CN106295474A - The fatigue detection method of deck officer, system and server - Google Patents
The fatigue detection method of deck officer, system and server Download PDFInfo
- Publication number
- CN106295474A CN106295474A CN201510279711.1A CN201510279711A CN106295474A CN 106295474 A CN106295474 A CN 106295474A CN 201510279711 A CN201510279711 A CN 201510279711A CN 106295474 A CN106295474 A CN 106295474A
- Authority
- CN
- China
- Prior art keywords
- human eye
- fatigue
- eye area
- image information
- deck
- 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.)
- Granted
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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The present invention proposes the fatigue detection method of a kind of deck officer, system and server.Wherein, this fatigue detection method includes: receive the video flowing that wearable device gathers；Multiple frame of video in video flowing are converted to multiple image information；Obtain the human eye area in multiple image information；And the human eye area in multiple image informations is carried out analysis of fatigue to judge whether deck officer is in fatigue state, and analysis result is sent to wearable device.The fatigue detection method of the embodiment of the present invention, obtains the video image of deck officer by wearable device, can avoid the impact of various objective factor, ensures the quality of the video image gathered, thus improves the reliability of server fatigue detecting.Further, it is reminded by wearable device when deck officer is in fatigue state, thus substantially increase deck officer's safety when driving boats and ships, it is to avoid the generation of accident, it is ensured that the security of the lives and property of deck officer.
Description
Technical field
The present invention relates to shipping technical field, particularly relate to the fatigue detection method of a kind of deck officer, system and service
Device.
Background technology
In recent years, along with the fast development of China's shipping industry, see its comprehensive strength have been obtained for significantly on the whole and improve,
Safety, the reach of science condition are prepared.But, during shipping business fast development, yet suffer from a lot of problem,
Safety problem ratio is more prominent, and in the case of major safety risks cannot be effectively controlled, security incident happens occasionally, ship
The security of the lives and property of oceangoing ship driver receives threat greatly.
Compared with the tired identification technology of development more quickly vehicle driver, for deck officer tired identification technology also
It is in the budding stage.At present, the tired identification technology for vehicle driver mainly has, and such as, Volvo motor corporation releases
Driver alert's system assist driver to improve traffic safety, warn in time before driver enters sleep state；
The PERCLOS system researched and developed by Carnegie Mellon University can be by analyzing position and the aperture of driver's eyes, to driving
Member's fatigue state judges；FaceLAB system by monitoring driver head's attitude, eyes open and-shut mode, gaze-direction,
The characteristic parameters such as pupil diameter, monitor in real time to the fatigue state of driver；The AWAKE system of European Union can be passed through
Comprehensively monitoring to driving behavior, by utilizing the multiple sensors such as image, pressure, to driver eye's state, sight line
The information such as direction, steering wheel grip are monitored in real time.
But, for vehicle driver's fatigue identification technology, for the development of the tired identification technology of deck officer
Mainly affected by following several respects:
(1) the driving cabin area of boats and ships is relatively big, and deck officer generally to take the behavior sights such as prestige of leaning to one side in driving procedure
Examine aquatic environment.Therefore, range of activity when deck officer drives is relatively big, and existing based on the knowledge of vehicle driver's fatigue
Other technology is difficult to comprehensively, gathers boats and ships driver status information exactly.
(2) there is the features such as simple to operate, single in deck officer in driving procedure, and the speed of boats and ships is relatively slow in addition, makes
Obtain deck officer's fault-tolerance in driving procedure relatively strong, cause deck officer's meaning of regularized operation in driving procedure
Know more weak.
(3) environment of ship-handling is mainly affected by natural environment and ship environment two aspect.Due to offshore environment by dense fog,
The impact of multiple factor such as water level fluctuation, usual aquatic environment is more changeable than circumstance complication on road.Furthermore, the equipment in boats and ships is made an uproar
Sound, vibrations degree are more complicated, and many such environmental effects cause the work degree of deck officer and mental pressure to strengthen, pole
Easily cause the fatigue of deck officer.
Therefore, for the fatigue detecting technology for deck officer is compared to the fatigue detecting technology of vehicle driver relatively
Complexity, and need the many factors considered.Appearance but ship's speed is relatively slow for speed, to boats and ships fatigue driving
Wrong ability is relatively strong, therefore the highest to the requirement of real-time of boats and ships fatigue detecting.
Summary of the invention
It is contemplated that one of technical problem solved the most to a certain extent in correlation technique.
To this end, the first of the present invention purpose is to propose the fatigue detection method of a kind of deck officer, this fatigue detecting side
Method obtains the video image of deck officer by wearable device, can avoid the impact of various objective factor, ensures
The quality of the video image gathered, thus improve the reliability of server fatigue detecting.Further, deck officer
By wearable device, it is reminded when being in fatigue state, thus substantially increase deck officer and driving ship
Safety during oceangoing ship, it is to avoid the generation of accident, it is ensured that the security of the lives and property of deck officer.
Second object of the present invention is to propose the fatigue detecting system of a kind of deck officer.
Third object of the present invention is to propose a kind of server.
For reaching above-mentioned purpose, first aspect present invention embodiment proposes the fatigue detection method of a kind of deck officer, including
Following steps: receive the video flowing that wearable device gathers；Multiple frame of video in described video flowing are converted to multiple
Image information；Obtain the human eye area in the plurality of image information；And to the human eye in the plurality of image information
Region carries out analysis of fatigue to judge whether deck officer is in fatigue state, and analysis result is sent to the most described can
Wearable device.
The fatigue detection method of the deck officer of the embodiment of the present invention, obtains regarding of deck officer by wearable device
Frequently image, can avoid the impact of various objective factor, including photoenvironment, water level fluctuation environment, operating environment with
And the water surface visual field, deck officer front etc., front-end collection system based on wearable device, it is possible to gather and regard clearly
Frequently image, shakes, illumination is not enough etc. under mal-condition, it is also possible to ensure the quality of the video image gathered in boats and ships,
Thus improve the reliability of server fatigue detecting.Further, machine vision technique is fused in fatigue detection method,
By wearable device by video image send to server, by server, video image is processed, human eye fixed
Position and fatigue detecting, reminded it by wearable device when deck officer is in fatigue state, in order to alert
Show deck officer, thus substantially increase deck officer's safety when driving boats and ships, it is to avoid sending out of accident
Raw, it is ensured that the security of the lives and property of deck officer.
For reaching above-mentioned purpose, second aspect present invention embodiment proposes the fatigue detecting system of a kind of deck officer, including
Server and wearable device, wherein, described wearable device is used for gathering video flowing, and by described video stream
To described server, and receive the analysis result that described server sends；Described server is used for receiving described can be worn
Wear the video flowing that equipment gathers, and the multiple frame of video in described video flowing are converted to multiple image information, and obtain
Take the human eye area in the plurality of image information, and the human eye area in the plurality of image information is carried out tired point
Analyse to judge whether deck officer is in fatigue state, and analysis result is sent to described wearable device.
