CN103839379B - Automobile and driver fatigue early warning detecting method and system for automobile - Google Patents
Automobile and driver fatigue early warning detecting method and system for automobile Download PDFInfo
- Publication number
- CN103839379B CN103839379B CN201410069318.5A CN201410069318A CN103839379B CN 103839379 B CN103839379 B CN 103839379B CN 201410069318 A CN201410069318 A CN 201410069318A CN 103839379 B CN103839379 B CN 103839379B
- Authority
- CN
- China
- Prior art keywords
- driver
- information
- eye
- face
- fatigue
- 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.)
- Active
Links
Abstract
The invention discloses an automobile and a driver fatigue early warning detecting method and system for the automobile. The driver fatigue early warning detecting system comprises a video collecting module, a facial recognition module, an eye detecting module and a judging module, wherein the video collecting module is used for collecting image information of passengers inside the automobile, the facial recognition module is used for processing the image information of the passengers inside the automobile to obtain facial characteristic parameters of a driver and obtaining facial position information of the driver according to the facial characteristic parameters of the driver, the eye detecting module is used for obtaining eye area information of the driver according to the facial position information of the driver and obtaining eye characteristic parameters of the driver according to the eye area information of the driver to detect eye characteristic information of the driver, and the judging module is used for judging whether the driver is in the fatigue state or not according to the eye characteristic information of the driver. According to the driver fatigue early warning detecting system, the driver inside the automobile can be automatically recognized, fatigue processes and movements of different drivers can be studied, the fatigued characteristics of the driver are memorized, and whether the driver is in the fatigue state or not is judged.
Description
Technical field
The present invention relates to automobile technical field, more particularly to a kind of driver fatigue early warning detecting system, with the system
Automobile and a kind of driver fatigue early-warning detection method.
Background technology
At present, the Related product for eye state being based on market is mostly according to perclos (Percentage of
EyeIid CIosure over the PupiI, over Time, tolerance fatigue/drowsiness physical quantity) standard, will eyelid close
The conjunction time accounts for the measurement index of the percentage ratio as physiological fatigue of a period of time.Using facial recognition techniques positioning eyes, nose
With corners of the mouth position, eyes, nose and corners of the mouth position are combined, can obtain driver further according to the tracking to eyeball and note
Force direction, and by judging whether the attention of driver disperses being reported to the police, such that it is able to realize preventing driver fatigue
The purpose of driving.
But, in correlation technique the face of a people is judged just in certain area, but generally vapour
In-car can not possibly only have one people of driver, so require to judge fatigue first to put forward condition very strict, therefore can not have
Effect ground reaches the effect of early warning, and there is also erroneous judgement, it is impossible to meet the needs of user.
The content of the invention
The purpose of the present invention is intended at least solve one of above-mentioned technological deficiency.
For this purpose, first purpose of the present invention is essence when proposing that one kind is capable of identify that in-car driver and detector drives
The driver fatigue early warning detecting system of refreshing state.
The second object of the present invention is to propose a kind of automobile with above-mentioned driver fatigue early warning detecting system.This
The 3rd bright purpose is to propose a kind of driver fatigue early-warning detection method.
To reach above-mentioned purpose, a kind of driver fatigue early warning detecting system that first aspect present invention embodiment is proposed,
Including：Video acquisition module, for gathering the image information of passenger；Facial recognition modules, the facial recognition modules are led to
Cross gray level co-occurrence matrixes, color space transformation and skin color modeling the image information of the passenger is processed with obtain drive
The face characteristic parameter of the person of sailing, and according to the face characteristic gain of parameter of the driver driver face location letter
Breath；Human eye detection module, for the ocular information of driver described in the face location acquisition of information according to the driver,
And according to the ocular acquisition of information of the driver driver eye characteristic parameter detecting the driver's
Eye feature information；Judge module, for being entered with default eye feature threshold value according to the eye feature information of the driver
Whether row compares to judge the driver in fatigue state.
Driver fatigue early warning detecting system according to embodiments of the present invention, by video acquisition module passenger is gathered
Image information, and the image information analysis of passenger are processed by facial recognition modules identify the face of driver
Characteristic parameter, then obtains the face location information of driver, human eye detection module according to the face location information of driver most
The eye characteristic parameter of driver is obtained eventually to detect the eye feature information of driver, and last judge module is according to driver's
Whether eye feature information judges driver in fatigue state.Therefore, the driver fatigue early warning detection of the embodiment of the present invention
System can automatic identification in-car driver, and fatigue process and the action of different drivers can be learnt, and memory is driven
Whether the fatigue characteristic of the person of sailing, judge driver in fatigue state, so as to realize that not being absorbed in the mental status to driver enters
Row early warning, it is ensured that driving safety, meets the needs of user, and reduces False Rate.
According to one embodiment of present invention, the facial recognition modules according to the gray level co-occurrence matrixes to the in-car
The image information of occupant is processed to obtain the face texture characteristic information of the passenger, and according to the passenger
Face texture characteristic information identify the driver.
According to one embodiment of present invention, the human eye detection module is entered according to the face location information of the driver
Column hisgram equilibrium treatment and multidomain treat-ment to lock the ocular of the driver, and to the ocular of the driver
Carry out Connected area disposal$ to obtain the eye position of the driver, and according to the eye position of the driver is extracted
The eye characteristic parameter of driver.
According to one embodiment of present invention, the judge module is sentenced according to perclos standards and BP neural network algorithm
Whether the driver of breaking is in the fatigue state.
According to one embodiment of present invention, described driver fatigue early warning detecting system also includes：Warning module,
The warning module sends alarm when the driver is in the fatigue state.
A kind of automobile that second aspect present invention embodiment is proposed includes above-mentioned driver fatigue early warning detecting system.
Automobile according to embodiments of the present invention, can automatic identification in-car driver, and different drivers can be learnt
Fatigue process and action, and the fatigue characteristic of memory driver, driver is judged whether in fatigue state, so as to reality
Now not being absorbed in the mental status to driver carries out early warning, it is ensured that driving safety, meets the needs of user.
