CN108154095A - A kind of method, apparatus and vehicle of determining fatigue driving - Google Patents
A kind of method, apparatus and vehicle of determining fatigue driving Download PDFInfo
- Publication number
- CN108154095A CN108154095A CN201711340449.2A CN201711340449A CN108154095A CN 108154095 A CN108154095 A CN 108154095A CN 201711340449 A CN201711340449 A CN 201711340449A CN 108154095 A CN108154095 A CN 108154095A
- Authority
- CN
- China
- Prior art keywords
- fatigue
- probability
- tired
- condition probability
- expression
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/59—Context or environment of the image inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions
- G06V20/597—Recognising the driver's state or behaviour, e.g. attention or drowsiness
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
- A61B5/14542—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue for measuring blood gases
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
- A61B5/18—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state for vehicle drivers or machine operators
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/6802—Sensor mounted on worn items
- A61B5/6803—Head-worn items, e.g. helmets, masks, headphones or goggles
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/74—Details of notification to user or communication with user or patient ; user input means
- A61B5/746—Alarms related to a physiological condition, e.g. details of setting alarm thresholds or avoiding false alarms
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/08—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to drivers or passengers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/174—Facial expression recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/18—Eye characteristics, e.g. of the iris
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/06—Alarms for ensuring the safety of persons indicating a condition of sleep, e.g. anti-dozing alarms
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/08—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to drivers or passengers
- B60W2040/0818—Inactivity or incapacity of driver
- B60W2040/0827—Inactivity or incapacity of driver due to sleepiness
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Surgery (AREA)
- Pathology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Animal Behavior & Ethology (AREA)
- Biomedical Technology (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Physiology (AREA)
- Psychiatry (AREA)
- Child & Adolescent Psychology (AREA)
- Educational Technology (AREA)
- Hospice & Palliative Care (AREA)
- Psychology (AREA)
- Social Psychology (AREA)
- Automation & Control Theory (AREA)
- Mathematical Physics (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Optics & Photonics (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Developmental Disabilities (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Ophthalmology & Optometry (AREA)
- Artificial Intelligence (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
This disclosure relates to a kind of method, apparatus and vehicle of determining fatigue driving, including:Obtain the brain blood oxygen scaling exponent of driver, face fatigue expression and coreclisis degree;First fatigue condition probability is determined according to the brain blood oxygen scaling exponent, the second fatigue condition probability is determined according to the tired expression of the face, third fatigue condition probability is determined according to the coreclisis degree;According to fatigue driving prior probability, the first fatigue condition probability, the second fatigue condition probability, which determines fatigue driving posterior probability;Wherein, which is preceding once to test the driver and drive the tired probability that fatigue conditions obtain.Since the disclosure determines that driver is in the probability of fatigue driving using three kinds of indexs simultaneously, thus judge more comprehensively, more accurately.
Description
Technical field
This disclosure relates to field of traffic, more particularly to a kind of method, apparatus and vehicle of determining fatigue driving.
Background technology
Vehicle drive is a kind of potentially dangerous behavior, and there are many kinds of the factors for causing traffic accident, for example, common
Factor include it is following several:
1st, vehicle is popularized, and road conditions are more complicated;
2nd, vehicle performance is promoted, and speed is faster;
3rd, city life rhythm faster, causes tired driver;
4th, aging trend, person for driving a car's average age rise, and reaction declines;
5th, carrier dog-eat-dog, a large amount of practitioners are in fatigue driving state.
Wherein, fatigue driving becomes the first main road killer.When driver drives, when situation occurs in road, need one
Fix time it is interior make operation in time, be otherwise susceptible to accident.When driver is in fatigue driving state, under reagency is serious
Drop, when reacting, often accident is inevitable.
Therefore, judge that driver whether in fatigue driving state, becomes a kind of demand.Existing judgment method judges driver
The state of mind, frequently with single index, such as blood oxygen saturation, tired Expression analysis.Single method judges driver's spirit shape
State, the probability of erroneous judgement are bigger.
Invention content
The purpose of the disclosure is to provide a kind of method, apparatus and vehicle of determining fatigue driving.
To achieve these goals, the disclosure provides a kind of method of determining fatigue driving, and this method includes:It obtains and drives
The brain blood oxygen scaling exponent of member, face fatigue expression and coreclisis degree;Determine that first is tired according to the brain blood oxygen scaling exponent
Labor conditional probability determines the second fatigue condition probability according to facial tired expression, determines that third is tired according to the coreclisis degree
Labor conditional probability;According to fatigue driving prior probability, the first fatigue condition probability, the second fatigue condition probability and
The third fatigue condition probability, determines fatigue driving posterior probability wherein, and the fatigue prior probability is preceding primary test institute
Driver is stated to drive the tired probability that fatigue conditions obtain.
