CN109740512A - A kind of method for recognizing human eye state for fatigue driving judgement - Google Patents
A kind of method for recognizing human eye state for fatigue driving judgement Download PDFInfo
- Publication number
- CN109740512A CN109740512A CN201811637535.4A CN201811637535A CN109740512A CN 109740512 A CN109740512 A CN 109740512A CN 201811637535 A CN201811637535 A CN 201811637535A CN 109740512 A CN109740512 A CN 109740512A
- Authority
- CN
- China
- Prior art keywords
- eye
- state
- value
- image
- eyes
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Eye Examination Apparatus (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
Method for recognizing human eye state for fatigue driving judgement of the invention, comprising: a) acquires the n2 under the n1 under eyes-open state images, closed-eye states image of driver；B) eye image is converted gray-value image by；C) image binaryzation；D) establishes projection histogram: f) divides human eye state region；G) human eye state identifies.Method for recognizing human eye state of the invention, it is opened eyes (half closes one's eyes) according to the eye closing of the sum of the difference absolute value of the human eye two-value projection histogram provided, half, the value range of eyes-open state, for eye image to be identified, personnel state is judged by judging the interval range that it falls into, and solves the technical issues of existing method for recognizing human eye state cannot identify or cannot preferably identify half eye opening (half closes one's eyes) state.
Description
Technical field
The present invention relates to a kind of method for recognizing human eye state for fatigue driving judgement, more specifically, more particularly to
A kind of human eye shape judged for fatigue driving that can effectively come out half eye opening of driver (half closes one's eyes) state recognition
State recognition methods.
Background technique
Harmful influence transportation scale constantly expands in recent years, and the traffic accident of generation also constantly increases.Most of traffic accidents
Thing is caused by realizing shallow, fatigue driving by driver safety, therefore carries out fatigue detecting to harmful influence driver, is to keep away
Exempt from one of the means of harmful influence traffic accident generation.The method quantified at present for degree of fatigue is divided into two major classes, subjective
Evaluation assessment and objective evaluation.Subjective estimate method mainly passes through fatigue scale and gives a mark to type Acquisition Librarian, than more typical
" the fatigue symptom Self-assessment Scale " for thering is Japanese industries Society of Public Health to develop.But subjective estimate method is since its subjectivity is bigger,
Interviewee's fatigue state in certain time can only be counted, real-time detection is unable to, so being applied in fatigue driving recognition detection field
It is less.
Objective evaluation is that detection interviewee's fatigue state is gone using objective detection technique, mainly passes through information collecting device
Some fatigue characteristics of interviewee are objectively detected.For example, pass through the physiological characteristic of contact device measuring interviewee,
Such as brain electricity, electrocardiosignal, pulse detection, electromyography signal detection.Or interviewee's behavior spy is measured by contactless device
Sign, such as head, eye feature detection etc..The method avoids the strong problem of subjectivity, reliability improves a lot.And it is right
In objective detection technique, by acquiring the video image of driver in real time, and then the fatigue state for analyzing driver is more often
One of method, this method wear any aided-detection device without driver, one need to be only set up immediately ahead of driver
Common camera, this analysis method will not impact driver not only not vulnerable to the interference of human factor,
Have many advantages, such as that operability is simple, controllability is strong.
After position of human eye is accurately positioned, method is generally mainly the following for method for recognizing human eye state:
Based on objective contour method: since profile can change a lot when eyes open eyes, close one's eyes, can use this change
Change the state to judge eyes.When eye opening, eye contour shape approximate ellipse can be fitted human eye profile with ellipse, pass through meter
The short axle and long axis ratio for calculating fitted ellipse, obtain current human eye state.Human eye iris is detected, if inspection does not measure circle, it is believed that
It is to close one's eyes.Such method advantage is that principle is simple, but calculation amount is very big.
Based on template matching method: foundation is opened eyes, eye closing template goes to match target to be detected, provides current human eye state.
This method advantage is higher to single environment accuracy, but under environment complicated and changeable, this method effect is general, and by template
Picture is affected.
