CN108596977A - A kind of outer continuous parameters scaling method of bionic eye binocular vision - Google Patents

A kind of outer continuous parameters scaling method of bionic eye binocular vision Download PDF

Info

Publication number
CN108596977A
CN108596977A CN201810185637.0A CN201810185637A CN108596977A CN 108596977 A CN108596977 A CN 108596977A CN 201810185637 A CN201810185637 A CN 201810185637A CN 108596977 A CN108596977 A CN 108596977A
Authority
CN
China
Prior art keywords
bionic eye
bit
zero
coordinate system
joint
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
Application number
CN201810185637.0A
Other languages
Chinese (zh)
Other versions
CN108596977B (en
Inventor
陈晓鹏
张陶然
汪常进
田野
金鉉優
程炜
陈曦
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.)
Beijing Institute of Technology BIT
Original Assignee
BEIJING HUAKAIHUI INFORMATION TECHNOLOGY Co Ltd
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 BEIJING HUAKAIHUI INFORMATION TECHNOLOGY Co Ltd filed Critical BEIJING HUAKAIHUI INFORMATION TECHNOLOGY Co Ltd
Priority to CN201810185637.0A priority Critical patent/CN108596977B/en
Publication of CN108596977A publication Critical patent/CN108596977A/en
Application granted granted Critical
Publication of CN108596977B publication Critical patent/CN108596977B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/80Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
    • G06T7/85Stereo camera calibration
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/20Analysis of motion
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10004Still image; Photographic image
    • G06T2207/10012Stereo images
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30244Camera pose

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Eye Examination Apparatus (AREA)
  • Length Measuring Devices By Optical Means (AREA)

Abstract

The invention discloses continuous parameters scaling method outside a kind of bionic eye binocular vision, this method reads the specific location of each monocular-camera by controlling the motor with position feedback that binocular camera moves;Then the relative matrix of the binocular camera at current time is obtained by calculation;With this method, only it is to be understood that the initial relative position parameter matrix of binocular camera, the relative position relation of the binocular camera of any time can be obtained using position feedback, to realize continuous calibration.

