CN104688251A - Method for detecting fatigue driving and driving in abnormal posture under multiple postures - Google Patents

Method for detecting fatigue driving and driving in abnormal posture under multiple postures Download PDF

Info

Publication number
CN104688251A
CN104688251A CN201510092886.1A CN201510092886A CN104688251A CN 104688251 A CN104688251 A CN 104688251A CN 201510092886 A CN201510092886 A CN 201510092886A CN 104688251 A CN104688251 A CN 104688251A
Authority
CN
China
Prior art keywords
point
eyes
driver
pose
detection
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
Application number
CN201510092886.1A
Other languages
Chinese (zh)
Inventor
刘哲
于涛
王鹏
冯仁委
苗政委
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xi'an Bang Wei Electronic Science And Technology Co Ltd
Original Assignee
Xi'an Bang Wei Electronic Science And Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Xi'an Bang Wei Electronic Science And Technology Co Ltd filed Critical Xi'an Bang Wei Electronic Science And Technology Co Ltd
Priority to CN201510092886.1A priority Critical patent/CN104688251A/en
Publication of CN104688251A publication Critical patent/CN104688251A/en
Pending legal-status Critical Current

Links

Classifications

    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/16Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
    • A61B5/18Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state for vehicle drivers or machine operators
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B2503/00Evaluating a particular growth phase or type of persons or animals
    • A61B2503/20Workers
    • A61B2503/22Motor vehicles operators, e.g. drivers, pilots, captains

Abstract

The invention relates to a method for detecting fatigue driving and driving in an abnormal posture under multiple postures. The method comprises the following steps: step 1, image acquisition: acquiring driver image information; step 2, image recognition: carrying out real-time face detection on the acquired images; step 3, face feature point calibration: calibrating feature points of eyes and a mouth of the detected face; step 4, judgment of the states of the eyes and the mouth: for the feature points, analyzing the closing states of the eyes and the mouth by adopting an eye state detection method on the basis of face width information; step 5, detecting of three-dimensional posture of the head: detecting the three-dimensional posture of the head through geometrical characteristics; step 6, judging of the states of fatigue driving and driving in the abnormal posture: judging whether the driver is in fatigue or not according to the closing states of the eyes and the mouth, and judging whether the driver is in the abnormal driving state or not according to the three-dimensional posture of the head.

