CN202463683U - Intelligent automobile lamp system with driver sight following function - Google Patents

Intelligent automobile lamp system with driver sight following function Download PDF

Info

Publication number
CN202463683U
CN202463683U CN2011205665304U CN201120566530U CN202463683U CN 202463683 U CN202463683 U CN 202463683U CN 2011205665304 U CN2011205665304 U CN 2011205665304U CN 201120566530 U CN201120566530 U CN 201120566530U CN 202463683 U CN202463683 U CN 202463683U
Authority
CN
China
Prior art keywords
chaufeur
transceiver
optical coupling
automobile lamp
car light
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.)
Expired - Fee Related
Application number
CN2011205665304U
Other languages
Chinese (zh)
Inventor
陈泽宇
赵广耀
杨英
周淑文
唐传茵
廖治
王迪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Northeastern University China
Original Assignee
Northeastern University China
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Northeastern University China filed Critical Northeastern University China
Priority to CN2011205665304U priority Critical patent/CN202463683U/en
Application granted granted Critical
Publication of CN202463683U publication Critical patent/CN202463683U/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Images

Landscapes

  • Image Processing (AREA)

Abstract

The utility model relates to an intelligent automobile lamp system with a driver sight following function, belongs to the field of an automobile active safety technology and in particular relates to an intelligent automobile lamp system with a driver sight following function. By the system, the light of an automobile lamp intelligently follows the sight of a driver to eliminate a visual blind zone and improve the safety of driving the automobile at night. The intelligent automobile lamp system comprises driver image acquisition equipment, a control system, a signal acquisition unit, an angle sensor and an automobile lamp motor arranged below the automobile lamp, wherein the driver image acquisition equipment is connected with the control system through a serial communication data line; the control system is connected with a motor controller and the signal acquisition unit through a controller area network (CAN) bus; the automobile lamp motor is connected with the motor controller; an output shaft of the automobile lamp motor is fixedly connected with the automobile lamp; the angle sensor is arranged on the output shaft of the automobile lamp motor; and the angle sensor is connected with the signal acquisition unit.

