Data smoothing processing.The system of the present invention gathers direction by acceleration information acquisition module using acceleration transducer first
The acceleration of motion of disk, as one of foundation of driving condition for judging driver；Pulse is utilized by pulse data acquisition module
Collection of the sensor for pulse data, pulse transducer are fixed at the wrist radial artery of human body；Acceleration information transmit with
Pretreatment module and pulse data storage are smoothed with pretreatment module using the method for weighted moving average to data, more
Point weights close to smooth window edge are smaller, and the point into smooth window is all little by little to be included in average value, gradually eliminate
Influence to overall smoothness：
Wherein
In formula (1), siRepresent i-th point of smooth value；xi+jRepresent data point；wjWeights are represented, close to middle weights
Larger, the weights close to edge are smaller, and weights summation is 1.In order to compactly obtain more preferable data smoothing treatment effect, this hair
Method of the bright selection method of weighted moving average as acceleration information smoothing processing, such as uses (1/4,1/2,1/4) to be used as weights,
For the recent movement acceleration information collected, adjacent three o'clock as a processing item, the result after smoothing processing
Update to data item corresponding to centre, circular treatment, until having handled data.
Pulse signal has the characteristics of signal is weak, frequency is low and noise is strong, and the interference of body state and external environment all can be right
The collection of physiology signal produces bigger influence.These noise jammings may result in pulse signal distortion, can cause
Larger detection error before pulse signal progress feature extraction, it is necessary to carrying out denoising.Most of pulse signal of human body
Low frequency region is distributed in, and noise signal is generally uniformly distributed in high-frequency region, the larger Wavelet Component of amplitude generally occurs within
In sign mutation region.Threshold method of the present invention based on wavelet transformation removes noise.
Firstly, it is necessary to wavelet transform is carried out to pulse signal.One typical wavelet is：
Assuming thatIt isFourier transformation,Referred to as wavelet mother function.By to small echo
Generating functionDiscrete wavelet race can be obtained by translating and stretching：
In formula (3), a is contraction-expansion factor, and b is shift factor.
Assuming that pulse signal is signal (t), signal (t)=start (t)+noise (t), start (t) represents original
Pulse signal, noise (t) represents noise.Discrete sampling is carried out to pulse signal：Signal (t), t=0,1,2 ..., N-
1。
The coefficient of wavelet transformation is：
Wavelet coefficient W is obtained by formula (4)signalAfter (a, b), handled with threshold value, determine the estimate of wavelet coefficientEnsureValue it is minimum, threshold value uses generic threshold value choosing method：
In formula (5), med represents the intermediate value of high frequency orthogonal wavelet coefficient.
The estimate of wavelet coefficient is obtainedWith wavelet inverse transformation to wavelet reconstruction, obtain estimating signalIt is exactly the pulse signal after denoising.
2nd, dynamic threshold is trained
(1) acceleration dynamic threshold
The moving acceleration data of steering wheel is a kind of data flow of time series, and the present invention is ground based on sliding window model
Study carefully the acceleration of motion of steering wheel.Sliding window model is the common model of processing time sequence data stream.Sliding window model
The characteristics of be that window size where the data of its processing is fixed, and the terminal of sliding window is always current time, this limit
Surely the valid data that also ensure that sliding window model processing are the data in data flow in most newly arrived window forever.
The method of present invention detection driver fatigue state is based on the motionless theories of steering wheel 4s, and carries out on this basis
Improve, the method for adding dynamic threshold, remove influence of change of the steering wheel with car body angle to testing result, more accurately
Judge the state of driver.
Because the angle of steering wheel in actual conditions and the situation on the road surface residing for vehicle are different, it is impossible to using fixation
Foundation of the threshold value as detection steering wheel position, the present invention propose a kind of side of dynamic training recent movement acceleration rate threshold
Method：Analyze whether vehicle is in metastable transport condition by the acceleration information of steering wheel first, during this period of time
Monitoring direction is faced left right continuous time of the fluctuation less than 15 degree, and this continuous time of decile is multiple continuous periods, is obtained
The weighted average of the acceleration information of steering wheel in these periods；Then the average value of these weighted averages is taken, is obtained
Driver obtains obtaining the waving interval of steering wheel by comparing acceleration information after threshold value in the dynamic threshold of current road segment,
The motionless theories of application direction disk 4s, judge that driver is in fatigue driving state when the continuous 4s of steering wheel is motionless.