The fatigue detecting system of the deck officer of the embodiment of the present invention, obtains regarding of deck officer by wearable device
Frequently image, can avoid the impact of various objective factor, including photoenvironment, water level fluctuation environment, operating environment with
And the water surface visual field, deck officer front etc., front-end collection system based on wearable device, it is possible to gather and regard clearly
Frequently image, shakes, illumination is not enough etc. under mal-condition, it is also possible to ensure the quality of the video image gathered in boats and ships,
Thus improve the reliability of server fatigue detecting.Further, machine vision technique is fused in fatigue detection method,
By wearable device by video image send to server, by server, video image is processed, human eye fixed
Position and fatigue detecting, reminded it by wearable device when deck officer is in fatigue state, in order to alert
Show deck officer, thus substantially increase deck officer's safety when driving boats and ships, it is to avoid sending out of accident
Raw, it is ensured that the security of the lives and property of deck officer.
For reaching above-mentioned purpose, first aspect present invention embodiment proposes a kind of server, including: receiver module, it is used for connecing
Receive the video flowing that wearable device gathers；Modular converter, for being converted to many by the multiple frame of video in described video flowing
Individual image information；Acquisition module, for obtaining the human eye area in the plurality of image information；And analysis module,
For the human eye area in the plurality of image information being carried out analysis of fatigue to judge whether deck officer is in fatigue
State, and analysis result is sent to described wearable device.
The server of the embodiment of the present invention, video image is processed, human eye location and fatigue detecting, in ship-handling
By wearable device, it is reminded when member is in fatigue state, in order to warning deck officer, thus significantly carry
High deck officer's safety when driving boats and ships, it is to avoid the generation of accident, it is ensured that the life of deck officer
Life property safety.
Aspect and advantage that the present invention adds will part be given in the following description, and part will become bright from the following description
Aobvious, or recognized by the practice of the present invention.
Accompanying drawing explanation
The present invention above-mentioned and/or that add aspect and advantage will be apparent from from the following description of the accompanying drawings of embodiments
With easy to understand, wherein:
Fig. 1 is the flow chart of the fatigue detection method of the deck officer of one embodiment of the invention；
Fig. 2 is the flow chart of the fatigue detection method of the deck officer of one specific embodiment of the present invention；
Fig. 3 is the schematic diagram of Haar feature in the embodiment of the present invention；
Fig. 4 is the schematic diagram of integrogram in the embodiment of the present invention；
Fig. 5 is the structural representation of the fatigue detecting system of the deck officer of one embodiment of the invention；And
Fig. 6 is the structural representation of the server of one embodiment of the invention.
Detailed description of the invention
Embodiments of the invention are described below in detail, and the example of described embodiment is shown in the drawings, the most identical
Or similar label represents same or similar element or has the element of same or like function.Retouch below with reference to accompanying drawing
The embodiment stated is exemplary, it is intended to is used for explaining the present invention, and is not considered as limiting the invention.
Additionally, term " first ", " second " are only used for describing purpose, and it is not intended that instruction or hint relative importance or
The implicit quantity indicating indicated technical characteristic.Thus, define " first ", the feature of " second " can be expressed or hidden
One or more this feature are included containing ground.In describing the invention, " multiple " are meant that two or more,
Unless otherwise expressly limited specifically.
In flow chart or at this, any process described otherwise above or method description are construed as, and represent and include one
Or the module of code, fragment or the part of the executable instruction of the more step for realizing specific logical function or process,
And the scope of the preferred embodiment of the present invention includes other realization, wherein can not press order that is shown or that discuss,
Including according to involved function by basic mode simultaneously or in the opposite order, performing function, this should be by the present invention's
Embodiment person of ordinary skill in the field understood.
Fig. 1 is the flow chart of the fatigue detection method of the deck officer of one embodiment of the invention, and Fig. 2 is the present invention
The flow chart of the fatigue detection method of the deck officer of one specific embodiment.
As depicted in figs. 1 and 2, the fatigue detection method of deck officer includes:
S101, receives the video flowing that wearable device gathers.
In one embodiment of the invention, wearable device can be glasses.Specifically, the present invention based on can
The fatigue detection method of the deck officer of wearable device is to use machine vision, sets based on Raspberry Pi B+ hardware
Standby platform is developed, and the equipment of front-end collection video flowing can use the form of glasses, is provided with RPi on glasses
Camera infrared camera.The mode using glasses can be prevented effectively from because of crewman's range of activity, driving habits and boats and ships
The impact of the factors such as environment.It is to say, can directly be obtained the eye image of deck officer by wearable glasses,
It is possible not only to avoid the impact of various objective factor, it is also possible to improve the quality of the eye image collected, for follow-up
Human eye location and fatigue detecting provide the image information of high-quality, reduce picture noise.Additionally, employing infrared camera can
To meet the demand collecting clear eye image under conditions of illumination at night deficiency.
Furthermore, wearable device, after collecting video image, first can carry out pretreatment to video image,
Such as, video image is compressed processing or setting the frame per second etc. of video image, thus can meet video
On the premise of the prescription of image, improve video image transfer rate.Then, wearable device by video image with
The mode of video flowing is sent on the image processing server of rear end, and wherein, server and wearable device can pass through
Wireless network communicates, and the mode of communication may include but be not limited to the one in Wifi, infrared, bluetooth, 3G network.
Video flowing, after receiving the video flowing that wearable device gathers, is backed up by server on the server.
Multiple frame of video in video flowing are converted to multiple image information by S102.
Specifically, server gets multiple frame of video from the video flowing received, and according to server is preset
Multiple frame of video are converted to image information by threshold value.Such as, wearable device gathers the video image of continuous 10 minutes,
Owing to normal person's wink time is about the 0.2-0.4 second, and speed of blinking under fatigue state is universal the slowest, be one by
The process gradually closed one's eyes, eyes typically at least need the time of about 1 second from opening up into Guan Bi.Therefore, server is permissible
Video frame rate is set as, and 10 (i.e. FPS=10) just can meet the captured in real time of human eye state, the most permissible
Produce 6000 sample image information.
It should be appreciated that server needs in advance the eye information of deck officer to be carried out feature learning, to improve
Human eye detection and the accuracy of fatigue detecting.
S103, obtains the human eye area in multiple image information.
In one embodiment of the invention, before the human eye area in server obtains multiple image informations, also may be used
So that multiple image informations to carry out pretreatment, thus server can obtain the image information of good quality.Wherein, in advance
Process includes one or more in image denoising process, equalization processing, contrast process.
Specifically, the human eye area in multiple image informations is positioned by server, i.e. positions position of human eye, with from
Image information obtains the image that part comprises human eye area accurately, removes information useless in image information.
