CN108009495A - Fatigue driving method for early warning - Google Patents

Fatigue driving method for early warning Download PDF

Info

Publication number
CN108009495A
CN108009495A CN201711237773.1A CN201711237773A CN108009495A CN 108009495 A CN108009495 A CN 108009495A CN 201711237773 A CN201711237773 A CN 201711237773A CN 108009495 A CN108009495 A CN 108009495A
Authority
CN
China
Prior art keywords
driver
eyes
image
aperture
fatigue driving
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.)
Withdrawn
Application number
CN201711237773.1A
Other languages
Chinese (zh)
Inventor
左瑜
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.)
Xian Cresun Innovation Technology Co Ltd
Original Assignee
Xian Cresun Innovation 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 Xian Cresun Innovation Technology Co Ltd filed Critical Xian Cresun Innovation Technology Co Ltd
Priority to CN201711237773.1A priority Critical patent/CN108009495A/en
Publication of CN108009495A publication Critical patent/CN108009495A/en
Withdrawn legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/00624Recognising scenes, i.e. recognition of a whole field of perception; recognising scene-specific objects
    • G06K9/00832Recognising scenes inside a vehicle, e.g. related to occupancy, driver state, inner lighting conditions
    • G06K9/00845Recognising the driver's state or behaviour, e.g. attention, drowsiness
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/00221Acquiring or recognising human faces, facial parts, facial sketches, facial expressions
    • G06K9/00268Feature extraction; Face representation
    • G06K9/00281Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/36Image preprocessing, i.e. processing the image information without deciding about the identity of the image
    • G06K9/46Extraction of features or characteristics of the image
    • G06K9/4642Extraction of features or characteristics of the image by performing operations within image blocks or by using histograms
    • GPHYSICS
    • G08SIGNALLING
    • G08BSIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00Alarms responsive to a single specified undesired or abnormal operating condition and not elsewhere provided for
    • G08B21/02Alarms for ensuring the safety of persons
    • G08B21/06Alarms for ensuring the safety of persons indicating a condition of sleep, e.g. anti-dozing alarms

Abstract

The present invention relates to a kind of fatigue driving method for early warning, including:The characteristic image of driver is obtained in real time;The characteristic image is handled, to obtain the fatigue strength of the driver;If the fatigue strength exceedes fatigue strength threshold value, judge that the driver is in fatigue driving state, and generate alerting signal.Fatigue driving method for early warning provided by the invention, the judgement promptness of fatigue driving is good, and accuracy is high, occurs to remind in time during fatigue driving, substantially increases drive safety.