Method based on Gray Projection: for iris relative to other positions of eyes, gray value is smaller, so using threshold
It is worth cutting techniques for the projection in eye image progress horizontal or vertical direction, analyzes its projected image, judge eye state,
The robustness of the party is stronger.The existing method for recognizing human eye state based on Gray Projection can relatively accurately identify driving
The eye opening of member, closed-eye state, but it is poor for the identification of half eyes-open state of driver, and half eyes-open state is precisely to reflect driver
At the time of being in or will enter fatigue driving state, therefore the half eyes-open state detection of driver is particularly important.
Summary of the invention
The present invention in order to overcome the shortcomings of the above technical problems, provides a kind of human eye state for fatigue driving judgement
Recognition methods.
The method for recognizing human eye state for fatigue driving judgement of the invention, which is characterized in that by following steps come
It realizes:
A) eye image acquires, and by image collecting device, acquires the n1 under eyes-open state image of driver,
N2 under closed-eye state images, and human eye area is identified in acquired image, using as eye image, if obtain
The width and height of eye image are respectively w pixel, h pixel；
The n1 that step a) is obtained is opened eye opening images, n2 an eye closing images and is converted into gray value figure by b) image preprocessing
Picture；
C) all eye images of the gray value format obtained in step b) are converted binary map by image binaryzation
Picture；
D) establishes projection histogram, by throwing of the eye image pixel quantity that numerical value is 1 in each column on axis of abscissas
Shadow, being formed by abscissa, black picture element quantity of picture traverse is n1 eye opening histogram of ordinate, is then opened eyes to n1
Histogram corresponds to the black picture element quantity on abscissa and averages and as new ordinate value, and it is straight to form the projection of eye opening two-value
Fang Tu；
Eye closing two-value projection histogram is obtained using identical method；If the function expression of eye closing two-value projection histogram
For f1(x), the function expression of eye opening two-value projection histogram is f2(x), wherein x=0,1,2 ... w；
E) seeks the sum of difference absolute value, if g1(x)=f1(x+1)-f1(x)、g2(x)=f2(x+1)-f2(x), pass through public affairs
Formula (1) seeks the difference absolute value of eye closing two-value projection histogram and S1:
Seek the difference absolute value of eye opening two-value projection histogram by formula (2) and S2:
F) divides human eye state region, the eye closing value range of the sum of the difference absolute value of human eye two-value projection histogram
ForHalf eye opening value range isEye opening value range isWherein k > 2, S2> S1；
G) human eye state identifies, in subsequent process, for the eye image of acquisition, first, in accordance with step b) to e) phase
Same method, the function that successively eye image is pre-processed, binaryzation, projection histogram is established, obtains projection histogram
Expression formula, finally seek the difference absolute value of the two-value projection histogram of eye image to be identified and S '；Then judge that S ' is fallen
Enter that section in step f), the corresponding state in fallen into section is denoted as current human eye state；
If it is judged that human eye state is eye closing or half eyes-open state, show that driver is currently at fatigue driving state；
If human eye state is eyes-open state, show that driver status is normal.
The method for recognizing human eye state for fatigue driving judgement of the invention, image binaryzation described in step c) use
Eye image is divided into the bianry image that prospect and background are constituted, eyes peripheral outline and rainbow by Otsu maximum variance between clusters
Membrane part is divided into prospect, and the pixel value of corresponding region is 1, and remaining area is divided into background, the pixel value of corresponding region
It is 0.
The beneficial effects of the present invention are: the method for recognizing human eye state for fatigue driving judgement of the invention, first will
Eye image under the eye opening of acquisition, closed-eye state carries out gray processing, binary conversion treatment, then using picture traverse as abscissa,
Black picture element quantity is that ordinate establishes the two-value projection histogram of eye image, then seeks eye closing, eye opening two-value projection histogram
The difference absolute value of figure and S1, S2, finally provide the eye closing of the sum of the difference absolute value of human eye two-value projection histogram, partly open
The value range of eye (half close one's eyes), eyes-open state, for eye image to be identified, by judge the interval range that it falls into come
Judge personnel state, when judging human eye state is eye closing or half eyes-open state, then shows that driver is fatigue driving, solve
Existing method for recognizing human eye state cannot identify or cannot preferably identify the technical issues of half eye opening (half closes one's eyes) state.