In one embodiment of the invention, the human eye area during server obtains multiple image informations specifically includes:
S1031, server carries out human eye location according to Adaboost algorithm based on Haar feature to multiple image informations,
And obtain the first human eye area.
S1032, server carries out binary conversion treatment to multiple image informations, and according to Adaboost based on Haar feature
Algorithm carries out human eye location to the image information after binary conversion treatment, to obtain the second human eye area.
S1033, server judges whether the first human eye area and the second human eye area mate, and when coupling by the first
Eye region and/or the second human eye area are as the human eye area in multiple image informations.
Specifically, server combines learning outcome to deck officer's eye feature according to image information, utilize based on
Human eye is carried out just positioning by the Adaboost algorithm of Haar feature, it is thus achieved that the first human eye area.Then, server by utilizing
Image information is analyzed and processes by image processing techniques, it is thus achieved that the bianry image of this image information, and according to generation
Bianry image utilize Adaboost algorithm based on Haar feature human eye is carried out location again, it is thus achieved that the second people
Eye region.Then, the first human eye area and the second human eye area are mated by server, if the first human eye area
Image collection comprise the image collection of the second human eye area, then judge human eye detection success, otherwise delete this image letter
Breath.
Specifically, server first carries out human eye feature based on Haar extraction to image information.Wherein, believe at image
In breath, human eye feature can be expressed as the information such as coordinate, distance, color, brightness, shape.It is special that Haar feature belongs to matrix
Levy, therefore can by its abstract be with point, line.The simple graph that the basic sets such as face are elementary composition.Wherein, such as Fig. 3
Shown in, Haar feature can be divided three classes: edge feature, profile and ring characteristics.The basic thought of Haar feature
It is exactly first by rectangle frame piecemeal, a kind of feature analysis side analyzed that the gray-scale pixels of piecemeal is combined with edge feature
Method.In the target image can be by abstract for the rectangular image area of ad-hoc location for Haar feature, can be by mesh by the method
Mark area image feature carries out quantification treatment.In image the grey scale pixel value of white portion with deduct black region pixel ash
Angle value sum, obtained numerical value is exactly the eigenvalue of institute overlay area.
Calculate by the way of using integrogram, feature calculation speed can be improved.Integrogram is that one can describe overall situation letter
The matrix method for expressing of breath, it is defined as:
Wherein, (x is y) original image (x, y) integral image at place, g (x ', y ') is at (x, y) original image at place to f.Therefore, as
Shown in Fig. 4, (x, y) at some integral image equal in this upper left side gray area all pixel values sum.
And then, the position of human eye in image information is identified by server according to Adaboost algorithm.For capture
For the image of 24*24 pixel, its Haar feature is in the number of images match is the most up to ten thousand, and wherein only exists few
Number available feature.By using Adaboost algorithm to realize quick human eye detection in the present invention, its basic thought is profit
Train Weak Classifier by a large amount of training sets, finally constitute strong classifier by algorithm superposition.
If human eye area image has k feature, then can be expressed as fj(xi), wherein, 1≤j≤k, xiIt is expressed as i-th
Individual sample image.The feature set of the most each image is represented by { f1(xi),f2(xi),f3(xi),…fj(xi),…fk(xi), its
In, the corresponding Weak Classifier of each feature.
Server is by a Weak Classifier hjX the composition of () comprises feature fj(x), threshold θjWith symbol pjThree parts, wherein,
One corresponding Weak Classifier of feature, classification thresholds is an eigenvalue to all Classification of Matrixes, classifier
It number it is then the expression symbol that has positive negative direction.The Weak Classifier of jth feature is expressed as by server:
Wherein, hjX () is the value of Weak Classifier, θjFor threshold value, pjFor controlling sign of inequality direction, value is+1 or-1, fj(x) be
Eigenvalue.
Based on Adaboost algorithm, to known n training sample (x1,y1),(x2,y2),…,(xn,yn) carry out following steps fortune
Calculate, wherein yi={ the true and false of 0,1} correspondence sample.
(1) n training sample, wherein m human eye sample are taken, l non-human eye sample, it is expressed as
(x1,y1),(x2,y2),…,(xn,yn), wherein, yi=0, yi=1 the most corresponding human eye sample and non-human eye sample.
(2) initialization error weight, for yiThe sample of=0,For yiThe sample of=1,
(3) initializing t=1, wherein t≤T, T is training sample grader number.
(4) weight is normalized to
(5) each feature f is trained a Weak Classifier h (x, f, p, θ), calculate the weighting (q of the Weak Classifier of its correspondencei)
Error rate εf=∑ | hj(xj)-qi|, and select error εfMinimum grader ht, and update weight
Wherein, ei=0 represents by correct classification, ei=1 represents by the classification of mistake,
(6) another t=t+1, repeats step (4), until t ＞ T.
(7) finally obtaining strong classifier is:
S104, carries out analysis of fatigue to judge whether deck officer is in fatigue to the human eye area in multiple image informations
State, and analysis result is sent to wearable device.
In one embodiment of the invention, server carries out analysis of fatigue to judge that deck officer is to human eye area
The no fatigue state that is in specifically includes: server calculates the human eye district in multiple image informations according to PERCLOS algorithm
The PERCLOS value in territory, and the threshold value that PERCLOS value and degree of fatigue differentiate is compared, Yi Ji
Judge that deck officer is in fatigue state during the threshold value that PERCLOS value differentiates more than or equal to degree of fatigue.Wherein, clothes
Business device is according to the below equation described PERCLOS value of calculating:
Wherein, N is human eye area sampling sum in continuous time,
Specifically, due to the state of eyes and the degree of fatigue of deck officer, there is the highest dependency, PERCLOS
Algorithm (i.e. Percentage of Eyelid Closure Over the Pupil Over time) is the opening and closing by analyzing eyes
A kind of method that situation detection is tired.Wherein, P80 standard is the highest with the dependency of degree of fatigue, " the gold being well recognized as
Judge " standard.
After the human eye area in image information is positioned by server, by image processing techniques to human eye area
The opening and closing degree of middle human eye judges.It is to say, after server calculates P (i), server can be by P (i)
Threshold value T differentiated with degree of fatigue compares, and wherein, threshold value T is to combine ship-handling environment according to experiment
The ideal numerical parameter obtained after closing evaluation, if P (i) >=T, then judges that human eye is in closure state, i.e. sentences
Disconnected deck officer is in fatigue state.If P (i) is ＜ T, then judges that human eye is in open configuration, i.e. judge that boats and ships are driven
The person of sailing is not in fatigue state.Then, server will determine that whether deck officer is in the analysis result of fatigue state
Send to wearable device.
In one embodiment of the invention, at server, analysis result is sent after wearable device, if clothes
Business device judges when deck officer is in fatigue state, and wearable device carries out alarm.Wherein, alarm bag
Include in light prompt, voice message and vibration prompt one or more.