Description

Fatigue driving method for early warning
Technical field
The present invention relates to field of image recognition, more particularly to a kind of fatigue driving method for early warning.
Background technology
As transportation is quickly grown, the trend that road traffic accident quantity rises is more obvious, wherein, due to tired Please the ratio that traffic accident caused by sailing accounts for total number of accident is higher always, causes huge economic loss and casualties, The strong interest of people is caused, how the state of mind of driver has been grasped and promptly responded with weight in time The meaning wanted.
In spite of legally providing, should stop rest when continuous operating motor vehicles are small more than 4, and the parking time of having a rest is no less than 20 minutes, but this actual implementation dynamics deficiency.Also some prior arts are proposed by detecting driver's pulse with to driving The physical condition for the person of sailing is monitored, but because the individual difference of driver is obvious, the accuracy of detection is not high, it is difficult to play and When, the effect of accurate detection driver fatigue state.
The content of the invention
Therefore, to solve technological deficiency and deficiency existing in the prior art, the present invention proposes a kind of pre- police of fatigue driving Method, this method include:
The characteristic image of driver is obtained in real time;
Processing feature image, to obtain the fatigue strength of driver;
If fatigue strength exceedes fatigue strength threshold value, judge that driver is in fatigue driving state, and generate alerting signal.
In one embodiment of the invention, characteristic image includes head image.
In one embodiment of the invention, processing feature image, including:
Edge extracting is carried out to head image, to extract the contouring head information of driver;
Judge the deviation range between the position of contouring head and default contouring head position is within the period of setting It is no to exceed contour offset threshold value.
In one embodiment of the invention, before the head image of driver is obtained in real time, further include:
The position-scheduled position of face contour is carried out in initial drive to driver, to obtain driver in non-fatigue driving shape Face contour position during state;
Face contour position of the driver in non-fatigue driving state is arranged to default contouring head position.
In one embodiment of the invention, characteristic image includes eyes image.
In one embodiment of the invention, processing feature image, including:
The eyes image of driver is handled, obtains driver's eyes aperture;
Judge whether driver's eyes aperture and the difference of default eyes aperture are below eyes within the period of setting Aperture deviation threshold.
In one embodiment of the invention, before the eyes image of driver is obtained in real time, further include:
Eyes aperture pre-determined bit is carried out in initial drive to driver, to obtain driver in non-fatigue driving state Eyes aperture;
Eyes aperture of the driver in non-fatigue driving state is arranged to default eyes aperture.
In one embodiment of the invention, processing feature image, including:
The eyes image of driver is handled, obtains the eyes aperture of driver;
If the eyes aperture of driver is higher than aperture threshold value, driver's eyes are set to open;Otherwise, driver is judged Eyes are closure;Within the period of setting, if driver's eyes are not only once opened, but also once close, then setting drives Member's eyes are blink;
Judge whether the frequency of wink of driver's eyes is higher than frequency of wink threshold value.
In one embodiment of the invention, the eyes aperture of driver is obtained, including:
Binaryzation eye gray-scale map is formed according to eyes image;
Canthus point is determined according to binaryzation eye gray-scale map;
Eye state is determined according to canthus point.
In one embodiment of the invention, binaryzation eye gray-scale map is formed according to eyes image, including:
According to the contrast of area of skin color, eyeball region and white of the eye region in eyes image, calculated using three color shade watersheds Method, extracts informer's profile, forms binaryzation eye gray-scale map.
A kind of the advantages of fatigue driving method for early warning provided by the invention, has:
1) fatigue state of driver can be found in time;
2) detection method accuracy provided by the invention is high;
3) detection method provided by the invention is implemented without equipment costly, it is of low cost.
By the detailed description below with reference to attached drawing, other side of the invention and feature become obvious.But it should know Road, attached drawing and explanation are only the purpose design explained, not as the restriction of the scope of the present invention, this is because it should With reference to appended claims.It should also be noted that unless otherwise noted, it is not necessary to which scale attached drawing, they only try hard to Conceptually illustrate structure and flow described herein.
Brief description of the drawings
Below in conjunction with attached drawing, the embodiment of the present invention is described in detail.
Fig. 1 is a kind of fatigue driving method for early warning flow diagram that the embodiment of the present invention proposes;
Fig. 2 is another fatigue driving method for early warning flow diagram that the embodiment of the present invention proposes;
Fig. 3 is another fatigue driving method for early warning flow diagram that the embodiment of the present invention proposes;
Fig. 4 is a kind of eyes image schematic diagram provided in an embodiment of the present invention;
Fig. 5 is a kind of eyes image grey level histogram provided in an embodiment of the present invention;
Fig. 6 is a kind of eyes image binaryzation grey level histogram provided in an embodiment of the present invention;