Description

A kind of outer continuous parameters scaling method of bionic eye binocular vision
Technical field
The invention belongs to technical field of computer vision, and in particular to a kind of outer continuous parameters calibration of bionic eye binocular vision Method.
Background technology
Traditional binocular vision calibration algorithm is based on Zhang Zhengyou monocular vision calibration algorithms, after being marked using gridiron pattern method Internal reference matrix to each camera and outer ginseng matrix.Outer ginseng matrix establishes the connection between world coordinate system and camera coordinate system System.Internal reference matrix establishes contacting between camera coordinate system and pixel coordinate system.Two such video camera is sat by the world Mark system can arrive between relational matrix.Include spin matrix and translation matrix.This binocular vision calibration is calculated Method is very effective in the case that two video camera relative positions are fixed.
But in the case where binocular camera moves, the Camera extrinsic that initial alignment obtains can change.If no It re-scales, then severe deviations will occur in the image depth information obtained using original formula.It needs to re-scale.It is moving Middle demarcated with gridiron pattern is unpractical, because gridiron pattern method is only applicable to static demarcating under original state and obtains initial internal reference square Battle array and outer ginseng matrix.Therefore using we need one can be with a kind of continuous calibration algorithm of camera motion.
Invention content
In view of this, the object of the present invention is to provide continuous parameters scaling methods outside a kind of bionic eye binocular vision so that Binocular camera external parameter matrix relativeness in dynamic movement process is able to Real-time solution.
A kind of outer continuous parameters scaling method of bionic eye binocular vision, includes the following steps:
The original state of two video cameras in step 1, arbitrary selection bionic eye binocular vision system, and remember that the two are initial State is respectively the first zero-bit q0With the first zero-bit q1
In the first zero-bit q0, demarcate to obtain in the first zero-bit q by Zhang Shi standardizations0When bionic eye binocular vision system it is outer Parameter matrixCLTCR(q0);Record three joint position information in bionic eye binocular vision system middle left and right eye joint, and foundation Joint position information is calculated in the first zero-bit q0When, transformation matrix of the right camera pedestal relative to bionic eye base coordinate systemNTR (q0) and transformation matrix of the left camera pedestal relative to bionic eye base coordinate systemNTL(q0);
Step 2, in the first zero-bit q1, demarcate to obtain in the first zero-bit q by Zhang Shi standardizations1When bionic eye binocular vision system The outer parameter matrix of systemCLTCR(q1);Three joint position information in images of left and right eyes joint are recorded, and according to joint position information meter It calculates in the first zero-bit q1When, transformation matrix of the right camera pedestal relative to bionic eye base coordinate systemNTR(q1) and left camera base Transformation matrix of the seat relative to bionic eye base coordinate systemNTL(q1);
Step 3 enables A=RTN(q0)NTL(q0)LTN(q1)NTR(q1), B=CRTCL(q0)CLTCR(q1), solve equation AX= XB can obtain X=RTCRSolution;Then basisLTCL=LTN(q0)NTR(q0)RTCR CRTCL(q0) solve obtainLTCLSolution.
Step 4, in free-position, read the value of each eyeball joint position, and record the position vector be state q, meter It calculates under the pose, transformation matrix of the bionic eye base coordinate system relative to left camera pedestalLTN(q) and bionic eye pedestal coordinate It is the transformation matrix relative to right camera pedestalRTN(q), and according toCLTCR(q)=CLTL LTN(q)NTR(q)RTCR, it is calculated The outer parameter matrix of free-positionCLTCR(q)。
The present invention has the advantages that:
Proposed by the present invention a kind of for proposing the outer continuous parameters scaling method of bionic eye binocular vision, this method passes through The motor with position feedback of control binocular camera movement reads the specific location of each monocular-camera.Then pass through calculating Obtain the relative matrix of the binocular camera at current time.With this method, only it is to be understood that binocular camera initial phase aligns Parameter matrix is set, the relative position relation of the binocular camera of any time can be obtained using position feedback, to realize Continuous calibration.
Description of the drawings
Fig. 1 is the calibration schematic diagram of the present invention;
Fig. 2 is flow chart of the method for the present invention;
Fig. 3 is the geometric coordinate schematic diagram of the bionic eye binocular stereo vision of the present invention.
Specific implementation mode
The present invention will now be described in detail with reference to the accompanying drawings and examples.
Bionic eye binocular vision system includes there are two eyeball, and two monocular-cameras are separately fixed in two eyeballs, Two eyeballs are set there are one bionic eye pedestal, and each monocular-camera further respectively has a camera pedestal;Each research is by revolving Turn, pitching, yaw three motor drivings, each motor is defined as a joint, and the location information in joint is by the biography on motor shaft Sensor obtains.
As shown in Figure 1 schematic diagram is demarcated for bionic eye binocular stereo vision.Wherein each eyeball has a rolling, pitching, partially The degree of freedom in three directions of boat.There is position feedback detection in each degree of freedom.Eyeball is obtained by the reading of position feedback to work as Preceding each shaft position.Wherein, coordinate system R (x, y, z) indicates that right eye eyeball coordinate system, coordinate system L (x, y, z) indicate left eye eye Spherical coordinate system.W (x, y, z) indicates world coordinate system.