To reach above-mentioned purpose, a kind of driver fatigue early warning for automobile that third aspect present invention embodiment is proposed
Detection method, comprises the following steps：S1, gathers the image information of passenger；S2, by gray level co-occurrence matrixes, color space
Conversion and skin color modeling are processed the image information of the passenger to obtain the face characteristic parameter of driver, and root
According to the driver face characteristic gain of parameter described in driver face location information；S3, according to the people of the driver
Face positional information obtains the ocular information of the driver, and according to the ocular acquisition of information of the driver
The eye characteristic parameter of driver is detecting the eye feature information of the driver；S4, it is special according to the eyes of the driver
Whether reference breath is compared to judge the driver in fatigue state with default eye feature threshold value.
Driver fatigue early-warning detection method for automobile according to embodiments of the present invention, by gathering passenger's
Image information, and the image information analysis of passenger are processed to identify the face characteristic parameter of driver, then obtain
The face location information of driver, the eye characteristic parameter of driver is finally obtained to examine according to the face location information of driver
The eye feature information of driver is surveyed, judges driver whether in tired shape finally according to the eye feature information of driver
State.Therefore, the driver fatigue early-warning detection method for automobile of the embodiment of the present invention can automatic identification in-car driver,
And fatigue process and the action of different drivers, and the fatigue characteristic of memory driver can be learnt, driver is judged
Whether fatigue state is in, so as to realize that not being absorbed in the mental status to driver carries out early warning, it is ensured that driving safety, and can be subtracted
Few False Rate.
According to one embodiment of present invention, in step s 2, according to the gray level co-occurrence matrixes to the passenger
Image information processed to obtain the face texture characteristic information of the passenger, and according to the face of the passenger
Portion's texture feature information identifies the driver.
According to one embodiment of present invention, in step s3, carried out directly according to the face location information of the driver
Square figure equilibrium treatment and multidomain treat-ment are to lock the ocular of the driver, and the ocular to the driver is carried out
Connected area disposal$ is obtaining the eye position of the driver, and extracts the driving according to the eye position of the driver
The eye characteristic parameter of member.
According to one embodiment of present invention, in step s 4, judged according to perclos standards and BP neural network algorithm
Whether the driver is in the fatigue state.
The additional aspect of the present invention and advantage will be set forth in part in the description, and partly will become from the following description
Obtain substantially, or recognized by the practice of the present invention.
Description of the drawings
The above-mentioned and/or additional aspect of the present invention and advantage will become from the following description of the accompanying drawings of embodiments
It is substantially and easy to understand, wherein：
Fig. 1 is the block diagram of the driver fatigue early warning detecting system according to the embodiment of the present invention；
Fig. 2A is the schematic diagram of textural characteristics face complexion area segmentation；
Fig. 2 B are the schematic diagram by textural characteristics Face detection result；
Fig. 3 is the flow chart that people face skin skin regions and the division in other regions are tentatively realized according to textural characteristics；
Fig. 4 A are the schematic diagram of face hair zones segmentation result；
Fig. 4 B are the schematic diagram that result is accurately positioned using color space skin color modeling method face；
Fig. 5 is to distinguish area of skin color and hair zones according to the foundation colour of skin space of one embodiment of the invention and accurately determine
The flow chart of position face；
Fig. 6 is " three front yards " standard schematic diagram of face；
Fig. 7 is image denoising and face subregion (1/2 and 1/3) locking flow figure；
Fig. 8 is the algorithm flow chart of ocular locking；
Fig. 9 A are the schematic diagram of night facial image situation；
Fig. 9 B are the schematic diagram of facial image situation on daytime；
Figure 10 is the not bending moment calculation flow charts of HUShi seven；
Figure 11 is BP neural network basic structure schematic diagram；
Figure 12 is the integration flow chart of integrated part；
Figure 13 is the top level diagram of the driver fatigue early warning detecting system internal module according to one embodiment of the invention；And
Figure 14 is the flow chart of the driver fatigue early-warning detection method for automobile according to the embodiment of the present invention.
Specific embodiment
Embodiments of the invention are described below in detail, the example of the embodiment is shown in the drawings, wherein from start to finish
Same or similar label represents same or similar element or the element with same or like function.Below with reference to attached
The embodiment of figure description is exemplary, is only used for explaining the present invention, and is not construed as limiting the claims.
Following disclosure provides many different embodiments or example is used for realizing the different structure of the present invention.For letter
Change disclosure of the invention, hereinafter the part and setting of specific examples are described.Certainly, they are only merely illustrative, and
Purpose does not lie in the restriction present invention.Additionally, the present invention can in different examples repeat reference numerals and/or letter.It is this heavy
It is again the relation between itself not indicating discussed various embodiments and/or arranging for purposes of simplicity and clarity.This
Outward, the invention provides various specific technique and material example, but those of ordinary skill in the art can be appreciated that
The applicable property of other techniques and/or the use of other materials.In addition, fisrt feature described below second feature it
" on " structure can include that the first and second features be formed as the embodiment of directly contact, it is also possible to including other feature shape
Into the embodiment between the first and second features, such first and second feature may not be directly contact.
In describing the invention, it should be noted that unless otherwise prescribed and limit, term " installation ", " connected ",
" connection " should be interpreted broadly, for example, it may be mechanically connected or electrical connection, or the connection of two element internals, can
Being to be joined directly together, it is also possible to be indirectly connected to by intermediary, for the ordinary skill in the art, can basis
Concrete condition understands the concrete meaning of above-mentioned term.
With reference to the accompanying drawings to describe the driver fatigue early warning detecting system of proposition according to embodiments of the present invention, with this
The automobile of system and the driver fatigue early-warning detection method for automobile.
Fig. 1 is the block diagram of the driver fatigue early warning detecting system according to the embodiment of the present invention.As shown in figure 1,
The driver fatigue early warning detecting system includes video acquisition module 10, facial recognition modules 20, human eye detection module 30 and sentences
Disconnected module 40.
Wherein, video acquisition module 10 is used to gather the image information of passenger.In one embodiment of the invention,
Video acquisition module 10 can be shooting importation.
Facial recognition modules 20 are by gray level co-occurrence matrixes, color space transformation and skin color modeling to the passenger's
Image information is processed to obtain the face characteristic parameter of driver, and according to the face characteristic gain of parameter of the driver
The face location information of the driver.
Wherein, according to one embodiment of present invention, facial recognition modules 20 according to the gray level co-occurrence matrixes to described
The image information of passenger is processed to obtain the face texture characteristic information of the passenger, and according to the in-car
The face texture characteristic information of occupant identifies the driver.