Fig. 7 is a kind of binaryzation eye gray-scale map schematic diagram provided in an embodiment of the present invention;
Fig. 8 is a kind of eye angle point grid schematic diagram provided in an embodiment of the present invention;
Fig. 9 is a kind of canthus angle schematic diagram provided in an embodiment of the present invention.
Embodiment
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, below in conjunction with the accompanying drawings to the present invention Embodiment be described in detail.
Embodiment one
Please refer to Fig.1, a kind of fatigue driving method for early warning flow diagram that Fig. 1 proposes for the embodiment of the present invention, the party Method includes:
The characteristic image of driver is obtained in real time;
Processing feature image, to obtain the fatigue strength of driver;
If fatigue strength exceedes fatigue strength threshold value, judge that driver is in fatigue driving state, and generate alerting signal.
The characteristic image acquisition device of driver in the present embodiment can be mounted in the shooting of in-car specific location Head, or the terminal device of driver, such as possess and take pictures, the mobile phone of recording function.For example, it can be incited somebody to action by wireless network The video of camera shooting is sent in cloud server, is detected by the video image of cloud server driver, to every Two field picture is detected and the feature to occurring in image positions.For example, cloud server in analysis diagram picture by driving Member be both hands whether off-direction disk, whether have the continuous serious fatigue driving behavior such as yawn.Certainly, adjacent two field pictures Between difference it is smaller, while in order to save computing resource, cloud server can only handle the figure at interval of certain frame number Picture, for example, only handling the 1st frame, the 11st frame, the image of the 21st frame.
Embodiment two
Further, on the basis of above-described embodiment, characteristic image includes head image.Correspondingly, processing feature figure Picture, is specially:
Edge extracting is carried out to head image, to extract the contouring head information of driver;
Judge the deviation range between the position of contouring head and default contouring head position is within the period of setting It is no to exceed contour offset threshold value.
Embodiment three
Further, on the basis of above-described embodiment, Fig. 2 is referred to, Fig. 2 is the another kind that the embodiment of the present invention proposes Fatigue driving method for early warning flow diagram;In the present embodiment, before the head image of driver is obtained in real time, also wrap Include:The position-scheduled position of face contour is carried out in initial drive to driver, to obtain driver in non-fatigue driving state Face contour position;
Face contour position of the driver in non-fatigue driving state is arranged to default contouring head position.
Driver is easy to doze off and bow in fatigue driving, when its face contour position is relative to non-fatigue driving There is obvious offset.The drift condition of face contour can be determined by the edge extracting of face-image, can also be by fixed Whether the characteristic image in the face-image of position, such as the position of lip move down, and then infer whether the face contour of driver is inclined Move past big.
In the present embodiment, contour offset threshold value can by after the driving habit of the long-term acquisition driver by modeling Arrive.Because the driving style of different drivers is variant, for example, for getting used to the driver rocked, may be set in 3s Face contour offset then judges its fatigue driving more than 5cm all the time;Driver for getting used to being absorbed in driving, may be set to Face contour offset, which is missing, in 3s then judges its fatigue driving more than 3cm.
Example IV
On the basis of above-described embodiment, in the present embodiment, characteristic image includes eyes image.Then correspondingly, handle Characteristic image, including:
The eyes image of driver is handled, obtains the eyes aperture of driver;
Judge whether the eyes aperture of driver is below with the difference of default eyes aperture within the period of setting Aperture deviation threshold.
Embodiment five
In one embodiment of the invention, Fig. 3 is referred to, Fig. 3 is another fatigue that the embodiment of the present invention proposes Driving warning method flow diagram;In the present embodiment, before the eyes image of driver is obtained in real time, further include:
Eyes aperture pre-determined bit is carried out in initial drive to driver, to obtain driver in non-fatigue driving state Eyes aperture;
Eyes aperture of the driver in non-fatigue driving state is arranged to default eyes aperture.
The present embodiment by judge the eyes aperture of driver judge driver whether can fatigue driving, this determination methods Conclusion can be provided much sooner, send alerting signal much sooner, avoid the generation of dangerous situation.
Embodiment six
In the present embodiment, processing feature image, including:
The eyes image of driver is handled, obtains the eyes aperture of driver;
If the eyes aperture of driver is higher than aperture threshold value, driver's eyes are judged to open;Otherwise, driver is judged Eyes are closure;Within the period of setting, if driver's eyes are not only once opened, but also once close, then setting drives Member's eyes are blink;
Judge whether the frequency of wink of driver's eyes is higher than frequency of wink threshold value.
Driver automatic can enter the rapid eye movements stage in fatigue driving.By obtaining driver in advance non- Frequency of wink in the case of fatigue driving, for example, continuing to monitor the blink feelings of driver within a certain period of time by capture apparatus Condition, counts the normal blink custom of the driver, and then sets frequency of wink threshold value.
Embodiment seven