Description

Fatigue under a kind of multi-pose and improper attitude drive detection method
Technical field
The present invention relates to and be a kind ofly applied to fatigue under multi-pose and improper attitude drives the method detected, particularly use multi-pose human face characteristic point positioning method to judge the state of human eye and face simultaneously, and add the detection of head 3 d pose, thus effectively judge whether driver belongs to fatigue driving and improper attitude is driven more accurately.
Background technology
Safety trip to have become in current field of traffic a very important problem, along with expanding economy, the recoverable amount of automobile continues to increase, and the vehicle accident amount meanwhile brought also rises year by year, be one of important hidden danger of current traffic safety in the multifactor middle fatigue driving of weight of impact safety trip.
Driving is one and not only expends bodily strength but also take mental heavy load work.It is almost in closed driver's cabin that driver is sitting in for a long time, the whole-body muscle moment is all in nervous duty, ideological plane is concentrated, if the time has grown physiology and the psychological energy consumed can not get immediately recovering and adjusting, just fatigue can be caused, generation mood is worried, easy ignition, phenomenon weak all over, fatigue can make the sensory organ of driver occur obstacle, its perception to surrounding, situation judgement and the manipulation ability to vehicle have decline in various degree, are therefore easy to vehicle accident occurs.Statistical data shows, the death toll directly caused by fatigue driving in 2010 to 2011 China accounts for 11.35% and 12.5% of the total death toll of vehicle driver traffic accident respectively, has 9000 people to die from fatigue driving approximately every year.
Meanwhile, U.S. government's release mechanism result of study shows, the driver is absent minded is the main reason occurred that causes a traffic accident, and sends short messages, reads newspaper, looks in the mirror and make up etc. when such as driving.The a investigation jointly completed by Virginia science and technology transport association and National expressway safety office in addition, draw by carrying out research to the Video Document of thousands of hours, such as collide in the vehicle accident occurred or some, to swipe and in other friction accidents, driver's distractibility is most important reason.In the traffic accident occurred, belong to absent minded cause account for 80%, and in the vehicle accident almost got into an accident, have 65% to be all because the absent minded of the driver causes.At driving condition, scatterbrained objective evidence and driver drive with the improper attitude of not watching front attentively.
Therefore, research and develop high performance driver fatigue and improper attitude driving condition Real-Time Monitoring and early warning technology, to improving, China's traffic safety status is significant.
The research of current driving fatigue is divided into subjectivity and objectivity two kinds of methods, and subjective research method has subjective survey table, driver oneself record, sleep habit application form, Stamford sleep yardstick table four kind.Objective method has the measuring method such as temperature and electrocardiogram when electroencephalogram, electro-oculogram, electromyogram, respiratory air flow, effect of breathing, arterial blood oxygen saturation.Although these methods are more accurately, because these methods are generally measure before driving or after driving, be advanced or delayed, non real-time.And in actual applications, people are more prone to use vehicle-mounted, contactless in real time fatigue detection device.
Along with computer vision technique and image processing techniques are applied to actual life more and more, fatigue detecting system based on machine vision relies on lower research and development, manufacturing cost and powerful image processing techniques, obtain the favor of more researcheres, in addition it is a kind of contactless system, harmful effect is not had to driver's work, is more easily esthetically acceptable to the consumers.Further, do not have at present a driver attention not to be concentrated, carry out the detection system of driving with improper attitude, but not normal attitude drives but one of important root that vehicle accident exactly takes place frequently.Therefore, face features is explored based on machine vision, the facial characteristics being in tired and improper driving condition is analysed in depth, set up a set of fatigue driving to detect and improper driving condition detects and combines, the detection system of effective work, vehicle accident caused thus must be greatly reduced, avoid the economic loss caused thus.
Summary of the invention
Technical problem to be solved by this invention is to provide fatigue under a kind of multi-pose and improper attitude drives detection method, with realizing the effect reducing vehicle accident.
It is as follows that the present invention solves the problems of the technologies described above taked technical scheme:
Fatigue under multi-pose and improper attitude drive a detection method, it is characterized in that, comprising:
Step 1: image acquisition, gathers driver's image information;
Step 2: image recognition, carries out real-time face detection to the image collected;
Step 3: face feature point is demarcated, and comprising: the characteristic point of the face detected being carried out to eyes and face is demarcated;
Step 4: eyes and mouth states judge, comprising: adopt based on the eyes detection methods analyst human eye of facial width information and the closure state of face characteristic point;
Step 5: head 3 d pose detects, and comprising: utilize geometric properties to detect head 3 d pose; Step 6: tired and improper attitude driving condition judges, comprising: the closure state according to human eye and face judges whether driver is in fatigue, and judges whether driver is in improper driving condition according to head 3 d pose.
Further, preferably, in step 2, comprising: adopt the method for statistical analysis and machine learning to summarize face sample and non-face sample statistical nature separately;
Build the grader distinguishing respective feature, realize Face detection with grader and detect.
Further, preferably, AdaBoost algorithm is used to detect face location.
Further, preferably, in step 3, adopt supervision gradient descent method to carry out face feature point demarcation, supervision gradient descent method is mainly divided into training and detects two links;
Training part was carried out operating and is calculated before system cloud gray model, and obtained the regression iterative parameter of facial modeling part, to realize positioning eyes, nose, face characteristic point in face, comprising:
To characteristic points such as the facial image manual markings eyes face noses in all training storehouses, and obtain an average face;
Obtain the disturbance parameter of average face, i.e. the average of zooming and panning and standard deviation, do Gauss distribution sampling for every piece image with this average and standard deviation, obtain the training initial value x of eigenvalue point 0, and calculate the sift feature φ of all initial values point 0,