(2) pulse dynamic threshold
The pulse of normal person is consistent with heartbeat, is 60-100 times per minute, usually per minute 70-80 times, is put down
It is about per minute 72 times.The elderly is slower, is 55 to 60 beats/min.Normal person's pulse frequency rule, be not in pulse interval time length
The short phenomenon to differ.Normal person's pulse is strong and weak impartial, is not in strong and weak alternate phenomenon.The frequency of pulse is by age and sex
Influence, in addition, motion and it is excited when can speed pulse, and rest, sleeping then slows down pulse.
In order to obtain the relation of pulse frequency and fatigue, the present invention has carried out contrast experiment's test.Experimenter amounts to 12, is
The adult of health, the age is between 23 to 26 years old.Experimenter measures 5 groups of pulse values when being in waking state；It is real
The person of testing is in when having fatigue state and measures 5 groups of pulse values again；Experimenter continues 5 groups of measurement when being in more tired
Pulse value.According to the pulse value of measurement, corresponding pulse frequency is calculated, pulse frequency of the experimenter under different conditions is counted and becomes
Change situation, as shown in table 1.
Table 1：The pulse frequency change of experimenter under different conditions
As shown in Table 1, pulse frequency and the fatigue of human body have close relation.When experimenter, which is in, fatigue state,
The amount of decrease of pulse frequency is between 8.57% -12.50%, and major part is more than 10%；When experimenter is in fatigue state, arteries and veins
The amount of decrease of rate is between 19.44% -24.66, and major part is more than 20%.The conclusion drawn according to experiment, the present invention propose
A kind of method based on dynamic threshold detection driver tired driving state, the degree declined according to the Variation of Drivers ' Heart Rate cycle judge
The degree of fatigue of driver：A period of time that driver just starts to drive is usually relatively more clear-headed, is detected during this period
Go out the pulse data of driver and calculate its corresponding heart rate periodic quantity, this heart rate periodic quantity is normal as driver
Heart rate periodic quantity, cross after this section of recovery time and continuously detected the pulse of driver and analyze the heart rate cycle, with the normal heart
The rate cycle compares, if declining degree has exceeded more than 10%, system judges that driver is in slight fatigue state, declines degree
More than 20%, system judges that driver is in fatigue state.
3rd, the fatigue driving state detection algorithm based on decision making level data fusion
The present invention is tired by the driver fatigue testing result based on recent movement acceleration and the driver based on pulse
The carry out decision level fusion of labor testing result.
Theory of the invention based on finite aggregate Θ, what Θ was represented is a framework of identification, complete comprising wanting system to be detected
Body object, it is the relation of mutual exclusion between object；Θ represents the set of tired and not tired two objects in the present invention.That is Θ=
{ fatigue, not tired }.
If θ subset is 2θ, f is 2θTo the mapping function of [0,1], and meet f (Φ)=0, to arbitrary S ∈ 2Θ, there is f
>=0 and ∑ f (s)=1 (s).F (s) represents the elementary probability value of an identification framework, reflects the size to s reliabilities.f1() and f2
() represents Basic probability assignment function derived from two independent evidence sources, is examined based on acceleration corresponding in the present invention
The driver tired driving state measured and driver tired driving state the two the evidence bodies gone out based on pulse detection.Meter
A Basic probability assignment function is calculated, to reflect two coefficient fuse informations of evidence body：
Order
Then
The present invention uses the decision rule based on probability assignments, can be expressed as the decision rule based on probability assignments
It is as follows：
For any set M, ifMeet： If there are f (S1)-f(S2) ＞ θ1, f (M) ＜ θ2, f (S1) ＞ f (M), then S1It is
To the result of decision of event, wherein θ1And θ2Represent the threshold value of setting.
As shown in figure 4, the invention provides a kind of fatigue driving state detecting system based on decision making level data fusion
Implementation method, this method comprise the following steps：
Step 1 builds Basic probability assignment function, for two evidence bodies of acceleration and pulse, calculates both respectively
Probability distribution function f, while to ensure to be independent of each other between the two evidence bodies, independently of each other.
The rule of combination of step 2 application evidence theory obtains a new evidence body, and this new evidence body is by accelerating
Degree and pulse the two evidence bodies are combined into what is come, and the basic probability assignment that new evidence body shows shows pair closer to 1
The accuracy that proposition judges is higher.
Step 3 application decision rule, obtains the judgement result of decision on fatigue state and exports.