Since the disclosure using three kinds of indexs with determining that driver is in the probability of fatigue driving simultaneously, thus judge is more comprehensively
With it is accurate.
Optionally, the first fatigue condition probability is determined according to the brain blood oxygen scaling exponent, including:Obtain the driver
Strontium dioxide index in specific time generates strontium dioxide time series;By the strontium dioxide time series average mark
For the isometric subinterval of the first quantity;By the way that trend fluction analysis method is gone to handle each isometric subinterval, obtain
Obtain each corresponding brain blood oxygen scaling exponent in the isometric subinterval;By the corresponding brain blood oxygen in each isometric subinterval
Scaling exponent is compared respectively with cognitive difficulties threshold value, determines the corresponding tired interval number in all tired subintervals;Wherein, institute
The corresponding brain blood oxygen scaling exponent in tired subinterval is stated less than cognitive difficulties threshold value;First fatigue is determined according to following manner
Conditional probability:
Wherein, p1For the first fatigue condition probability, m1For the tired interval number, n1For first quantity.
Due to present disclose provides a kind of more accurate method of the prior probability of the brain blood oxygen index of determining driver, thus
The driver tired driving posterior probability obtained by the prior probability of brain blood oxygen index is more accurate.
Optionally, the cognitive difficulties threshold value is 1.6.
Present disclose provides it is a kind of by actual test verification for judging whether brain blood oxygen scaling exponent represents fatigue
Threshold value, thus judge fatigue accuracy higher.
Optionally, it is described that second fatigue condition probability is determined according to facial tired expression, including:Obtain the driver's
Expression real-time described in every frame and the tired expression in tired expression library are compared, really by the real-time expression of the second quantity frame number
The fixed tired frame number being consistent with the tired expression;The second fatigue condition probability is determined according to following manner:
Wherein, p2For the second fatigue condition probability, m2For the tired frame number, n2For second quantity.
Due to present disclose provides a kind of more accurate method of the prior probability of the tired expression of the face of determining driver, because
It is and more accurate by the driver tired driving posterior probability that the prior probability of facial tired expression obtains.
Optionally, it is described that third fatigue condition probability is determined according to the coreclisis degree, including:Obtain third number of frames
The ocular image of several drivers;By ocular image described in bright pupil Algorithm Analysis, the driver is determined
Eye be closed closure frame number;The third fatigue condition probability is determined according to following manner:
Wherein, p3For the third fatigue condition probability, m3For the tired frame number, n3For second quantity.
Due to present disclose provides a kind of more accurate method of the prior probability of the coreclisis degree of determining driver, thus
The driver tired driving posterior probability obtained by the prior probability of coreclisis degree is more accurate.
Optionally, according to fatigue driving prior probability, the first fatigue condition probability, the second fatigue condition probability
And the third fatigue condition probability, determine fatigue driving posterior probability, including:The fatigue is calculated according to equation below to drive
Sail posterior probability:
f1=f0*p1*p2*p3/(p1*p2*p3+(1-p1)*(1-p2)*(1-p3));
Wherein, f0For the fatigue driving prior probability, f1For the fatigue driving posterior probability, p1It is tired for described first
Labor conditional probability, p2For the first fatigue condition probability, p3For the first fatigue condition probability, p1For the described first fatigue
Conditional probability.
It is general that the disclosure provides a kind of prior probability calculating driver tired driving posteriority more fully by three kinds of indexs
The method of rate, calculating more fatigue driving posterior probability is accurate, whether can more accurately obtain driver in fatigue driving shape
The conclusion of state.
The disclosure provides a kind of device of determining fatigue driving, which includes:Acquisition module, for obtaining driver's
Brain blood oxygen scaling exponent, face fatigue expression and coreclisis degree;Processing module, for true according to the brain blood oxygen scaling exponent
Fixed first fatigue condition probability determines the second fatigue condition probability according to facial tired expression, true according to the coreclisis degree
Determine third fatigue condition probability;Determination module, for according to fatigue driving prior probability, the first fatigue condition probability, institute
The second fatigue condition probability is stated, the third fatigue condition probability determines fatigue driving posterior probability;Wherein, the fatigue is first
It is preceding once to test the driver and drive the tired probability that fatigue conditions obtain to test probability.
Optionally, the processing module, is used for:Strontium dioxide index of the driver in specific time is obtained, it is raw
Into strontium dioxide time series;The strontium dioxide time series is equally divided into the isometric subinterval of the first quantity;By going
Trend fluction analysis method handles each isometric subinterval, obtains the corresponding brain in each isometric subinterval
Blood oxygen scaling exponent;The corresponding brain blood oxygen scaling exponent in each isometric subinterval is carried out respectively with cognitive difficulties threshold value
Compare, determine the corresponding tired interval number in all tired subintervals;Wherein, the corresponding brain blood oxygen scale in the tired subinterval refers to
Number is less than cognitive difficulties threshold value;The first fatigue condition probability is determined according to following manner:
Wherein, p1For the first fatigue condition probability, m1For the tired interval number, n1For first quantity.