Gradient Descent direction and the deviation factors thereof of eigenvalue point can be obtained by formula (1), wherein, { d itraining storehouse in face image set, it is the characteristic point true value that facial image concentrates all manual markings. training characteristics value point x iwith true value point between matrix of differences, φ ibe the sift feature of eigenvalue point, R is Gradient Descent direction, and b is deviation factors, and k represents iterations;
arg min R k , b k Σ d i Σ x k i | | Δx * ki - R k φ k i - b k | | 2 - - - ( 1 )
By formula (2), the characteristic point x in each width facial image is upgraded, and recalculates the sift feature upgrading rear characteristic point,
x k=x k-1+R k-1φ k-1+b k-1(2)
Finally, iterative is carried out to formula 1 and formula 2, eigenvalue point x kconverge on true value point x *, now train end, the Gradient Descent direction R in the middle of each iterative process of finally trying to achieve kwith deviation factors b knamely required regression iterative parameter is detected.
Further, preferably, in step 3, at detection-phase, specifically comprise:
First, average for the standard obtained in training process face sample is navigated to photographic head and detect initial coordinate as characteristic point in the middle of the facial image that obtains;
Secondly, calculate the sift feature of all initial coordinate point, be designated as φ 0;
Finally, carry out regression iterative by formula (3) and calculate final facial characteristics point coordinates,
x k=x k-1+R k-1φ k-1+b k-1(3)
Wherein, R kand b kthe regression iterative parameter obtained the training stage, φ kfor the sift feature of each iterative characteristic point.
Further, preferably, in step 4, eyes mouth states judges mainly to be divided into two parts, and namely learn and detect, first 4 seconds is learning phase, enters detection-phase afterwards;
At learning phase, pass through face feature point, obtain the eye-level eyeHeight in each two field picture and face width faceWidth, calculate the ratio r ate of each frame eyeHeight and faceWidth, finally obtain the meansigma methods rateMean of rate value in all frames, as the facial characteristics of this driver, the closure state of eyes will be used for judging at detection-phase;
At detection-phase, by face feature point, obtain left eye width leftEyeWidth and right eye width rightEyeWidth, compare the size of leftEyeWidth and rightEyeWidth, obtain larger width maxWidth and less width minWidth; When the ratio of minWidth and maxWidth is greater than 0.8, illustrate that two width values are close, now, face width faceWidth be two width and 3.5 times.Namely
faceWidth=(maxWidth+minWidth)*3.5
When the ratio of minWidth and maxWidth is less than 0.8, illustrate that two width value difference are comparatively large, now face width faceWdith is 7 times of maxWidth, that is:
faceWidth=7*maxWidth;
Finally, after known eye-level eyeHeight and face width faceWidth, by the rateMean that learning phase calculates, the size now comparing ratio r ate and the rateMean of eyeHeight and faceWidth of each frame opens to what judge eyes the state of closing, as rate<rateMean/2 for closing one's eyes, otherwise for opening eyes.
Further, preferably, in step 4, for the judgement of mouth states, comprising:
First obtain height mouthHeight that face opens and face height faceHeight by face feature point, when the ratio of mouthHeight and faceHeight is greater than 8%, then thinks and be in the state of yawning.
Further, preferably, in step 5, specifically comprise: detect driver and whether be in the deflection of head and upper and lower pitch attitude, wherein, deflection calculates by following methods:
When face is in front, subnasal point is equal with the angle of eyespot outside left and right,
When face deflection, after namely the side degree of depth rotates, outside subnasal point and left and right, the angle difference of eyespot is β eye_out, subnasal point β poor with the angle of eyespot in left and right can be calculated to obtain simultaneously eye_in, the angle difference β of subnasal point and left and right corners of the mouth point mouth, in order to reduce error, the side degree of depth anglec of rotation of desirable face is the average of these three angles, that is: &beta; 0 = 1 3 ( &beta; eye _ out + &beta; eye _ in + &beta; mouth )
Elevation angle can calculate by the following method, comprising:
The side view of face can be regarded as an ellipse, and y-axis is oval middle separated time, and x-axis is the perpendicular bisector of eyes and mouth line, does not so have upper and lower pitching, has α when namely vertical depth rotates 12.After vertical depth rotates, x-axis is no longer the perpendicular bisector of eyes and mouth line, and according to the character of isosceles triangle, the computing formula of the side degree of depth anglec of rotation is:
α 0, β 0for approximate Attitude estimation value, utilize α 0, β 0as initial value, use quasi-Newton method to human face posture Exact Solution, can in the hope of the accurate deflection angle of human face posture.
Further, preferably, in step 6, specifically comprise:
; In the middle of driving procedure, detect the PERCLOS value of driver's eyes in real time, if PERCLOS value is greater than thresholding, get PERCLOS>0.35, then carry out tired audible alarm;
When eye-closing period is greater than safety value and the eye-closing period T of setting c> 1.2 seconds, also can carry out fatigue warning;
As yawn frequency M>0.8 per minute, be judged as fatigue state, carry out voice message.
Further, preferably, in step 6, utilize driver head's 3 d pose data further, whether real-time judge driver watches front attentively, and whether attention of driving is concentrated, and comprising:
When attitude left avertence >poseleft or right avertence >poseright being detected or the >poseup or nutation >posedown that faces upward, and continue duration >Tpose, then think that driver attention is not concentrated, Poseleft, poseright can according to the putting position of photographic head slightly differences;
When photographic head is placed at driver's front dead center, poseleft, poseright are 20 degree, and poseup, posedown are 20 degree, and Tpose is 1.5 seconds; When driver attention is not concentrated, carry out attitude warning, prompting driver attention is concentrated and is driven.
After this invention takes such scheme, can judge whether driver is in fatigue state and whether is in improper driving condition according to the characteristic point in analyst's face portion, thus play and remind driver in time, the effect of guarantee driving safety.
Other features and advantages of the present invention will be set forth in the following description, and, partly become apparent from description, or understand by implementing the present invention.Object of the present invention and other advantages realize by structure specifically noted in write description, claims and accompanying drawing and obtain.
Accompanying drawing explanation