Description

A kind of have an intelligent automotive light system that chaufeur sight is followed function
Technical field
The utility model belongs to the automobile active safety technical field; Particularly relate to a kind of intelligent automotive light system that chaufeur sight is followed function that has; The vision blind area is eliminated through making car light light intelligence follow chaufeur sight by this system, improves the safety of vehicle night running.
Background technology
During night travel, chaufeur need rely on car light to distinguish road environment, thus the safety traffic of control vehicle; But the light majority of the common car light of installing on the modern automobile can only straight trend of purchasing the place ahead, has only a fixing range of exposures.And vehicle ' is when strange four corners, the pedestrian and the road environment of the vehicular sideview often that chaufeur is more paid close attention to; In addition, also similar situation can take place in dim and narrow and small environment such as parking area, as shown in Figure 1.In these cases, the light in straight trend of purchasing the place ahead obviously can't satisfy the observation demand of chaufeur, and chaufeur can only rely on the surplus light of faint scattering to grope condition of road surface, has serious potential safety hazard.
The utility model content
To the problem that prior art exists, the utility model provides a kind of intelligent automotive light system that chaufeur sight is followed function that has.The vehicle light illumination direction of this system is always followed the sight of chaufeur, has guaranteed that chaufeur always can see the zone of oneself hope observing clearly, thereby has improved the safety that vehicle at night goes.
To achieve these goals; The utility model adopts following technical scheme; A kind of have an intelligent automotive light system that chaufeur sight is followed function; The car light motor that comprises chaufeur image capture device, control system, signal gathering unit, angular transducer and be arranged on the car light below, said chaufeur image capture device is connected with control system through the serial communication data line; Said control system is connected with signal gathering unit with electric machine controller respectively through the CAN bus, and said car light motor is connected with electric machine controller; The output shaft of said car light motor is captiveed joint with car light, and said angular transducer is arranged on the output shaft of car light motor, and said angular transducer is connected with signal gathering unit.
Said signal gathering unit comprises AD converter, optical coupling isolator, first micro controller system and a CAN transceiver; The input end of said AD converter is connected with the signal output part of angular transducer, and the mouth of said AD converter is connected with first micro controller system through first optical coupling isolator; Said first micro controller system is connected with a CAN transceiver through second optical coupling isolator, the 3rd optical coupling isolator, and a said CAN transceiver is connected with the CAN bus.
Said control system comprises RS232 interface, RS232 transceiver, optical coupling isolator, second singlechip and the 2nd CAN transceiver; Said RS232 interface is connected with the input end of RS232 transceiver, and the mouth of RS232 transceiver is connected with second singlechip through the 4th optical coupling isolator; Said second singlechip is connected with the 2nd CAN transceiver through the 5th optical coupling isolator, the 6th optical coupling isolator, and said the 2nd CAN transceiver is connected with the CAN bus.
Said chaufeur image capture device adopts has the post processing of image function, and can shine or the unglazed infrared camera of working under the situation that shines at low light.
The beneficial effect of the utility model:
In the vehicle traveling process, the direction that chaufeur sight is pointed to is that chaufeur hopes to observe clearly zone most, so this direction also is the optimal direction of illumination of car light light.The car light of the intelligent automotive light system of the utility model can be intended to automatically according to the vision of chaufeur regulates light angle, makes the projecting direction of car light always follow driver vision, as shown in Figure 2; So just can eliminate the irradiation blind area of the light in the straight trend of purchasing of traditional car light the place ahead effectively.Guaranteed that chaufeur always can clearly distinguish the regional road conditions of oneself hope observing, thus the safety that has greatly improved the chaufeur night travel.
Description of drawings
Fig. 1 is the vision blind area scheme drawing of existing common car light;
Fig. 2 is the control effect scheme drawing that chaufeur sight is followed the intelligent automotive light system of function that has of the utility model;
Fig. 3 is the structural representation that chaufeur sight is followed the intelligent automotive light system of function that has of the utility model;
Fig. 4 is the schematic block circuit diagram of the signal gathering unit of the utility model;
Fig. 5 is the schematic circuit diagram of the signal gathering unit of the utility model;
Fig. 6 is the schematic block circuit diagram of the control system of the utility model;
Fig. 7 is the schematic circuit diagram of the control system of the utility model;
Fig. 8 is the program flow diagram of control method that chaufeur sight is followed the intelligent automotive light system of function that has of the utility model;
Fig. 9 is the geometric relationship figure that the facial corner of chaufeur calculates;
Among Fig. 3,1-chaufeur image capture device, 2-serial communication data line, 3-control system, 4-CAN bus, 5-signal gathering unit, 6-electric machine controller, 7-car light motor, 8-angular transducer, 9-car light;
Among Fig. 4,10-AD conv, 11-first optical coupling isolator, 12-first micro controller system, 13-second optical coupling isolator, 14-the 3rd optical coupling isolator, 15-the one CAN transceiver;
Among Fig. 6,16-RS232 transceiver, 17-the 4th optical coupling isolator, 18-second singlechip, 19-the 5th optical coupling isolator, 20-the 6th optical coupling isolator, 21-the 2nd CAN transceiver.
The specific embodiment