Optionally, the cognitive difficulties threshold value is 1.6.
Optionally, the processing module, is used for:The real-time expression of the second quantity frame number of the driver is obtained, it will be every
Real-time expression described in frame is compared with the tired expression in tired expression library, determines the tired frame being consistent with the tired expression
Number;The second fatigue condition probability is determined according to following manner:
Wherein, p2For the second fatigue condition probability, m2For the tired frame number, n2For second quantity.
Optionally, the processing module, is used for:Obtain the ocular image of the driver of third quantity frame number;
By ocular image described in bright pupil Algorithm Analysis, the eye for determining the driver is the closure frame number being closed;According to such as
Under type determines the third fatigue condition probability:
Wherein, p3For the second fatigue condition probability, m3For the tired frame number, n3For second quantity.
Optionally, the determination module, is used for:The fatigue driving posterior probability is calculated according to equation below:
f1=f0*p1*p2*p3/(p1*p2*p3+(1-p1)*(1-p2)*(1-p3));
Wherein, f0For the fatigue driving prior probability, f1For the fatigue driving posterior probability, p1It is tired for described first
Labor conditional probability, p2For the first fatigue condition probability, p3For the first fatigue condition probability, p1For the described first fatigue
Conditional probability.
The disclosure also provides a kind of vehicle, includes the device of determining fatigue driving described above.
Other feature and advantage of the disclosure will be described in detail in subsequent specific embodiment part.
Description of the drawings
Attached drawing is for providing further understanding of the disclosure, and a part for constitution instruction, with following tool
Body embodiment is used to explain the disclosure, but do not form the limitation to the disclosure together.In the accompanying drawings:
Fig. 1 is a kind of method flow diagram for determining fatigue driving that the disclosure provides;
Fig. 2 is the Bayesian Structure that the disclosure merges three kinds of detection mode detection fatigue drivings;
Fig. 3 is a kind of device block diagram for determining fatigue driving that the disclosure provides;
Fig. 4 is the device block diagram that the another kind that the disclosure provides determines fatigue driving.
Reference sign
310 --- signal acquisition module, 320 --- embedded host, 330 --- human-computer interaction interface
311 --- camera.312 --- brain blood oxygen collecting unit, 320 --- embedded host,
321 --- brain blood oxygen analysis algorithm, 322 --- facial expression recognition,
323 --- bright pupil algorithm, 324 --- control unit, 330 --- human-computer interaction interface.
Specific embodiment
The specific embodiment of the disclosure is described in detail below in conjunction with attached drawing.It should be understood that this place is retouched
The specific embodiment stated is only used for describing and explaining the disclosure, is not limited to the disclosure.
As shown in Figure 1, present disclose provides a kind of method of determining fatigue driving, this method includes:
In a step 101, the brain blood oxygen scaling exponent of driver, face fatigue expression and coreclisis degree are obtained;
In a step 102, the first fatigue condition probability is determined according to the brain blood oxygen scaling exponent, according to facial tired table
Feelings determine the second fatigue condition probability, and third fatigue condition probability is determined according to the coreclisis degree;
In step 103, according to fatigue driving prior probability, the first fatigue condition probability, the described second tired item
Part probability and the third fatigue condition probability, determine fatigue driving posterior probability;Wherein, before the tired prior probability is
The driver is once tested to drive the tired probability that fatigue conditions obtain.
In the present embodiment, can the brain blood oxygen mark be obtained by brain blood oxygen parameter acquisition equipment (such as head-type detector)
Spend index.
The disclosure obtains and is used to judge that the data of fatigue driving can be divided into three kinds:
1st, brain blood oxygen parameter
Optionally, the first fatigue condition probability is determined according to the brain blood oxygen scaling exponent, including:Obtain the driver
Strontium dioxide index in specific time generates strontium dioxide time series;By the strontium dioxide time series average mark
For the isometric subinterval of the first quantity;By the way that trend fluction analysis method is gone to handle each isometric subinterval, obtain
Obtain each corresponding brain blood oxygen scaling exponent in the isometric subinterval;By the corresponding brain blood oxygen in each isometric subinterval
Scaling exponent is compared respectively with cognitive difficulties threshold value, determines the corresponding tired interval number in all tired subintervals;Wherein, institute
The corresponding brain blood oxygen scaling exponent in tired subinterval is stated less than cognitive difficulties threshold value;First fatigue is determined according to following manner
Conditional probability:
Wherein, p1For the first fatigue condition probability, m1For the tired interval number, n1For first quantity.