Further, on the basis of above-described embodiment, the eyes aperture of driver, Ke Yiwei are obtained:
Binaryzation eye gray-scale map is formed according to eyes image;
Canthus point is determined according to binaryzation eye gray-scale map;
Eye state is determined according to canthus point.
Further, on the basis of the above embodiment, binaryzation eye gray-scale map is formed according to eyes image, can be with For:
According to the contrast of area of skin color, eyeball region and white of the eye region in eyes image, calculated using three color shade watersheds Method, extracts informer's profile, forms binaryzation eye gray-scale map.
Embodiment eight
The present embodiment specifically provides a kind of definite eyes aperture, and then judges eye opening, the method for closed-eye state, this method bag Include following steps:
Step 1, obtain eyes image;
Step 2, according to eyes image formed binaryzation eye gray-scale map;
Step 3, according to binaryzation eye gray-scale map determine canthus point;
Step 4, according to canthus point determine eye state.
Wherein, for step 2, can include:
According to the contrast of area of skin color, eyeball region and white of the eye region in eyes image, calculated using three color shade watersheds Method, extracts informer's profile, forms binaryzation eye gray-scale map.
Wherein, for using three color shade watershed algorithms in step 2, extraction informer profile forms binaryzation eye gray scale Figure, can include:
The gray value of all pixels, forms grey level histogram in step 21, extraction eyes image;
Step 22, the grey value difference according to area of skin color, eyeball region and white of the eye region, by grey level histogram two-value Change, form binaryzation grey level histogram;
Step 23, extract informer's profile formation binaryzation eye gray-scale map according to binaryzation grey level histogram.
Wherein, for binaryzation grey level histogram in step 22 include the first gray value interval, the second gray value interval and 3rd gray value interval;
Wherein, for step 23, can include:
Determine that the first adjusting threshold value, second adjust threshold value and the 3rd adjusting threshold value according to binaryzation grey level histogram;
Threshold value, the second adjusting threshold value and the 3rd adjusting threshold value extraction informer profile, which are adjusted, according to first forms binaryzation eye Gray-scale map.
Wherein, in step 23 according to binaryzation grey level histogram determine the first adjusting threshold value, second adjust threshold value and 3rd adjusts threshold value, can include:
Average is taken to adjust threshold as first the minimum value of the maximum of first gray value interval and the second gray value interval Value;
Average is taken to adjust threshold as second the minimum value of the maximum of second gray value interval and the 3rd gray value interval Value;
Threshold value is adjusted using the maximum of the 3rd gray value interval as the 3rd.
Wherein, for adjusting threshold value, the second adjusting threshold value and the 3rd adjusting threshold value extraction informer according to first in step 23 Profile forms binaryzation eye gray-scale map, can include:
The eyes image of binaryzation is divided into the according to the first adjusting threshold value, the second adjusting threshold value and the 3rd adjusting threshold value One pixel region, the second pixel region and the 3rd pixel region;
Set the first pixel region identical with the gray value of the second pixel region;
Informer's profile is extracted according to the first pixel region, the second pixel region and the 3rd pixel region and forms binaryzation eye Portion's gray-scale map.
Wherein, for step 3, can include:
The first straight line using eyeball center as the center of circle is rotated in binaryzation eye gray-scale map, when in first straight line When the pixel of 3rd pixel region is most, first straight line will form two with the second pixel region and the 3rd pixel region intersection A first intersection point, using two the first intersection points as First view angle point and the second canthus point;
The second straight line using eyeball center as the center of circle is rotated in binaryzation eye gray-scale map, when in second straight line When the pixel of 3rd pixel region is minimum, second straight line will form two with the second pixel region and the 3rd pixel region intersection A second intersection point, using two the second intersection points as the 3rd canthus point and four eyed angle point.Wherein, for step 4, can wrap Include:
Connect First view angle point, the second canthus point, the 3rd canthus point and four eyed angle point and form the first canthus angle, second Canthus angle, the 3rd canthus angle and the 4th canthus angle;
According to the first canthus angle, the second canthus angle, the 3rd canthus angle and the 4th canthus angle, pass through eye state Equations eye state value;
Eye state is determined according to the size of eye state value.
Wherein, it is for eye state equation in step 4:
Wherein, offset0 and offset1 is error compensation, b1, b2, b3, b4 be respectively the first canthus angle, second Angle angle, the 3rd canthus angle and the 4th canthus angle.
Method proposed by the present invention based on canthus angle-determining eye state, learns mould without substantial amounts of high-definition image Plate, can preferably reduce computational complexity, improve real-time, and reliability is high, is with a wide range of applications, of the invention in addition It is of low cost without equipment costly.
Embodiment nine
On the basis of above-described embodiment, the present embodiment to the method based on canthus angle-determining eye state furtherly It is bright.
This method comprises the following steps:
Step 1, obtain eyes image
By the means such as take pictures, the image of simple eye is obtained, as shown in figure 4, Fig. 4 is provided in an embodiment of the present invention one Kind eyes image schematic diagram.
Step 2, extraction informer's profile