Claims (4)
1. a kind of fatigue driving state detecting system based on decision making level data fusion, it is characterised in that the system includes adding
Speed data acquisition module, acceleration information transmission with pretreatment module, acceleration information dynamic threshold training module, based on adding
The speed data detection algorithm application module of driver tired driving state, pulse data acquisition module, pulse data storage with
Pretreatment module, pulse data dynamic threshold training module, the algorithm based on pulse data detection driver tired driving state
Application module, data fusion module, the system efficiently can easily detect driver fatigue state, and the system is logical first
The moving acceleration data of acceleration transducer collection steering wheel is crossed, driver's pulse data is gathered by pulse transducer, point
It is other that both data are pre-processed, dynamic threshold is calculated respectively to pretreated two kinds of data, obtained on driving
Whether member is in the Preliminary detection result of fatigue state, by carrying out decision level fusion to two kinds of testing results, is more defined
Testing result after true fusion；
The function of acceleration information acquisition module is：Recent movement acceleration information is gathered using acceleration transducer；
Acceleration information transmits is with the function of pretreatment module：To the recent movement acceleration gathered with acceleration transducer
The initial data application weighting method of moving average is smoothed；
The function of acceleration information dynamic threshold training module is：Treated to being transmitted by acceleration information with pretreatment module
Data calculate average value again, obtain dynamic threshold of the driver in current road segment, then obtained by comparing acceleration information
The waving interval of steering wheel；
The function of the algorithm application module of driver tired driving state is detected based on acceleration information is：Using steering wheel 4s not
Dynamic theory, the preliminary testing result for judging fatigue driving state；
The function of pulse data acquisition module is：Gathered and driven based on the pulse transducer being fixed at the wrist radial artery of human body
Pulse data in member's driving procedure；
Pulse data stores is with the function of pretreatment module：The pulse data collected is stored, is then based on wavelet transformation
Threshold method removes pulse signal noise, reuses the method for weighted moving average and data are smoothed；
The function of pulse data dynamic threshold training module is：Analyze and calculate the pulse frequency change of driver, for different
Individual establishes the threshold value of corresponding normal driving state；
The function of the algorithm application module of driver tired driving state is detected based on pulse data is：By with normal driving shape
The threshold value of state relatively judges whether currently available driver's pulse frequency is normal, and then judges whether driver drives in fatigue
Sail state；
The function of data fusion module is：To the algorithm application module based on acceleration information detection driver tired driving state
Carried out with the recognition result application evidence theory of the algorithm application module based on pulse data detection driver tired driving state
Decision level fusion, whether driver is obtained in fatigue by the fatigue driving state detection algorithm based on decision making level data fusion
The testing result of driving condition.
2. a kind of fatigue driving state detecting system based on decision making level data fusion according to claim 1, its feature
It is：The system is accelerated by acceleration information acquisition module using acceleration transducer to gather the motion of steering wheel first
Degree, as one of foundation of driving condition for judging driver；By pulse data acquisition module using pulse transducer to pulse
Data are acquired, and pulse transducer is fixed at the wrist radial artery of human body；Acceleration information transmit with pretreatment module and
Pulse data storage is smoothed with pretreatment module using the method for weighted moving average to data, closer to smooth window
The point weights at edge are smaller, and the point into smooth window is all little by little to be included in average value, gradually eliminate to overall smooth journey
The influence of degree：
Wherein
In formula (1), siRepresent i-th point of smooth value；xi+jRepresent data point；wjWeights are represented, it is larger close to middle weights,
Weights close to edge are smaller, and weights summation is 1；Select side of the method for weighted moving average as acceleration information smoothing processing
Method, for the recent movement acceleration information collected, adjacent three o'clock as a processing item, the knot after smoothing processing
Data item corresponding to centre, circular treatment, until having handled data are arrived in fruit renewal.