Below in conjunction with accompanying drawing, the present invention is described in detail, to make above-mentioned advantage of the present invention definitely.Wherein:
Fig. 1 is the schematic diagram carrying out elevation angle test in the present invention;
Fig. 2 is the schematic flow sheet of specific embodiment of the invention mode.
Detailed description of the invention
Describe embodiments of the present invention in detail below with reference to drawings and Examples, to the present invention, how application technology means solve technical problem whereby, and the implementation procedure reaching technique effect can fully understand and implement according to this.It should be noted that, only otherwise form conflict, each embodiment in the present invention and each feature in each embodiment can be combined with each other, and the technical scheme formed is all within protection scope of the present invention.
Specifically, the object of the present invention is to provide and be a kind ofly applied to the method for detecting fatigue driving under multi-pose and judge whether driver is in fatigue state and whether is in improper driving condition according to the characteristic point in analyst's face portion, thus play and remind driver in time, the effect of guarantee driving safety.
To achieve these goals, technical scheme of the present invention is:
Fatigue under multi-pose and improper attitude drive a detection method, comprising: step 1: image acquisition, gather driver's image information;
Step 2: image recognition, carries out real-time face detection to the image collected;
Step 3: face feature point is demarcated, and comprising: the characteristic point of the face detected being carried out to eyes and face is demarcated;
Step 4: eyes and mouth states judge, comprising: adopt based on the eyes detection methods analyst human eye of facial width information and the closure state of face characteristic point;
Step 5: head 3 d pose detects, and comprising: utilize geometric properties to detect head 3 d pose;
Step 6: tired and improper attitude driving condition judges, comprising: the closure state according to human eye and face judges whether driver is in fatigue, and judges whether driver is in improper driving condition according to head 3 d pose.
In other words, the present invention mainly comprises following several key steps:
1. collection driver image information 2. uses AdaBoost algorithm to carry out real-time face detection 3. to the image collected and supervises to the face employing detected the characteristic point demarcation that gradient descent method carries out eyes and face.4. adopted based on the eyes detection methods analyst human eye of facial width information and the closure state of face by characteristic point position again.5. utilize geometric properties to carry out detection 6. to head 3 d pose after and judge whether driver is in fatigue according to the closure state of human eye and face, and judge whether driver is in improper driving condition according to head 3 d pose.
Compared with existing fatigue driving detecting system, the present invention has following beneficial effect:
The present invention adds driver head's 3 d pose detection module while fatigue driving detects, especially carry out recording and prompt alarm for driver's rhembasmus, bradykinesia, the improper driving condition such as absent minded, compensate for that some fatigue driving detecting systems on market are single detects tired function.
In addition, in fatigue driving context of detection, present invention employs supervision gradient descent method to carry out facial modeling, and propose a kind of eyes detection method based on facial width information and carry out eyes detection, can have when very large deflection angle is spent at face and carry out face feature point location and carry out eyes opening the detection of closing, and then carry out driver fatigue judgement, this method is more suitable for the practical situation of face multi-pose, solves the shortcoming of current fatigue driving detecting system to human face posture bad adaptability.
The fatigue state that this method more adapts under the situation of actual Driving Scene and multi-pose detects, and makes fatigue driving detecting system humanized.The method has real-time, practicality, reliability, adaptability, round-the-clock, contactlessly in real time can carry out driver fatigue state and improper driving condition supervision, for safe driving provides effective automobile active safety guarantee.
With reference to Fig. 2, the main calculation procedure of method of the present invention is as follows:
Image acquisition:
Realize driver's face capture function mainly through mark/high-definition camera, from the acquisition video flowing that photographic head is real-time, and data flow is delivered to CPU carry out follow-up process.
Face datection:
Adopt the method for detecting human face of Corpus--based Method study, its essence is and adopt the method for statistical analysis and machine learning to summarize face sample and non-face sample statistical nature separately, then build the grader distinguishing respective feature, realize Face detection with grader and detect.In method, the concrete AdaBoost algorithm that uses detects face location.
Face feature point is demarcated:
System adopts supervision gradient descent method to carry out face feature point demarcation, and supervision gradient descent method is mainly divided into training and detects two links.
Training part was carried out operating and is calculated before system cloud gray model, mainly for obtaining the regression iterative parameter of facial modeling part, algorithm operationally can be positioned accurately to characteristic points such as eyes, nose, faces in face.
First, to characteristic points such as the facial image manual markings eyes face noses in all training storehouses, and an average face is obtained.
Secondly, obtain the disturbance parameter of average face, i.e. the average of zooming and panning and standard deviation.Do Gauss distribution sampling for every piece image with this average and standard deviation, obtain the training initial value x of eigenvalue point 0, and calculate the sift feature φ of all initial values point 0.
Again, Gradient Descent direction and the deviation factors thereof of eigenvalue point can be obtained by formula (1).Wherein, { d itraining storehouse in face image set, it is the characteristic point true value that facial image concentrates all manual markings. training characteristics value point x iwith true value point between matrix of differences, φ ibe the sift feature of eigenvalue point, R is Gradient Descent direction, and b is deviation factors, and k represents iterations.
arg min R k , b k &Sigma; d i &Sigma; x k i | | &Delta;x * ki - R k &phi; k i - b k | | 2 - - - ( 1 )
Again, by formula (2), the characteristic point x in each width facial image is upgraded, and recalculate the sift feature upgrading rear characteristic point.
x k=x k-1+R k-1φ k-1+b k-1(2)
Finally, iterative is carried out to formula 1 and formula 2, eigenvalue point x kconverge on true value point x *, now train end, the Gradient Descent direction R in the middle of each iterative process of finally trying to achieve kwith deviation factors b knamely required regression iterative parameter is detected.