According to the contrast of area of skin color, eyeball region and white of the eye region in eyes image, calculated using three color shade watersheds Method, extracts informer's profile.It is specific as follows:
Step a), the gray value for extracting pixel
Extract the gray value of all pixels in eyes image, because area of skin color, eyeball region and this 3 areas of white of the eye region The gray value in domain has notable difference, and in any 1 region in 3 regions, containing a greater number pixel, these pixels Gray value be nearly at same scope.
Step b), set adjusting threshold value
Due to photo accuracy reason, compensation is set;According to the different colours of skin, area of skin color is set;Area of skin color, eyeball area The corresponding threshold value that adjusts in domain and these three regions of white of the eye region is respectively the first adjusting threshold value, the second adjusting threshold value and the 3rd adjusting Threshold value.Regulative mode can be that manually or automatically automatic adjustment mode is as follows:
Step b11), draw histogram curve, as shown in figure 5, Fig. 5 is a kind of eyes image provided in an embodiment of the present invention Grey level histogram.Since white of the eye area grayscale value is maximum, so its gray value region is 200 or so;Eyeball area grayscale value is most It is small, so its gray value region is 25 or so;The gray value of skin area is placed in the middle, so its gray value region is 75 or so.It is false If other gray value area pixel numbers are essentially 0.For three kinds of regions of automatic distinguishing.It is 0 to take all pixels quantity in Fig. 5 Point, shown in equation below:
Wherein i is abscissa, and D is abscissa distance, and H is ordinate value.I.e. using i as starting point, using i+D as all of terminal The average of the sum of ordinate, less than offset k, then this D point, is 0;Otherwise it is 1.
Step b12), Fig. 5 carried out shown in the result figure 6 that is formed after the fortran in step b1, Fig. 6 is real for the present invention A kind of eyes image binaryzation grey level histogram of example offer is provided.Fig. 6 includes the first gray value interval 11, the second gray value interval 12 and the 3rd gray value interval 13.It is final to obtain the first adjusting threshold value, the second adjusting threshold value and the 3rd adjusting threshold value, such as following public affairs Shown in formula:
THR1=(POS1+POS2)/2
THR2=(POS3+POS4)/2
THR3=POS5
Wherein, THR1 adjusts threshold value for first, and THR2 adjusts threshold value for second, and THR3 adjusts threshold value for the 3rd;POS1 is The terminal (i.e. the maximum of the first gray value interval) of first gray value interval 11, POS2 are the starting point of the second gray value interval 12 (i.e. the minimum value of the second gray value interval), POS3 be the second gray value interval 12 terminal (i.e. the second gray value interval is most Big value), POS4 is the starting point (i.e. the minimum value of the 3rd gray value interval) of the 3rd gray value interval 13, and POS5 is the 3rd gray value The terminal (i.e. the maximum of the 3rd gray value interval) in section 13.
Step c), remember for binaryzation grey level histogram, all pixels of 0~THR1 of extraction as the first pixel region For PIX1;The all pixels of THR1~THR2 are extracted as the second pixel region, are denoted as PIX2, extraction THR2~THR3's is all Pixel is denoted as PIX3 as the 3rd pixel region.The gray value size in wherein this 3 regions is ordered as PIX1<PIX2<PIX3. In order to simplify the operation of extraction informer's profile, it is assumed that PIX1=PIX2.Finally formed binaryzation eye gray-scale map such as Fig. 7 institutes Show, Fig. 7 is a kind of binaryzation eye gray-scale map schematic diagram provided in an embodiment of the present invention.
Step 3, determine canthus point
In the binaryzation eye gray-scale map that Fig. 7 is obtained, using eyeball center as the center of circle, use straight line with angle a into Row rotation, until 360 degree of rotations finish.When white pixel point (i.e. the pixel of the 3rd pixel region) is most on final straight line, The point that the straight line has a common boundary with black and white (i.e. the second pixel region and the 3rd pixel region) is First view angle point and the second canthus point, is remembered For T1 and T2;Similarly, when white pixel point is minimum, point that the straight line and black and white are had a common boundary is the 3rd canthus point and four eyed angle point, It is denoted as T3 and T4;As shown in figure 8, Fig. 8 is a kind of eye angle point grid schematic diagram provided in an embodiment of the present invention.
Step 4, judge eye state
As shown in figure 9, Fig. 9 is a kind of canthus angle schematic diagram provided in an embodiment of the present invention.TI, T3, T2, T4 are connected, The first canthus angle, the second canthus angle, the 3rd canthus angle and the 4th canthus angle are obtained, is denoted as b1, b2, b3, b4.According to Eye state equation judges eye state.Eye state equation is as follows:
Wherein, offset0 and offset1 is error compensation, different according to the different error compensations of photographing device.When θ is got over It is small, illustrate eyes closer to closure, otherwise θ is bigger, illustrates that the amplitude that eyes are opened is bigger.As θ=0, eyes close completely, Due to individual difference, eyes, which open state, can define threshold decision.
To sum up, specific case used herein is set forth method provided by the invention, and above example is said It is bright to be only intended to help the method and its core concept for understanding the present invention;Meanwhile for those of ordinary skill in the art, foundation The thought of the present invention, there will be changes in specific embodiments and applications, to sum up, this specification content should not manage Solve as limitation of the present invention, protection scope of the present invention should be subject to appended claim.