The fatigue detection method of the deck officer of the embodiment of the present invention, obtains regarding of deck officer by wearable device
Frequently image, can avoid the impact of various objective factor, including photoenvironment, water level fluctuation environment, operating environment with
And the water surface visual field, deck officer front etc., front-end collection system based on wearable device, it is possible to gather and regard clearly
Frequently image, shakes, illumination is not enough etc. under mal-condition, it is also possible to ensure the quality of the video image gathered in boats and ships,
Thus improve the reliability of server fatigue detecting.
Further, machine vision technique is fused in fatigue detection method, by wearable device, video image is sent
To server, by server, video image is processed, human eye location and fatigue detecting, at deck officer
By wearable device, it is reminded when fatigue state, in order to warning deck officer, thus substantially increase
Deck officer's safety when driving boats and ships, it is to avoid the generation of accident, it is ensured that the lives and properties of deck officer
Safety.
In order to realize above-described embodiment, the present invention also proposes the fatigue detecting system of a kind of deck officer.
Fig. 5 is the structural representation of the fatigue detecting system of the deck officer of one embodiment of the invention, such as Fig. 5 institute
Showing, the fatigue detecting system of deck officer includes server 10 and wearable device 20.
Specifically, wearable device 20 is used for gathering video flowing, and by video stream to server 10, and receive
The analysis result that server sends.Wherein, wearable device 20 can be glasses.Wearable device 20 is collecting
After video image, first video image can be carried out pretreatment, such as, be compressed video image processing or
Set the frame per second etc. of video image, thus can improve video figure on the premise of meeting the prescription of video image
As transfer rate.Then, video image is sent to the image procossing of rear end in the way of video flowing by wearable device 20
On server 10, wherein, server 10 and wearable device 20 can be communicated by wireless network, communication
Mode may include but be not limited to the one in Wifi, infrared, bluetooth, 3G network.
Server 10 is for receiving the video flowing that wearable device 20 gathers, and the multiple frame of video in video flowing is turned
It is changed to multiple image information, and obtains the human eye area in multiple image information, and to the people in multiple image informations
Eye region carries out analysis of fatigue to judge whether deck officer is in fatigue state, and analysis result sent extremely can
Wearable device 20.Specifically, server 10, after receiving the video flowing that wearable device 20 gathers, will regard
Frequency stream backs up on server 10.Then, server 10 gets multiple video from the video flowing received
Frame, and according to the threshold value preset in server 10, multiple frame of video are converted to image information.Such as, wearable device
20 video images gathering continuous 10 minutes, owing to normal person's wink time is about the 0.2-0.4 second, and at tired shape
Speed of blinking under state is universal relatively slow, is a process gradually closed one's eyes, and eyes typically at least need 1 from opening up into Guan Bi
Time about Miao.Therefore, video frame rate can be set as that 10 (i.e. FPS=10) just can meet by server 10
The captured in real time of human eye state, can produce 6000 sample image information under this condition.
Wherein, before server 10 is additionally operable to the human eye area in obtaining multiple image informations, to multiple image informations
Carry out pretreatment, thus server 10 can obtain the image information of good quality.Wherein, pretreatment includes that image goes
Make an uproar process, equalization processing, contrast process in one or more.
Then, the human eye area in multiple image informations is positioned by server 10, i.e. positions position of human eye, with from
Image information obtains the image that part comprises human eye area accurately, removes information useless in image information.
In one embodiment of the invention, server 10 is specifically for calculating according to Adaboost based on Haar feature
Method carries out human eye location to multiple image informations, and obtains the first human eye area, and multiple image informations are carried out two
Value processes, and according to Adaboost algorithm based on Haar feature, the image information after binary conversion treatment is carried out human eye calmly
Position, to obtain the second human eye area, and judge whether the first human eye area and the second human eye area mate, and
Timing using the first human eye area or the second human eye area as the human eye area in multiple image informations.Specifically, service
Device 10 combines the learning outcome to deck officer's eye feature according to image information, utilizes based on Haar feature
Human eye is carried out just positioning by Adaboost algorithm, it is thus achieved that the first human eye area.Then, server 10 utilizes image procossing
Image information is analyzed and processes by technology, it is thus achieved that the bianry image of this image information, and according to the binary map generated
As utilizing Adaboost algorithm based on Haar feature that human eye is carried out location again, it is thus achieved that the second human eye area.
Then, the first human eye area and the second human eye area are mated by server 10, if the image of the first human eye area
Set comprises the image collection of the second human eye area, then judge human eye detection success, otherwise delete this image information.
Furthermore, server 10 first carries out human eye feature based on Haar extraction to image information.Wherein, exist
In image information, human eye feature can be expressed as the information such as coordinate, distance, color, brightness, shape.Haar feature belongs to
Matrix character, therefore can by its abstract be with point, line.The simple graph that the basic sets such as face are elementary composition.Wherein,
As it is shown on figure 3, Haar feature can be divided three classes: edge feature, profile and ring characteristics.The base of Haar feature
This thought is exactly first by rectangle frame piecemeal, a kind of feature analyzed that the gray-scale pixels of piecemeal combined with edge feature
Analysis method.The method can be passed through in the target image by abstract for the rectangular image area of ad-hoc location for Haar feature
Target area image feature can be carried out quantification treatment.In image the grey scale pixel value of white portion with deduct black region
Grey scale pixel value sum, obtained numerical value is exactly the eigenvalue of institute overlay area.
Server 10 calculates by the way of using integrogram, can improve feature calculation speed.Integrogram is that one can be retouched
Stating the matrix method for expressing of global information, it is defined as:
Wherein, (x is y) original image (x, y) integral image at place, g (x ', y ') is at (x, y) original image at place to f.Therefore, as
Shown in Fig. 4, (x, y) at some integral image equal in this upper left side gray area all pixel values sum.
And then, the position of human eye in image information is identified by server 10 according to Adaboost algorithm.For capture
24*24 pixel image for, its Haar feature is in the number of images match is the most up to ten thousand, and wherein only exists
Minority available feature.By using Adaboost algorithm to realize quick human eye detection in the present invention, its basic thought is
Utilize a large amount of training set to train Weak Classifier, finally constitute strong classifier by algorithm superposition.
If human eye area image has k feature, then can be expressed as fj(xi), wherein, 1≤j≤k, xiIt is expressed as i-th
Individual sample image.The feature set of the most each image is represented by { f1(xi),f2(xi),f3(xi),…fj(xi),…fk(xi), its
In, the corresponding Weak Classifier of each feature.
Server 10 is by a Weak Classifier hjX the composition of () comprises feature fj(x), threshold θjWith symbol pjThree parts, wherein,
One corresponding Weak Classifier of feature, classification thresholds is an eigenvalue to all Classification of Matrixes, classifier
It number it is then the expression symbol that has positive negative direction.The Weak Classifier of jth feature is expressed as by server 10:
Wherein, hjX () is the value of Weak Classifier, θjFor threshold value, pjFor controlling sign of inequality direction, value is+1 or-1, fj(x) be
Eigenvalue.
Based on Adaboost algorithm, to known n training sample (x1,y1),(x2,y2),…,(xn,yn) walk as follows
Rapid computing, wherein yi={ the true and false of 0,1} correspondence sample.
(1) n training sample, wherein m human eye sample are taken, l non-human eye sample, it is expressed as
(x1,y1),(x2,y2),…,(xn,yn), wherein, yi=0, yi=1 the most corresponding human eye sample and non-human eye sample.
(2) initialization error weight, for yiThe sample of=0,For yiThe sample of=1,
(3) initializing t=1, wherein t≤T, T is training sample grader number.
(4) weight is normalized to
(5) each feature f is trained a Weak Classifier h (x, f, p, θ), calculate the weighting (q of the Weak Classifier of its correspondencei)
Error rate εf=∑ | hj(xj)-qi|, and select error εfMinimum grader ht, and update weight
Wherein, ei=0 represents by correct classification, ei=1 represents by the classification of mistake,
(6) another t=t+1, repeats step (4), until t ＞ T.
(7) finally obtaining strong classifier is:
In one embodiment of the invention, server 10 is specifically for calculating multiple images according to PERCLOS algorithm
The PERCLOS value of the human eye area in information, and the threshold value that PERCLOS value and degree of fatigue differentiate is compared,
And judge that deck officer is in fatigue state when the threshold value that PERCLOS value differentiates more than or equal to degree of fatigue.
Wherein, server 10 is according to the below equation described PERCLOS value of calculating:Its
In, N is human eye area sampling sum in continuous time,
Owing to the state of eyes and the degree of fatigue of deck officer have the highest dependency, PERCLOS algorithm is by dividing
A kind of method of the opening and closing situation detection fatigue of analysis eyes.Wherein, P80 standard is the highest with the dependency of degree of fatigue, is
" gold judgement " standard generally acknowledged.
After the human eye area in image information is positioned by server 10, by image processing techniques to human eye district
In territory, the opening and closing degree of human eye judges.It is to say, after server 10 calculates P (i), server 10
Can threshold value T that P (i) and degree of fatigue differentiate be compared, wherein, threshold value T is to ship-handling ring according to experiment
The ideal numerical parameter that border obtains after carrying out Comprehensive Assessment, if P (i) >=T, then judges that human eye is in closed form
State, i.e. judges that deck officer is in fatigue state.If P (i) is ＜ T, then judge that human eye is in open configuration, i.e. sentences
Disconnected deck officer is not in fatigue state.Then, server 10 will determine that whether deck officer is in fatigue state
Analysis result send to wearable device 20.
In one embodiment of the invention, wearable device 20 is additionally operable to when server 10 judges at deck officer
When fatigue state, carry out alarm.Wherein, alarm includes light prompt, voice message and vibration prompt
In one or more.
The fatigue detecting system of the deck officer of the embodiment of the present invention, obtains regarding of deck officer by wearable device
Frequently image, can avoid the impact of various objective factor, including photoenvironment, water level fluctuation environment, operating environment with
And the water surface visual field, deck officer front etc., front-end collection system based on wearable device, it is possible to gather and regard clearly
Frequently image, shakes, illumination is not enough etc. under mal-condition, it is also possible to ensure the quality of the video image gathered in boats and ships,
Thus improve the reliability of server fatigue detecting.
Further, machine vision technique is fused in fatigue detection method, by wearable device, video image is sent
To server, by server, video image is processed, human eye location and fatigue detecting, at deck officer
By wearable device, it is reminded when fatigue state, in order to warning deck officer, thus substantially increase
Deck officer's safety when driving boats and ships, it is to avoid the generation of accident, it is ensured that the lives and properties of deck officer
Safety.
In order to realize above-described embodiment, the present invention also proposes a kind of server.
Fig. 6 is the structural representation of the server of one embodiment of the invention, and as shown in Figure 6, server includes receiving
Module 110, modular converter 120, acquisition module 130, analysis module 140 and pretreatment module 150, wherein, obtain
Delivery block 130 includes the first acquiring unit 131, second acquisition unit 132 and judging unit 133, analyzes module 140
Including computing unit 141, comparing unit 142 and judging unit 143.
Specifically, receiver module 110 is for receiving the video flowing that wearable device gathers.
Modular converter 120 is for being converted to multiple image information by the multiple frame of video in video flowing.Specifically, conversion
Module 120 gets multiple frame of video from the video flowing that receiver module 110 receives, and will according to default threshold value
Multiple frame of video are converted to image information.Such as, wearable device gathers the video image of continuous 10 minutes, due to just
Ordinary person's wink time is about the 0.2-0.4 second, and speed of blinking under fatigue state is universal relatively slow, is one and gradually closes one's eyes
Process, eyes typically at least need the time of about 1 second from opening up into Guan Bi.Therefore, modular converter 120 is permissible
Video frame rate is set as, and 10 (i.e. FPS=10) just can meet the captured in real time of human eye state, the most permissible
Produce 6000 sample image information.
Acquisition module 130 is for obtaining the human eye area in multiple image information.
In one embodiment of the invention, server also includes pretreatment module 150, and pretreatment module 150 is for right
Multiple image informations carry out pretreatment, and wherein, pretreatment includes at image denoising process, equalization processing, contrast
One or more in reason.
In one embodiment of the invention, acquisition module 130 includes the first acquiring unit 131, second acquisition unit
132 and judging unit 133.Wherein, the first acquiring unit 131 is for calculating according to Adaboost based on Haar feature
Method carries out human eye location to multiple image informations, and obtains the first human eye area.Second acquisition unit 132 is for many
Individual image information carries out binary conversion treatment, and according to Adaboost algorithm based on Haar feature to the figure after binary conversion treatment
As information carries out human eye location, to obtain the second human eye area.Judging unit 133 for judge the first human eye area with
Whether the second human eye area mates, and when coupling using the first human eye area and/or the second human eye area as multiple images
Human eye area in information.Specifically, the first acquiring unit 131 combines deck officer's eye according to image information
The learning outcome of feature, utilizes Adaboost algorithm based on Haar feature to carry out human eye just and positions, it is thus achieved that be the first
Eye region.Then, second acquisition unit 132 utilizes image processing techniques be analyzed image information and process, and obtains
Obtain the bianry image of this image information, and utilize Adaboost algorithm based on Haar feature according to the bianry image generated
Human eye is carried out location again, it is thus achieved that the second human eye area.Next, it is determined that unit 133 is by the first human eye area
Mate with the second human eye area, if the image collection of the first human eye area comprises the image set of the second human eye area
Close, then judge human eye detection success, otherwise delete this image information.