The SrO of driver can be obtained in the disclosure according to certain time interval (such as 1 second, driver can set)2
(strontium dioxide) parameter obtains multiple SrO of driver in a period of time2Parameter, and these parameters are put into a set,
Generate strontium dioxide time series.
It is divided into multiple isometric subintervals according to will gather later, each subinterval corresponds to congenial duration (such as 5 seconds).Profit
The data in each subinterval are obtained with DFA (Detrended Fluctuation Analysis, remove trend fluction analysis) method
Corresponding COSSE (Celebral Oxygenation Saturation Scaling Exponents, brain blood oxygen scale) ginseng
Number.
It, can be in accordance with the following steps with the data of DFA methods processing strontium dioxide time series:
If SrO2Time series is the time series { x that length is Nk, k=1,2 ..., N }.
1st, construction accumulated deviation sequence Y (i) eliminates the interference of the trend such as period, noise;
Wherein
2nd, by the positive N for being divided into length as s of Y (i)sA disjoint isometric subinterval.Wherein, if N cannot be whole by s
It removes, then the last one section is not calculated within Ns.
3rd, to each subinterval v (k=1,2 ..., Ns) data carry out polynomial regression fit, obtain local trend letter
Number pv(i).Wherein, pv(i) it can be primary (being denoted as DFA1), secondary (being denoted as DFA2) or more high-order moment (being denoted as DFAn),
Then trend in each subinterval is eliminated, calculates its mean variance:
4th, the q rank wave functions of complete sequence are calculated by the mean variance in each subinterval.
Q is 0 or positive integer.Wherein, if q is for 0, q rank wave function calculation formula:
If q is positive integer, q rank wave function calculation formula are:
Different goes trend order n to correspond to different wave functions
5th, it using each q value, is drawn by log-log coordinateWith the scatter plot of each q value, with minimum two
Multiplication is fitted these scatterplots, the slope of the straight line of acquisition, the COSSE values for a subinterval.
It can be compared by being used for cognitive difficulties threshold value and the COSSE parameters in each subinterval, judge this sub-district
Between whether driver tired in the corresponding time, i.e., whether the subinterval is tired subinterval.
Quantity for tired subinterval accounts for the proportion of all subinterval quantity, and as driver is in fatigue state
First fatigue condition probability.
Optionally, the cognitive difficulties threshold value is 1.6.
2nd, the tired facial expression image of face
It is described that second fatigue condition probability is determined according to facial tired expression, including:Obtain the second number of the driver
The real-time expression of frame number is measured, the tired expression in expression real-time described in every frame and tired expression library is compared, determining and institute
State the tired frame number that tired expression is consistent;The second fatigue condition probability is determined according to following manner:
Wherein, p2For the third fatigue condition probability, m2For the tired frame number, n2For second quantity.
The disclosure can establish tired expression library by training.It trains and establishes expression sample database, can include following several
A step:
1st, for given training sample, the tracking of face contour is realized to every frame image therein by track algorithm,
To eliminate the influence of head movement.
2nd, the movable information of consecutive frame face cortex picture point is calculated by optical flow method.
3rd, after obtaining the information, by OMPP, (Orthogonal Manifold Preserving Projection are orthogonal
Prevalence keeps projection) dimension-reduction treatment is carried out, to obtain higher-dimension prevalence in low dimension projective space characteristics, the low-dimensional of tired expression is thrown
Shadow space characteristics collect, and form tired expression library.
It, can be according to intervals, driver's face of continuous acquisition multiframe after having ready-made tired expression library
Real-time expression, and in accordance with the following steps to determine per the real-time expression of frame whether be tired expression:
1st, the real-time expression of driver's face is obtained by camera, processing early period is carried out to the real-time expression of every frame reduces dimension
Number;For example, dimension can be reduced to 64*64.
2nd, the tracking of face contour is realized using track algorithm.
3rd, the movable information of adjacent real-time expression frame is obtained by optical flow approach, this movable information is relative to preceding a burst of
Facial expression frame, dimension increase 2 times of dimension of the data frame after step 1 dimensionality reduction;For example, dimension increases to 2*64*64.
4th, the low dimensional manifold feature of real-time expression frame is obtained using OMPP dimensionality reductions.
5th, with reference to the Fatigue pattern library established in training process, by K- near neighbor methods determine the real-time expression of each frame be with
Tired expression is consistent, and determines the quantity being consistent, i.e., tired frame number.
3rd, coreclisis image
Optionally, it is described that third fatigue condition probability is determined according to the coreclisis degree, including:Obtain third number of frames
The ocular image of several drivers;By ocular image described in bright pupil Algorithm Analysis, the driver is determined
Eye be closed closure frame number;The third fatigue condition probability is determined according to following manner:
Wherein, p3For the third fatigue condition probability, m3For the tired frame number, n3For second quantity.
The judgement of disclosure coreclisis situation can be detected by the recognizer of " bright pupil ".