It is illustrated in figure 3 this method and calculates each joint geometric coordinate schematic diagram.Wherein, coordinate system R (x, y, z) is indicated Right eye eyeball coordinate system, is abbreviated as R.Coordinate system L (x, y, z) indicates left eye eyeball coordinate system, is abbreviated as L.Coordinate system CR (x, y, Z) it indicates right eye camera coordinates system, is abbreviated as CR.Coordinate system CL (x, y, z) indicates left eye camera coordinates system, is abbreviated as CL.It sits Mark system N (x, y, z) indicates right eye camera coordinates system, is abbreviated as N.
In the present invention, coordinate system and transformation matrix are defined first, specially:
CL, left camera coordinates system;
L, left camera base coordinate system;
CR, right camera coordinates system;
R, right camera base coordinate system;
N, bionic eye neck finish coordinate system;
LTCL, transformation matrix of the left camera coordinates system relative to left camera base coordinate system, fixed but unknown, systematic error It can also sum up in the point that in this matrix;
RTCR, transformation matrix of the right camera coordinates system relative to right camera base coordinate system, fixed but unknown, systematic error It can also sum up in the point that in this matrix;
CLTCR, outer parameter matrix of the right camera coordinates system relative to transformation matrix, that is, bionic eye of left camera coordinates system, by It can change in right and left eyes relative pose, which is changing always.
CRTCL, transformation matrix of the left camera coordinates system relative to right camera coordinates system, since right and left eyes relative pose can be sent out Changing, the parameter are changing always.
NTL, transformation matrix of the left camera pedestal relative to bionic eye base coordinate system, the position for depending on left eye joint is believed Breath, the location information can be obtained by epibulbar corresponding joint position sensor;
LTN, transformation matrix of the bionic eye base coordinate system relative to left camera pedestal, the position for depending on left eye joint is believed Breath, the location information can be obtained by epibulbar corresponding joint position sensor;
NTR, transformation matrix of the right camera pedestal relative to bionic eye base coordinate system, the position for depending on right eye joint is believed Breath, the location information can be obtained by epibulbar corresponding joint position sensor;
RTN, transformation matrix of the bionic eye base coordinate system relative to right camera pedestal, the position for depending on right eye joint is believed Breath, the location information can be obtained by epibulbar corresponding joint position sensor;
As shown in Fig. 2, the outer continuous parameters scaling method of bionic eye binocular vision in the present invention, is suitable for binocular camera The continually changing vision system in position each other.This algorithm is initial dependent on being obtained using Zhang Zhengyou binocular camera calibration algorithms Binocular camera relative position relation.This method is read each by controlling the motor with position feedback of binocular camera movement The specific location of monocular-camera.Then the binocular camera and eye base coordinate system and neck at current time is obtained by calculation The relative position relation matrix of joint coordinate system.With this method, only it is to be understood that the initial relative position parameter of binocular camera Matrix can obtain the relative position relation of the binocular camera of any time using position feedback.To realize continuous mark It is fixed.In this way, the outer ginseng matrix of the binocular camera in movement is allow to obtain Real-time solution.It is continuous suitable for binocular position The system of variation.
This method is divided to be made of two steps, is manual demarcating steps and automatic Calibration step respectively.In manual calibration step Suddenly, external parameters calibration is carried out in the first zero-bit, the second zero-bit respectively, while records the joint position of the first zero-bit, the second zero-bit Feedback information.According to the outer parameter and joint position feedback information of above-mentioned calibration, can settle accounts around each camera relative to camera base The transformation matrix of seat.In automatic Calibration step, bionic eye is obtained when free-position is continuously moved using manual demarcating steps As a result, in conjunction with current joint location information, it can real-time resolving binocular camera external parameter matrix in dynamic movement process.
It is as follows:
Step 1, the original state for arbitrarily choosing two binocular cameras carry out static mark using the two original states It is fixed.Remember that the two original states are respectively the first zero-bit q0 and the first zero-bit q1.In the first zero-bit q0, demarcated by Zhang Shi standardizations Obtain outer parameter matrixCLTCR(q0);Three joint position information in the images of left and right eyes joint of position recording sensor are calculated the The transformation matrix of one zero-bit q0NTR(q0),NTL(q0);
Step 2, in the first zero-bit q1, demarcated to obtain outer parameter matrix by Zhang Shi standardizationsCLTCR(q1);Record position senses Three joints position location information in the right and left eyes joint of device, calculates transformation matrixNTR(q1),NTL(q1);
Step 3 enables A=RTN(q0)NTL(q0)LTN(q1)NTR(q1), B=CRTCL(q0)CLTCR(q1), solve equation AX= XB can obtain X=RTCRSolution;Then basisLTCL=LTN(q0)NTR(q0)RTCR CRTCL(q0) solve obtainLTCLSolution.
Step 4, in free-position, read the value of each joint position, and record the position vector be state q, calculateLTN (q),RTN(q), and according toCLTCR(q)=CLTL LTN(q)NTR(q)RTCR, the outer parameter matrix of free-position is calculated.
In conclusion the above is merely preferred embodiments of the present invention, being not intended to limit the scope of the present invention. All within the spirits and principles of the present invention, any modification, equivalent replacement, improvement and so on should be included in the present invention's Within protection domain.