Specifically, image information is first carried out based on Haar by the first acquiring unit 131 and second acquisition unit 132
Human eye feature extracts.Wherein, in image information, human eye feature can be expressed as coordinate, distance, color, brightness, shape
The information such as shape.Haar feature belongs to matrix character, therefore can by its abstract be with point, line.The basic set units such as face
The simple graph of element composition.Wherein, as it is shown on figure 3, Haar feature can be divided three classes: edge feature, profile
And ring characteristics.The basic thought of Haar feature is exactly first by rectangle frame piecemeal, then by the gray-scale pixels of piecemeal and edge
Feature combine analyze a kind of characteristic analysis method.The rectangular image area of ad-hoc location can be taken out in the target image
As for Haar feature, target area image feature being carried out quantification treatment by the method.White portion in image
Grey scale pixel value with deduct black region grey scale pixel value sum, obtained numerical value is exactly the eigenvalue of institute overlay area.
Calculate by the way of using integrogram, feature calculation speed can be improved.Integrogram is that one can describe overall situation letter
The matrix method for expressing of breath, it is defined as:
Wherein, (x is y) original image (x, y) integral image at place, g (x ', y ') is at (x, y) original image at place to f.Therefore, as
Shown in Fig. 4, (x, y) at some integral image equal in this upper left side gray area all pixel values sum.
And then, the first acquiring unit 131 and second acquisition unit 132 according to Adaboost algorithm in image information
Position of human eye is identified.For capture 24*24 pixel image for, its Haar feature images match
During number is the most up to ten thousand, and wherein only exist minority available feature.By using Adaboost algorithm to realize soon in the present invention
The human eye detection of speed, its basic thought is to utilize a large amount of training set to train Weak Classifier, is finally constituted by algorithm superposition
Strong classifier.
If human eye area image has k feature, then can be expressed as fj(xi), wherein, 1≤j≤k, xiIt is expressed as i-th
Individual sample image.The feature set of the most each image is represented by { f1(xi),f2(xi),f3(xi),…fj(xi),…fk(xi), its
In, the corresponding Weak Classifier of each feature.
First acquiring unit 131 and second acquisition unit 132 are by a Weak Classifier hjX the composition of () comprises feature fj(x),
Threshold θjWith symbol pjThree parts, wherein, the corresponding Weak Classifier of feature, classification thresholds be one to all squares
Battle array carries out the eigenvalue classified, and class symbol is then the symbol that an expression has positive negative direction.Server is special by jth
The Weak Classifier levied is expressed as:
Wherein, hjX () is the value of Weak Classifier, θjFor threshold value, pjFor controlling sign of inequality direction, value is+1 or-1, fj(x) be
Eigenvalue.
Based on Adaboost algorithm, to known n training sample (x1,y1),(x2,y2),…,(xn,yn) carry out following steps fortune
Calculate, wherein yi={ the true and false of 0,1} correspondence sample.
(1) n training sample, wherein m human eye sample are taken, l non-human eye sample, it is expressed as
(x1,y1),(x2,y2),…,(xn,yn), wherein, yi=0, yi=1 the most corresponding human eye sample and non-human eye sample.
(2) initialization error weight, for yiThe sample of=0,For yiThe sample of=1,
(3) initializing t=1, wherein t≤T, T is training sample grader number.
(4) weight is normalized to
(5) each feature f is trained a Weak Classifier h (x, f, p, θ), calculate the weighting (q of the Weak Classifier of its correspondencei)
Error rate εf=∑ | hj(xj)-qi|, and select error εfMinimum grader ht, and update weight
Wherein, ei=0 represents by correct classification, ei=1 represents by the classification of mistake,
(6) another t=t+1, repeats step (4), until t ＞ T.
(7) finally obtaining strong classifier is:
Analyze module 140 for the human eye area in multiple image informations is carried out analysis of fatigue to judge deck officer
Whether it is in fatigue state, and analysis result is sent to wearable device.
In one embodiment of the invention, analyze module 140 include computing unit 141, comparing unit 142 and judge
Unit 143.Wherein, computing unit 141 for calculating the human eye district in multiple image informations according to PERCLOS algorithm
The PERCLOS value in territory.Wherein, computing unit 141 is according to below equation calculating PERCLOS value:Wherein, N is human eye area sampling sum in continuous time, Comparing unit 142 is for entering PERCLOS value with the threshold value that degree of fatigue differentiates
Row compares.Judging unit 143 is when the threshold value differentiated more than or equal to degree of fatigue in PERCLOS value, it is judged that ship
Oceangoing ship driver is in fatigue state.
The server of the embodiment of the present invention, video image is processed, human eye location and fatigue detecting, in ship-handling
By wearable device, it is reminded when member is in fatigue state, in order to warning deck officer, thus significantly carry
High deck officer's safety when driving boats and ships, it is to avoid the generation of accident, it is ensured that the life of deck officer
Life property safety.
Should be appreciated that each several part of the present invention can realize by hardware, software, firmware or combinations thereof.In above-mentioned reality
Execute in mode, software that multiple steps or method in memory and can be performed by suitable instruction execution system with storage or
Firmware realizes.Such as, if realized with hardware, with the most the same, available well known in the art under
Any one or their combination in row technology realize: have the logic gates for data signal realizes logic function
Discrete logic, there is the special IC of suitable combination logic gate circuit, programmable gate array (PGA), existing
Field programmable gate array (FPGA) etc..
In the present invention, unless otherwise clearly defined and limited, term " install ", " being connected ", " connection ", etc. term should do
Broadly understood, connect for example, it may be fixing, it is also possible to be to removably connect, or integral；Can be to be mechanically connected,
It can also be electrical connection；Can be to be joined directly together, it is also possible to be indirectly connected to by intermediary, can be two element internals
Connection or the interaction relationship of two elements, unless otherwise clear and definite restriction.For those of ordinary skill in the art
Speech, can understand above-mentioned term concrete meaning in the present invention as the case may be.
In the description of this specification, reference term " embodiment ", " some embodiments ", " example ", " concrete example ",
Or specific features, structure, material or the feature bag that the description of " some examples " etc. means to combine this embodiment or example describes
It is contained at least one embodiment or the example of the present invention.In this manual, to the schematic representation of above-mentioned term necessarily
It is directed to identical embodiment or example.And, the specific features of description, structure, material or feature can be arbitrary
Individual or multiple embodiment or example combine in an appropriate manner.Additionally, in the case of the most conflicting, the skill of this area
The feature of the different embodiments described in this specification or example and different embodiment or example can be combined by art personnel
And combination.
Although above it has been shown and described that embodiments of the invention, it is to be understood that above-described embodiment is exemplary,
Being not considered as limiting the invention, those of ordinary skill in the art within the scope of the invention can be to above-described embodiment
It is changed, revises, replaces and modification.