Specifically, in an embodiment of the present invention, during drive r's face recognition, first by textural characteristics
Difference, can preferably distinguish the face under night.For example with gray level co-occurrence matrixes, and extract four features change of correlation
Amount, according to the value of characteristic variable, so as to realize the division in people face skin skin regions and other regions, so as to tentatively judge face
Position, and night effect becomes apparent from.Wherein, the schematic diagram that textural characteristics face complexion area is split is as shown in Figure 2 A.
Wherein, it is to the mathematical definition of gray level co-occurrence matrixes：Gray level co-occurrence matrixes are the pixels from image intensity value for i
(x, y) sets out, and statistics is d with its distance, and gray value is the pixel (x+a, y+b) of j, while the frequency P (i, j, d, θ) for occurring,
It is mathematically represented as：
P (i, j, d, θ)=[(x, y), (x+a, y+b) | f (x, y)=i；F (x+a, y+b)=j] } (1)
In formula (1), gray level co-occurrence matrixes are described on θ directions, and a pair of pixels for being separated by d pixel distances have respectively
There is the probability of activity i and j.
According to gray level co-occurrence matrixes, 14 kinds of characteristic quantities can be obtained, wherein energy, entropy, contrast, local stationary these four
Feature can preferably distinguish people face skin skin regions.
Wherein, energy is mathematically represented as：
Wherein, L is gray level.
Entropy is mathematically represented as：
Contrast is mathematically represented as：
Wherein, k=(i-j)2。
Local stationary is mathematically represented as：
In an example of the present invention, by textural characteristics Face detection result as shown in Figure 2 B.
In one embodiment of the invention, as shown in figure 3, tentatively realizing people face skin skin regions according to textural characteristics
It is with the flow process of the division in other regions：
S301, initialization.
S302, image Compression.
S303, is loaded into image parameter after compression is processed.
S304, generates gray level co-occurrence matrixes.
S305, calculates texture eigenvalue (such as energy, entropy, four kinds of features of contrast and local stationary).
S306, judges whether face skin area.If it is, execution step S307；If not, execution step S308.Its
In, energy, entropy, four kinds of eigenvalues of contrast and local stationary have a threshold value for face complexion, and when more than the threshold value,
Then it is judged to face.
S307, grey scale pixel value is set to 255, into step S309.
S308, grey scale pixel value is set to 0, into step S309.
S309, generates new figure, shows segmentation result.
S310, terminates.
Then, using color space transformation, complexion model is set up, and colour of skin Similarity Measure face location coordinates hair position
Judge so as to accurately judge the position of face.
Wherein, the conversion of color space, is exactly mapped to YCbCr color spaces by RGB color, because human skin
The Clustering features of the colour of skin can be preferably embodied on YCbCr color spaces.Conversion formula is：
After complexion model YCbCr space normalization chroma histogram, the colour of skin meets dimensional Gaussian model M=(m, C), wherein
M is average, and C is covariance.Face and non-face can preferably be distinguished by this complexion model.
According to complexion model, it may be determined that class face complexion area.It is similar that class is determined according to the Y-component of YCbCr color spaces
Hair zones.It is distributed according to the position relationship of area of skin color and hair zones, the position of comprehensive descision face.Wherein, face head
Send out region segmentation result as shown in Figure 4 A, result is accurately positioned as shown in Figure 4 B using color space skin color modeling method face.
In one embodiment of the invention, as shown in figure 5, according to colour of skin space distinguish area of skin color and hair zones and
The flow process for being accurately positioned face is：
S501, initialization.
S502, obtains image parameter.
S503, YCbCr color spaces are mapped to by image from rgb color space.
S504, sets up complexion model.
S505, reads current pixel value.
S506, according to current pixel color component relation skin or hair zones are judged whether.If it is, execution step
S507；If not, execution step S508.
S507, judges whether to belong to skin area according to current pixel color component relation.If it is, execution step
S509；If not, execution step S510.
S508, by pixel value 2 are set to, into step S511.
S509, by pixel value 0 is set to, into step S511.
S510, by pixel value 1 is set to, into step S511.
S511, image result after being split.
S512, according to hair face location relation locating human face position.
S513, terminates.
Therefore say, in an embodiment of the present invention, using gray level co-occurrence matrixes and colored colour of skin space orientation face process
Extractable driver's face characteristic value, so as to recognize driver's face, if driver may be selected for fixed human user it is antitheft
Pattern, plays automobile burglar effect.That is, the driver fatigue early warning detecting system of the embodiment of the present invention is using intelligent face
Portion's technology of identification, can be with automatic identification car owner, such that it is able to prevent automobile stolen.
In an embodiment of the present invention, human eye detection module 30 is used for the face location acquisition of information according to the driver
The ocular information of the driver, and the eye of driver according to the ocular acquisition of information of the driver is special
Levy parameter to detect the eye feature information of the driver.
Wherein, according to one embodiment of present invention, human eye detection module 30 is believed according to the face location of the driver
Breath carries out histogram equalization process and multidomain treat-ment to lock the ocular of the driver, and to the eye of the driver
Region carries out Connected area disposal$ to obtain the eye position of the driver, and is extracted according to the eye position of the driver
The eye characteristic parameter of the driver.
Specifically, the main purpose that characteristic area is extracted is to extract eye feature region, and characteristic area is carried out point
Analyse using obtain can as differentiate driver's mental status feature octuple data.Wherein, whole characteristic area is extracted and is divided into three
Individual part：Further locking, the extraction of eye areas, the part of feature extraction three.The extraction of characteristic area is mainly with people's face
Portion's region division and facial face position relation, magnitude relationship are main modeling foundation, while be aided with UNICOM domain algorithm carry out,
UNICOM domain algorithm is protection algorithm.
The further locking of characteristic area first.For one the more face of standard, face have respective solid
Positioning is put, and for eyes, its position has been fixed on centre three/part of face.If obtaining more accurate
Human face region after, program can be scanned in a less scope.Specifically way is：First according to the figure for collecting
Picture, carries out regional compartmentalization, as shown in fig. 6, the part that can typically choose centre 1/3 is extracted and extracted as a further feature
Region.