Detailed description of the invention
Fig. 1 is gray processing, the binary conversion treatment schematic diagram of eye opening eye image of the invention；
Fig. 2 is gray processing, the binary conversion treatment schematic diagram of half eye opening eye image of the invention；
Fig. 3 is gray processing, the binary conversion treatment schematic diagram of eye closing eye image of the invention；
Fig. 4 is that eye opening bianry image of the invention is formed by projection histogram；
Fig. 5 is the function curve of eye opening projection histogram of the invention；
Fig. 6 is that half eye opening bianry image of the invention is formed by projection histogram；
Fig. 7 is the function curve of half eye opening projection histogram of the invention；
Fig. 8 is that eye closing bianry image of the invention is formed by projection histogram；
Fig. 9 is the function curve of eye closing projection histogram of the invention；
Figure 10 is the accuracy rate that existing Two-peak method identifies human eye state under dark, moderate and stronger state；
Figure 11 is for recognition methods of the invention to the accurate of human eye state identification under dark, moderate and stronger state
Rate.
Specific embodiment
The invention will be further described with embodiment with reference to the accompanying drawing.
As shown in Figure 1, giving the gray processing of eye opening eye image of the invention, binary conversion treatment schematic diagram, Fig. 2 is provided
Gray processing, the binary conversion treatment schematic diagram of half eye opening eye image of the invention, it can be clearly seen that when in opening eyes or half
When eyes-open state, pupil region pixel carries out image so Threshold sementation can be used obviously secretly in other position pixels
Pupil region is converted black picture element by binarization segmentation, other regioinvertions are white pixel.Common image binaryzation point
It has cut: adaptive threshold, histogram thresholding, Otsu maximum variance between clusters etc..Using Otsu adaptive threshold point in the present invention
It cuts technology to separate target (pupil) and background (other regions), gray level image is converted into bianry image at this time.Such as Fig. 3 institute
Show, gives gray processing, the binary conversion treatment schematic diagram of eye closing eye image of the invention.
Otsu algorithm principle is prospect and two images of background to be divided the image into using Threshold sementation, and work as and obtain most
When good segmentation threshold, the inter-class variance between background and prospect is maximum.Because variance is a kind of measurement of intensity profile uniformity,
Inter-class variance between background and prospect is bigger, illustrates that the two-part difference for constituting image is bigger, prospect mistake is divided into when part
Background or part background mistake, which are divided into prospect, all can cause two parts difference to become smaller.Therefore, make the maximum segmentation meaning of inter-class variance
Misclassification probability it is minimum.
Fig. 1, Fig. 2 and it is shown in Fig. 3 open eyes, half opens eyes and eye closing image, eyes peripheral outline and iris after binaryzation
Part is divided into prospect (numerical value 1), and other regions are divided into background (numerical value 0).At this point, we count binaryzation
Black picture element quantity projection to axis of abscissas on of the eye image in each column, projection histogram such as Fig. 4, Fig. 6 and Fig. 8 institute
Show.
As shown in figure 4, giving eye opening bianry image of the invention is formed by projection histogram, ground for ease of observation
Study carefully, is the function curve of eye opening projection histogram of the invention shown in Fig. 5 after serialization, in bimodal by its serialization
Feature.Fig. 8 gives eye closing bianry image of the invention and is formed by projection histogram, will be to close one's eyes after its serialization in Fig. 9
The function curve of projection histogram, in the quadratic function shape that Open Side Down.Fig. 6 gives half eye opening bianry image institute of the invention
The projection histogram of formation will be the function curve of half eye opening projection histogram shown in fig. 7 after its serialization, in opening to
Under quadratic function shape, be also in double-peak feature.So if merely in shape from the function curve of human eye projection histogram
Judge, then cannot effectively distinguish opening eyes with half eyes-open state.
There is the bimodal distance by counting human eye histogram to judge eye state at present, however, the method can only judge
Eyes fully closed condition can not correctly judge semi-closure conjunction state.Because driver, in fatigue, eyes are typically in
Semi-closure conjunction state, at this point, detecting that eyes are in semi-closure conjunction state and can be considered as fatigue driving.As shown in Figure 10, it gives existing
Have Two-peak method under dark, moderate and stronger state to human eye state identification accuracy rate, it is seen that no matter intensity of illumination such as
What, when eyes are in semi-closure conjunction state, Two-peak method cannot accurately identify eye state, and when dark, can reduce eye opening
State recognition rate.
Method for recognizing human eye state for fatigue driving judgement of the invention, is realized by following steps:
A) eye image acquires, and by image collecting device, acquires the n1 under eyes-open state image of driver,
N2 under closed-eye state images, and human eye area is identified in acquired image, using as eye image, if obtain
The width and height of eye image are respectively w pixel, h pixel；
The n1 that step a) is obtained is opened eye opening images, n2 an eye closing images and is converted into gray value figure by b) image preprocessing
Picture；
C) all eye images of the gray value format obtained in step b) are converted binary map by image binaryzation
Picture；
D) establishes projection histogram, by throwing of the eye image pixel quantity that numerical value is 1 in each column on axis of abscissas
Shadow, being formed by abscissa, black picture element quantity of picture traverse is n1 eye opening histogram of ordinate, is then opened eyes to n1
Histogram corresponds to the black picture element quantity on abscissa and averages and as new ordinate value, and it is straight to form the projection of eye opening two-value
Fang Tu；
Eye closing two-value projection histogram is obtained using identical method；If the function expression of eye closing two-value projection histogram
For f1(x), the function expression of eye opening two-value projection histogram is f2(x), wherein x=0,1,2 ... w；
E) seeks the sum of difference absolute value, if g1(x)=f1(x+1)-f1(x)、g2(x)=f2(x+1)-f2(x), pass through public affairs
Formula (1) seeks the difference absolute value of eye closing two-value projection histogram and S1:
Seek the difference absolute value of eye opening two-value projection histogram by formula (2) and S2:
F) divides human eye state region, the eye closing value range of the sum of the difference absolute value of human eye two-value projection histogram
ForHalf eye opening value range isEye opening value range isWherein k > 2, S2> S1；
G) human eye state identifies, in subsequent process, for the eye image of acquisition, first, in accordance with step b) to e) phase
Same method, the function that successively eye image is pre-processed, binaryzation, projection histogram is established, obtains projection histogram
Expression formula, finally seek the difference absolute value of the two-value projection histogram of eye image to be identified and S '；Then judge that S ' is fallen
Enter that section in step f), the corresponding state in fallen into section is denoted as current human eye state；
If it is judged that human eye state is eye closing or half eyes-open state, show that driver is currently at fatigue driving state；
If human eye state is eyes-open state, show that driver status is normal.
In order to illustrate the feasibility of the method for recognizing human eye state for fatigue driving judgement of the invention, if Fig. 5, Fig. 7
Function curve with the projection histogram under eye opening given by Fig. 9, half eye opening and closed-eye state is continuous function, and everywhere may be used
It leads, the sum for seeking the difference absolute value of Gray Projection histogram is converted into the exhausted of the integral for seeking Gray Projection histogram derived function
To the sum of value.For setting the function f of eye closing two-value projection histogram1(x), its maximum value might as well be set as h1, and f1(a)=h1, f1
(0)=f1(w)=0.Then f1(x) difference absolute value and calculation method are as follows:
For eye opening Gray Projection histogram functions, if its expression formula is f2It (x), is bimodal function, first wave peak value is
h2, secondary peak value is h3, valley value h4, and f2(b)=h2, f2(c)=h3,
f2(d)=h4, f2(0)=f2(w)=0,0 < b < d < c < w, h3> h2> h4>=0 has:
Due to f under eyes-open state1(x) first wave peak value h3Greater than f under closed-eye state2(x) crest value h1, so can
Obtain S2> S1, i.e., eye opening Histogram Difference and be greater than eye closing Histogram Difference and.For half closed-eye state Histogram Difference and Ying Jie
In S1With S2Between.
As shown in figure 11, recognition methods of the invention is given under dark, moderate and stronger state to human eye shape
The accuracy rate of state identification judges the state of eyes photo using eye opening Histogram Difference and feature under three kinds of different conditions
Accurate picture, and joined illumination factor, experiment, which acquires, 450 opens one's eyes portion's photo totally, wherein lower 50 photos of each state.It can
, it is evident that being under semi-closure conjunction state for human eye, its accuracy rate can be greatly improved using Histogram Difference and feature, and
And in the case where dark, accuracy rate also has different degrees of promotion in contrast to Two-peak method.
Claims (2)
1. a kind of method for recognizing human eye state for fatigue driving judgement, which is characterized in that realized by following steps:
A) eye image acquires, and by image collecting device, acquires the n1 under eyes-open state image of driver, is closing one's eyes
N2 under state images, and human eye area is identified in acquired image, using as eye image, if the human eye obtained
The width and height of image are respectively w pixel, h pixel；
The n1 that step a) is obtained is opened eye opening images, n2 an eye closing images and is converted into gray-value image by b) image preprocessing；
C) all eye images of the gray value format obtained in step b) are converted bianry image by image binaryzation；
D) establishes projection histogram, projection of the pixel quantity on axis of abscissas that numerical value is 1 in each column by eye image,
Being formed by abscissa, black picture element quantity of picture traverse is n1 eye opening histogram of ordinate, is then opened eyes to n1 straight
The black picture element quantity that side schemes on corresponding abscissa averages and as new ordinate value, forms eye opening two-value and projects histogram
Figure；
Eye closing two-value projection histogram is obtained using identical method；If the function expression of eye closing two-value projection histogram is f1
(x), the function expression of eye opening two-value projection histogram is f2(x), wherein x=0,1,2 ... w；
E) seeks the sum of difference absolute value, if g1(x)=f1(x+1)-f1(x)、g2(x)=f2(x+1)-f2(x), pass through formula (1)
Seek the difference absolute value of eye closing two-value projection histogram and S1:
Seek the difference absolute value of eye opening two-value projection histogram by formula (2) and S2:
F) divides human eye state region, and the eye closing value range of the sum of the difference absolute value of human eye two-value projection histogram isHalf eye opening value range isEye opening value range isWherein k > 2, S2> S1；
G) human eye state identifies, in subsequent process, for the eye image of acquisition, first, in accordance with step b) to e) identical
Method, the function representation that successively eye image is pre-processed, binaryzation, projection histogram is established, obtains projection histogram
Formula, finally seek the difference absolute value of the two-value projection histogram of eye image to be identified and S '；Then judge that S ' falls into step
It is rapid f) in that section, the corresponding state in fallen into section is denoted as current human eye state；
If it is judged that human eye state is eye closing or half eyes-open state, show that driver is currently at fatigue driving state；If
Human eye state is eyes-open state, shows that driver status is normal.
2. the method for recognizing human eye state according to claim 1 for fatigue driving judgement, it is characterised in that: step c)
The image binaryzation uses Otsu maximum variance between clusters, and eye image is divided into the binary map that prospect and background are constituted
Picture, eyes peripheral outline and iris portion are divided into prospect, and the pixel value of corresponding region is 1, and remaining area is divided into
Background, the pixel value of corresponding region are 0.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811637535.4A CN109740512A (en) | 2018-12-29 | 2018-12-29 | A kind of method for recognizing human eye state for fatigue driving judgement |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811637535.4A CN109740512A (en) | 2018-12-29 | 2018-12-29 | A kind of method for recognizing human eye state for fatigue driving judgement |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109740512A true CN109740512A (en) | 2019-05-10 |
Family
ID=66362285
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811637535.4A Pending CN109740512A (en) | 2018-12-29 | 2018-12-29 | A kind of method for recognizing human eye state for fatigue driving judgement |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109740512A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111078000A (en) * | 2019-11-18 | 2020-04-28 | 中北大学 | Method, device and system for performing eye-machine interaction according to eye behavior characteristics |
| CN117082665A (en) * | 2023-10-17 | 2023-11-17 | 深圳市帝狼光电有限公司 | LED eye-protection desk lamp illumination control method and system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101739548A (en) * | 2009-02-11 | 2010-06-16 | 北京智安邦科技有限公司 | Eye detection method and system |