As shown in Figure 3; A kind of have an intelligent automotive light system that chaufeur sight is followed function; Comprise chaufeur image capture device 1, control system 3, signal gathering unit 5, angular transducer 8 and the car light motor 7 that is arranged on car light 9 belows, said chaufeur image capture device 1 is connected with control system 3 through serial communication data line 2; Said control system 3 is connected with signal gathering unit 5 with electric machine controller 6 respectively through CAN bus 4, and said car light motor 7 is connected with electric machine controller 6; The output shaft of said car light motor 7 is captiveed joint with car light 9, and said angular transducer 8 is arranged on the output shaft of car light motor 7, and said angular transducer 8 is connected with signal gathering unit 5.
Said chaufeur image capture device 1 adopts has the post processing of image function; And can shine or the unglazed infrared camera of working under the situation that shines at low light; What said infrared camera adopted is that permanent Feng Wei looks the infrared CCD camera that company produces, and its model is: HF-6540S.What said angular transducer 8 adopted is the rotary encoder that peace lattice wound Electronics Co., Ltd. produces, and its model is: Angtron-RE-38 is used to measure the actual rotational angle value of car light, and the actual rotational angle value that records is converted into voltage signal.What said car light motor 7 adopted is brshless DC motor, and its model is: 57BL52-230.The model of said electric machine controller 6 is: ZD-6716-V2, and it possesses the CAN transmission-receiving function, the control command of can receiving control system 3 sending, and control car light motor 7 drives car lights 9 and rotates.
As shown in Figure 4; Said signal gathering unit 5 comprises AD converter 10, optical coupling isolator, first micro controller system 12 and a CAN transceiver 15; The input end of said AD converter 10 is connected with the signal output part of angular transducer 8, and the mouth of said AD converter 10 is connected with first micro controller system 12 through first optical coupling isolator 11; Said first micro controller system 12 is connected with a CAN transceiver 15 through second optical coupling isolator 13, the 3rd optical coupling isolator 14, and a said CAN transceiver 15 is connected with CAN bus 4.Its physical circuit annexation is as shown in Figure 5.
As shown in Figure 6; Said control system 3 adopts embedded controller; Comprise RS232 interface, RS232 transceiver 16, optical coupling isolator, second singlechip 18 and the 2nd CAN transceiver 21; Said RS232 interface is connected with the input end of RS232 transceiver 16, and the mouth of RS232 transceiver 16 is connected with second singlechip 18 through the 4th optical coupling isolator 17; Said second singlechip 18 is connected with the 2nd CAN transceiver 21 through the 5th optical coupling isolator 19, the 6th optical coupling isolator 20, and said the 2nd CAN transceiver 21 is connected with CAN bus 4.Its physical circuit annexation is as shown in Figure 7.
A kind of have a control method that chaufeur sight is followed the intelligent automotive light system of function, and its control process can be divided into three parts: the one, image acquisition, the 2nd, image recognition, the 3rd, control with carry out, as shown in Figure 8, it specifically comprises the steps:
Step 1: obtain the chaufeur face-image through infrared camera;
Step 2: the chaufeur face-image that obtains is changed into gray level image, and carry out image filtering, noise reduction;
Step 3: based on the Otsu algorithm image is carried out binaryzation and cut apart;
Step 4: utilize the infrared face image region brightness higher, and the black partially characteristics of background image brightness are rejected the background image in the image based on grey level histogram;
Step 5: based on the Sobel operator image is carried out rim detection, identification and location chaufeur face contour; Zone with chaufeur face contour top 1/3rd is an area-of-interest, in area-of-interest, adopts Sobel operator extraction eye marginal information, identification and location chaufeur binocular profile;
Step 6: the recognition result according to the chaufeur face contour simulates line of centers A-A 1, and calculate facial width L;
Step 7: calculate binocular center-point C point to A-A 1Apart from d;
Step 8: calculate the facial corner ξ of chaufeur;
Step 9: with chaufeur binocular profile is area-of-interest, and the geometric center point of selecting the binocular profile based on gray value similarity criterion between pixel, is carried out chaufeur pupil location based on region-growing method as seed points in binocular profile scope;
Step 10:, confirm chaufeur eyeball corner β based on the BP neural network according to binocular profile and pupil position;
Step 11: confirm chaufeur sight corner α according to facial corner ξ and eyeball corner β 1
Step 12: calculate the current controling signal of car light motor based on PI control, and said current controling signal sent to electric machine controller through the CAN bus, said PI control be input as chaufeur sight corner α 1With car light corner α 2Difference;
Step 13: electric machine controller makes it drive car light and rotates according to the current controling signal control car light motor that is received, and changes illumination direction;
Step 14: measure car light corner α through angular transducer 2
Step 15: judge chaufeur sight corner α 1Whether equal car light corner α 2If,, then execution in step 16; Otherwise, with car light corner α 2As feedback signal, return execution in step 12;
Step 10 six: finish.
One action process below in conjunction with description of drawings the utility model:
The infrared camera of the utility model is arranged at the chaufeur front lower place.During use, at first obtain the chaufeur face-image through infrared camera, infrared camera converts the physical image of obtaining to digital image, and through the serial communication data line in real time with the chaufeur Digital Image Transmission to control system.In control system, have graphics processing unit module and control unit module two parts program, in the graphics processing unit module, at first the chaufeur digital image that collects is changed into gray level image, and carry out image filtering, noise reduction; Based on the Otsu algorithm image is carried out binaryzation then and cut apart, utilize the infrared face image region brightness higher, and the black partially characteristics of background image brightness are rejected the background image in the image based on grey level histogram; Pixel can be distributed as two parts zone in grey level histogram, through a rational threshold value is set, can reject the background image in the image, and obtain the chaufeur facial zone.In the chaufeur facial zone, based on the Sobel operator image is carried out rim detection, identification and location chaufeur face contour; Zone with chaufeur face contour top 1/3rd is an area-of-interest, in area-of-interest, adopts Sobel operator extraction eye marginal information, identification and location chaufeur binocular profile.