For the image of input, histogram equalization is carried out first, increased picture contrast, be conducive to follow-up
Binaryzation.Subsequently algorithm carries out binaryzation to image or so piecemeal.According to the general location of human eye carry out in height draw
Point, realize the further locking of characteristic area, i.e., human eye is looked for by face subregion.Specifically, as shown in fig. 7, image denoising
And face subregion (1/2 and 1/3) locking flow is：
S701, our department's branch starts.
S702, histogram equalization process.
S703, to whole face figure left and right multidomain treat-ment is carried out.
Binary conversion treatment is done respectively in S704, the area of left and right two.
S705, merges and two-value expansion process.
S706, analytic function connect1 (or 2), such as into step S708.
Terminate S707, this part.
The algorithm of S708, analytic function connect2 in this part.
S709, judges whether face region is near square.If it is, execution step S710；If not, execution step S711.
S710, takes in the middle of face 1/2 part, into step S712.
S711, takes in the middle of face 1/3 part, into step S712.
S712, counts often row Gray Projection value.
S713, the from bottom to top every row Gray Projection value of sequential scanning.
S714, judges whether that Gray Projection value is reached minimum.If it is, execution step S715；If not, execution step
S721。
S715, judges whether minimum Gray Projection value is border.If it is, execution step S716；If not, execution step
S717。
S716, end mark its highest behavior lowermost row+5.
S717, judges whether to reach doubtful peak.If it is, execution step S718；If not, execution step S719.
S718, judges whether to be unsatisfactory for persistence.If it is, return to step S713；If not, execution step S720.
S719, return to step S717 after the line number that adds up.
S720, labelling highest line.
Whether S721, judge Gray Projection value more than 5.If it is, execution step S722；If not, execution step S723.
S722, labelling projection behavior lower limit, execution step S723.
S723, continues to scan on rear return to step S714.
Then, eye areas are extracted.In an embodiment of the present invention, precondition is had been obtained for through further lock
Human eye after fixed extracts region.Ensuing work is exactly that standby when being verified and being extracted unsuccessfully is processed.Whatsoever hold
The people of looks, as long as human eye spacing (is such as seated on driver's seat) in the case of fixation always in a fixed scope, two
Difference in height compared to being a less side for eyebrow to the difference in height of two.Especially it should be pointed out that due to above-mentioned
In further reducing the operation of hunting zone, the operation for scanning screening be from bottom to top, therefore can be from largely keeping away
Exempt from the interference from eyebrow.For the above-mentioned reasons, these parameters are suitable to judge, extract human eye.And algorithm is in compiling procedure
The impact of situations such as take into account the difference on evening, daytime, spectacle-frame, eyebrow, torticollis, wherein：Due to mainly with facial anti-
Based on light, scope of the evening compared to Extract eyes on daytime out is confined near pupil, therefore size can be compared and wanted daytime
It is less；Mainly be embodied in spectacle-frame " can produce " connected domain between two for the impact of spectacle-frame, so can compare
It is easier to cause identification chaotic.These are all embodied in one by one in the specific parameter of algorithm, mechanism.Wherein, as shown in figure 8, eye
The algorithm flow of area locking is：
S801, algorithm starts.
S802, extracts according to previous stage result (positional information) to the two-value data being input into.
S803, detects whether also new connected domain.If it is, execution step S804；If not, execution step S807.
S804, the current connected domain all the elements of labelling.
S805, judges whether to meet glasses condition.If it is, execution step S806；If not, execution step S807.
S806, deletion record, return to step S803.
S807, typing information collects array.
S808, judges whether to travel through all information elements.If it is, execution step S811；If not, execution step
S809。
S809, judges whether eyes condition.If it is, execution step S810；If not, return to step S808.
S810, the output first order is found successfully.
S811, finds failure.
S812, expands hunting zone.
S813, is extracted in the two-value data of input.
S814, detects whether also new connected domain.If it is, execution step S815；If not, execution step S818.
S815, the current connected domain all the elements of labelling.
S816, judges whether to meet glasses condition.If it is, execution step S817；If not, execution step S818.
S817, deletion record, return to step S814.
S818, typing information collects array.
S819, judges whether to travel through all information elements.If it is, execution step S822；If not, execution step
S820。
S820, judges whether eyes condition.If it is, execution step S821；If not, return to step S819.
Find successfully S821, the output second level.
S822, finds utter failure.
Wherein, connected domain analysis I is that eye areas are further ensured that, with the step for Gray Projection sequentially
Perform, connected domain analysis here pair as if the eye areas that determined based on Gray Projection carry out main binaryzation, position
Whether the analysis of confidence breath, the information being mainly concerned with is unified for the height of eyes, and whether spacing is in given area.
If it is determined that correctly will be marked at two eye positions immediately, and intercept picture and carry out feature analysiss.If
It is out of one's reckoning, immediately enters connected domain analysis II.
As safety measures, the mechanism of connected domain analysis II be 1/3 human face region obtained in recognition of face basis it
On, binaryzation is carried out, find then and meet " whether the height of eyes is unified, and spacing is whether in given area " this condition
's.If still can not find so algorithm thinks that face finds failure, find carries out subsequent treatment by connected domain analysis I.
Then, feature extraction is carried out.In terms of feature extraction, it is considered to which the different time in evening on daytime uses two kinds of spies
Levy, i.e. the size of HUShi not bending moment and pupil, wherein：
1st, HUShi not bending moments
Moment function has a wide range of applications in graphical analyses, and such as pattern recognition, target classification, target recognition and orientation is estimated
Meter, picture coding and reconstruct etc..One square collection calculated from a width digital picture, generally describes the picture shape
Global characteristics, and there is provided substantial amounts of with regard to the different types of geometrical property information of the image, such as size, position, direction and
Shape etc..This characteristic descriptive power of image moment is widely used in various image procossings, computer vision and robot
During the target recognition of technical field is estimated with orientation.For example, first moment is relevant with shape, and second moment shows that curve is flat around straight line
The degree of expansion of average, third moment is then the symmetric measurement with regard to meansigma methodss.One can be derived by second moment and third moment
Group totally seven not bending moments.Process can be carried out simplifying to image, reservation can most reflect the information of target property, then with the figure after simplification
As calculating invariant moment features, amount of calculation can be reduced.