| CN103268479A (en) * | 2013-05-29 | 2013-08-28 | 电子科技大学 | Method for detecting fatigue driving around clock |
| CN103886307A (en) * | 2014-04-15 | 2014-06-25 | 王东强 | Sight tracking and fatigue early warning method |
| CN105224285A (en) * | 2014-05-27 | 2016-01-06 | 北京三星通信技术研究有限公司 | Eyes open and-shut mode pick-up unit and method |
| CN106250801A (en) * | 2015-11-20 | 2016-12-21 | 北汽银翔汽车有限公司 | Based on Face datection and the fatigue detection method of human eye state identification |
| CN108256390A (en) * | 2016-12-29 | 2018-07-06 | 广州映博智能科技有限公司 | Eye motion method for catching based on projecting integral and iris recognition |
-
2018
- 2018-12-29 CN CN201811637535.4A patent/CN109740512A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101739548A (en) * | 2009-02-11 | 2010-06-16 | 北京智安邦科技有限公司 | Eye detection method and system |
| CN103268479A (en) * | 2013-05-29 | 2013-08-28 | 电子科技大学 | Method for detecting fatigue driving around clock |
| CN103886307A (en) * | 2014-04-15 | 2014-06-25 | 王东强 | Sight tracking and fatigue early warning method |
| CN105224285A (en) * | 2014-05-27 | 2016-01-06 | 北京三星通信技术研究有限公司 | Eyes open and-shut mode pick-up unit and method |
| CN106250801A (en) * | 2015-11-20 | 2016-12-21 | 北汽银翔汽车有限公司 | Based on Face datection and the fatigue detection method of human eye state identification |
| CN108256390A (en) * | 2016-12-29 | 2018-07-06 | 广州映博智能科技有限公司 | Eye motion method for catching based on projecting integral and iris recognition |
Non-Patent Citations (4)
| Title |
|---|
| GARCIA I ET AL: "Vision-based drowsiness detector for real driving conditions", 《 2012 IEEE INTELLIGENT VEHICLES SYMPOSIUM》 * |
| 侯向丹 等: "基于积分投影和差分投影的人眼定位", 《计算机工程与科学》 * |
| 朱喜燕 等: "Gentle-Adaboost在红外视频驾驶员疲劳检测中的应用研究", 《计算机测量与控制》 * |
| 邢益良 等: "积分投影与活动轮廓相结合的人眼疲劳识别", 《计算机工程》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111078000A (en) * | 2019-11-18 | 2020-04-28 | 中北大学 | Method, device and system for performing eye-machine interaction according to eye behavior characteristics |
| CN117082665A (en) * | 2023-10-17 | 2023-11-17 | 深圳市帝狼光电有限公司 | LED eye-protection desk lamp illumination control method and system |
| CN117082665B (en) * | 2023-10-17 | 2023-12-15 | 深圳市帝狼光电有限公司 | LED eye-protection desk lamp illumination control method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Alshaqaqi et al. | Driver drowsiness detection system | |
| KR100912746B1 (en) | Method for traffic sign detection | |
| CN108053615B (en) | Method for detecting fatigue driving state of driver based on micro-expression | |
| CN104091147A (en) | Near infrared eye positioning and eye state identification method | |
| US8233670B2 (en) | System and method for traffic sign recognition | |
| CN100373397C (en) | Pre-processing method for iris image | |
| CN110097034A (en) | A kind of identification and appraisal procedure of Intelligent human-face health degree | |
| CN106250801A (en) | Based on Face datection and the fatigue detection method of human eye state identification | |
| CN103263278B (en) | Image processing method for automatically measuring thickness of fetal nuchal translucency from ultrasonic image | |
| CN101339603A (en) | Method for selecting qualified iris image from video frequency stream | |
| CN102054163A (en) | Method for testing driver fatigue based on monocular vision | |
| CN106650669A (en) | Face recognition method for identifying counterfeit photo deception | |
| WO2020098038A1 (en) | Pupil tracking image processing method | |
| CN101593352A (en) | Driving safety monitoring system based on face orientation and visual focus | |
| WO2010037332A1 (en) | Method and device for training classifier, method and device for recognizing picture | |
| CN101375796A (en) | Real-time detection system of fatigue driving | |
| CN105787929A (en) | Skin rash point extraction method based on spot detection | |
| Liu et al. | A practical driver fatigue detection algorithm based on eye state | |
| CN109740512A (en) | A kind of method for recognizing human eye state for fatigue driving judgement | |
| CN106022278A (en) | Method and system for detecting people wearing burka in video images | |
| CN106203338B (en) | Human eye state method for quickly identifying based on net region segmentation and threshold adaptive | |
| CN113140093A (en) | Fatigue driving detection method based on AdaBoost algorithm | |
| TWI427545B (en) | Face recognition method based on sift features and head pose estimation | |
| CN103729646A (en) | Eye image validity detection method | |
| KR101523765B1 (en) | Enhanced Method for Detecting Iris from Smartphone Images in Real-Time |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190510 |