After obtaining the binocular profile, be area-of-interest with chaufeur binocular profile, the geometric center point of at first selecting the binocular profile is stipulated gray value similarity criterion between a pixel then as seed points, carries out chaufeur pupil location based on region-growing method.For example based on the criterion of grey scale pixel value, if the absolute value of the difference of the gray value between adjacent two pixels, thinks then that they are similar less than given threshold value.From the seed points of eyes geometric centre, check seed points whole points on every side one by one, if a point is similar to seed points, then incorporated in this seed region, in whole binocular profile, repeat this process, can realize chaufeur pupil location.
Behind the identification of graphics processing unit module and location chaufeur face contour and the binocular profile, the control unit module is calculated the facial corner of chaufeur according to processing result image.The computation process of the facial corner of said chaufeur, as shown in Figure 9, specifically comprise the steps:
At first simulate the line of centers A-A of a face contour according to the recognition result of chaufeur face contour 1, and calculate facial width L; Simulate binocular center-point C point according to chaufeur binocular profile, calculate binocular center-point C point to face contour line of centers A-A 1Apart from d; Calculate the facial corner ξ of chaufeur according to facial width L and d then, its computing formula is following:
ξ = arctan ( 2 d L )
In the formula, d-binocular center-point is to the distance of face contour line of centers;
The facial width of L-.
According to binocular profile and the pupil position that the graphics processing unit module is extracted, confirm chaufeur eyeball corner β based on the BP neural network; Use the position of pupil in the binocular profile as training input data, chaufeur eyeball corner β exports as network, is shown below:
β = Σ j = 1 l f ( Σ i = 1 n w ij x i - a j ) w jk - b k
In the formula, β-chaufeur eyeball corner;
x iThe position coordinate of-pupil in the binocular profile;
L-neural network input layer number;
N-neural network hidden layer node number;
w Ij-neural network hidden layer connects weights;
w Jk-neural network output layer connects weights;
a j-neural network hidden layer threshold value;
b k-neural network output layer threshold value;
Before use BP neural network is carried out the prediction of eyeball corner, at first need carry out network training, promptly carry out the renewal of weights and threshold value through a part known pupil position and eyeball corner test data, be shown below:
w ij = w ij + ηH j ( 1 - H j ) x i Σ k = 1 m w jk e k w jk = w jk + ηH j e k a j = a j + ηH j ( 1 - H j ) x i Σ k = 1 m w jk e k b k = b k + e k
In the formula, w Ij-neural network hidden layer connects weights;
w Jk--the neural network output layer connects weights;
a j-neural network hidden layer threshold value;
b k-neural network output layer threshold value;
η-neural network learning speed;
H jThe output of j node of-neural network hidden layer;
e k-neural network prediction error;
M-neural network output layer node number;
x iThe position coordinate of-pupil in the binocular profile.
Confirm chaufeur sight corner α according to facial corner ξ of chaufeur and eyeball corner β 1, its detailed process is following:
At first, set a corner threshold value Ψ, Ψ can be taken as 3 °, i.e. 0.05rad.If the facial corner ξ of chaufeur surpasses corner threshold value Ψ, explain that then the chaufeur head rotates, and at this moment, makes α 1=ξ; If the facial corner ξ of chaufeur is lower than corner threshold value Ψ; Explain that then the chaufeur head does not rotate, judge further then whether eyeball corner β surpasses corner threshold value Ψ, if eyeball corner β surpasses corner threshold value Ψ; Though then explanation chaufeur head this moment does not rotate; But eyes at this moment, make α to the side stravismus 1=β; If eyeball corner β is lower than corner threshold value Ψ, then think it is slightly rocking naturally of chaufeur head at this moment, be not to have a mind to observe lateral side regions, at this moment, make α 1=0.
Confirming chaufeur sight corner α 1Afterwards, calculate the current controling signal of car light motor, and said current controling signal sent to electric machine controller through the CAN bus based on PI control, said PI control be input as chaufeur sight corner α 1With car light corner α 2Difference α 1-α 2, its computing formula is following:
I m＝P D(α 1-α 2)+P I∫(α 1-α 2)dt
In the formula, I m-current of electric control signal;
P DThe proportionality coefficient of-PI control;
P IThe integral coefficient of-PI control;
α 1-chaufeur sight corner;
α 2-car light corner.
Electric machine controller makes it drive car light and rotates according to the current controling signal control car light motor that is received, and changes illumination direction.The angular transducer that is arranged on the car light motor shaft is converted into the voltage signal of 0～5V with the car light angular signal, and signal gathering unit is gathered this voltage signal, and this voltage signal is converted into car light corner α 2Then, send to control system through the CAN bus.The control unit module is according to the actual measurement value of feedback of car light corner and the deviation between the chaufeur sight corner; Control correcting current control signal in real time based on PI; Actual rotational angle and the chaufeur sight corner of guaranteeing car light coincide; Thereby the corner that makes the car light motor can be followed the sight of chaufeur, even car light light-illuminating direction is followed pilot's line of vision.