For square, it is frequently utilized for differentiating the larger object of morphological differencess, so opening for eyes is closed in characteristic area
May be considered under this resolution and there is huge change, therefore algorithm employs this stack features.The following is HUShi seven
The computing formula of individual not bending moment：
Wherein, μpqFor p+q rank central moments,
2. the size of pupil
According to perclos standards, fatigue state can be determined by calculating the ratio of eyes closed in a period of time.Knot
Close the restriction (figure) of actual imaging means, then for pupil image quality preferable night use this feature and with this
Standard is criterion.
Wherein, the measuring method of perclos is：The face image of driver is captured with video camera, by image processing method
Method obtains eye image, determines that eyes are opened or closed through the means of graphical analyses and identification.Define eye pupil
Aperture is to open more than 20%, and pupil aperture is equal to 20% or less for closure.Eyes make and break ratio is eyes in time of measuring
The ratio of the time that the time of closure opens with eyes, the time is that the Time Calculation processed by every two field picture is gone out.
Herein, due to not there is a problem of calculating area, with the mental status classification value of neutral net output as from change
Amount, statistics within a certain period of time, is opened, shared ratio of closing one's eyes, and such as ratio provides feedback more than threshold value, then system.
The parameter specifically chosen when using perclos criterions be the width of pupil region as the 8th characteristic quantity, because
It is not Scale invariant for this parameter, it is therefore desirable to gather the width of the pupil region of a few minutes in advance in practical operation
In the hope of maximum so as to realizing follow-up real-time monitoring.
In a specific example of the present invention, night facial image situation is distinguished by the contrast of pupil image definition
With facial image situation on daytime.Wherein, night facial image situation as shown in Figure 9 A, facial image situation on daytime such as Fig. 9 B institutes
Show.
3rd, algorithm is realized
This part algorithm is mainly in principal function, in the case of being have in the result that principal function is found according to human eye, enters
Row is processed.The locking of human eye area is sequentially carried out, binaryzation determines pupil width, finally by the image input HU after binaryzation
The constant moment function of family name carries out process and obtains remaining 7 degree of freedom scale invariant feature.Wherein, as shown in Figure 10, the not bending moment meters of HUShi seven
Calculating flow process is：
S1001, algorithm starts.
S1002, extracts ocular from colour original.
S1003, reduces region of search to original 1/4.
S1004, binary conversion treatment.
S1005, travels through all rows in the range of from h/5 to 4h/5.
S1006, tries to achieve Breadth Maximum.
S1007, by binary image HUShi moment functions are input into.
S1008, obtains result.
In an embodiment of the present invention, judge module 40 be used for according to the eye feature information of the driver with it is default
Whether eye feature threshold value is compared to judge the driver in fatigue state.
Wherein, according to one embodiment of present invention, judge module 40 is according to perclos standards and BP (Back
Propagation, back propagation) whether neural network algorithm judge the driver in the fatigue state.
Last judge module 40 carries out Classification and Identification according to perclos and BP neural network.According in the result that BP judges
Correct result carries out examining for octuple data, i.e., entered using the pupil width maximum that pupil width and first few minutes are measured
Row judges, if width is less than the 0.7 of standard maximum value, continues the situation for being judged to close one's eyes.Wherein, BP neural network is tied substantially
Structure is as shown in figure 11.
According to one embodiment of present invention, as shown in figure 1, above-mentioned driver fatigue early warning detecting system also includes：
Warning module 50, when the driver is in the fatigue state, warning module 50 sends alarm.Meanwhile, warning module
50 can with the engine controller ECU both-way communications of automobile, in emergency situations pro-active intervention ECU (for example brake, subtract
Speed, bright light etc.).
Wherein, the integrated part in the driver fatigue early warning detecting system of the embodiment of the present invention can be divided into video acquisition and
Two parts of feedback of reporting to the police, although they do not play a key effect in a program, as realizing being threaded a needle in whole algorithm
The part of lead, needs to illustrate.
As shown in figure 12, the integration flow process of integrated part includes：
S1201, algorithm starts.
S1202, photographic head is affiliated on interface.
S1203, arranges the parameter of shooting image.
S1204, judges whether photographic head is online.If it is, execution step S1205；If not, return to step S1204, after
It is continuous to judge.
S1205, arranges video recording parameter.
S1206, carries out call back function.
S1207, disassembles data.
S1208, YUV2 image format conversion is into RGB image form.
S1209, stores picture.
S1210, shows in real time feedback interface.
Wherein, algorithm employs the mode of mark bit array in execution, is carried out according to all of flag bit of 60 frame ins comprehensive
Evaluation is closed, the judgement of state is carried out according in mark bit array 0/1/2 number：' 0 ' account for more than 40 frames be exactly nobody onboard
State；' 1 ' account for more than 30 frames be exactly current driver's state be abnormal condition；It is exactly currently to drive that ' 2 ' account for more than 30 frames
The state of the person of sailing is normal condition.The state of driver is fed back based on result above.Have to what is illustrated in addition
It is that photographic head needs initialization, therefore whole program is not processed image during photographic head is initialized.
After the driver fatigue early warning detecting system study of the embodiment of the present invention terminates and is stable, user may be selected to pass through
CAN communication accesses ECU special interfaces, in case of emergency pro-active intervention ECU (brake, deceleration, double sudden strains of a muscle etc.).
Therefore, the driver fatigue early warning detecting system of the embodiment of the present invention is interactive software system, relies on DSP to put down
Platform, with detecting driver's face state, the function of mental status during detecting its driving.Wherein, as shown in figure 1, of the invention
The driver fatigue early warning detecting system of embodiment can be by video acquisition, recognition of face, human eye detection, the part of condition adjudgement four
Composition.Also, the top level diagram of the driver fatigue early warning detecting system internal module is as shown in figure 13.
In sum, the driver fatigue early warning detecting system of the embodiment of the present invention employs autonomous in-car light conditions
Monitoring amendment, can carry out the automatic amendment of in-car human face light；And intelligent facial recognition techniques of sampling, can be with automatic identification car
It is main, such that it is able to anti-theft feature；And implant new human eye detection mechanism, the fatigue process of the different drivers of study and
Action simultaneously remembers its fatigue characteristic with neutral net, is allowed to more adaptability, intelligent；The communication with ECU is realized simultaneously,
Client may be selected and ECU both-way communications, in case of emergency pro-active intervention ECU (brake, deceleration, double sudden strains of a muscle etc.)；Additionally, can pin
Different automobile types are carried out with demarcation, and to be allowed to accuracy rate higher, is highly suitable for main engine plants and coordinates to demarcate to concentrate assembling.Therefore, the present invention
The driver fatigue early warning detecting system of embodiment makes giving fatigue pre-warning more intelligent, improves adaptability, and is capable of identify that car
Owner identification is antitheft to carry out, and may participate in car networking, makes automobile more intelligent.