Optionally, it is described according to fatigue driving prior probability, the first fatigue condition probability, second fatigue condition
Probability and the third fatigue condition probability, determine fatigue driving posterior probability, including:
The fatigue driving posterior probability is calculated according to equation below:
f1=f0*p1*p2*p3/(p1*p2*p3+(1-p1)*(1-p2)*(1-p3));
Wherein, f0For the fatigue driving prior probability, f1For the fatigue driving posterior probability, p1It is tired for described first
Labor conditional probability, p2For the first fatigue condition probability, p3For the first fatigue condition probability, p1For the described first fatigue
Conditional probability.
The Bayesian Structure that the disclosure merges three kinds of detection mode detection fatigue drivings is as shown in Figure 2.Wherein, fatigue driving
Prior probability f0, can be empirical data, or the result of successive ignition.
Determining that (user can set fatigue driving posterior probability, such as 0.6) can pass through HMI later higher than certain numerical value
(Human Machine Interface, human-computer interaction interface) alarms.HMI can be Music centre loudspeaker, buzzer, body light,
It can be the graphical interfaces lamp of vehicle mounted guidance.
As shown in figure 3, the disclosure provides a kind of apparatus structure block diagram of determining fatigue driving, which includes:
Signal acquisition module 310, embedded host 320, human-computer interaction interface 330;
Signal acquisition module 310, including camera 311 and brain blood oxygen collecting unit 312, for acquiring facial expression image
With brain blood sample parameter;
Embedded host 320, built-in brain blood oxygen analysis algorithm 321, facial expression recognition 322,323 He of bright pupil algorithm
Control unit 324 can obtain the first fatigue condition probability, the second fatigue condition probability and third with three kinds of analyticals
Fatigue condition probability.And pass through Bayesian Fusion the first fatigue condition probability, the second fatigue condition probability and third fatigue condition
Probability and fatigue driving prior probability obtain fatigue driving posterior probability.Control unit 324 is more than in fatigue driving posterior probability
During danger threshold, by CAN (Controller Area Network, controller local area network) buses to human-computer interaction interface
330 alarms
Human-computer interaction interface 330 receives alarm signal, and alarms to driver.
As shown in figure 4, the disclosure provides the device of another determining fatigue driving, which includes:
Acquisition module 410, for obtaining the brain blood oxygen scaling exponent of driver, face fatigue expression and coreclisis degree;
Processing module 420, it is tired according to face for determining the first fatigue condition probability according to the brain blood oxygen scaling exponent
Labor expression determines the second fatigue condition probability, and third fatigue condition probability is determined according to the coreclisis degree;
Determination module 430, for according to fatigue driving prior probability, the first fatigue condition probability, described second is tired
Labor conditional probability, the third fatigue condition probability, determines fatigue driving posterior probability;Wherein, the tired prior probability is
It is preceding once to test the driver and drive the tired probability that fatigue conditions obtain.
Optionally, the processing module 420, is used for:Strontium dioxide index of the driver in specific time is obtained,
Generate strontium dioxide time series;The strontium dioxide time series is equally divided into the isometric subinterval of the first quantity;Pass through
Trend fluction analysis method is gone to handle each isometric subinterval, it is corresponding to obtain each isometric subinterval
Brain blood oxygen scaling exponent;By the corresponding brain blood oxygen scaling exponent in each isometric subinterval respectively with cognitive difficulties threshold value into
Row compares, and determines the corresponding tired interval number in all tired subintervals;Wherein, the corresponding brain blood oxygen scale in the tired subinterval
Index is less than cognitive difficulties threshold value;The first fatigue condition probability is determined according to following manner:
Wherein, p1For the first fatigue condition probability, m1For the tired interval number, n1For first quantity.
Optionally, the cognitive difficulties threshold value is 1.6.
Optionally, the processing module 420, is used for:The real-time expression of the second quantity frame number of the driver is obtained, it will
Expression is compared with the tired expression in tired expression library in real time described in per frame, determines the fatigue being consistent with the tired expression
Frame number;The second fatigue condition probability is determined according to following manner:
Wherein, p2For the second fatigue condition probability, m2For the tired frame number, n2For second quantity.
Optionally, the processing module 420, is used for:Obtain the ocular figure of the driver of third quantity frame number
Picture;By ocular image described in bright pupil Algorithm Analysis, the eye for determining the driver is the closure frame number being closed;According to
Following manner determines the third fatigue condition probability:
Wherein, p3For the third fatigue condition probability, m3For the tired frame number, n3For second quantity, k3It is
Three fatigue thresholds.
Optionally, the determination module 430, is used for:The fatigue driving posterior probability is calculated according to equation below:
f1=f0*p1*p2*p3/(p1*p2*p3+(1-p1)*(1-p2)*(1-p3));
Wherein, f0For the fatigue driving prior probability, f1For the fatigue driving posterior probability, p1It is tired for described first
Labor conditional probability, p2For the first fatigue condition probability, p3For the first fatigue condition probability, p1For the described first fatigue
Conditional probability.