Claims (1)

1. a kind of outer continuous parameters scaling method of bionic eye binocular vision, which is characterized in that include the following steps:
The original state of two video cameras in step 1, arbitrary selection bionic eye binocular vision system, and remember the two original states Respectively the first zero-bit q0With the first zero-bit q1
In the first zero-bit q0, demarcate to obtain in the first zero-bit q by Zhang Shi standardizations0When bionic eye binocular vision system outer parameter MatrixCLTCR(q0);Three joint position information in bionic eye binocular vision system middle left and right eye joint are recorded, and according to joint Positional information calculation is in the first zero-bit q0, transformation matrix of the right camera pedestal relative to bionic eye base coordinate systemNTR(q0) with And transformation matrix of the left camera pedestal relative to bionic eye base coordinate systemNTL(q0);
Step 2, in the first zero-bit q1, demarcate to obtain in the first zero-bit q by Zhang Shi standardizations1When bionic eye binocular vision system Outer parameter matrixCLTCR(q1);Three joint position information in images of left and right eyes joint are recorded, and are calculated according to joint position information First zero-bit q1When, transformation matrix of the right camera pedestal relative to bionic eye base coordinate systemNTR(q1) and left camera pedestal phase For the transformation matrix of bionic eye base coordinate systemNTL(q1);
Step 3 enables A=RTN(q0)NTL(q0)LTN(q1)NTR(q1), B=CRTCL(q0)CLTCR(q1), equation AX=XB is solved, can be obtained Go out X=RTCRSolution;Then basisLTCL=LTN(q0)NTR(q0)RTCR CRTCL(q0) solve obtainLTCLSolution.
Step 4, in free-position, read the value of each eyeball joint position, and record the position vector be state q, calculate exist Under the pose, transformation matrix of the bionic eye base coordinate system relative to left camera pedestalLTN(q) and bionic eye base coordinate system phase For the transformation matrix of right camera pedestalRTN(q), and according toCLTCR(q)=CLTL LTN(q)NTR(q)RTCR, arbitrary position is calculated The outer parameter matrix of appearanceCLTCR(q)。
CN201810185637.0A 2018-03-07 2018-03-07 Bionic eye binocular vision external parameter continuous calibration method Expired - Fee Related CN108596977B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810185637.0A CN108596977B (en) 2018-03-07 2018-03-07 Bionic eye binocular vision external parameter continuous calibration method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810185637.0A CN108596977B (en) 2018-03-07 2018-03-07 Bionic eye binocular vision external parameter continuous calibration method

Publications (2)

Publication Number Publication Date
CN108596977A true CN108596977A (en) 2018-09-28
CN108596977B CN108596977B (en) 2021-06-11

Family

ID=63625806

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810185637.0A Expired - Fee Related CN108596977B (en) 2018-03-07 2018-03-07 Bionic eye binocular vision external parameter continuous calibration method

Country Status (1)

Country Link
CN (1) CN108596977B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110189382A (en) * 2019-05-31 2019-08-30 东北大学 A kind of more binocular cameras movement scaling method based on no zone of mutual visibility domain
CN112102401A (en) * 2020-09-21 2020-12-18 中国科学院上海微***与信息技术研究所 Target positioning method, device, system, equipment and storage medium

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103991006A (en) * 2014-04-01 2014-08-20 浙江大学 Calibration method and device for robot hole forming platform vision measurement system
CN104154875A (en) * 2014-08-20 2014-11-19 深圳大学 Three-dimensional data acquisition system and acquisition method based on two-axis rotation platform
CN107610086A (en) * 2016-07-12 2018-01-19 天津超众机器人科技有限公司 Industrial parallel robot rapid visual detection algorithm based on bionic compound eye structure

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103991006A (en) * 2014-04-01 2014-08-20 浙江大学 Calibration method and device for robot hole forming platform vision measurement system
CN104154875A (en) * 2014-08-20 2014-11-19 深圳大学 Three-dimensional data acquisition system and acquisition method based on two-axis rotation platform
CN107610086A (en) * 2016-07-12 2018-01-19 天津超众机器人科技有限公司 Industrial parallel robot rapid visual detection algorithm based on bionic compound eye structure

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
DI FAN ET AL.: "Design of Anthropomorphic Robot Bionic Eyes", 《PROCEEDINGS OF THE 2017 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND BIOMIMETICS》 *
KAI WELKE ET AL.: "Kinematic Calibration for Saccadic Eye Movements", 《ROBOTICS: SCIENCE AND SYSTEMS CONFERENCE》 *
M.KENDAL ACKERMAN AND GREGORY S. CHIRIKJIAN: "A Probabilistic Solution to the AX=XB Problem: Sensor CalibrationWithout Correspondence", 《INTERNATIONAL CONFERENCE ON GEOMETRIC SCIENCE OF INFORMATION》 *
QINGBIN WANG ET AL.: "Binocular Initial Location and Extrinsic Parameters Real-time Calculation for Bionic Eye System", 《PROCEEDING OF THE 11 TH WORLD CONGRESS ON INTELLIGENT CONTROL AND AUTOMATION》 *
李明超: "基于OpenCV的机器人双目手眼标定***的研究与实现", 《中国学位论文全文数据库》 *
李龙 等: "结构光视觉引导的轨迹跟踪***的标定技术", 《计算机工程与应用》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110189382A (en) * 2019-05-31 2019-08-30 东北大学 A kind of more binocular cameras movement scaling method based on no zone of mutual visibility domain
CN112102401A (en) * 2020-09-21 2020-12-18 中国科学院上海微***与信息技术研究所 Target positioning method, device, system, equipment and storage medium
CN112102401B (en) * 2020-09-21 2024-05-28 中国科学院上海微***与信息技术研究所 Target positioning method, device, system, equipment and storage medium

