CN109165630A - A kind of fatigue monitoring method based on two-dimentional eye recognition - Google Patents
A kind of fatigue monitoring method based on two-dimentional eye recognition Download PDFInfo
- Publication number
- CN109165630A CN109165630A CN201811093844.XA CN201811093844A CN109165630A CN 109165630 A CN109165630 A CN 109165630A CN 201811093844 A CN201811093844 A CN 201811093844A CN 109165630 A CN109165630 A CN 109165630A
- Authority
- CN
- China
- Prior art keywords
- eye
- image
- face
- algorithm
- method based
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/59—Context or environment of the image inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions
- G06V20/597—Recognising the driver's state or behaviour, e.g. attention or drowsiness
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/165—Detection; Localisation; Normalisation using facial parts and geometric relationships
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
Abstract
The present invention discloses a kind of fatigue monitoring method based on two-dimentional eye recognition, and by eye recognition algorithm and face feature point algorithm for estimating carries out eye recognition to image information and human eye state detects, and can be used for carrying out fatigue monitoring in automobile.The driving situation of Real-time Feedback driver and tired situation, and send the tired situation of driver on App in real time and realize detection of the household to tired driver situation.Realize household supervision, in real time remind, thus realize reduction because tired driver drive and caused by accident casualty.The present invention has good effect on discrimination and safety and without a large amount of human face data, can greatly improve fatigue monitoring speed in the case where guaranteeing safety.
Description
Technical field
The invention belongs to image processing method law technology more particularly to a kind of fatigue monitoring sides based on two-dimentional eye recognition
Method.
Background technique
With society be constantly progressive and an urgent demand of the various aspects for quickly and effectively auto authentication, biology
Feature identification technique has obtained development at full speed in recent decades.As a kind of inherent attribute of people, and have very strong
Self stability and individual difference, biological characteristic become the most ideal foundation of auto authentication.Current biological characteristic
Identification technology mainly includes: fingerprint recognition, retina identification, iris recognition, Gait Recognition, hand vein recognition, recognition of face etc..
Compared with other recognition methods, eye recognition is due to having directly, friendly, convenient feature, user without any mental handicape,
It is easy to be received by user, to obtain extensive research and application.
Chinese CN200510027371.X patent is related to the recognition methods of a kind of human eye positioning and human eye opening and closing.This
Invention is mainly solved the problems, such as to the human eye of dynamic image into identification.It the steps include: the frame figure for arriving camera Dynamic Extraction
As carrying out automatic gray scale balance using grey level histogram, the face of people is revealed from background convexity, recycles adjustable half window
Thresholding extracts face from background, according to the human eye pixel block size of estimation, removes non-human eye area, then in conjunction with people
The two-dimensional geometry relationship of eye automatically determines the eyes of people, is shown on the original image with black surround, if do not detected double
Eye, system voice prompt；The size for recycling eyes pixel, judges the opening and closure of eyes；If eyes open, original graph
There to be black surround to show on picture, program does not issue prompt tone；If eyes are closed, will be shown without black surround on original image, program
Issue prompt tone.The recognition methods needs data set amount big, and detection speed is slow.
Chinese CN201710315119.1 patent is related to a kind of naked-eye stereoscopic display device of combination visual fatigue detection
And method.The present apparatus is mainly by infrared binocular camera, eye image analytical unit, eye space position feedback unit, human eye
Vision area adjustment unit, left and right view show adjustment unit, human eye fatigue detecting unit and video output unit composition.This method is adopted
The space coordinate for positioning human eye in real time with human eye eye tracking technology, using programmable logic array technology real-time tracking eyes position
It sets movement and effectively reduces human eye with corresponding liquid crystal grating image sub-pixels progress rearrangement to change best view region
Picture crosstalk in watching process improves user and sees view experience；Meanwhile it being used by above-mentioned technology real-time detection, measurement and evaluation
Eye strain state of family during seeing view, determines its bad physiological reaction degree, pre- to prompt user to take timely measure
Anti- or alleviation visual fatigue.This method needs a large amount of data set to be trained, and recognition speed is slower.