Claims (21)
1. the fatigue detection method of a deck officer, it is characterised in that comprise the following steps:
Receive the video flowing that wearable device gathers；
Multiple frame of video in described video flowing are converted to multiple image information；
Obtain the human eye area in the plurality of image information；And
Human eye area in the plurality of image information is carried out analysis of fatigue to judge whether deck officer is in fatigue
State, and analysis result is sent to described wearable device.
2. the fatigue detection method of deck officer as claimed in claim 1, it is characterised in that obtain the plurality of
Human eye area in image information specifically includes:
According to Adaboost algorithm based on Haar feature, the plurality of image information carried out human eye location, and obtain
One human eye area；
The plurality of image information is carried out binary conversion treatment, and according to Adaboost algorithm based on Haar feature to two-value
Image information after change processes carries out human eye location, to obtain the second human eye area；
Judge whether described first human eye area and described second human eye area mate, and when coupling by described the first
Eye region and/or described second human eye area are as the human eye area in the plurality of image information.
3. the fatigue detection method of deck officer as claimed in claim 1, it is characterised in that to described human eye district
Territory carries out analysis of fatigue to judge whether deck officer is in fatigue state and specifically includes:
According to the PERCLOS value of the human eye area in the PERCLOS algorithm the plurality of image information of calculating, and by institute
State the threshold value that PERCLOS value and degree of fatigue differentiate to compare, and in described PERCLOS value more than or equal to institute
Judge that described deck officer is in fatigue state when stating the threshold value of degree of fatigue differentiation.
4. the fatigue detection method of deck officer as claimed in claim 3, it is characterised in that according to below equation
Calculate described PERCLOS value:
Wherein, N is human eye area sampling sum in continuous time,
5. the fatigue detection method of deck officer as claimed in claim 1, it is characterised in that by analysis result
Send after described wearable device, also include:
When judging that described deck officer is in fatigue state, described wearable device carries out alarm.
6. the fatigue detection method of deck officer as claimed in claim 5, it is characterised in that described alarm
Including one or more in light prompt, voice message and vibration prompt.
7. the fatigue detection method of deck officer as claimed in claim 1, it is characterised in that described many obtaining
Before human eye area in individual image information, also include:
The plurality of image information carries out pretreatment, and wherein, described pretreatment includes image denoising process, equalization
Process, contrast process in one or more.
8. the fatigue detection method of the deck officer as described in any one of claim 1-7, it is characterised in that described
Wearable device is glasses.
9. the fatigue detecting system of a deck officer, it is characterised in that including: server and wearable device,
Wherein,
Described wearable device is used for gathering video flowing, and by described video stream to described server, and receive
The analysis result that described server sends；And
Described server is for receiving the video flowing that described wearable device gathers, and multiple in described video flowing is regarded
Frequently frame is converted to multiple image information, and obtains the human eye area in the plurality of image information, and to the plurality of
Human eye area in image information carries out analysis of fatigue to judge whether deck officer is in fatigue state, and general point
Analysis result sends to described wearable device.
10. the fatigue detecting system of deck officer as claimed in claim 9, it is characterised in that described server
Specifically for:
According to Adaboost algorithm based on Haar feature, the plurality of image information carried out human eye location, and obtain
One human eye area, and the plurality of image information is carried out binary conversion treatment, and according to Adaboost based on Haar feature
Algorithm carries out human eye location to the image information after binary conversion treatment, to obtain the second human eye area, and judges described
Whether the first human eye area and described second human eye area mate, and when coupling by described first human eye area or described
Second human eye area is as the human eye area in the plurality of image information.
The fatigue detecting system of 11. deck officers as claimed in claim 9, it is characterised in that described server
Specifically for:
According to the PERCLOS value of the human eye area in the PERCLOS algorithm the plurality of image information of calculating, and by institute
State the threshold value that PERCLOS value and degree of fatigue differentiate to compare, and in described PERCLOS value more than or equal to institute
Judge that described deck officer is in fatigue state when stating the threshold value of degree of fatigue differentiation.
The fatigue detecting system of 12. deck officers as claimed in claim 11, it is characterised in that server according to
The below equation described PERCLOS value of calculating:
Wherein, N is human eye area sampling sum in continuous time,
The fatigue detecting system of 13. deck officers as claimed in claim 9, it is characterised in that described wearable
Equipment is additionally operable to:
When described server judges that described deck officer is in fatigue state, carry out alarm.
The fatigue detecting system of 14. deck officers as claimed in claim 13, it is characterised in that described warning carries
Show one or more that include in light prompt, voice message and vibration prompt.
The fatigue detecting system of 15. deck officers as claimed in claim 1, it is characterised in that described server
It is additionally operable to:
The plurality of image information carries out pretreatment, and wherein, described pretreatment includes image denoising process, equalization
Process, contrast process in one or more.
The fatigue detecting system of 16. deck officers as described in any one of claim 9-15, it is characterised in that institute
Stating wearable device is glasses.
17. 1 kinds of servers, it is characterised in that including:
Receiver module, for receiving the video flowing that wearable device gathers；
Modular converter, for being converted to multiple image information by the multiple frame of video in described video flowing；
Acquisition module, for obtaining the human eye area in the plurality of image information；And
Analyze module, for the human eye area in the plurality of image information is carried out analysis of fatigue to judge ship-handling
Whether member is in fatigue state, and sends analysis result to described wearable device.
18. servers as claimed in claim 17, it is characterised in that described acquisition module includes:
First acquiring unit, for carrying out the plurality of image information according to Adaboost algorithm based on Haar feature
Human eye positions, and obtains the first human eye area；
Second acquisition unit, for carrying out binary conversion treatment, and according to based on Haar feature to the plurality of image information
Adaboost algorithm carries out human eye location to the image information after binary conversion treatment, to obtain the second human eye area；
Judging unit, is used for judging whether described first human eye area and described second human eye area mate, and in coupling
Time using described first human eye area and/or described second human eye area as the human eye area in the plurality of image information.
19. servers as claimed in claim 17, it is characterised in that described analysis module includes:
Computing unit, for according to the human eye area in the PERCLOS algorithm the plurality of image information of calculating
PERCLOS value；
Comparing unit, for comparing the threshold value that described PERCLOS value and degree of fatigue differentiate；And
Judging unit, when the threshold value differentiated more than or equal to described degree of fatigue in described PERCLOS value, it is judged that institute
State deck officer and be in fatigue state.