Claims (10)

  1. A kind of 1. fatigue driving method for early warning, it is characterised in that including:
    The characteristic image of driver is obtained in real time;
    The characteristic image is handled, to obtain the fatigue strength of the driver;
    If the fatigue strength exceedes fatigue strength threshold value, judge that the driver is in fatigue driving state, and generate prompting letter Number.
  2. 2. the method as described in claim 1, it is characterised in that the characteristic image includes head image.
  3. 3. method as claimed in claim 2, it is characterised in that the characteristic image is handled, including:
    Edge extracting is carried out to the head image, to extract the contouring head information of the driver;
    Judge the deviation range between the position of the contouring head and default contouring head position is within the period of setting It is no to exceed contour offset threshold value.
  4. 4. method as claimed in claim 3, it is characterised in that before the head image of driver is obtained in real time, further include:
    The position-scheduled position of face contour is carried out in initial drive to the driver, is driven with obtaining the driver in non-fatigue Sail face contour position during state;
    Face contour position of the driver in non-fatigue driving state is arranged to default contouring head position.
  5. 5. the method as described in claim 1, it is characterised in that the characteristic image includes eyes image.
  6. 6. method as claimed in claim 5, it is characterised in that the characteristic image is handled, including:
    The eyes image of the driver is handled, obtains the driver's eyes aperture;
    Judge whether the driver's eyes aperture and the difference of default eyes aperture are below eyes within the period of setting Aperture deviation threshold.
  7. 7. method as claimed in claim 6, it is characterised in that before the eyes image of driver is obtained in real time, further include:
    Eyes aperture pre-determined bit is carried out in initial drive to the driver, to obtain the driver in non-fatigue driving shape Eyes aperture during state;
    Eyes aperture of the driver in non-fatigue driving state is arranged to default eyes aperture.
  8. 8. method as claimed in claim 5, it is characterised in that the characteristic image is handled, including:
    The eyes image of the driver is handled, obtains the eyes aperture of the driver;
    If the eyes aperture of the driver is higher than aperture threshold value, the driver's eyes are set to open;Otherwise, institute is judged Driver's eyes are stated as closure;Within the period of setting, if the driver's eyes are not only once opened, but also once close Close, then set the driver's eyes as blink;
    Judge whether the frequency of wink of the driver's eyes is higher than frequency of wink threshold value.
  9. 9. such as claim 6-8 any one of them methods, it is characterised in that the eyes aperture of the driver is obtained, including:
    Binaryzation eye gray-scale map is formed according to the eyes image;
    Canthus point is determined according to the binaryzation eye gray-scale map;
    Eye state is determined according to the canthus point.
  10. 10. method as claimed in claim 9, it is characterised in that binaryzation eye gray-scale map is formed according to the eyes image, Including:
    According to the contrast of area of skin color, eyeball region and white of the eye region in the eyes image, calculated using three color shade watersheds Method, extracts informer's profile, forms the binaryzation eye gray-scale map.
CN201711237773.1A 2017-11-30 2017-11-30 Fatigue driving method for early warning Withdrawn CN108009495A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711237773.1A CN108009495A (en) 2017-11-30 2017-11-30 Fatigue driving method for early warning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711237773.1A CN108009495A (en) 2017-11-30 2017-11-30 Fatigue driving method for early warning