At detection-phase, first, average for the standard obtained in training process face sample is navigated to photographic head and detect initial coordinate as characteristic point in the middle of the facial image that obtains.
Secondly, calculate the sift feature of all initial coordinate point, be designated as φ 0.
Finally, carry out regression iterative by formula (3) and calculate final facial characteristics point coordinates.Wherein, R kand b kthe regression iterative parameter obtained the training stage, φ kfor the sift feature of each iterative characteristic point.
x k=x k-1+R k-1φ k-1+b k-1(3)
Eyes mouth states judges:
Eyes mouth states judges mainly to be divided into two parts, namely learns and detects, and system brings into operation first 4 seconds for learning phase, enters detection-phase afterwards.
At learning phase, pass through face feature point, obtain the eye-level eyeHeight in each two field picture and face width faceWidth, calculate the ratio r ate of each frame eyeHeight and faceWidth, finally obtain the meansigma methods rateMean of rate value in all frames, as the facial characteristics of this driver.The closure state of eyes will be used for judging at detection-phase.
At detection-phase, by face feature point, obtain left eye width leftEyeWidth and right eye width rightEyeWidth, compare the size of leftEyeWidth and rightEyeWidth, obtain larger width maxWidth and less width minWidth; When the ratio of minWidth and maxWidth is greater than 0.8, illustrate that two width values are close to (namely face towards comparatively front), now, face width faceWidth be two width and 3.5 times.Namely
faceWidth=(maxWidth+minWidth)*3.5。
When the ratio of minWidth and maxWidth is less than 0.8, two width value difference comparatively large (namely facial orientation comparatively side) are described, now face width faceWdith is 7 times of maxWidth, that is: faceWidth=7*maxWidth.
Finally, after known eye-level eyeHeight and face width faceWidth, by the rateMean that learning phase calculates, the size now comparing ratio r ate and the rateMean of eyeHeight and faceWidth of each frame opens to what judge eyes the state of closing.As rate<rateMean/2 for closing one's eyes, otherwise for opening eyes.
For the judgement of mouth states, first obtain height mouthHeight that face opens and face height faceHeight by face feature point, when the ratio of mouthHeight and faceHeight is greater than 8%, then thinks and be in the state of yawning.
Head 3 d pose detects:
In the middle of driver drives process, the improper head pose affecting traffic safety is mainly deflection and the pitching up and down of head.
Deflection calculates by following methods.When face is in front, subnasal point is equal with the angle of eyespot outside left and right.When face deflection, after namely the side degree of depth rotates, outside subnasal point and left and right, the angle difference of eyespot is β eye_out, subnasal point β poor with the angle of eyespot in left and right can be calculated to obtain simultaneously eye_in, the angle difference β of subnasal point and left and right corners of the mouth point mouth, in order to reduce error, the side degree of depth anglec of rotation of desirable face is the average of these three angles, that is: &beta; 0 = 1 3 ( &beta; eye _ out + &beta; eye _ in + &beta; mouth ) .
Elevation angle can calculate by the following method.The side view of face can be regarded as an ellipse, and y-axis is oval middle separated time, and x-axis is the perpendicular bisector of eyes and mouth line, does not so have upper and lower pitching, has α when namely vertical depth rotates 12.After vertical depth rotates, x-axis is no longer the perpendicular bisector of eyes and mouth line, and according to the character of isosceles triangle, the computing formula of the side degree of depth anglec of rotation is: &alpha; 0 = &alpha; 1 - &alpha; 2 2 , As shown in Figure 1.
α 0, β 0for approximate Attitude estimation value, utilize α 0, β 0as initial value, use quasi-Newton method to human face posture Exact Solution, can in the hope of the accurate deflection angle of human face posture.
Tired and improper attitude driving condition judges:
System adopts PERCLOS principle, yawn frequency combines to judge whether driver is in fatigue.Time scale shared when PERCLOS principle refers to eyes closed in certain hour.In the middle of driving procedure, detect the PERCLOS value of driver's eyes in real time, if PERCLOS value is greater than thresholding, in native system, get PERCLOS>0.35, then carry out tired audible alarm.
Meanwhile, it is also one of outward manifestation of fatigue state that eyes are crossed closed for a long time, and has important impact to safe driving, therefore, when eye-closing period is greater than safety value and the eye-closing period T of setting c> 1.2 seconds, also can carry out fatigue warning.
Finally, people is that yawning number of times also can increase in fatigue, and within a period of time, yawning number of times exceedes predetermined threshold, also thinks there is fatigue state.As yawn frequency M>0.8 per minute, be judged as fatigue state, carry out voice message.
The driver head's 3 d pose data obtained before utilization, whether real-time judge driver watches front attentively, and whether attention of driving is concentrated.When attitude left avertence >poseleft or right avertence >poseright being detected or the >poseup or nutation >posedown that faces upward, and continue duration >Tpose, then think that driver attention is not concentrated.Poseleft, poseright can according to the putting position of photographic head slightly differences.When photographic head is placed at driver's front dead center, poseleft, poseright are 20 degree, and poseup, posedown are 20 degree, and Tpose is 1.5 seconds.
When driver attention is not concentrated, can carry out attitude warning, prompting driver attention is concentrated and is driven.
It should be noted that, for said method embodiment, in order to simple description, therefore it is all expressed as a series of combination of actions, but those skilled in the art should know, the application is not by the restriction of described sequence of movement, because according to the application, some step can adopt other orders or carry out simultaneously.Secondly, those skilled in the art also should know, the embodiment described in description all belongs to preferred embodiment, and involved action and module might not be that the application is necessary.
Those skilled in the art should understand, the embodiment of the application can be provided as method, system or computer program.Therefore, the application can adopt the form of complete hardware embodiment, completely software implementation or the embodiment in conjunction with software and hardware aspect.
Last it is noted that the foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, although with reference to previous embodiment to invention has been detailed description, for a person skilled in the art, it still can be modified to the technical scheme described in foregoing embodiments, or carries out equivalent replacement to wherein portion of techniques feature.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (10)