In recent years, there is also many drawbacks, detection speed for traditional human eye there is no relatively good computer face recognition software
Degree is slow, needs data set amount big, excessively high to environmental requirement.Currently, there are many algorithms of research recognition of face both at home and abroad, usual face
The face automatic mode identification technology of still image is divided into three categories: recognition methods based on geometrical characteristic, special based on algebra
The recognition methods of sign and recognition methods based on connection mechanism；A large amount of face's figure is needed using the identification method of neural network
Library identified using the method for self study, its requirement to face's picture library is relatively high.Therefore, it is badly in need of designing a kind of based on two
The fatigue monitoring method of eye recognition is tieed up to solve the above problems.
Summary of the invention
In response to the problems existing in the prior art, the purpose of the present invention is to provide a kind of, and the fatigue based on two-dimentional eye recognition is supervised
Survey method, low to environmental requirement during eye recognition by the eye recognition algorithm, recognition speed is fast, discrimination is high；
Mass data collection is not needed by the face feature point algorithm for estimating to be trained, and is solved training data and is collected difficult ask
Topic.
To achieve the above object, the technical solution adopted by the present invention is that:
The present invention provides a kind of fatigue monitoring method based on two-dimentional eye recognition, comprising the following steps:
(1) man face image acquiring is carried out to driver by image collecting device and finds face；
(2) eye recognition and human eye shape are carried out to image information by eye recognition algorithm and face feature point algorithm for estimating
State detection；
(3) according to human eye state delivering prompt tone, and App is sent status information to.
The working principle of fatigue detection method of the present invention are as follows: in driving procedure, driver is controllable to open fatigue
The device of detection is acquired the face-image of driver by the camera on automobile, then is obtained by eye recognition algorithm
To the location information of human eye, the position of human eye is monitored in real time；By the face feature point algorithm for estimating to eye spy
The fluctuation situation of sign point is recorded, and the driving condition of driver is determined by the variation of eye feature point, by specifically mentioning
Show sound to warn fatigue state, and status information is sent to App.Realize household supervision, in real time remind, thus realize reduction because
Accident casualty caused by for tired driver driving.The present invention has good effect and is not necessarily on discrimination and safety
A large amount of human face datas can greatly improve fatigue monitoring speed in the case where guaranteeing safety.
The eye recognition algorithm the following steps are included:
1) image is subjected to gray proces, judges the pixel depth of current pixel Yu its neighboring pixel, represented and schemed with arrow
As dimmed direction；
2) all gradients in each direction are calculated in the small cube for dividing the image into 16 × 16 pixels；With direction
Property most strong direction arrow replace original small cube, obtain processing image；
3) histograms of oriented gradients generated by a large amount of human face photos, by the side of processing image and generation obtained by step 2)
It is compared to histogram of gradients, finding similar place is face.
The principle of the eye recognition algorithm are as follows: picture is subjected to gray proces first, then to each picture in picture
Element is checked, while checking surrounding's element of individual element, it is therefore an objective to which the pixel for finding current pixel and its neighboring pixel is deep
Degree, and the direction dimmed with arrow representative image.If by all pixels being substituted with the method, final pixel meeting
It is replaced by arrow, but this operand is too big.In order to accomplish this point, inventor divides the image into the small of some 16 × 16 pixels
Square, in each small cube, we will calculate how many gradient in each principal direction, and (how many is directed toward, and is directed toward right
On, it is directed toward right etc.).Then that original small cube is replaced with the arrow in that most strong direction of directive property.It only needs later
It is compared with the histograms of oriented gradients generated from various facial photos, finding similar place is exactly face.
Present method solves the detections of traditional eye recognition to environmental requirement height, and detection speed is slow and the disadvantage of safety difference.
The face feature point algorithm for estimating the following steps are included:
A, by calling convolutional neural networks in dlib to be trained above-mentioned machine learning algorithm, 68 after being trained
Feature point detection algorithm obtains 68 characteristic points of any face using 68 feature point detection algorithm, so that it is determined that image
The position of middle eyes and mouth；
B, it determines in image behind the position of eyes and mouth, image is rotated, is scaled and mistake is cut, so that eyes and mouth
Bar as close to center；
C, eye locations are marked in real time according to the 6 of eye characteristic points, and the state of 6 characteristic points of human eye is carried out real
When monitor.