3. a kind of fatigue driving state detecting system based on decision making level data fusion according to claim 1, its feature
It is：Threshold method of the system based on wavelet transformation removes pulse signal noise, including：
Firstly, it is necessary to carry out wavelet transform to pulse signal, wavelet is：
Assuming that It isFourier transformation,Referred to as wavelet mother function, by wavelet mother functionDiscrete wavelet race can be obtained by translating and stretching：
In formula (3), a is contraction-expansion factor, and b is shift factor；
Assuming that pulse signal is signal (t), signal (t)=start (t)+noise (t), start (t) represents original arteries and veins
Fight signal, noise (t) represents noise, and discrete sampling is carried out to pulse signal：Signal (t), t=0,1,2 ..., N-1；
The coefficient of described wavelet transformation is：
<mrow>
<msub>
<mi>W</mi>
<mrow>
<mi>s</mi>
<mi>i</mi>
<mi>g</mi>
<mi>n</mi>
<mi>a</mi>
<mi>l</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>a</mi>
<mo>,</mo>
<mi>b</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msup>
<mn>2</mn>
<mfrac>
<mi>a</mi>
<mn>2</mn>
</mfrac>
</msup>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>n</mi>
<mo>=</mo>
<mn>0</mn>
</mrow>
<mrow>
<mi>N</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</munderover>
<mi>f</mi>
<mrow>
<mo>(</mo>
<mi>n</mi>
<mo>)</mo>
</mrow>
<mrow>
<mo>(</mo>
<msup>
<mn>2</mn>
<mi>a</mi>
</msup>
<mi>n</mi>
<mo>-</mo>
<mi>b</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>4</mn>
<mo>)</mo>
</mrow>
</mrow>
Wavelet coefficient W is obtained by formula (4)signalAfter (a, b), handled with threshold value, determine the estimate of wavelet coefficientEnsureValue it is minimum, threshold value uses generic threshold value choosing method, including：
<mrow>
<mi>T</mi>
<mo>=</mo>
<mi>m</mi>
<mi>e</mi>
<mi>d</mi>
<mo>/</mo>
<mn>0.6475</mn>
<msqrt>
<mrow>
<mn>2</mn>
<mi>l</mi>
<mi>n</mi>
<mi> </mi>
<mi>N</mi>
</mrow>
</msqrt>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>5</mn>
<mo>)</mo>
</mrow>
</mrow>
In formula (5), med represents the intermediate value of high frequency orthogonal wavelet coefficient；
The estimate of wavelet coefficient is obtainedWith wavelet inverse transformation to wavelet reconstruction, obtain estimating signalIt is exactly the pulse signal after denoising.
4. a kind of fatigue driving state detecting system based on decision making level data fusion according to claim 1, its feature
It is：Analyze whether vehicle is in phase by the acceleration information of the steering wheel collected by acceleration information acquisition module first
To stable transport condition, during this period of time monitoring direction is faced left right continuous time of the fluctuation less than 15 degree, decile this connect
The continuous time is multiple continuous periods, and steering wheel in these periods is obtained with pretreatment module by acceleration information transmission
The weighted average of acceleration information；Then being averaged for these weighted averages is taken by acceleration information dynamic threshold training module
Value, obtains dynamic threshold of the driver in current road segment, obtains obtaining the ripple of steering wheel by comparing acceleration information after threshold value
Dynamic section, then the algorithm application module by detecting driver tired driving state based on acceleration information are motionless according to steering wheel 4s
Theory, judge that driver is in fatigue driving state when the continuous 4s of steering wheel is motionless.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510249302.7A CN104952210B (en) | 2015-05-15 | 2015-05-15 | A kind of fatigue driving state detecting system and method based on decision making level data fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510249302.7A CN104952210B (en) | 2015-05-15 | 2015-05-15 | A kind of fatigue driving state detecting system and method based on decision making level data fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104952210A CN104952210A (en) | 2015-09-30 |
| CN104952210B true CN104952210B (en) | 2018-01-05 |
Family
ID=54166832
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510249302.7A Expired - Fee Related CN104952210B (en) | 2015-05-15 | 2015-05-15 | A kind of fatigue driving state detecting system and method based on decision making level data fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104952210B (en) |
Families Citing this family (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105261153A (en) * | 2015-11-03 | 2016-01-20 | 北京奇虎科技有限公司 | Vehicle running monitoring method and device |