The disclosure provides a kind of vehicle, the device of the determining fatigue driving described in including above-mentioned Fig. 3 and Fig. 4.
The preferred embodiment of the disclosure is described in detail above in association with attached drawing, still, the disclosure is not limited to above-mentioned reality
The detail in mode is applied, in the range of the technology design of the disclosure, a variety of letters can be carried out to the technical solution of the disclosure
Monotropic type, for example, the strontium dioxide in brain blood oxygen parameter detecting replaces with EEG (Electro Encephalography brains electricity
Figure).Alternatively, it is also possible to only do two kinds or the fatigue detecting more than three types, and determine driver's according to the result of detection
Fatigue driving posterior probability.These simple variants belong to the protection domain of the disclosure.
It is further to note that specific technical features described in the above specific embodiments, in not lance
In the case of shield, can be combined by any suitable means, in order to avoid unnecessary repetition, the disclosure to it is various can
The combination of energy no longer separately illustrates.
In addition, arbitrary combination can also be carried out between a variety of different embodiments of the disclosure, as long as it is without prejudice to originally
Disclosed thought should equally be considered as disclosure disclosure of that.
Claims (13)
- A kind of 1. method of determining fatigue driving, which is characterized in that including:Obtain the brain blood oxygen scaling exponent of driver, face fatigue expression and coreclisis degree;First fatigue condition probability is determined according to the brain blood oxygen scaling exponent, determines that second is tired according to the tired expression of the face Labor conditional probability determines third fatigue condition probability according to the coreclisis degree;According to fatigue driving prior probability, the first fatigue condition probability, the second fatigue condition probability and described Three fatigue condition probability, determine fatigue driving posterior probability;Wherein, the tired prior probability is preceding once to test the driving Member drives the tired probability that fatigue conditions obtain.
- 2. according to the method described in claim 1, it is characterized in that, described determine that first is tired according to the brain blood oxygen scaling exponent Labor conditional probability, including:Strontium dioxide index of the driver in specific time is obtained, generates strontium dioxide time series;The strontium dioxide time series is equally divided into the isometric subinterval of the first quantity;By the way that trend fluction analysis method is gone to handle each isometric subinterval, each isometric subinterval is obtained Corresponding brain blood oxygen scaling exponent;The corresponding brain blood oxygen scaling exponent in each isometric subinterval with cognitive difficulties threshold value is compared respectively, is determined The corresponding tired interval number in all fatigue subintervals;Wherein, the corresponding brain blood oxygen scaling exponent in the tired subinterval, which is less than, recognizes Know difficult threshold value;The first fatigue condition probability is determined according to following manner:Wherein, p1For the first fatigue condition probability, m1For the tired interval number, n1For first quantity.
- 3. according to the method described in claim 2, it is characterized in that, the cognitive difficulties threshold value is 1.6.
- 4. according to the method described in claim 3, it is characterized in that, described determine the second fatigue according to the tired expression of the face Conditional probability, including:The real-time expression of the second quantity frame number of the driver is obtained, it will be in expression real-time described in every frame and tired expression library Tired expression is compared, and determines the tired frame number being consistent with the tired expression;The second fatigue condition probability is determined according to following manner:Wherein, p2For the second fatigue condition probability, m2For the tired frame number, n2For second quantity.
- 5. according to the method described in claim 3, it is characterized in that, described determine third fatigue item according to the coreclisis degree Part probability, including:Obtain the ocular image of the driver of third quantity frame number;By ocular image described in bright pupil Algorithm Analysis, the eye for determining the driver is the closure frame number being closed;The third fatigue condition probability is determined according to following manner:Wherein, p3For the second fatigue condition probability, m3For the eye closing frame number, n3For the third quantity.
- 6. method according to claim 4 or 5, which is characterized in that it is described according to fatigue driving prior probability, described first Fatigue condition probability, the second fatigue condition probability and the third fatigue condition probability, determine that fatigue driving posteriority is general Rate, including:The fatigue driving posterior probability is calculated according to equation below:f1=f0*p1*p2*p3/(p1*p2*p3+(1-p1)*(1-p2)*(1-p3));Wherein, f0For the fatigue driving prior probability, f1For the fatigue driving posterior probability, p1For the described first tired item Part probability, p2For the first fatigue condition probability, p3For the first fatigue condition probability, p1For first fatigue condition Probability.
- 7. a kind of device of determining fatigue driving, which is characterized in that including:Acquisition module, for obtaining the brain blood oxygen scaling exponent of driver, face fatigue expression and coreclisis degree;Processing module, for determining the first fatigue condition probability according to the brain blood oxygen scaling exponent, according to the face fatigue Expression determines the second fatigue condition probability, and third fatigue condition probability is determined according to the coreclisis degree;Determination module, for according to fatigue driving prior probability, the first fatigue condition probability, second fatigue condition is general Rate and the third fatigue condition probability, determine fatigue driving posterior probability;Wherein, the tired prior probability is preceding primary The driver is tested to drive the tired probability that fatigue conditions obtain.
- 8. device according to claim 7, which is characterized in that the processing module is used for:Strontium dioxide index of the driver in specific time is obtained, generates strontium dioxide time series;The strontium dioxide time series is equally divided into the isometric subinterval of the first quantity;By the way that trend fluction analysis method is gone to handle each isometric subinterval, each isometric subinterval is obtained Corresponding brain blood oxygen scaling exponent;The corresponding brain blood oxygen scaling exponent in each isometric subinterval with cognitive difficulties threshold value is compared respectively, is determined The corresponding tired interval number in all fatigue subintervals;Wherein, the corresponding brain blood oxygen scaling exponent in the tired subinterval, which is less than, recognizes Know difficult threshold value;The first fatigue condition probability is determined according to following manner:Wherein, p1For the first fatigue condition probability, m1For the tired interval number, n1For first quantity.
- 9. device according to claim 8, which is characterized in that the cognitive difficulties threshold value is 1.6.
- 10. device according to claim 9, which is characterized in that the processing module is used for:The real-time expression of the second quantity frame number of the driver is obtained, it will be in expression real-time described in every frame and tired expression library Tired expression is compared, and determines the tired frame number being consistent with the tired expression;The second fatigue condition probability is determined according to following manner:Wherein, p2For the second fatigue condition probability, m2For the tired frame number, n2For second quantity.
- 11. device according to claim 9, which is characterized in that the processing module is used for:Obtain the ocular image of the driver of third quantity frame number;By ocular image described in bright pupil Algorithm Analysis, the eye for determining the driver is the closure frame number being closed;The third fatigue condition probability is determined according to following manner:Wherein, p3For the second fatigue condition probability, m3For the tired frame number, n3For second quantity.
- 12. the device according to claim 10 or 11, which is characterized in that the determination module is used for:The fatigue driving posterior probability is calculated according to equation below:f1=f0*p1*p2*p3/(p1*p2*p3+(1-p1)*(1-p2)*(1-p3));Wherein, f0For the fatigue driving prior probability, f1For the fatigue driving posterior probability, p1For the described first tired item Part probability, p2For the first fatigue condition probability, p3For the first fatigue condition probability, p1For first fatigue condition Probability.
- 13. a kind of vehicle, which is characterized in that including:7 to 12 any one of them device of the claims.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711340449.2A CN108154095A (en) | 2017-12-14 | 2017-12-14 | A kind of method, apparatus and vehicle of determining fatigue driving |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711340449.2A CN108154095A (en) | 2017-12-14 | 2017-12-14 | A kind of method, apparatus and vehicle of determining fatigue driving |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108154095A true CN108154095A (en) | 2018-06-12 |
Family
ID=62467266
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711340449.2A Pending CN108154095A (en) | 2017-12-14 | 2017-12-14 | A kind of method, apparatus and vehicle of determining fatigue driving |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108154095A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109740625A (en) * | 2018-11-22 | 2019-05-10 | 深圳市三诺数字科技有限公司 | A kind of safe driving method of discrimination, system and vehicle based on Fusion Features |
CN110786869A (en) * | 2019-10-29 | 2020-02-14 | 浙江工业大学 | Method for detecting fatigue degree of programmer |
CN111084610A (en) * | 2019-12-20 | 2020-05-01 | 东南大学 | Time-space characteristic analysis method for near-infrared brain imaging signals of autism children |
GB2578764A (en) * | 2018-11-07 | 2020-05-27 | Jaguar Land Rover Ltd | Apparatus and method for controlling vehicle system operation |
CN112389444A (en) * | 2020-10-16 | 2021-02-23 | 爱驰汽车(上海)有限公司 | Vehicle early warning method and device based on heart rate detection of driver |
CN113569699A (en) * | 2021-07-22 | 2021-10-29 | 上汽通用五菱汽车股份有限公司 | Attention analysis method, vehicle, and storage medium |
CN114283559A (en) * | 2022-03-04 | 2022-04-05 | 西南交通大学 | Driver fatigue early warning method, device, equipment and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101574260A (en) * | 2009-06-12 | 2009-11-11 | 大连微龙软件有限公司 | Vehicle-mounted fatigue early warning device and method thereof |
GB2487888A (en) * | 2009-10-30 | 2012-08-08 | Shenzhen Safdao Technology Corp Ltd | Method, device and car for fatigue driving detection |
TW201608532A (en) * | 2014-08-26 | 2016-03-01 | 國立臺南大學 | Method and device for warning driver |
-
2017
- 2017-12-14 CN CN201711340449.2A patent/CN108154095A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101574260A (en) * | 2009-06-12 | 2009-11-11 | 大连微龙软件有限公司 | Vehicle-mounted fatigue early warning device and method thereof |
GB2487888A (en) * | 2009-10-30 | 2012-08-08 | Shenzhen Safdao Technology Corp Ltd | Method, device and car for fatigue driving detection |
TW201608532A (en) * | 2014-08-26 | 2016-03-01 | 國立臺南大學 | Method and device for warning driver |
Non-Patent Citations (1)
Title |
---|
张伟: "融合脑血氧饱和度及面部表情的多诱因非合适驾驶状态识别", 《中国博士学位论文全文数据库 工程科技II辑》 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2578764A (en) * | 2018-11-07 | 2020-05-27 | Jaguar Land Rover Ltd | Apparatus and method for controlling vehicle system operation |
GB2578764B (en) * | 2018-11-07 | 2021-10-27 | Jaguar Land Rover Ltd | Apparatus and method for controlling vehicle system operation |
CN109740625A (en) * | 2018-11-22 | 2019-05-10 | 深圳市三诺数字科技有限公司 | A kind of safe driving method of discrimination, system and vehicle based on Fusion Features |
CN110786869A (en) * | 2019-10-29 | 2020-02-14 | 浙江工业大学 | Method for detecting fatigue degree of programmer |
CN110786869B (en) * | 2019-10-29 | 2021-12-21 | 浙江工业大学 | Method for detecting fatigue degree of programmer |
CN111084610A (en) * | 2019-12-20 | 2020-05-01 | 东南大学 | Time-space characteristic analysis method for near-infrared brain imaging signals of autism children |
CN112389444A (en) * | 2020-10-16 | 2021-02-23 | 爱驰汽车(上海)有限公司 | Vehicle early warning method and device based on heart rate detection of driver |
CN112389444B (en) * | 2020-10-16 | 2022-04-12 | 爱驰汽车(上海)有限公司 | Vehicle early warning method and device based on heart rate detection of driver |
CN113569699A (en) * | 2021-07-22 | 2021-10-29 | 上汽通用五菱汽车股份有限公司 | Attention analysis method, vehicle, and storage medium |
CN113569699B (en) * | 2021-07-22 | 2024-03-08 | 上汽通用五菱汽车股份有限公司 | Attention analysis method, vehicle, and storage medium |
CN114283559A (en) * | 2022-03-04 | 2022-04-05 | 西南交通大学 | Driver fatigue early warning method, device, equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108154095A (en) | A kind of method, apparatus and vehicle of determining fatigue driving | |
US9286515B2 (en) | Doze detection method and apparatus thereof | |
US7839292B2 (en) | Real-time driving danger level prediction | |
CN105286802B (en) | Driver Fatigue Detection based on video information | |
EP2351524A1 (en) | Drowsiness determining device and program | |
CN112434611B (en) | Early fatigue detection method and system based on eye movement subtle features | |
BRPI0712837B1 (en) | method for determining and analyzing a location of visual interest | |
CN106913333B (en) | Method for selecting sensitivity characteristic index of continuous attention level | |
JP2008099884A (en) | Condition estimating apparatus | |
CN110855934A (en) | Fatigue driving identification method, device and system, vehicle-mounted terminal and server | |
Wang et al. | A method of detecting driver drowsiness state based on multi-features of face | |
CN107563346A (en) | One kind realizes that driver fatigue sentences method for distinguishing based on eye image processing | |
CN107832792A (en) | A kind of method for detecting fatigue driving and device | |
CN113989788A (en) | Fatigue detection method based on deep learning and multi-index fusion | |
JP5139470B2 (en) | Sleepiness level estimation device and sleepiness level estimation method | |
CN110097012B (en) | Fatigue detection method for monitoring eye movement parameters based on N-range image processing algorithm | |
JP2007236488A (en) | Vigilance degree estimating device, system, and method | |
CN117876807A (en) | Personnel state adjustment training method and device based on simulation situation multi-mode data | |
CN111160071B (en) | Fatigue driving detection method and device | |
CN114399752A (en) | Eye movement multi-feature fusion fatigue detection system and method based on micro eye jump characteristics | |
Wang et al. | Enhancing YOLOv7-Based Fatigue Driving Detection through the Integration of Coordinate Attention Mechanism | |
Hayashi et al. | Prediction of stopping maneuver considering driver's state | |
CN110021147A (en) | A kind of method for detecting fatigue driving demarcated based on machine learning and numerical value | |
JP2004089272A (en) | Method and device for biological state estimation, and program | |
CN117077086A (en) | Advanced brain fatigue early warning system, equipment and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180612 |
|
RJ01 | Rejection of invention patent application after publication |