Publications (1)

Publication Number Publication Date
CN108009495A true CN108009495A (en) 2018-05-08

Family

ID=62055479

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711237773.1A Withdrawn CN108009495A (en) 2017-11-30 2017-11-30 Fatigue driving method for early warning

Country Status (1)

Country Link
CN (1) CN108009495A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109002817A (en) * 2018-08-31 2018-12-14 武汉理工大学 Adapter tube performance monitoring early warning system based on automatic driving vehicle driving fatigue temporal behavior
CN109409331A (en) * 2018-11-27 2019-03-01 惠州华阳通用电子有限公司 A kind of anti-fatigue-driving method based on radar
CN109741630A (en) * 2019-01-18 2019-05-10 爱驰汽车有限公司 Away from threshold value of warning adjusting method, system, equipment and storage medium when automobile
CN109886199A (en) * 2019-02-21 2019-06-14 百度在线网络技术(北京)有限公司 Information processing method, device, vehicle and mobile terminal

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101923645A (en) * 2009-06-09 2010-12-22 黑龙江大学 Iris splitting method suitable for low-quality iris image in complex application context
CN103345619A (en) * 2013-06-26 2013-10-09 上海永畅信息科技有限公司 Self-adaption correcting method of human eye natural contact in video chat
CN103390153A (en) * 2012-08-10 2013-11-13 眼验有限责任公司 Method and system of texture features for biometric authentication
CN104361332A (en) * 2014-12-08 2015-02-18 重庆市科学技术研究院 Human face eye region positioning method for fatigue driving detection
CN105205480A (en) * 2015-10-31 2015-12-30 潍坊学院 Complex scene human eye locating method and system
CN105488957A (en) * 2015-12-15 2016-04-13 小米科技有限责任公司 Fatigue driving detection method and apparatus
CN107368778A (en) * 2017-06-02 2017-11-21 深圳奥比中光科技有限公司 Method for catching, device and the storage device of human face expression