Claims (4)

1. one kind has the intelligent automotive light system that chaufeur sight is followed function; The car light motor that it is characterized in that comprising chaufeur image capture device, control system, signal gathering unit, angular transducer and be arranged on the car light below, said chaufeur image capture device is connected with control system through the serial communication data line; Said control system is connected with signal gathering unit with electric machine controller respectively through the CAN bus, and said car light motor is connected with electric machine controller; The output shaft of said car light motor is captiveed joint with car light, and said angular transducer is arranged on the output shaft of car light motor, and said angular transducer is connected with signal gathering unit.
2. a kind of intelligent automotive light system that chaufeur sight is followed function that has according to claim 1; It is characterized in that said signal gathering unit comprises AD converter, optical coupling isolator, first micro controller system and a CAN transceiver; The input end of said AD converter is connected with the signal output part of angular transducer, and the mouth of said AD converter is connected with first micro controller system through first optical coupling isolator; Said first micro controller system is connected with a CAN transceiver through second optical coupling isolator, the 3rd optical coupling isolator, and a said CAN transceiver is connected with the CAN bus.
3. a kind of intelligent automotive light system that chaufeur sight is followed function that has according to claim 1; It is characterized in that said control system comprises RS232 interface, RS232 transceiver, optical coupling isolator, second singlechip and the 2nd CAN transceiver; Said RS232 interface is connected with the input end of RS232 transceiver, and the mouth of RS232 transceiver is connected with second singlechip through the 4th optical coupling isolator; Said second singlechip is connected with the 2nd CAN transceiver through the 5th optical coupling isolator, the 6th optical coupling isolator, and said the 2nd CAN transceiver is connected with the CAN bus.
4. a kind of intelligent automotive light system that chaufeur sight is followed function that has according to claim 1; It is characterized in that said chaufeur image capture device employing has the post processing of image function, and can shine or the unglazed infrared camera of working under the situation that shines at low light.
CN2011205665304U 2011-12-30 2011-12-30 Intelligent automobile lamp system with driver sight following function Expired - Fee Related CN202463683U (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2011205665304U CN202463683U (en) 2011-12-30 2011-12-30 Intelligent automobile lamp system with driver sight following function

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2011205665304U CN202463683U (en) 2011-12-30 2011-12-30 Intelligent automobile lamp system with driver sight following function

Publications (1)

Publication Number Publication Date
CN202463683U true CN202463683U (en) 2012-10-03

Family

ID=46911869

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2011205665304U Expired - Fee Related CN202463683U (en) 2011-12-30 2011-12-30 Intelligent automobile lamp system with driver sight following function

Country Status (1)

Country Link
CN (1) CN202463683U (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102490646A (en) * 2011-12-30 2012-06-13 东北大学 Intelligent vehicle lamp system with driver eyesight following function and control method for intelligent vehicle lamp system
CN105398369A (en) * 2014-09-04 2016-03-16 通用汽车环球科技运作有限责任公司 Motor Vehicle And Method For Operating A Motor Vehicle
CN106828285A (en) * 2017-03-09 2017-06-13 李良杰 The car light and control method rotated with pilot's line of vision
CN110623629A (en) * 2019-07-31 2019-12-31 毕宏生 Visual attention detection method and system based on eyeball motion

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102490646A (en) * 2011-12-30 2012-06-13 东北大学 Intelligent vehicle lamp system with driver eyesight following function and control method for intelligent vehicle lamp system
CN105398369A (en) * 2014-09-04 2016-03-16 通用汽车环球科技运作有限责任公司 Motor Vehicle And Method For Operating A Motor Vehicle
CN106828285A (en) * 2017-03-09 2017-06-13 李良杰 The car light and control method rotated with pilot's line of vision
CN110623629A (en) * 2019-07-31 2019-12-31 毕宏生 Visual attention detection method and system based on eyeball motion

Similar Documents

Publication Publication Date Title
CN102490646B (en) Intelligent vehicle lamp system with driver eyesight following function and control method for intelligent vehicle lamp system
CN105196910B (en) Safe driving assistant system and its control method under a kind of misty rain weather
CN101941399B (en) Vehicle lane change auxiliary device and auxiliary judgment method thereof
CN104881955B (en) A kind of driver tired driving detection method and system
CN110517521A (en) A kind of lane departure warning method based on road car fusion perception
CN103057470A (en) Pre-prompt device and pre-prompt method for regulation violation line-rolling of vehicles
CN110085042B (en) Vehicle driving early warning system and method based on information fusion
CN104751600B (en) Anti-fatigue-driving safety means and its application method based on iris recognition
CN107323459B (en) Method for identifying driver intention identifying sensor device
CN105620489A (en) Driving assistance system and real-time warning and prompting method for vehicle
CN202463683U (en) Intelligent automobile lamp system with driver sight following function
CN107972671A (en) A kind of driving behavior analysis system
CN109145719B (en) Driver fatigue state identification method and system
CN104129389A (en) Method for effectively judging and recognizing vehicle travelling conditions and device thereof
CN104742912A (en) Lane deviation detection method and device
CN105303830A (en) Driving behavior analysis system and analysis method
CN104021370A (en) Driver state monitoring method based on vision information fusion and driver state monitoring system based on vision information fusion
CN110472556A (en) Driver attention's state analysis system and analysis method based on monocular vision
CN105404862A (en) Hand tracking based safe driving detection method
CN111891129A (en) Intelligent driving system of electric automobile
CN103692974B (en) A kind of vehicle driving safety method for early warning based on environmental monitoring and system
CN115027484A (en) Human-computer fusion perception method for high-degree automatic driving
CN108447305A (en) A kind of new friendship rule are lower to give precedence to pedestrian's method for early warning and early warning system
CN116030662B (en) Intelligent safety detection system and method based on big data
CN112863244A (en) Method and device for promoting safe driving of vehicle

Legal Events

Date Code Title Description
C14 Grant of patent or utility model
GR01 Patent grant
C17 Cessation of patent right
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20121003

Termination date: 20131230