Driver fatigue early warning detecting system according to embodiments of the present invention, by video acquisition module passenger is gathered
Image information, and the image information analysis of passenger are processed by facial recognition modules identify the face of driver
Characteristic parameter, then obtains the face location information of driver, human eye detection module according to the face location information of driver most
The eye characteristic parameter of driver is obtained eventually to detect the eye feature information of driver, and last judge module is according to driver's
Whether eye feature information judges driver in fatigue state.Therefore, the driver fatigue early warning detection of the embodiment of the present invention
System can automatic identification in-car driver, and fatigue process and the action of different drivers can be learnt, and memory is driven
Whether the fatigue characteristic of the person of sailing, judge driver in fatigue state, so as to realize that not being absorbed in the mental status to driver enters
Row early warning, it is ensured that driving safety, meets the needs of user, and reduces False Rate.
Additionally, embodiments of the invention also proposed a kind of automobile, it includes above-mentioned driver fatigue early warning detection system
System.
Automobile according to embodiments of the present invention, can automatic identification in-car driver, and different drivers can be learnt
Fatigue process and action, and the fatigue characteristic of memory driver, driver is judged whether in fatigue state, so as to reality
Now not being absorbed in the mental status to driver carries out early warning, it is ensured that driving safety, meets the needs of user.
Figure 14 is the flow chart of the driver fatigue early-warning detection method for automobile according to the embodiment of the present invention.As schemed
Shown in 14, the driver fatigue early-warning detection method for being used for automobile is comprised the following steps：
S1, gathers the image information of passenger.
S2, at the image information of gray level co-occurrence matrixes, color space transformation and skin color modeling to passenger
Manage to obtain the face characteristic parameter of driver, and the face location letter of the face characteristic gain of parameter driver according to driver
Breath.
According to one embodiment of present invention, in step s 2, the image of passenger is believed according to gray level co-occurrence matrixes
Breath is processed to obtain the face texture characteristic information of the passenger, and is believed according to the face texture feature of passenger
Breath identifies driver.
Wherein, specific facial recognition, Face detection are as described above, be just no longer described in detail here.
S3, according to the ocular information of the face location acquisition of information driver of driver, and according to the eye of driver
Portion area information obtains the eye characteristic parameter of driver to detect the eye feature information of driver.
According to one embodiment of present invention, in step s3, column hisgram is entered according to the face location information of driver
Equilibrium treatment and multidomain treat-ment to lock the ocular of driver, and Connected area disposal$ is carried out to the ocular of driver with
The eye position of driver is obtained, and the eye characteristic parameter of driver is extracted according to the eye position of driver.
Wherein, specific human eye detection process is as described above, be just no longer described in detail here.
S4, is compared to judge that driver is according to the eye feature information of driver with default eye feature threshold value
It is no in fatigue state.
According to one embodiment of present invention, in step s 4, judged according to perclos standards and BP neural network algorithm
Whether driver is in fatigue state.
Similarly, according to perclos and BP neural network the process of Classification and Identification is carried out as described above, no longer going to live in the household of one's in-laws on getting married here
State.
Driver fatigue early-warning detection method for automobile according to embodiments of the present invention, by gathering passenger's
Image information, and the image information analysis of passenger are processed to identify the face characteristic parameter of driver, then obtain
The face location information of driver, the eye characteristic parameter of driver is finally obtained to examine according to the face location information of driver
The eye feature information of driver is surveyed, judges driver whether in tired shape finally according to the eye feature information of driver
State.Therefore, the driver fatigue early-warning detection method for automobile of the embodiment of the present invention can automatic identification in-car driver,
And fatigue process and the action of different drivers, and the fatigue characteristic of memory driver can be learnt, driver is judged
Whether fatigue state is in, so as to realize that not being absorbed in the mental status to driver carries out early warning, it is ensured that driving safety, and can be subtracted
Few False Rate.
In flow chart or here any process described otherwise above or method description are construed as, expression includes
It is one or more for realizing specific logical function or process the step of the module of code of executable instruction, fragment or portion
Point, and the scope of the preferred embodiment of the present invention includes other realization, wherein can not press shown or discussion suitable
Sequence, including according to involved function by it is basic simultaneously in the way of or in the opposite order, carry out perform function, this should be of the invention
Embodiment person of ordinary skill in the field understood.
In flow charts expression or here logic described otherwise above and/or step, for example, are considered use
In the order list of the executable instruction for realizing logic function, in may be embodied in any computer-readable medium, for
Instruction execution system, device or equipment (as computer based system, the system including processor or other can hold from instruction
The system of row system, device or equipment instruction fetch and execute instruction) use, or with reference to these instruction execution systems, device or set
It is standby and use.For the purpose of this specification, " computer-readable medium " can any can be included, store, communicate, propagate or pass
The dress that defeated program is used for instruction execution system, device or equipment or with reference to these instruction execution systems, device or equipment
Put.The more specifically example (non-exhaustive list) of computer-readable medium includes following：With the electricity that one or more are connected up
Connecting portion (electronic installation), portable computer diskette box (magnetic device), random access memory (RAM), read only memory
(ROM), erasable edit read-only storage (EPROM or flash memory), fiber device, and portable optic disk is read-only deposits
Reservoir (CDROM).In addition, computer-readable medium can even is that the paper that can thereon print described program or other are suitable
Medium, because for example by carrying out optical scanning to paper or other media edlin, interpretation can then be entered or if necessary with it
His suitable method is processed to electronically obtain described program, in being then stored in computer storage.
It should be appreciated that each several part of the present invention can be realized with hardware, software, firmware or combinations thereof.Above-mentioned
In embodiment, the software that multiple steps or method can in memory and by suitable instruction execution system be performed with storage
Or firmware is realizing.For example, if realized with hardware, and in another embodiment, can be with well known in the art
Any one of row technology or their combination are realizing：With for realizing the logic gates of logic function to data signal
Discrete logic, the special IC with suitable combinational logic gate circuit, programmable gate array (PGA), scene
Programmable gate array (FPGA) etc..