Also Published As

Publication number Publication date
CN108596977B (en) 2021-06-11

Similar Documents

Publication Publication Date Title
US10859840B2 (en) Graphics rendering method and apparatus of virtual reality
CN105118055B (en) Camera position amendment scaling method and system
CN110136208A (en) A kind of the joint automatic calibration method and device of Visual Servoing System
CN107862720A (en) Pose optimization method and pose optimization system based on the fusion of more maps
CN108436909A (en) A kind of hand and eye calibrating method of camera and robot based on ROS
US20070132863A1 (en) Image taking apparatus, image processing method, and image processing program
CN104202527B (en) For adjusting the device and method of camera lens in portable terminal
CN110764613A (en) Eye movement tracking calibration method based on head-mounted eye movement module
Ude et al. Foveated vision systems with two cameras per eye
WO2004039077A1 (en) Bionic vision and sight automatic control system and method
CN110188728A (en) A kind of method and system of head pose estimation
JP2016173313A (en) Visual line direction estimation system, visual line direction estimation method and visual line direction estimation program
CN106646903A (en) Follow-up head display device and method
CN108596977A (en) A kind of outer continuous parameters scaling method of bionic eye binocular vision
CN113793389B (en) Virtual-real fusion calibration method and device for augmented reality system
Atienza et al. Active gaze tracking for human-robot interaction
WO2020048461A1 (en) Three-dimensional stereoscopic display method, terminal device and storage medium
CN109885169A (en) Eyeball parameter calibration and direction of visual lines tracking based on three-dimensional eyeball phantom
JP4865065B1 (en) Stereo imaging device control system
CN110928311B (en) Indoor mobile robot navigation method based on linear features under panoramic camera
US20220365594A1 (en) Control method and apparatus for virtual reality device
Marchand et al. Visual servoing through mirror reflection
CN112288801A (en) Four-in-one self-adaptive tracking shooting method and device applied to inspection robot
CN110445982A (en) A kind of tracking image pickup method based on six degree of freedom equipment
CN111915739A (en) Real-time three-dimensional panoramic information interactive information system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
CB03 Change of inventor or designer information
CB03 Change of inventor or designer information

Inventor after: Chen Xiaopeng

Inventor after: Wang Changjin

Inventor after: Zhang Taoran

Inventor after: Tian Ye

Inventor after: Jin Xuanyou

Inventor after: Cheng Wei

Inventor after: Chen Xi

Inventor before: Chen Xiaopeng

Inventor before: Zhang Taoran

Inventor before: Wang Changjin

Inventor before: Tian Ye

Inventor before: Jin Xuanyou

Inventor before: Cheng Wei

Inventor before: Chen Xi

TA01 Transfer of patent application right
TA01 Transfer of patent application right

Effective date of registration: 20190201

Address after: 100081 No. 5, Zhongguancun South Street, Haidian District, Beijing

Applicant after: BEIJING INSTITUTE OF TECHNOLOGY

Address before: Room 1713, Block C, Weber Age Center, 17 South Street, Zhongguancun, Haidian District, Beijing, 100081

Applicant before: BEIJING HUAKAIHUI INFORMATION TECHNOLOGY Co.,Ltd.

CB03 Change of inventor or designer information

Inventor after: Chen Xiaopeng

Inventor after: Huang Qiang

Inventor after: Zhang Taoran

Inventor after: Wang Changjin

Inventor after: Tian Ye

Inventor after: Jin Xuanyou

Inventor after: Cheng Wei

Inventor after: Chen Xi

Inventor before: Chen Xiaopeng

Inventor before: Wang Changjin

Inventor before: Zhang Taoran

Inventor before: Tian Ye

Inventor before: Jin Xuanyou

Inventor before: Cheng Wei

Inventor before: Chen Xi

CB03 Change of inventor or designer information
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20210611