The basic ideas of the face feature point algorithm for estimating be call dlib in convolutional neural networks to above-mentioned engineering
It practises algorithm to be trained, 68 feature point detection algorithms after being trained are obtained arbitrarily using 68 feature point detection algorithm
68 characteristic points of face, so that it is determined that in image eyes and mouth position；After knowing the position of eyes and mouth, it will scheme
As being rotated, being scaled and mistake is cut so that eyes and mouth be as close to center, so no matter facial orientation which side, we
Eyes and mouth can be moved to centre to roughly the same position, so that in next step to the identification of eye locations and eye
The judgement of state is more accurate；Then eye locations are marked in real time according to the 6 of eye characteristic points, and to 6 features of human eye
The state of point carries out real-time monitoring.
In the face feature point algorithm for estimating step C, the side of real-time monitoring is carried out to the state of 6 characteristic points of human eye
Method are as follows: the concept for introducing eyes aspect ratio EAR is respectively defined as p to 6 characteristic point coordinates of human eye1,p2,p3,p4,p5,p6,
EAR formula is
Determine that human eye state is effective blink or fatigue state by the variation fluctuation of EAR.
The length-width ratio of eyes is generally constant when eyes open, but can quickly fall to when blinking
Zero, the detection to blink can be realized by the monitoring to EAR.
The judgment method effectively blinked are as follows: when the EAR value fluctuates, and record frame number is greater than 4 frame picture, note
Record is this time effectively blink.
The judgment method of the fatigue state are as follows: within a specified time detect that EAR most 0 or change frequency are too small, remember
Record is fatigue driving.To issue alarm sound prompt driver and will send information on App.
Compared with prior art, the beneficial effects of the present invention are:
(1) eye recognition algorithm of the present invention divides the image into the small cube of some 16 × 16 pixels, each small
In square, we will calculate how many gradient in each principal direction, then with the arrow in that most strong direction of directive property come
It instead of that original small cube, is compared with histograms of oriented gradients and finds face location, overcome traditional eye recognition
Detection is to environmental requirement height, and detection speed is slow and the disadvantage of safety difference；
(2) face feature point algorithm for estimating of the present invention organically blends computer vision and machine learning algorithm, solution
It is certainly existing that the predicament that Face datection is faced with training data collection difficulty is carried out using neural network；
(3) a kind of fatigue monitoring method based on two-dimentional eye recognition of the present invention, passes through the frequency of wink to driver
With eye closing situation, driver is reminded, and is transmitted information on App, realizes household to the real-time monitoring of driver；
(4) present invention provide it is a kind of quickly accurately Eye Recognition not only solve traditional human eye detection to environment
It is required that it is high, data set is required to do, detects the slow limitation of speed, and realize household to the remote prompting of driver.
Detailed description of the invention
Fig. 1 is a kind of flow chart of the fatigue monitoring method based on two-dimentional eye recognition of the present invention；
Fig. 2 is 68 feature point diagrams of face of the present invention；
Fig. 3 (a) is the coordinate diagram of 6 characteristic points of human eye of the present invention；
The coordinate diagram of 6 characteristic points of human eye when Fig. 3 (b) is present invention blink；
Fig. 3 (c) is fluctuation situation map of the present invention in fatigue state EAR.
Specific embodiment
Technical solution of the present invention will be clearly and completely described below, it is clear that described embodiment is only
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
All other embodiment obtained, shall fall within the protection scope of the present invention under the conditions of not making creative work.
A kind of fatigue monitoring method based on two-dimentional eye recognition, comprising the following steps:
(1) man face image acquiring is carried out to driver by image collecting device；
(2) eye recognition and human eye shape are carried out to image information by eye recognition algorithm and face feature point algorithm for estimating
State detection；
(3) according to human eye state delivering prompt tone, and App is sent status information to.
As shown in Figure 1, opening automotive interior camera, face is found, is detected by 68 characteristic point of face and determines face position
It sets and eye locations, blink detections is carried out to 6 characteristic points of eye, if frequency of wink is excessively high or closes one's eyes, system will be recorded as
Fatigue state, system, which will sound an alarm, reminds driver, and pushes current state to App, realizes real time monitoring and reminds；It is monitoring
During, the frequency information of blink detection is sent to App by system, draws the driving status that line summarizes driver.
The eye recognition algorithm the following steps are included:
Picture is subjected to gray proces first, then each pixel in picture is checked, while checking single member
Surrounding's element of element, it is therefore an objective to find the pixel depth of current pixel Yu its neighboring pixel, and dimmed with arrow representative image
Direction.If by all pixels being substituted with the method, final pixel can be replaced by arrow, but this operand is too
Greatly.In order to accomplish this point, inventor divides the image into the small cube of some 16 × 16 pixels, in each small cube, I
Will calculate how many gradient in each principal direction (how many is directed toward, and is directed toward upper right, is directed toward right etc.).Then with direction
Property most strong that direction arrow replace that original small cube.It only needs later and from various facial photos
The histograms of oriented gradients of generation is compared, and finding similar place is exactly face.
The face feature point algorithm for estimating the following steps are included:
As shown in Fig. 2, the basic ideas of the face feature point algorithm for estimating are to call convolutional neural networks pair in dlib
Above-mentioned machine learning algorithm is trained, 68 feature point detection algorithms after being trained, and is detected and is calculated using 68 characteristic point
Method obtains 68 characteristic points of any face, so that it is determined that in image eyes and mouth position；Knowing eyes and mouth
Behind position, image is rotated, is scaled and mistake is cut so that eyes and mouth be as close to center, so no matter face court
To which side, we can move eyes and mouth to roughly the same position to centre, so that in next step to eye locations
Identification and the judgement of eye state are more accurate；Then eye locations are marked in real time according to the 6 of eye characteristic points, and to people
The state of 6 characteristic points of eye carries out real-time monitoring.
In the face feature point algorithm for estimating step C, the side of real-time monitoring is carried out to the state of 6 characteristic points of human eye
Method are as follows: the concept for introducing eyes aspect ratio EAR is respectively defined as p to 6 characteristic point coordinates of human eye as shown in Fig. 3 (a)1,
p2,p3,p4,p5,p6, EAR formula is
Determine that human eye state is effective blink or fatigue state by the variation fluctuation of EAR.
The length-width ratio of eyes is generally constant when eyes open, as shown in Fig. 3 (b), the meeting when blinking
Zero is quickly fallen to, the detection to blink can be realized by the monitoring to EAR.When the EAR value fluctuates, record
When frame number is greater than 4 frame picture, record is this time effectively blink.
It is the fluctuation situation of EAR value as shown in Fig. 3 (c), within a specified time detects that EAR value is 0 or change frequency
It is too small, then it is considered that driver is in a state of fatigue.To issue alarm sound prompt driver and will send information on App.
It although an embodiment of the present invention has been shown and described, for the ordinary skill in the art, can be with
A variety of variations, modification, replacement can be carried out to these embodiments without departing from the principles and spirit of the present invention by understanding
And modification, the scope of the present invention is defined by the appended.
Claims (6)
1. a kind of fatigue monitoring method based on two-dimentional eye recognition, which comprises the following steps:
(1) man face image acquiring is carried out to driver by image collecting device and finds face；
(2) eye recognition and human eye state inspection are carried out to image information by eye recognition algorithm and face feature point algorithm for estimating
It surveys；
(3) according to human eye state delivering prompt tone, and App is sent status information to.
2. a kind of fatigue monitoring method based on two-dimentional eye recognition according to claim 1, which is characterized in that the people
Eye recognizer the following steps are included:
1) image is subjected to gray proces, judges the pixel depth of current pixel Yu its neighboring pixel, become with arrow representative image
Dark direction；
2) all gradients in each direction are calculated in the small cube for dividing the image into 16 × 16 pixels；Most with directive property
The arrow in strong direction replaces original small cube, obtains processing image；
3) histograms of oriented gradients generated by a large amount of human face photos, by the ladder of the direction of processing image and generation obtained by step 2)
Degree histogram compares, and finding similar place is face.
3. a kind of fatigue monitoring method based on two-dimentional eye recognition according to claim 1, which is characterized in that the face
Portion's facial feature estimation algorithm the following steps are included:
A, by calling convolutional neural networks in dlib to be trained above-mentioned machine learning algorithm, 68 features after being trained
Point detection algorithm, obtains 68 characteristic points of any face using 68 feature point detection algorithm, so that it is determined that eye in image
The position of eyeball and mouth；
B, it determines in image behind the position of eyes and mouth, image is rotated, is scaled and mistake is cut, so that eyes and mouth are most
It may be close to center；
C, eye locations are marked in real time according to the 6 of eye characteristic points, and the state of 6 characteristic points of human eye is supervised in real time
It surveys.