Those skilled in the art are appreciated that to realize all or part of step that above-described embodiment method is carried
Suddenly the hardware that can be by program to instruct correlation is completed, and described program can be stored in a kind of computer-readable storage medium
In matter, the program upon execution, including one or a combination set of the step of embodiment of the method.
Additionally, each functional unit in each embodiment of the invention can be integrated in a processing module, it is also possible to
It is that unit is individually physically present, it is also possible to which two or more units are integrated in a module.Above-mentioned integrated mould
Block both can be realized in the form of hardware, it would however also be possible to employ the form of software function module is realized.The integrated module is such as
Fruit is realized and as independent production marketing or when using using in the form of software function module, it is also possible to be stored in a computer
In read/write memory medium.
Storage medium mentioned above can be read only memory, disk or CD etc..
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means to combine specific features, structure, material or spy that the embodiment or example are described
Point is contained at least one embodiment of the present invention or example.In this manual, to the schematic representation of above-mentioned term not
Necessarily refer to identical embodiment or example.And, the specific features of description, structure, material or feature can be any
One or more embodiments or example in combine in an appropriate manner.
Although an embodiment of the present invention has been shown and described, for the ordinary skill in the art, can be with
Understanding can carry out various changes, modification, replacement to these embodiments without departing from the principles and spirit of the present invention
And modification, the scope of the present invention is by claims and its is equal to limit.
Claims (8)
1. a kind of driver fatigue early warning detecting system, it is characterised in that include：
Video acquisition module, for gathering the image information of passenger；
Facial recognition modules, the facial recognition modules are by gray level co-occurrence matrixes, color space transformation and skin color modeling to institute
The image information for stating passenger is processed to obtain the face characteristic parameter of driver, and according to the face of the driver
Characteristic parameter obtains the face location information of the driver；
Human eye detection module, the ocular for driver described in the face location acquisition of information according to the driver is believed
Breath, and according to the ocular acquisition of information of the driver driver eye characteristic parameter detecting the driver
Eye feature information, wherein, it is equal that the human eye detection module enters column hisgram according to the face location information of the driver
Process and multidomain treat-ment weigh to lock the ocular of the driver, and the ocular to the driver carries out connected domain
Process to obtain the eye position of the driver, and the eye of the driver is extracted according to the eye position of the driver
Portion's characteristic parameter；
Judge module, for being compared to sentence with default eye feature threshold value according to the eye feature information of the driver
Whether the driver of breaking is in fatigue state.
2. driver fatigue early warning detecting system as claimed in claim 1, it is characterised in that the facial recognition modules according to
The gray level co-occurrence matrixes are processed the image information of the passenger to obtain the face texture of the passenger
Characteristic information, and the driver is identified according to the face texture characteristic information of the passenger.
3. driver fatigue early warning detecting system as claimed in claim 1, it is characterised in that the judge module according to
Whether perclos standards and BP neural network algorithm judge the driver in the fatigue state.
4. the driver fatigue early warning detecting system as described in any one of claim 1-3, it is characterised in that also include：
Warning module, when the driver is in the fatigue state, the warning module sends alarm.
5. a kind of automobile, it is characterised in that include the driver fatigue early warning detection system as described in any one of claim 1-4
System.
6. a kind of driver fatigue early-warning detection method for automobile, it is characterised in that comprise the following steps：
S1, gathers the image information of passenger；
S2, at the image information of gray level co-occurrence matrixes, color space transformation and skin color modeling to the passenger
Manage to obtain the face characteristic parameter of driver, and according to the face characteristic gain of parameter of the driver driver people
Face positional information；
S3, the ocular information of driver according to the face location acquisition of information of the driver, and driven according to described
The eye characteristic parameter of driver described in the ocular acquisition of information of the person of sailing to detect the eye feature information of the driver,
Wherein, carry out histogram equalization according to the face location information of the driver to process with multidomain treat-ment to lock the driver
Ocular, and the ocular to the driver carries out Connected area disposal$ to obtain the eye position of the driver,
And the eye characteristic parameter of the driver is extracted according to the eye position of the driver；
S4, is compared to judge the driving according to the eye feature information of the driver with default eye feature threshold value
Whether member is in fatigue state.
7. the driver fatigue early-warning detection method of automobile is used for as claimed in claim 6, it is characterised in that in step S2
In, the image information of the passenger is processed according to the gray level co-occurrence matrixes obtain the face of the passenger
Portion's texture feature information, and the driver is identified according to the face texture characteristic information of the passenger.