1. the fatigue under multi-pose and improper attitude drive a detection method, it is characterized in that, comprising:
Step 1: image acquisition, gathers driver's image information;
Step 2: image recognition, carries out real-time face detection to the image collected;
Step 3: face feature point is demarcated, and comprising: the characteristic point of the face detected being carried out to eyes and face is demarcated;
Step 4: eyes and mouth states judge, comprising: adopt based on the eyes detection methods analyst human eye of facial width information and the closure state of face characteristic point;
Step 5: head 3 d pose detects, and comprising: utilize geometric properties to detect head 3 d pose;
Step 6: tired and improper attitude driving condition judges, comprising: the closure state according to human eye and face judges whether driver is in fatigue, and judges whether driver is in improper driving condition according to head 3 d pose.
2. the fatigue under multi-pose according to claim 1 and improper attitude drive detection method, it is characterized in that, in step 2, comprising: adopt the method for statistical analysis and machine learning to summarize face sample and non-face sample statistical nature separately;
Build the grader distinguishing respective feature, realize Face detection with grader and detect.
3. the fatigue under multi-pose according to claim 2 and improper attitude drive detection method, it is characterized in that, use AdaBoost algorithm to detect face location.
4. the fatigue under multi-pose according to claim 1 and improper attitude drive detection method, it is characterized in that, in step 3, adopt supervision gradient descent method to carry out face feature point demarcation, supervision gradient descent method is mainly divided into training and detects two links;
Training part was carried out operating and is calculated before system cloud gray model, and obtained the regression iterative parameter of facial modeling part, to realize positioning eyes, nose, face characteristic point in face, comprising:
To characteristic points such as the facial image manual markings eyes face noses in all training storehouses, and obtain an average face;
Obtain the disturbance parameter of average face, i.e. the average of zooming and panning and standard deviation, do Gauss distribution sampling for every piece image with this average and standard deviation, obtain the training initial value x of eigenvalue point 0, and calculate the sift feature φ of all initial values point 0,
Gradient Descent direction and the deviation factors thereof of eigenvalue point can be obtained by formula (1), wherein, { d itraining storehouse in face image set, it is the characteristic point true value that facial image concentrates all manual markings. training characteristics value point x iwith true value point between matrix of differences, φ ibe the sift feature of eigenvalue point, R is Gradient Descent direction, and b is deviation factors, and k represents iterations;
arg min R k , b k &Sigma; d i &Sigma; x k i | | &Delta; x * ki - R k &phi; k i - b k | | 2 - - - ( 1 )
By formula (2), the characteristic point x in each width facial image is upgraded, and recalculates the sift feature upgrading rear characteristic point,
x k=x k-1+R k-1φ k-1+b k-1(2)
Finally, iterative is carried out to formula 1 and formula 2, eigenvalue point x kconverge on true value point x *, now train end, the Gradient Descent direction R in the middle of each iterative process of finally trying to achieve kwith deviation factors b knamely required regression iterative parameter is detected.
5. the fatigue under multi-pose according to claim 4 and improper attitude drive detection method, it is characterized in that, in step 3, at detection-phase, specifically comprise:
First, average for the standard obtained in training process face sample is navigated to photographic head and detect initial coordinate as characteristic point in the middle of the facial image that obtains;
Secondly, calculate the sift feature of all initial coordinate point, be designated as φ 0;
Finally, carry out regression iterative by formula (3) and calculate final facial characteristics point coordinates,
x k=x k-1+R k-1φ k-1+b k-1(3)
Wherein, R kand b kthe regression iterative parameter obtained the training stage, φ kfor the sift feature of each iterative characteristic point.
6. the fatigue under multi-pose according to claim 1 and improper attitude drive detection method, it is characterized in that, in step 4, eyes mouth states judges mainly to be divided into two parts, and namely learn and detect, first 4 seconds is learning phase, enters detection-phase afterwards;
At learning phase, pass through face feature point, obtain the eye-level eyeHeight in each two field picture and face width faceWidth, calculate the ratio r ate of each frame eyeHeight and faceWidth, finally obtain the meansigma methods rateMean of rate value in all frames, as the facial characteristics of this driver, the closure state of eyes will be used for judging at detection-phase;
At detection-phase, by face feature point, obtain left eye width leftEyeWidth and right eye width rightEyeWidth, compare the size of leftEyeWidth and rightEyeWidth, obtain larger width maxWidth and less width minWidth; When the ratio of minWidth and maxWidth is greater than 0.8, illustrate that two width values are close, now, face width faceWidth be two width and 3.5 times.Namely
faceWidth=(maxWidth+minWidth)*3.5;
When the ratio of minWidth and maxWidth is less than 0.8, illustrate that two width value difference are comparatively large, now face width faceWdith is 7 times of maxWidth, that is:
faceWidth=7*maxWidth;
Finally, after known eye-level eyeHeight and face width faceWidth, by the rateMean that learning phase calculates, the size now comparing ratio r ate and the rateMean of eyeHeight and faceWidth of each frame opens to what judge eyes the state of closing, as rate<rateMean/2 for closing one's eyes, otherwise for opening eyes.
7. the fatigue under multi-pose according to claim 6 and improper attitude drive detection method, it is characterized in that, in step 4, for the judgement of mouth states, comprising:
First obtain height mouthHeight that face opens and face height faceHeight by face feature point, when the ratio of mouthHeight and faceHeight is greater than 8%, then thinks and be in the state of yawning.
8. the fatigue under multi-pose according to claim 1 and improper attitude drive detection method, it is characterized in that, in step 5, specifically comprise: detect driver and whether be in the deflection of head and upper and lower pitch attitude, wherein, deflection calculates by following methods:
When face is in front, subnasal point is equal with the angle of eyespot outside left and right,
When face deflection, after namely the side degree of depth rotates, outside subnasal point and left and right, the angle difference of eyespot is β eye_out, subnasal point β poor with the angle of eyespot in left and right can be calculated to obtain simultaneously eye_in, subnasal point and the left and right corners of the mouth
Elevation angle can calculate by the following method, comprising:
The side view of face can be regarded as an ellipse, and y-axis is oval middle separated time, and x-axis is the perpendicular bisector of eyes and mouth line, does not so have upper and lower pitching, has α when namely vertical depth rotates 12.After vertical depth rotates, x-axis is no longer the perpendicular bisector of eyes and mouth line, and according to the character of isosceles triangle, the computing formula of the side degree of depth anglec of rotation is:
α 0, β 0for approximate Attitude estimation value, utilize α 0, β 0as initial value, use quasi-Newton method to human face posture Exact Solution, can in the hope of the accurate deflection angle of human face posture.
9. the fatigue under multi-pose according to claim 1 and improper attitude drive detection method, it is characterized in that, in step 6, specifically comprise:
In the middle of driving procedure, detect the PERCLOS value of driver's eyes in real time, if PERCLOS value is greater than thresholding, get PERCLOS>0.35, then carry out tired audible alarm;
When eye-closing period is greater than safety value and the eye-closing period T of setting c> 1.2 seconds, also can carry out fatigue warning;
As yawn frequency M>0.8 per minute, be judged as fatigue state, carry out voice message.
10. the fatigue under multi-pose according to claim 9 and improper attitude drive detection method, it is characterized in that, in step 6, utilize driver head's 3 d pose data further, whether real-time judge driver watches front attentively, and whether attention of driving is concentrated, and comprising:
When attitude left avertence >poseleft or right avertence >poseright being detected or the >poseup or nutation >posedown that faces upward, and continue duration >Tpose, then think that driver attention is not concentrated, Poseleft, poseright can according to the putting position of photographic head slightly differences;
When photographic head is placed at driver's front dead center, poseleft, poseright are 20 degree, and poseup, posedown are 20 degree, and Tpose is 1.5 seconds; When driver attention is not concentrated, carry out attitude warning, prompting driver attention is concentrated and is driven.
CN201510092886.1A 2015-03-02 2015-03-02 Method for detecting fatigue driving and driving in abnormal posture under multiple postures Pending CN104688251A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510092886.1A CN104688251A (en) 2015-03-02 2015-03-02 Method for detecting fatigue driving and driving in abnormal posture under multiple postures

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510092886.1A CN104688251A (en) 2015-03-02 2015-03-02 Method for detecting fatigue driving and driving in abnormal posture under multiple postures

Publications (1)

Publication Number Publication Date
CN104688251A true CN104688251A (en) 2015-06-10

Family

ID=53336119

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510092886.1A Pending CN104688251A (en) 2015-03-02 2015-03-02 Method for detecting fatigue driving and driving in abnormal posture under multiple postures

Country Status (1)

Country Link
CN (1) CN104688251A (en)

Cited By (39)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105117681A (en) * 2015-06-29 2015-12-02 电子科技大学 Multi-characteristic fatigue real-time detection method based on Android
CN105216690A (en) * 2015-09-22 2016-01-06 何承鹏 A kind of vehicle security drive responds to system for prompting automatically
CN105354986A (en) * 2015-11-12 2016-02-24 熊强 Driving state monitoring system and method for automobile driver
CN105488957A (en) * 2015-12-15 2016-04-13 小米科技有限责任公司 Fatigue driving detection method and apparatus
CN105632104A (en) * 2016-03-18 2016-06-01 内蒙古大学 Fatigue driving detection system and method
CN105769120A (en) * 2016-01-27 2016-07-20 深圳地平线机器人科技有限公司 Fatigue driving detection method and device
CN106125643A (en) * 2016-06-22 2016-11-16 华东师范大学 A kind of industry control safety protection method based on machine learning techniques
CN106326912A (en) * 2015-07-01 2017-01-11 埃西勒国际通用光学公司 A user fatigue level analysis component
CN106355147A (en) * 2016-08-26 2017-01-25 张艳 Acquiring method and detecting method of live face head pose detection regression apparatus
CN106355838A (en) * 2016-10-28 2017-01-25 深圳市美通视讯科技有限公司 Fatigue driving detection method and system
CN106485191A (en) * 2015-09-02 2017-03-08 腾讯科技(深圳)有限公司 A kind of method for detecting fatigue state of driver and system
WO2017129306A1 (en) * 2016-01-29 2017-08-03 Robert Bosch Gmbh Method and device for locating a head of a driver in a vehicle and head localization system
CN107194346A (en) * 2017-05-19 2017-09-22 福建师范大学 A kind of fatigue drive of car Forecasting Methodology
CN107392159A (en) * 2017-07-27 2017-11-24 竹间智能科技(上海)有限公司 A kind of facial focus detecting system and method
CN107491746A (en) * 2017-08-02 2017-12-19 安徽慧视金瞳科技有限公司 A kind of face prescreening method based on the analysis of big gradient pixel
CN107491769A (en) * 2017-09-11 2017-12-19 中国地质大学(武汉) Method for detecting fatigue driving and system based on AdaBoost algorithms
CN107491302A (en) * 2017-07-31 2017-12-19 广东欧珀移动通信有限公司 terminal control method and device
CN107704836A (en) * 2017-10-17 2018-02-16 电子科技大学 Method for detecting fatigue driving based on object detection
CN107913058A (en) * 2017-12-07 2018-04-17 中车大连机车车辆有限公司 Locomotive driver and conductor's health monitor system
CN108108684A (en) * 2017-12-15 2018-06-01 杭州电子科技大学 A kind of attention detection method for merging line-of-sight detection
CN108108651A (en) * 2016-11-25 2018-06-01 广东亿迅科技有限公司 The non-wholwe-hearted driving detection method of driver and system based on video human face analysis
CN108162755A (en) * 2017-12-25 2018-06-15 芜湖皖江知识产权运营中心有限公司 A kind of fatigue driving identification control method applied in intelligent vehicle
CN108304745A (en) * 2017-01-10 2018-07-20 普天信息技术有限公司 A kind of driver's driving behavior detection method, device
CN108363968A (en) * 2018-01-31 2018-08-03 上海瀚所信息技术有限公司 A kind of tired driver driving monitoring system and method based on key point extraction
CN108647616A (en) * 2018-05-01 2018-10-12 南京理工大学 Real-time drowsiness detection method based on facial characteristics
CN109272764A (en) * 2018-09-30 2019-01-25 广州鹰瞰信息科技有限公司 A kind of based reminding method and system of dangerous driving
CN109409174A (en) * 2017-08-18 2019-03-01 安徽三联交通应用技术股份有限公司 Driving human face monitoring method, system, medium and equipment based on deep learning
CN109614892A (en) * 2018-11-26 2019-04-12 青岛小鸟看看科技有限公司 A kind of method for detecting fatigue driving, device and electronic equipment
CN109617967A (en) * 2018-12-14 2019-04-12 深圳市曹操货的科技有限公司 One kind being based on big data Intelligent internet of things control platform
CN109670421A (en) * 2018-12-04 2019-04-23 青岛小鸟看看科技有限公司 A kind of fatigue state detection method and device
CN109791432A (en) * 2016-09-22 2019-05-21 苹果公司 The state for postponing the information for influencing graphic user interface changes until not during absorbed situation
CN109919049A (en) * 2019-02-21 2019-06-21 北京以萨技术股份有限公司 Fatigue detection method based on deep learning human face modeling
CN110021147A (en) * 2019-05-07 2019-07-16 四川九洲视讯科技有限责任公司 A kind of method for detecting fatigue driving demarcated based on machine learning and numerical value
CN110287916A (en) * 2019-06-28 2019-09-27 浙江吉利控股集团有限公司 It is a kind of for judging the method and system of driver attention
CN110363052A (en) * 2018-04-11 2019-10-22 杭州海康威视数字技术股份有限公司 Determine the method, apparatus and computer equipment of the human face posture in image
CN110751011A (en) * 2019-05-23 2020-02-04 北京嘀嘀无限科技发展有限公司 Driving safety detection method, driving safety detection device and vehicle-mounted terminal
CN110766912A (en) * 2018-07-27 2020-02-07 长沙智能驾驶研究院有限公司 Driving early warning method, device and computer readable storage medium
WO2020237664A1 (en) * 2019-05-31 2020-12-03 驭势(上海)汽车科技有限公司 Driving prompt method, driving state detection method and computing device
WO2021159269A1 (en) * 2020-02-11 2021-08-19 云图技术有限公司 Vision-based safe driving early warning method and system, and storage medium

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1495658A (en) * 2002-06-30 2004-05-12 贺贵明 Driver's face image identification and alarm device and method
CN101877051A (en) * 2009-10-30 2010-11-03 江苏大学 Driver attention state monitoring method and device
CN104361716A (en) * 2014-10-31 2015-02-18 新疆宏开电子系统集成有限公司 Method for detecting and reminding fatigue in real time

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1495658A (en) * 2002-06-30 2004-05-12 贺贵明 Driver's face image identification and alarm device and method
CN101877051A (en) * 2009-10-30 2010-11-03 江苏大学 Driver attention state monitoring method and device
CN104361716A (en) * 2014-10-31 2015-02-18 新疆宏开电子系统集成有限公司 Method for detecting and reminding fatigue in real time

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
XUEHAN XIONG 等: "Supervised Descent Method and Its Applications to Face Alignment", 《PROCESS OF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION》 *

Cited By (48)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105117681A (en) * 2015-06-29 2015-12-02 电子科技大学 Multi-characteristic fatigue real-time detection method based on Android
CN105117681B (en) * 2015-06-29 2018-06-08 电子科技大学 Multiple features fatigue real-time detection method based on Android
CN106326912A (en) * 2015-07-01 2017-01-11 埃西勒国际通用光学公司 A user fatigue level analysis component
CN106485191B (en) * 2015-09-02 2018-12-11 腾讯科技(深圳)有限公司 A kind of method for detecting fatigue state of driver and system
CN106485191A (en) * 2015-09-02 2017-03-08 腾讯科技(深圳)有限公司 A kind of method for detecting fatigue state of driver and system
CN105216690A (en) * 2015-09-22 2016-01-06 何承鹏 A kind of vehicle security drive responds to system for prompting automatically
CN105354986A (en) * 2015-11-12 2016-02-24 熊强 Driving state monitoring system and method for automobile driver
CN105354986B (en) * 2015-11-12 2017-12-01 熊强 Driver's driving condition supervision system and method
CN105488957A (en) * 2015-12-15 2016-04-13 小米科技有限责任公司 Fatigue driving detection method and apparatus
CN105488957B (en) * 2015-12-15 2018-06-12 小米科技有限责任公司 Method for detecting fatigue driving and device
CN105769120B (en) * 2016-01-27 2019-01-22 深圳地平线机器人科技有限公司 Method for detecting fatigue driving and device
CN105769120A (en) * 2016-01-27 2016-07-20 深圳地平线机器人科技有限公司 Fatigue driving detection method and device
WO2017129306A1 (en) * 2016-01-29 2017-08-03 Robert Bosch Gmbh Method and device for locating a head of a driver in a vehicle and head localization system
CN105632104A (en) * 2016-03-18 2016-06-01 内蒙古大学 Fatigue driving detection system and method
CN105632104B (en) * 2016-03-18 2019-03-01 内蒙古大学 A kind of fatigue driving detecting system and method
CN106125643A (en) * 2016-06-22 2016-11-16 华东师范大学 A kind of industry control safety protection method based on machine learning techniques
CN106355147A (en) * 2016-08-26 2017-01-25 张艳 Acquiring method and detecting method of live face head pose detection regression apparatus
CN109791432A (en) * 2016-09-22 2019-05-21 苹果公司 The state for postponing the information for influencing graphic user interface changes until not during absorbed situation
CN106355838A (en) * 2016-10-28 2017-01-25 深圳市美通视讯科技有限公司 Fatigue driving detection method and system
CN108108651A (en) * 2016-11-25 2018-06-01 广东亿迅科技有限公司 The non-wholwe-hearted driving detection method of driver and system based on video human face analysis
CN108304745A (en) * 2017-01-10 2018-07-20 普天信息技术有限公司 A kind of driver's driving behavior detection method, device
CN107194346A (en) * 2017-05-19 2017-09-22 福建师范大学 A kind of fatigue drive of car Forecasting Methodology
CN107392159A (en) * 2017-07-27 2017-11-24 竹间智能科技(上海)有限公司 A kind of facial focus detecting system and method
CN107491302A (en) * 2017-07-31 2017-12-19 广东欧珀移动通信有限公司 terminal control method and device
CN107491746A (en) * 2017-08-02 2017-12-19 安徽慧视金瞳科技有限公司 A kind of face prescreening method based on the analysis of big gradient pixel
CN107491746B (en) * 2017-08-02 2020-07-17 安徽慧视金瞳科技有限公司 Face pre-screening method based on large gradient pixel analysis
CN109409174A (en) * 2017-08-18 2019-03-01 安徽三联交通应用技术股份有限公司 Driving human face monitoring method, system, medium and equipment based on deep learning
CN107491769A (en) * 2017-09-11 2017-12-19 中国地质大学(武汉) Method for detecting fatigue driving and system based on AdaBoost algorithms
CN107704836A (en) * 2017-10-17 2018-02-16 电子科技大学 Method for detecting fatigue driving based on object detection
CN107913058A (en) * 2017-12-07 2018-04-17 中车大连机车车辆有限公司 Locomotive driver and conductor's health monitor system
CN108108684B (en) * 2017-12-15 2020-07-17 杭州电子科技大学 Attention detection method integrating sight detection
CN108108684A (en) * 2017-12-15 2018-06-01 杭州电子科技大学 A kind of attention detection method for merging line-of-sight detection
CN108162755A (en) * 2017-12-25 2018-06-15 芜湖皖江知识产权运营中心有限公司 A kind of fatigue driving identification control method applied in intelligent vehicle
CN108363968A (en) * 2018-01-31 2018-08-03 上海瀚所信息技术有限公司 A kind of tired driver driving monitoring system and method based on key point extraction
CN110363052A (en) * 2018-04-11 2019-10-22 杭州海康威视数字技术股份有限公司 Determine the method, apparatus and computer equipment of the human face posture in image
CN108647616A (en) * 2018-05-01 2018-10-12 南京理工大学 Real-time drowsiness detection method based on facial characteristics
CN110766912A (en) * 2018-07-27 2020-02-07 长沙智能驾驶研究院有限公司 Driving early warning method, device and computer readable storage medium
CN109272764A (en) * 2018-09-30 2019-01-25 广州鹰瞰信息科技有限公司 A kind of based reminding method and system of dangerous driving
CN109614892A (en) * 2018-11-26 2019-04-12 青岛小鸟看看科技有限公司 A kind of method for detecting fatigue driving, device and electronic equipment
CN109670421B (en) * 2018-12-04 2021-04-16 青岛小鸟看看科技有限公司 Fatigue state detection method and device
CN109670421A (en) * 2018-12-04 2019-04-23 青岛小鸟看看科技有限公司 A kind of fatigue state detection method and device
CN109617967A (en) * 2018-12-14 2019-04-12 深圳市曹操货的科技有限公司 One kind being based on big data Intelligent internet of things control platform
CN109919049A (en) * 2019-02-21 2019-06-21 北京以萨技术股份有限公司 Fatigue detection method based on deep learning human face modeling
CN110021147A (en) * 2019-05-07 2019-07-16 四川九洲视讯科技有限责任公司 A kind of method for detecting fatigue driving demarcated based on machine learning and numerical value
CN110751011A (en) * 2019-05-23 2020-02-04 北京嘀嘀无限科技发展有限公司 Driving safety detection method, driving safety detection device and vehicle-mounted terminal
WO2020237664A1 (en) * 2019-05-31 2020-12-03 驭势(上海)汽车科技有限公司 Driving prompt method, driving state detection method and computing device
CN110287916A (en) * 2019-06-28 2019-09-27 浙江吉利控股集团有限公司 It is a kind of for judging the method and system of driver attention
WO2021159269A1 (en) * 2020-02-11 2021-08-19 云图技术有限公司 Vision-based safe driving early warning method and system, and storage medium

Similar Documents

Publication Publication Date Title
CN104688251A (en) Method for detecting fatigue driving and driving in abnormal posture under multiple postures
Fridman et al. Cognitive load estimation in the wild
Wang et al. Driver fatigue detection: a survey
CN104616438B (en) A kind of motion detection method of yawning for fatigue driving detection
CN105719431A (en) Fatigue driving detection system
CN104361716A (en) Method for detecting and reminding fatigue in real time
CN105117681B (en) Multiple features fatigue real-time detection method based on Android
CN101593425B (en) Machine vision based fatigue driving monitoring method and system
Vural et al. Automated drowsiness detection for improved driving safety
CN104574817A (en) Machine vision-based fatigue driving pre-warning system suitable for smart phone
CN103824420B (en) Fatigue driving identification system based on heart rate variability non-contact measurement
Batista A drowsiness and point of attention monitoring system for driver vigilance
CN103479367B (en) A kind of Driver Fatigue Detection based on facial movement unit identification
CN105260703B (en) A kind of driver&#39;s smoking behavioral value method suitable under multi-pose
Sukno et al. Automatic assessment of eye blinking patterns through statistical shape models
TW201140511A (en) Drowsiness detection method
Tadesse et al. Driver drowsiness detection through HMM based dynamic modeling
Hossain et al. IOT based real-time drowsy driving detection system for the prevention of road accidents
CN109308445B (en) A kind of fixation post personnel fatigue detection method based on information fusion
CN110103816A (en) A kind of driving condition detection method
Ursulescu et al. Driver drowsiness detection based on eye analysis
CN109543651A (en) A kind of driver&#39;s dangerous driving behavior detection method
Bhoyar et al. Implementation on visual analysis of eye state using image processing for driver fatigue detection
Li et al. Fatigue driving detection system based on face feature
Anumas et al. Driver fatigue monitoring system using video face images & physiological information

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20150610