4. a kind of fatigue monitoring method based on two-dimentional eye recognition according to claim 3, which is characterized in that the face
In portion facial feature estimation algorithm steps C, to the method for the state progress real-time monitoring of 6 characteristic points of human eye are as follows: introduce eyes
The concept of aspect ratio EAR is respectively defined as p to 6 characteristic point coordinates of human eye1,p2,p3,p4,p5,p6, EAR formula is
Determine that human eye state is effective blink or fatigue state by the variation fluctuation of EAR.
5. a kind of fatigue monitoring method based on two-dimentional eye recognition according to claim 4, which is characterized in that described to have
Imitate the judgment method blinked are as follows: when the EAR value fluctuates, and record frame number is greater than 4 frame picture, record is this time effectively to blink
Eye.
6. a kind of fatigue monitoring method based on two-dimentional eye recognition according to claim 4, which is characterized in that described tired
The judgment method of labor state are as follows: within a specified time detect that EAR most 0 or change frequency are too small, be recorded as fatigue driving.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811093844.XA CN109165630A (en) | 2018-09-19 | 2018-09-19 | A kind of fatigue monitoring method based on two-dimentional eye recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811093844.XA CN109165630A (en) | 2018-09-19 | 2018-09-19 | A kind of fatigue monitoring method based on two-dimentional eye recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109165630A true CN109165630A (en) | 2019-01-08 |
Family
ID=64879662
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811093844.XA Pending CN109165630A (en) | 2018-09-19 | 2018-09-19 | A kind of fatigue monitoring method based on two-dimentional eye recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109165630A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109784313A (en) * | 2019-02-18 | 2019-05-21 | 上海骏聿数码科技有限公司 | A kind of blink detection method and device |
| CN109871773A (en) * | 2019-01-21 | 2019-06-11 | 深圳市云眸科技有限公司 | Biopsy method, device and door access machine |
| CN109977820A (en) * | 2019-03-14 | 2019-07-05 | 重庆邮电大学 | A kind of fatigue driving determination method |
| CN113034851A (en) * | 2021-03-11 | 2021-06-25 | 中铁工程装备集团有限公司 | Tunnel boring machine driver fatigue driving monitoring device and method |
| CN113065804A (en) * | 2021-04-27 | 2021-07-02 | 山东交通学院 | Hazardous chemical substance road transportation risk assessment method and system |
| CN113198067A (en) * | 2021-06-03 | 2021-08-03 | 浙大城市学院 | Automatic medical care monitoring system |
| CN113844455A (en) * | 2021-08-27 | 2021-12-28 | 奇瑞汽车股份有限公司 | Method and device for braking a vehicle while the driver is tired |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104361716A (en) * | 2014-10-31 | 2015-02-18 | 新疆宏开电子系统集成有限公司 | Method for detecting and reminding fatigue in real time |
| CN107563346A (en) * | 2017-09-20 | 2018-01-09 | 南京栎树交通互联科技有限公司 | One kind realizes that driver fatigue sentences method for distinguishing based on eye image processing |
| CN107679468A (en) * | 2017-09-19 | 2018-02-09 | 浙江师范大学 | A kind of embedded computer vision detects fatigue driving method and device |
| CN107977607A (en) * | 2017-11-20 | 2018-05-01 | 安徽大学 | A kind of fatigue driving monitoring method based on machine vision |
| CN108363968A (en) * | 2018-01-31 | 2018-08-03 | 上海瀚所信息技术有限公司 | A kind of tired driver driving monitoring system and method based on key point extraction |
| CN108446600A (en) * | 2018-02-27 | 2018-08-24 | 上海汽车集团股份有限公司 | A kind of vehicle driver's fatigue monitoring early warning system and method |
-
2018
- 2018-09-19 CN CN201811093844.XA patent/CN109165630A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104361716A (en) * | 2014-10-31 | 2015-02-18 | 新疆宏开电子系统集成有限公司 | Method for detecting and reminding fatigue in real time |