20. servers as claimed in claim 19, it is characterised in that described computing unit calculates according to below equation
Described PERCLOS value:
Wherein, N is human eye area sampling sum in continuous time,
21. servers as claimed in claim 17, it is characterised in that also include:
Pretreatment module, for the plurality of image information is carried out pretreatment, wherein, described pretreatment includes image
Denoising, equalization processing, contrast process in one or more.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510279711.1A CN106295474B (en) | 2015-05-28 | 2015-05-28 | Fatigue detection method, system and the server of deck officer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510279711.1A CN106295474B (en) | 2015-05-28 | 2015-05-28 | Fatigue detection method, system and the server of deck officer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106295474A true CN106295474A (en) | 2017-01-04 |
| CN106295474B CN106295474B (en) | 2019-03-22 |
Family
ID=57634266
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510279711.1A Active CN106295474B (en) | 2015-05-28 | 2015-05-28 | Fatigue detection method, system and the server of deck officer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106295474B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108304764A (en) * | 2017-04-24 | 2018-07-20 | 中国民用航空局民用航空医学中心 | Fatigue state detection device and detection method in simulated flight driving procedure |
| CN109407609A (en) * | 2018-12-05 | 2019-03-01 | 江苏永钢集团有限公司 | A kind of facility information point detection system |
| CN110063736A (en) * | 2019-05-06 | 2019-07-30 | 苏州国科视清医疗科技有限公司 | The awake system of fatigue detecting and rush of eye movement parameter monitoring based on MOD-Net network |
| CN111353636A (en) * | 2020-02-24 | 2020-06-30 | 交通运输部水运科学研究所 | Multi-mode data based ship driving behavior prediction method and system |
| CN113947869A (en) * | 2021-10-18 | 2022-01-18 | 广州海事科技有限公司 | Alarm method, system, computer equipment and medium based on ship driving state |
| CN114537612A (en) * | 2021-12-31 | 2022-05-27 | 武汉理工大学 | Fatigue detection device and method for crew on duty at ship bridge |
| CN114663964A (en) * | 2022-05-24 | 2022-06-24 | 武汉理工大学 | Ship remote driving behavior state monitoring and early warning method and system and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102324166A (en) * | 2011-09-19 | 2012-01-18 | 深圳市汉华安道科技有限责任公司 | Fatigue driving detection method and device |
| CN103093215A (en) * | 2013-02-01 | 2013-05-08 | 北京天诚盛业科技有限公司 | Eye location method and device |
| CN104269028A (en) * | 2014-10-23 | 2015-01-07 | 深圳大学 | Fatigue driving detection method and system |
-
2015
- 2015-05-28 CN CN201510279711.1A patent/CN106295474B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102324166A (en) * | 2011-09-19 | 2012-01-18 | 深圳市汉华安道科技有限责任公司 | Fatigue driving detection method and device |
| CN103093215A (en) * | 2013-02-01 | 2013-05-08 | 北京天诚盛业科技有限公司 | Eye location method and device |
| CN104269028A (en) * | 2014-10-23 | 2015-01-07 | 深圳大学 | Fatigue driving detection method and system |
Non-Patent Citations (1)
| Title |
|---|
| 杨东: ""基于面部变化特征的驾驶疲劳监测方法研究"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108304764A (en) * | 2017-04-24 | 2018-07-20 | 中国民用航空局民用航空医学中心 | Fatigue state detection device and detection method in simulated flight driving procedure |
| CN108304764B (en) * | 2017-04-24 | 2021-12-24 | 中国民用航空局民用航空医学中心 | Fatigue state detection device and detection method in simulated flight driving process |
| CN109407609A (en) * | 2018-12-05 | 2019-03-01 | 江苏永钢集团有限公司 | A kind of facility information point detection system |
| CN110063736A (en) * | 2019-05-06 | 2019-07-30 | 苏州国科视清医疗科技有限公司 | The awake system of fatigue detecting and rush of eye movement parameter monitoring based on MOD-Net network |
| CN110063736B (en) * | 2019-05-06 | 2022-03-08 | 苏州国科视清医疗科技有限公司 | Eye movement parameter monitoring fatigue detection and wake-up promotion system based on MOD-Net network |
| CN111353636A (en) * | 2020-02-24 | 2020-06-30 | 交通运输部水运科学研究所 | Multi-mode data based ship driving behavior prediction method and system |
| CN113947869A (en) * | 2021-10-18 | 2022-01-18 | 广州海事科技有限公司 | Alarm method, system, computer equipment and medium based on ship driving state |
| CN113947869B (en) * | 2021-10-18 | 2023-09-01 | 广州海事科技有限公司 | Alarm method, system, computer equipment and medium based on ship driving state |
| CN114537612A (en) * | 2021-12-31 | 2022-05-27 | 武汉理工大学 | Fatigue detection device and method for crew on duty at ship bridge |
| CN114663964A (en) * | 2022-05-24 | 2022-06-24 | 武汉理工大学 | Ship remote driving behavior state monitoring and early warning method and system and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106295474B (en) | 2019-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106295474A (en) | The fatigue detection method of deck officer, system and server | |
| US10867195B2 (en) | Systems and methods for monitoring driver state | |
| CN101593425B (en) | Machine vision based fatigue driving monitoring method and system | |
| CN105844257B (en) | The early warning system and method for road sign are missed based on machine vision travelling in fog day | |
| CN105769120B (en) | Method for detecting fatigue driving and device | |
| CN104637246B (en) | Driver multi-behavior early warning system and danger evaluation method | |
| CN105354988B (en) | A kind of driver tired driving detecting system and detection method based on machine vision | |
| CN101639894B (en) | Method for detecting train driver behavior and fatigue state on line and detection system thereof | |
| CN101950355B (en) | Method for detecting fatigue state of driver based on digital video | |
| CN104751600B (en) | Anti-fatigue-driving safety means and its application method based on iris recognition | |
| CN108309311A (en) | A kind of real-time doze of train driver sleeps detection device and detection algorithm | |
| CN112016457A (en) | Driver distraction and dangerous driving behavior recognition method, device and storage medium | |
| CN105286802B (en) | Driver Fatigue Detection based on video information | |
| CN103824420A (en) | Fatigue driving identification system based on heart rate variability non-contact measuring | |
| CN102752458A (en) | Driver fatigue detection mobile phone and unit | |
| CN109977930A (en) | Method for detecting fatigue driving and device | |
| CN108647708A (en) | Driver evaluation's method, apparatus, equipment and storage medium | |
| CN103839046A (en) | Automatic driver attention identification system and identification method thereof | |
| CN102085099A (en) | Method and device for detecting fatigue driving | |
| CN105844245A (en) | Fake face detecting method and system for realizing same | |
| CN104574819A (en) | Fatigued drive detection method based on mouth features | |
| CN109543577A (en) | A kind of fatigue driving detection method for early warning based on facial expression feature | |
| CN108108651B (en) | Method and system for detecting driver non-attentive driving based on video face analysis | |
| CN113140093A (en) | Fatigue driving detection method based on AdaBoost algorithm | |
| Ribarić et al. | A neural-network-based system for monitoring driver fatigue |
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 |