| CN105405253B (en) * | 2015-12-18 | 2017-05-24 | 中交第一公路勘察设计研究院有限公司 | Method and apparatus for monitoring fatigue state of driver |
| CN105632103A (en) * | 2016-03-11 | 2016-06-01 | 张海涛 | Method and device for monitoring fatigue driving |
| CN106384129B (en) * | 2016-09-13 | 2018-07-20 | 西安科技大学 | A kind of driver's driving condition discrimination method based on closed loop driving model |
| CN106446849B (en) * | 2016-09-30 | 2019-08-23 | 福建省福信富通网络科技股份有限公司 | A kind of method for detecting fatigue driving |
| CN106548132A (en) * | 2016-10-16 | 2017-03-29 | 北海益生源农贸有限责任公司 | The method for detecting fatigue driving of fusion eye state and heart rate detection |
| CN106539581B (en) * | 2016-12-07 | 2019-08-20 | 中国民用航空总局第二研究所 | Controller's fatigue detection method and system based on probabilistic method |
| CN106851478A (en) * | 2017-02-10 | 2017-06-13 | 深圳市笨笨机器人有限公司 | Multi-channel information processing method and system |
| US20210101604A1 (en) * | 2017-03-28 | 2021-04-08 | Kyushu Institute Of Technology | Driver state detection device |
| CN106874900A (en) * | 2017-04-26 | 2017-06-20 | 桂林电子科技大学 | A kind of tired driver detection method and detection means based on steering wheel image |
| JP6509940B2 (en) * | 2017-05-10 | 2019-05-08 | 本田技研工業株式会社 | Driving support device and driving support method |
| CN108926352B (en) * | 2017-05-22 | 2020-10-09 | 北京大学 | Driving fatigue detection method and system |
| CN107233103B (en) * | 2017-05-27 | 2020-11-20 | 西南交通大学 | High-speed rail dispatcher fatigue state evaluation method and system |
| CN107212870A (en) * | 2017-06-28 | 2017-09-29 | 上海电力学院 | A kind of ear clip device for detecting driver status |
| CN107472130B (en) * | 2017-08-04 | 2020-09-11 | 山东理工大学 | Method and system for monitoring dangerous driving state of driver |
| DE112017008148B4 (en) * | 2017-11-21 | 2021-12-16 | Mitsubishi Electric Corporation | Anomaly detection device and anomaly detection method |
| CN108682119B (en) * | 2018-05-29 | 2020-05-26 | 重庆大学 | Driver fatigue state detection method based on smart phone and smart watch |
| CN108765876A (en) * | 2018-05-31 | 2018-11-06 | 东北大学 | Driving fatigue depth analysis early warning system based on multimode signal and method |
| CN108995531A (en) * | 2018-08-09 | 2018-12-14 | 爱驰汽车有限公司 | fatigue warning device and method for automobile |
| CN109389806B (en) * | 2018-11-08 | 2020-07-24 | 山东大学 | Fatigue driving detection early warning method, system and medium based on multi-information fusion |
| CN109993119A (en) * | 2019-03-31 | 2019-07-09 | 广东乐之康医疗技术有限公司 | A kind of data collection and learning method based on wearable device |
| CN110101372A (en) * | 2019-04-24 | 2019-08-09 | 上海工程技术大学 | A kind of municipal rail train driver physiological status monitoring system |
| CN110334592A (en) * | 2019-05-27 | 2019-10-15 | 天津科技大学 | A kind of monitoring of driver's abnormal behaviour and safety control system and safety control method |
| CN110464371A (en) * | 2019-08-29 | 2019-11-19 | 苏州中科先进技术研究院有限公司 | Method for detecting fatigue driving and system based on machine learning |
| CN114073513A (en) * | 2020-08-10 | 2022-02-22 | 安徽华米健康科技有限公司 | Detection method and device for getting up at night during sleep and intelligent wearable equipment |
| CN112233276B (en) * | 2020-10-13 | 2022-04-29 | 重庆科技学院 | Steering wheel corner statistical characteristic fusion method for fatigue state recognition |
| CN113642522B (en) * | 2021-09-01 | 2022-02-08 | 中国科学院自动化研究所 | Audio and video based fatigue state detection method and device |
| CN116382488B (en) * | 2023-06-01 | 2023-10-27 | 隽智生物医学研究实验室(佛山)有限公司 | Human-computer interaction intelligent regulation and control decision system and method based on human body state identification |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101090482A (en) * | 2006-06-13 | 2007-12-19 | 唐琎 | Driver fatigue monitoring system and method based on image process and information mixing technology |
| CN102509418A (en) * | 2011-10-11 | 2012-06-20 | 东华大学 | Fatigue driving estimation and early-warning method and device of multi-sensor information fusion |
| CN203252643U (en) * | 2013-05-20 | 2013-10-30 | 江西理工大学 | Intelligent sleep prevention device by detecting driver pulse |
| CN103714660A (en) * | 2013-12-26 | 2014-04-09 | 苏州清研微视电子科技有限公司 | System for achieving fatigue driving judgment on basis of image processing and fusion between heart rate characteristic and expression characteristic |
| CN104112335A (en) * | 2014-07-25 | 2014-10-22 | 北京机械设备研究所 | Multi-information fusion based fatigue driving detecting method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011008457A (en) * | 2009-06-25 | 2011-01-13 | Hitachi Ltd | Automobile driver doze prevention device |
-
2015
- 2015-05-15 CN CN201510249302.7A patent/CN104952210B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101090482A (en) * | 2006-06-13 | 2007-12-19 | 唐琎 | Driver fatigue monitoring system and method based on image process and information mixing technology |
| CN102509418A (en) * | 2011-10-11 | 2012-06-20 | 东华大学 | Fatigue driving estimation and early-warning method and device of multi-sensor information fusion |
| CN203252643U (en) * | 2013-05-20 | 2013-10-30 | 江西理工大学 | Intelligent sleep prevention device by detecting driver pulse |
| CN103714660A (en) * | 2013-12-26 | 2014-04-09 | 苏州清研微视电子科技有限公司 | System for achieving fatigue driving judgment on basis of image processing and fusion between heart rate characteristic and expression characteristic |
| CN104112335A (en) * | 2014-07-25 | 2014-10-22 | 北京机械设备研究所 | Multi-information fusion based fatigue driving detecting method |
Non-Patent Citations (1)
| Title |
|---|
| 基于多传感器数据融合的驾驶状态监控研究;李埃荣;《中国优秀硕士学位论文全文数据库 信息科技辑》;20080415(第4期);第12-57页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104952210A (en) | 2015-09-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104952210B (en) | A kind of fatigue driving state detecting system and method based on decision making level data fusion | |
| Mendonca et al. | A review of obstructive sleep apnea detection approaches | |
| US11311201B2 (en) | Feature selection for cardiac arrhythmia classification and screening | |
| Boonnithi et al. | Comparison of heart rate variability measures for mental stress detection | |
| US9655559B2 (en) | Automated sleep staging using wearable sensors | |
| CN108765876A (en) | Driving fatigue depth analysis early warning system based on multimode signal and method | |
| Vicente et al. | Detection of driver's drowsiness by means of HRV analysis | |
| CN102138789B (en) | Dynamic electrocardiogram and motion recording and analyzing system | |
| EP3843623B1 (en) | Photoplethysmography based detection of transitions between awake, drowsiness, and sleep phases of a subject | |
| US10582862B1 (en) | Determination and monitoring of basal heart rate | |
| US20110319724A1 (en) | Methods and systems for non-invasive, internal hemorrhage detection | |
| EP2765905A1 (en) | Seizure detection methods, apparatus, and systems using a wavelet transform maximum modulus or autoregression algorithm | |
| Singh et al. | An approach for real-time stress-trend detection using physiological signals in wearable computing systems for automotive drivers | |
| Singh et al. | Assessment of driver stress from physiological signals collected under real-time semi-urban driving scenarios | |
| Singh et al. | Biosignal based on-road stress monitoring for automotive drivers | |
| Nguyen et al. | Neural network approach for non-invasive detection of hyperglycemia using electrocardiographic signals | |
| Luguev et al. | Deep learning based affective sensing with remote photoplethysmography | |
| Radha et al. | LSTM knowledge transfer for HRV-based sleep staging | |
| EP3387994B1 (en) | Biological state estimation device, biological state estimation method, computer program, and recording medium | |
| Malykhina et al. | Digitalization of medical services for detecting violations of cerebrovascular regulation based on a neural network signal analysis algorithm | |
| Clark et al. | Machine learning based prediction of future stress events in a driving scenario | |
| Jegan et al. | Mental Stress Detection and Classification using SVM Classifier: A Pilot Study | |
| Widasari et al. | A new investigation of automatic sleep stage detection using decision-tree-based support vector machine and spectral features extraction of ecg signal | |
| Chen et al. | Beat-to-beat heart rate detection based on seismocardiogram using BiLSTM network | |
| Trardi et al. | Computationally efficient algorithm for atrial fibrillation detection using linear and geometric features of RR time-series derivatives |
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 | ||
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20150930 Assignee: NUPT INSTITUTE OF BIG DATA RESEARCH AT YANCHENG Assignor: NANJING University OF POSTS AND TELECOMMUNICATIONS Contract record no.: X2020980007071 Denomination of invention: A fatigue driving state detection system and method based on decision level data fusion Granted publication date: 20180105 License type: Common License Record date: 20201026 |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180105 |