| CN107679468A (en) * | 2017-09-19 | 2018-02-09 | 浙江师范大学 | A kind of embedded computer vision detects fatigue driving method and device |
| CN107563346A (en) * | 2017-09-20 | 2018-01-09 | 南京栎树交通互联科技有限公司 | One kind realizes that driver fatigue sentences method for distinguishing based on eye image processing |
| CN107977607A (en) * | 2017-11-20 | 2018-05-01 | 安徽大学 | A kind of fatigue driving monitoring method based on machine vision |
| CN108363968A (en) * | 2018-01-31 | 2018-08-03 | 上海瀚所信息技术有限公司 | A kind of tired driver driving monitoring system and method based on key point extraction |
| CN108446600A (en) * | 2018-02-27 | 2018-08-24 | 上海汽车集团股份有限公司 | A kind of vehicle driver's fatigue monitoring early warning system and method |
Non-Patent Citations (1)
| Title |
|---|
| 汪前进 等: "基于计算机视觉的危险驾驶预警系统研究与设计", 《长沙大学学报》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109871773A (en) * | 2019-01-21 | 2019-06-11 | 深圳市云眸科技有限公司 | Biopsy method, device and door access machine |
| CN109784313A (en) * | 2019-02-18 | 2019-05-21 | 上海骏聿数码科技有限公司 | A kind of blink detection method and device |
| CN109977820A (en) * | 2019-03-14 | 2019-07-05 | 重庆邮电大学 | A kind of fatigue driving determination method |
| CN113034851A (en) * | 2021-03-11 | 2021-06-25 | 中铁工程装备集团有限公司 | Tunnel boring machine driver fatigue driving monitoring device and method |
| CN113065804A (en) * | 2021-04-27 | 2021-07-02 | 山东交通学院 | Hazardous chemical substance road transportation risk assessment method and system |
| CN113198067A (en) * | 2021-06-03 | 2021-08-03 | 浙大城市学院 | Automatic medical care monitoring system |
| CN113844455A (en) * | 2021-08-27 | 2021-12-28 | 奇瑞汽车股份有限公司 | Method and device for braking a vehicle while the driver is tired |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109165630A (en) | A kind of fatigue monitoring method based on two-dimentional eye recognition | |
| CN108427503B (en) | Human eye tracking method and human eye tracking device | |
| CN108229362B (en) | Binocular face recognition living body detection method based on access control system | |
| CN101593425B (en) | Machine vision based fatigue driving monitoring method and system | |
| CN107292251B (en) | Driver fatigue detection method and system based on human eye state | |
| CN100592322C (en) | An automatic computer authentication method for photographic faces and living faces | |
| CN108764058B (en) | Double-camera face in-vivo detection method based on thermal imaging effect | |
| CN101523411B (en) | Eye opening detection system and method of detecting eye opening | |
| CN104123543B (en) | A kind of eye movement recognition methods based on recognition of face | |
| CN105224285A (en) | Eyes open and-shut mode pick-up unit and method | |
| CN108596087B (en) | Driving fatigue degree detection regression model based on double-network result | |
| CN105243386A (en) | Face living judgment method and system | |
| CN108549854A (en) | A kind of human face in-vivo detection method | |
| CN111062292B (en) | Fatigue driving detection device and method | |
| CN105913013A (en) | Binocular vision face recognition algorithm | |
| CN101872419A (en) | Method for detecting fatigue of automobile driver | |
| CN104881956A (en) | Fatigue driving early warning system | |
| US11682231B2 (en) | Living body detection method and device | |
| CN115171024A (en) | Face multi-feature fusion fatigue detection method and system based on video sequence | |
| CN108108651B (en) | Method and system for detecting driver non-attentive driving based on video face analysis | |
| CN114092985A (en) | Terminal control method, device, terminal and storage medium | |
| KR20190050551A (en) | Apparatus and method for recognizing body motion based on depth map information | |
| US20230237682A1 (en) | Method and system for binocular ranging for children | |
| CN112528767A (en) | Machine vision-based construction machinery operator fatigue operation detection system and method | |
| Zhang et al. | An approach of region of interest detection based on visual attention and gaze tracking |
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: 20190108 |