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101923645A (en) * 2009-06-09 2010-12-22 黑龙江大学 Iris splitting method suitable for low-quality iris image in complex application context
CN103390153A (en) * 2012-08-10 2013-11-13 眼验有限责任公司 Method and system of texture features for biometric authentication
CN103345619A (en) * 2013-06-26 2013-10-09 上海永畅信息科技有限公司 Self-adaption correcting method of human eye natural contact in video chat
CN104361332A (en) * 2014-12-08 2015-02-18 重庆市科学技术研究院 Human face eye region positioning method for fatigue driving detection
CN105205480A (en) * 2015-10-31 2015-12-30 潍坊学院 Complex scene human eye locating method and system
CN105488957A (en) * 2015-12-15 2016-04-13 小米科技有限责任公司 Fatigue driving detection method and apparatus
CN107368778A (en) * 2017-06-02 2017-11-21 深圳奥比中光科技有限公司 Method for catching, device and the storage device of human face expression

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
张海水: "基于红外差频和多特征的疲劳驾驶监测系统研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 *
赵文等: "基于红外图像的眼睛开闭检测方法", 《计算机工程与设计》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109002817A (en) * 2018-08-31 2018-12-14 武汉理工大学 Adapter tube performance monitoring early warning system based on automatic driving vehicle driving fatigue temporal behavior
CN109409331A (en) * 2018-11-27 2019-03-01 惠州华阳通用电子有限公司 A kind of anti-fatigue-driving method based on radar
CN109741630A (en) * 2019-01-18 2019-05-10 爱驰汽车有限公司 Away from threshold value of warning adjusting method, system, equipment and storage medium when automobile
CN109886199A (en) * 2019-02-21 2019-06-14 百度在线网络技术(北京)有限公司 Information processing method, device, vehicle and mobile terminal

Similar Documents

Publication Publication Date Title
CN108009495A (en) Fatigue driving method for early warning
Rongben et al. Monitoring mouth movement for driver fatigue or distraction with one camera
CN104809445B (en) method for detecting fatigue driving based on eye and mouth state
CN101593425B (en) Machine vision based fatigue driving monitoring method and system
US7253739B2 (en) System and method for determining eye closure state
CN104751600B (en) Anti-fatigue-driving safety means and its application method based on iris recognition
CN104778453B (en) A kind of night pedestrian detection method based on infrared pedestrian&#39;s brightness statistics feature
CN104361332B (en) A kind of face eye areas localization method for fatigue driving detection
US8154591B2 (en) Eyelid opening level determination device and computer readable medium storing computer program thereof
CN106781282A (en) A kind of intelligent travelling crane driver fatigue early warning system
CN106846734A (en) A kind of fatigue driving detection device and method
CN106965675B (en) A kind of lorry swarm intelligence safety work system
CN105389553A (en) Living body detection method and apparatus
WO2018218839A1 (en) Living body recognition method and system
CN107578008A (en) Fatigue state detection method based on blocking characteristic matrix algorithm and SVM
CN102306293A (en) Method for judging driver exam in actual road based on facial image identification technology
CN107895157B (en) Method for accurately positioning iris center of low-resolution image
CN110532976A (en) Method for detecting fatigue driving and system based on machine learning and multiple features fusion
CN101615241B (en) Method for screening certificate photos
CN104809482A (en) Fatigue detecting method based on individual learning
CN108647708A (en) Driver evaluation&#39;s method, apparatus, equipment and storage medium
JP4989249B2 (en) Eye detection device, dozing detection device, and method of eye detection device
CN111062292A (en) Fatigue driving detection device and method
CN1889093A (en) Recognition method for human eyes positioning and human eyes opening and closing
US20120189161A1 (en) Visual attention apparatus and control method based on mind awareness and display apparatus using the visual attention apparatus

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
WW01 Invention patent application withdrawn after publication

Application publication date: 20180508

WW01 Invention patent application withdrawn after publication