8. the driver fatigue early-warning detection method of automobile is used for as claimed in claim 6, it is characterised in that in step S4
In, judge the driver whether in the fatigue state according to perclos standards and BP neural network algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410069318.5A CN103839379B (en) | 2014-02-27 | 2014-02-27 | Automobile and driver fatigue early warning detecting method and system for automobile |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410069318.5A CN103839379B (en) | 2014-02-27 | 2014-02-27 | Automobile and driver fatigue early warning detecting method and system for automobile |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103839379A CN103839379A (en) | 2014-06-04 |
| CN103839379B true CN103839379B (en) | 2017-05-10 |
Family
ID=50802836
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410069318.5A Active CN103839379B (en) | 2014-02-27 | 2014-02-27 | Automobile and driver fatigue early warning detecting method and system for automobile |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103839379B (en) |
Families Citing this family (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101555444B1 (en) * | 2014-07-10 | 2015-10-06 | 현대모비스 주식회사 | An apparatus mounted in vehicle for situational awareness and a method thereof |
| CN104123549B (en) * | 2014-07-30 | 2017-05-03 | 中国人民解放军第三军医大学第二附属医院 | Eye positioning method for real-time monitoring of fatigue driving |
| TWI603293B (en) * | 2014-10-13 | 2017-10-21 | 由田新技股份有限公司 | Method and apparatus for detecting blink |
| CN104382607B (en) * | 2014-11-26 | 2016-08-24 | 重庆科技学院 | Driver's video image fatigue detection method towards real vehicle operating mode |
| CN104616437A (en) * | 2015-02-27 | 2015-05-13 | 浪潮集团有限公司 | Vehicle-mounted fatigue identification system and method |
| CN107464310A (en) * | 2015-04-17 | 2017-12-12 | 吴红平 | Drive recorder suitable for face, based on Internet of Things |
| CN105128669A (en) * | 2015-07-26 | 2015-12-09 | 张习梅 | Safe monitoring system for motor vehicle |
| CN105069966A (en) * | 2015-07-26 | 2015-11-18 | 张习梅 | Motor vehicle safety monitoring system |
| 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 |
| CN105139584B (en) * | 2015-09-30 | 2017-12-12 | 宇龙计算机通信科技(深圳)有限公司 | A kind of fatigue driving processing method and processing 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 |
| FR3047932B1 (en) * | 2016-02-19 | 2018-03-16 | Peugeot Citroen Automobiles Sa | DEVICE AND METHOD FOR ESTIMATING THE ATTENTION LEVEL OF A DRIVER OF A VEHICLE |
| US10449968B2 (en) | 2016-09-23 | 2019-10-22 | Ford Motor Company | Methods and apparatus for adaptively assisting developmentally disabled or cognitively impaired drivers |
| CN106599821A (en) * | 2016-12-07 | 2017-04-26 | 中国民用航空总局第二研究所 | Controller fatigue detection method and system based on BP neural network |
| JP6712545B2 (en) * | 2016-12-19 | 2020-06-24 | 日立オートモティブシステムズ株式会社 | Electronic control device, electronic control system, and electronic control method |
| CN107697069B (en) * | 2017-10-31 | 2020-07-28 | 上海汽车集团股份有限公司 | Intelligent control method for fatigue driving of automobile driver |
| CN108052920A (en) * | 2017-12-27 | 2018-05-18 | 百度在线网络技术（北京）有限公司 | For the method and apparatus of output information |
| CN108556629B (en) * | 2018-04-18 | 2021-10-01 | 张玉明 | Driving safety auxiliary system and method |
| CN109063686A (en) * | 2018-08-29 | 2018-12-21 | 安徽华元智控科技有限公司 | A kind of fatigue of automobile driver detection method and system |
| CN109243144A (en) * | 2018-10-16 | 2019-01-18 | 南京伊斯特机械设备有限公司 | A kind of recognition of face warning system and its method for fatigue driving |
| CN109697095A (en) * | 2018-11-26 | 2019-04-30 | 量子云未来（北京）信息科技有限公司 | A kind of method, apparatus and terminal device promoting user's sleep |
| CN112396805B (en) * | 2019-07-31 | 2022-10-25 | Oppo广东移动通信有限公司 | Fatigue driving reminding method, device, terminal and storage medium |
| CN112201008A (en) * | 2020-08-21 | 2021-01-08 | 深圳市豪恩汽车电子装备股份有限公司 | Fatigue driving monitoring system and method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101593352A (en) * | 2009-06-12 | 2009-12-02 | 浙江大学 | Driving safety monitoring system based on face orientation and visual focus |
| CN102201148A (en) * | 2011-05-25 | 2011-09-28 | 北京航空航天大学 | Driver fatigue detecting method and system based on vision |
| TW201140511A (en) * | 2010-05-11 | 2011-11-16 | Chunghwa Telecom Co Ltd | Drowsiness detection method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4967559B2 (en) * | 2006-09-19 | 2012-07-04 | 株式会社豊田中央研究所 | Doze driving prevention device and program |
-
2014
- 2014-02-27 CN CN201410069318.5A patent/CN103839379B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101593352A (en) * | 2009-06-12 | 2009-12-02 | 浙江大学 | Driving safety monitoring system based on face orientation and visual focus |
| TW201140511A (en) * | 2010-05-11 | 2011-11-16 | Chunghwa Telecom Co Ltd | Drowsiness detection method |
| CN102201148A (en) * | 2011-05-25 | 2011-09-28 | 北京航空航天大学 | Driver fatigue detecting method and system based on vision |
Non-Patent Citations (1)
| Title |
|---|
| 复杂光照条件下的肤色检测研究;刘佩如;《中国优秀硕士学位论文全文数据库 信息科技辑》;20110415;第42页，第46页，第53-54页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103839379A (en) | 2014-06-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103839379B (en) | Automobile and driver fatigue early warning detecting method and system for automobile | |
| US11783601B2 (en) | Driver fatigue detection method and system based on combining a pseudo-3D convolutional neural network and an attention mechanism | |
| CN108309311A (en) | A kind of real-time doze of train driver sleeps detection device and detection algorithm | |
| Lenskiy et al. | Driver’s eye blinking detection using novel color and texture segmentation algorithms | |
| Ji et al. | Fatigue state detection based on multi-index fusion and state recognition network | |
| Tipprasert et al. | A method of driver’s eyes closure and yawning detection for drowsiness analysis by infrared camera | |
| CN106250801A (en) | Based on Face datection and the fatigue detection method of human eye state identification | |
| CN109740477A (en) | Study in Driver Fatigue State Surveillance System and its fatigue detection method | |
| González-Ortega et al. | Real-time vision-based eye state detection for driver alertness monitoring | |
| CN111753674A (en) | Fatigue driving detection and identification method based on deep learning | |
| CN107563346A (en) | One kind realizes that driver fatigue sentences method for distinguishing based on eye image processing | |
| Anitha et al. | A two fold expert system for yawning detection | |
| CN104331160A (en) | Lip state recognition-based intelligent wheelchair human-computer interaction system and method | |
| Rezaee et al. | Real-time intelligent alarm system of driver fatigue based on video sequences | |
| Devi et al. | Fuzzy based driver fatigue detection | |
| CN113989788A (en) | Fatigue detection method based on deep learning and multi-index fusion | |
| Al Redhaei et al. | Realtime driver drowsiness detection using machine learning | |
| Panicker et al. | Open-eye detection using iris–sclera pattern analysis for driver drowsiness detection | |
| Liu et al. | 3dcnn-based real-time driver fatigue behavior detection in urban rail transit | |
| Rani et al. | Development of an Automated Tool for Driver Drowsiness Detection | |
| Jimenez et al. | Detection of the tiredness level of drivers using machine vision techniques | |
| Kumar | Morphology based facial feature extraction and facial expression recognition for driver vigilance | |
| Liu et al. | Design and implementation of multimodal fatigue detection system combining eye and yawn information | |
| CN115100704A (en) | Face recognition device and method for resisting spoofing attack by combining thermal infrared and visible light | |
| Keyvanara et al. | Robust real-time driver drowsiness detection based on image processing and feature extraction methods |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |