CN106485753B - The method and apparatus of camera calibration for pilotless automobile - Google Patents

The method and apparatus of camera calibration for pilotless automobile Download PDF

Info

Publication number
CN106485753B
CN106485753B CN201610811688.0A CN201610811688A CN106485753B CN 106485753 B CN106485753 B CN 106485753B CN 201610811688 A CN201610811688 A CN 201610811688A CN 106485753 B CN106485753 B CN 106485753B
Authority
CN
China
Prior art keywords
coordinate
coordinate system
camera
described image
video camera
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.)
Active
Application number
CN201610811688.0A
Other languages
Chinese (zh)
Other versions
CN106485753A (en
Inventor
曾庆喜
马杉
冯玉鹏
方啸
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.)
Dazhuo Intelligent Technology Co ltd
Dazhuo Quxing Intelligent Technology Shanghai Co ltd
Original Assignee
Chery Automobile 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 Chery Automobile Co Ltd filed Critical Chery Automobile Co Ltd
Priority to CN201610811688.0A priority Critical patent/CN106485753B/en
Publication of CN106485753A publication Critical patent/CN106485753A/en
Application granted granted Critical
Publication of CN106485753B publication Critical patent/CN106485753B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Studio Devices (AREA)
  • Closed-Circuit Television Systems (AREA)

Abstract

The invention discloses a kind of method and apparatus of camera calibration for pilotless automobile, belong to image procossing and computer vision skill field.Method includes: to acquire default working face by video camera, obtains image, and image includes the first known point and the second known point;Image pixel coordinates system is established in the picture;According to the optical axis and image of video camera, camera coordinate system is established;It determines the first position relationship between camera coordinate system and world coordinate system, the outer parameter of video camera is determined according to first position relationship;Obtain the first coordinate and the second coordinate of the first known point and the second known point respectively in world coordinate system;Determine the second position relationship of camera coordinate system and image pixel coordinates system;According to the first coordinate, the second coordinate and second position relationship, the intrinsic parameter of video camera is determined.Device includes: acquisition module, and first establishes module, and second establishes module, and the first determining module obtains module, the second determining module and third determining module.

Description

The method and apparatus of camera calibration for pilotless automobile
Technical field
It is the present invention relates to image procossing and technical field of computer vision, in particular to a kind of for pilotless automobile The method and apparatus of camera calibration.
Background technique
Continuous development and needs with human sciences' technology, video camera using more and more extensive, in social life side Face plays extremely important effect.For example, installing a video camera on pilotless automobile, car-mounted terminal carries out video camera mark After fixed, available pose of camera, and then determine the position of pilotless automobile.Wherein, camera calibration is from video camera The two dimensional image of shooting obtains the significant process of geological information in three-dimensional space.Geological information pair in two dimensional image and three-dimensional space Should be related to determined by the geometry and optical characteristics of video camera imaging, camera calibration is just to solve for the geometry and light of video camera Learn the process of characteristic, the i.e. intrinsic parameter of solution video camera and outer parameter.Wherein, camera intrinsic parameter describes inside video camera Geometry and optical characteristics, optical center, focal length including video camera etc., outer parameter, that is, video camera of video camera is in world coordinate system Position.
Currently, the method for camera calibration is that video camera needs to shoot at least two width figures of Same Scene from different perspectives Picture carries out SIFT (Scale-invariant feature transform, scale invariant feature at least two images respectively Transformation) characteristic point extraction and matching, obtain in two images correspond to the same three-dimensional space point multiple SIFT features Pixel coordinate calculated separately in multiple SIFT features according to the pixel coordinate and preset algorithm of multiple SIFT features The characteristic value of each SIFT feature, according to the characteristic value of each SIFT feature, determine the video camera intrinsic parameter and outer ginseng Number.
In the implementation of the present invention, the inventor finds that the existing technology has at least the following problems:
During above-mentioned camera calibration, video camera needs to shoot at least two images, also, to two images be intended into The extraction and matching of row SIFT feature, however the extraction of SIFT feature and matched computationally intensive is carried out to image, thus Cause camera calibration to take a long time, and then leads to camera calibration low efficiency.
Summary of the invention
In order to solve problems in the prior art, the present invention provides a kind of camera calibrations for pilotless automobile Method and apparatus.Technical solution is as follows:
In a first aspect, the present invention provides a kind of method of camera calibration for pilotless automobile, the method Include:
Default working face is acquired by video camera, obtains image, described image includes known to the first known point and second Point;
In described image, image pixel coordinates system is established;
According to the optical axis and described image of the video camera, camera coordinate system is established;
It determines the first position relationship between the camera coordinate system and world coordinate system, is closed according to the first position System, determines the outer parameter of the video camera;
Obtain first known point and second known point respectively the first coordinate in the world coordinate system and Second coordinate;
Determine the second position relationship between the camera coordinate system and described image pixel coordinate system;
According to first coordinate, second coordinate and the second position relationship, the internal reference of the video camera is determined Number.
In a kind of possible design, the optical axis and described image according to the video camera establishes camera coordinates System, comprising:
The optical axis center for obtaining the video camera, using the optical axis center as origin, with the horizontal direction right of described image To for X-axis positive direction, to be parallel to the optical axis of the video camera, and it is directed toward by the video camera side of the default working face The camera coordinates are established to for Z axis positive direction with direction is Y-axis positive direction perpendicular to described image, and straight up System.
It is described that the outer parameter of the video camera is determined according to the first position relationship in a kind of possible design, packet It includes:
Determine any first position point of described image in institute by following formula one according to the first position relationship State the relationship of the coordinate and coordinate of the first position point in the camera coordinate system in world coordinate system:
Formula one:
Wherein, MoutFor the outer parameter of the video camera, (Xw, Yw, Zw) for described image any first position point in institute State the coordinate in world coordinate system, (Xc, Yc, Zc) it is coordinate of the first position point in the camera coordinate system, R is Spin matrix, p are translation matrix.
In a kind of possible design, between the determination camera coordinate system and described image pixel coordinate system Second position relationship, comprising:
Image physical coordinates system is established, determines between described image pixel coordinate system and described image physical coordinates system Three positional relationships;
According to the third place relationship, by following formula two, determine that any second position point in described image exists The relationship of coordinate and coordinate of the second position point in the camera coordinate system in described image pixel coordinate system:
Formula two:
Wherein, (uc, vc) it is coordinate of any second position point in described image pixel coordinate system in described image, (Xc, Yc, Zc) it is coordinate of the second position point in the camera coordinate system, kxFor the first amplification coefficient, kyIt is second Amplification coefficient, (u0, v0) it is coordinate of the coordinate origin of described image physical coordinates system in described image pixel coordinate system.
It is described to be closed according to first coordinate, second coordinate and the second position in a kind of possible design System, determines the intrinsic parameter of the video camera, comprising:
According to first coordinate, second coordinate and the second position relationship, obtained known to described first respectively Point and third coordinate and 4-coordinate of second known point in described image pixel coordinate system;
According to first coordinate, second coordinate, the third coordinate and the 4-coordinate, pass through following formula Three, determine the intrinsic parameter of the video camera:
Formula three:
Wherein, (Xw1, Yw1, 0) and it is first coordinate of first known point in the world coordinate system, (Xw2, Yw2, 0) For second coordinate of second known point in the world coordinate system, (u1, v1) it is first known point in described image Third coordinate in pixel coordinate system, (u2, v2) it is second known point 4-coordinate in described image pixel coordinate system.
Second aspect, the present invention provides a kind of device of camera calibration for pilotless automobile, described devices Include:
Acquisition module obtains image, described image includes known to first for acquiring default working face by video camera Point and the second known point;
First establishes module, for establishing image pixel coordinates system in described image;
Second establishes module, for the optical axis and described image according to the video camera, establishes camera coordinate system;
First determining module, for determining the first position relationship between the camera coordinate system and world coordinate system, According to the first position relationship, the outer parameter of the video camera is determined;
Module is obtained, for obtaining first known point and second known point respectively in the world coordinate system The first coordinate and the second coordinate;
Second determining module, for determining the second between the camera coordinate system and described image pixel coordinate system Set relationship;
Third determining module, for determining according to first coordinate, second coordinate and the second position relationship The intrinsic parameter of the video camera.
In a kind of possible design, described second establishes module, is also used to obtain the optical axis center of the video camera, with The optical axis center is origin, using the direction horizontally to the right of described image as X-axis positive direction, to be parallel to the light of the video camera Axis, and being directed toward the direction of the default working face by the video camera is Z axis positive direction, with perpendicular to described image, and it is perpendicular Straight upward direction is Y-axis positive direction, establishes the camera coordinate system.
In a kind of possible design, first determining module is also used to according to the first position relationship, by with Lower formula one determines coordinate of any first position point of described image in the world coordinate system and the first position point The relationship of coordinate in the camera coordinate system:
Formula one:
Wherein, MoutFor the outer parameter of the video camera, (Xw, Yw, Zw) for described image any first position point in institute State the coordinate in world coordinate system, (Xc, Yc, Zc) it is coordinate of the first position point in the camera coordinate system, R is Spin matrix, p are translation matrix.
In a kind of possible design, second determining module, comprising:
First determination unit determines described image pixel coordinate system and described image for establishing image physical coordinates system The third place relationship between physical coordinates system;
Second determination unit, for being determined in described image according to the third place relationship by following formula two Coordinate of any second position point in described image pixel coordinate system and the second position point are in the camera coordinate system In coordinate relationship:
Formula two:
Wherein, (uc, vc) it is coordinate of any second position point in described image pixel coordinate system in described image, (Xc, Yc, Zc) it is coordinate of the second position point in the camera coordinate system, kxFor the first amplification coefficient, kyIt is second Amplification coefficient, (u0, v0) it is coordinate of the coordinate origin of described image physical coordinates system in described image pixel coordinate system.
In a kind of possible design, the third determining module, comprising:
Acquiring unit, for being obtained respectively according to first coordinate, second coordinate and the second position relationship The third coordinate and 4-coordinate of first known point and second known point in described image pixel coordinate system;
Third determination unit, for according to first coordinate, second coordinate, the third coordinate and the described 4th Coordinate determines the intrinsic parameter of the video camera by following formula three:
Formula three:
Wherein, (Xw1, Yw1, 0) and it is first coordinate of first known point in the world coordinate system, (Xw2, Yw2, 0) For second coordinate of second known point in the world coordinate system, (u1, v1) it is first known point in described image Third coordinate in pixel coordinate system, (u2, v2) it is second known point 4-coordinate in described image pixel coordinate system.
In the present invention, the image that computer is shot by obtaining video camera establishes camera coordinate system, image pixel Coordinate system and image physical coordinates system obtain the first coordinate of the first known point and the second known point in world coordinate system respectively With the third coordinate and 4-coordinate of the second coordinate and the first known point and the second known point in image pixel coordinates system, It will substitute into the formula determined according to the positional relationship between camera coordinate system, world coordinate system and image pixel coordinates system, The outer parameter and intrinsic parameter that the video camera can be found out, are omitted cumbersome calculating process, simple to operation, improve video camera mark Fixed working efficiency.
Detailed description of the invention
Fig. 1 is a kind of method flow diagram of camera calibration for pilotless automobile provided in an embodiment of the present invention;
Fig. 2 is a kind of method flow diagram of camera calibration for pilotless automobile provided in an embodiment of the present invention;
Fig. 3 is a kind of image pixel coordinates system provided in an embodiment of the present invention and image physical coordinates system schematic diagram;
Fig. 4 is a kind of camera coordinate system and world coordinate system schematic diagram provided in an embodiment of the present invention;
Fig. 5 is a kind of apparatus structure signal of camera calibration for pilotless automobile provided in an embodiment of the present invention Figure.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to embodiment party of the present invention Formula is described in further detail.
The embodiment of the invention provides a kind of method of camera calibration for pilotless automobile, the execution of this method Main body can be computer, and computer can be car-mounted terminal.
Referring to Fig. 1, this method comprises:
Step 101: default working face being acquired by video camera, obtains image, image includes the first known point and second Known point.
Step 102: in the images, establishing image pixel coordinates system.
Step 103: according to the optical axis and image of video camera, establishing camera coordinate system.
Step 104: the first position relationship between the camera coordinate system and world coordinate system is determined, according to this first Relationship is set, determines the outer parameter of the video camera.
Step 105: obtain first known point and second known point respectively the first coordinate in world coordinate system and Second coordinate.
Step 106: determining the second position relationship between the camera coordinate system and the image pixel coordinates system.
Step 107: according to first coordinate, second coordinate and the second position relationship, determining the intrinsic parameter of video camera.
Optionally, according to the optical axis of video camera and image, camera coordinate system is established, comprising:
The optical axis center for obtaining video camera, using optical axis center as origin, using the direction horizontally to the right of image as X-axis pros To, to be parallel to the optical axis of video camera, and be Z axis positive direction by the direction that video camera is directed toward default working face, with perpendicular to Image, and direction is Y-axis positive direction straight up, establishes camera coordinate system.
Optionally, according to first position relationship, the outer parameter of video camera is determined, comprising:
Determine any first position point of described image in institute by following formula one according to the first position relationship State the relationship of the coordinate and coordinate of the first position point in the camera coordinate system in world coordinate system:
Formula one:
Wherein, MoutFor the outer parameter of video camera, (Xw, Yw, Zw) it is any first position point of image in world coordinate system In coordinate, (Xc, Yc, Zc) it is coordinate of the first position point in camera coordinate system, R is spin matrix, and p is translation matrix.
Optionally, the second position relationship between camera coordinate system and image pixel coordinates system is determined, comprising:
Image physical coordinates system is established, determines that the third place between image pixel coordinates system and image physical coordinates system is closed System;
According to the third place relationship, by following formula two, determine any second position point in image in image pixel The relationship of coordinate and coordinate of the second position point in camera coordinate system in coordinate system:
Formula two:
Wherein, (uc, vc) it is coordinate of any second position point in image pixel coordinates system in image, (Xc, Yc, Zc) For coordinate of the second position point in camera coordinate system, kxFor the first amplification coefficient, kyFor the second amplification coefficient, (u0, v0) be Coordinate of the coordinate origin of image physical coordinates system in image pixel coordinates system.
Optionally, according to the first coordinate, the second coordinate and second position relationship, the intrinsic parameter of video camera is determined, comprising:
According to the first coordinate, the second coordinate and second position relationship, the first known point is obtained respectively and the second known point exists Third coordinate and 4-coordinate in image pixel coordinates system;
Video camera is determined by following formula three according to the first coordinate, the second coordinate, third coordinate and 4-coordinate Intrinsic parameter:
Formula three:
Wherein, (Xw1, Yw1, 0) and it is first coordinate of first known point in world coordinate system, (Xw2, Yw2, 0) and it is second Know the second coordinate a little in world coordinate system, (u1, v1) it is third coordinate of first known point in image pixel coordinates system, (u2, v2) it is the second known point 4-coordinate in image pixel coordinates system.
In the embodiment of the present invention, the image that computer is shot by obtaining video camera establishes camera coordinate system, figure As pixel coordinate system and image physical coordinates system, the of the first known point and the second known point in world coordinate system is obtained respectively The third coordinate and the 4th of one coordinate and the second coordinate and the first known point and the second known point in image pixel coordinates system Coordinate will substitute into the public affairs determined according to the positional relationship between camera coordinate system, world coordinate system and image pixel coordinates system In formula, the outer parameter and intrinsic parameter of the video camera can be found out, cumbersome calculating process is omitted, it is simple to operation, it improves and takes the photograph The working efficiency of camera calibration.
The embodiment of the invention provides a kind of method of camera calibration for pilotless automobile, the execution of this method Main body can be computer, and computer can be car-mounted terminal.
Referring to fig. 2, this method comprises:
Step 201: computer acquires default working face by video camera, obtains image, which includes known to first Point and the second known point.
It should be noted that default working face is the working face of the intrinsic parameter for determining video camera and outer parameter, Include at least two position known points in default working face, is referred to as the first scene point and the second scene point.Camera light Axis is vertical with the default working face, and the position of camera position and the default working face is fixed, and is shot by video camera The default working face, obtains image;Computer obtains the image from the video camera.
Wherein, the first scene point, the second scene point in working face are preset and respectively corresponds the first known point in the image With the second known point.
Step 202: in the images, computer establishes image pixel coordinates system.
Referring to Fig. 3, computer obtains image top left corner apex o0, with top left corner apex o0For image pixel coordinates system Coordinate origin, using direction horizontally to the right as the u axis positive direction of image pixel coordinates system, using direction vertically downward as image The v axis positive direction of pixel coordinate system, establishes image pixel coordinates system uo0v。
Further, since the position for presetting working face and video camera is fixed, the computer acquisition preset plane is being taken the photograph After image in camera, camera coordinate system need to be first established, according to the positional relationship of camera coordinate system and world coordinate system, really The outer parameter for determining video camera is realized especially by following steps 203-204.
Step 203: computer establishes camera coordinate system according to the optical axis and the image of video camera.
Specifically, computer obtains the optical axis center of video camera, using optical axis center as origin, with the horizontal direction right of image It to be parallel to the optical axis of video camera, and is that Z axis is square by the direction that video camera is directed toward default working face to for X-axis positive direction To with perpendicular to image, and direction is Y-axis positive direction straight up, establishes camera coordinate system.
For the ease of being distinguished with subsequent world coordinate system, the X-axis, Y-axis and Z axis of camera coordinate system are claimed respectively For XcAxis, YcAxis and ZcAxis;The X-axis, Y-axis and Z axis of world coordinate system are referred to as XwAxis, YwAxis and ZwAxis.
Referring to fig. 4, computer obtains the optical axis center O of the video cameracAnd pass through camera light axis center OcLight Axis, optical axis center OcThe center of lens i.e. in the video camera, computer is with OcFor the origin of camera coordinate system, with the water of image Flat right direction is the X of camera coordinate systemcAxis positive direction to be parallel to the optical axis of video camera, and is directed toward by video camera and is preset The direction of working face is the Z of camera coordinate systemcAxis positive direction, with direction is video camera perpendicular to image, and straight up The Y of coordinate systemcAxis positive direction establishes camera coordinate system OcXcYcZc
Referring to fig. 4, computer obtains the optical axis of video camera and the intersection point O of default working facew, obtain the camera coordinates The X of systemcAxis positive direction, YcAxis positive direction, ZcAxis positive direction chooses OwFor the origin of world coordinate system, camera coordinate system is chosen XcAxis positive direction, YcAxis positive direction, ZcAxis positive direction is the X of world coordinate systemwAxis positive direction, YwAxis positive direction, ZwAxis is square To determining world coordinate system OwXwYwZw
Wherein, the X in camera coordinate systemcAxis, YcAxis is respectively parallel to the X in world coordinate systemwAxis, YwAxis, video camera Z in coordinate systemcZ in axis and world coordinate systemwAxis is on camera optical axis, and direction is consistent.
Step 204: computer determines the first position relationship between camera coordinate system and world coordinate system, according to first Positional relationship determines the outer parameter of video camera.
It should be noted that since camera optical axis is perpendicular to default working face, and video camera and default working face Position immobilizes, and therefore, the outer parameter of video camera determines, step 204 can be with are as follows:
Computer determines any first position point of described image by following formula one according to the first position relationship In the relationship of coordinate and coordinate of the first position point in the camera coordinate system in the world coordinate system:
Formula one:
Wherein, MoutFor the outer parameter of video camera, (Xw, Yw, Zw) it is any first position point of image in world coordinate system In coordinate, (Xc, Yc, Zc) it is coordinate of the first position point in camera coordinate system, R is spin matrix, and p is translation matrix.
It should be noted that the outer parameter of video camera is imaged for describing position of the video camera in world coordinate system The first position relationship of machine coordinate system and world coordinate system, since camera coordinate system and world coordinate system determine, camera shooting The outer parameter of machine determines, specifically, the X known to step 203 in camera coordinate systemcAxis, YcAxis is respectively parallel to world coordinates X in systemwAxis, YwAxis, and the X in camera coordinate systemcAxis, YcX in axis and world coordinate systemwAxis, YwAxis direction is consistent, takes the photograph Z in camera coordinate systemcZ in axis and world coordinate systemwAxis is on camera optical axis, and direction is consistent, therefore, video camera Outer parameterWherein, R is spin matrix,P is translation Matrix, p=[00d], d are distance of the optical axis center to working face.
Therefore, coordinate (X of any first position point in world coordinate system in image can be obtainedw, Yw, Zw) and first Set the coordinate (X a little in camera coordinate systemc, Yc, Zc) relationship are as follows:
Further, after computer determines the outer parameter of video camera, the intrinsic parameter for determining video camera is also needed, in video camera Parameter describes the optics and geometrical property of video camera, the i.e. focal length of video camera, optical axis center etc., due to the internal reference of video camera Several related with the internal structure of video camera, and therefore, computer determines the second of camera coordinate system and image pixel coordinates system Positional relationship obtains coordinate of the first known point, the second known point in default working face in world coordinate system, according to this Second position relationship and the coordinate, and then determine the intrinsic parameter of video camera, specifically, can be real by following steps 205-207 It is existing.
Step 205: computer obtains the first known point and the second known point the first coordinate in world coordinate system respectively With the second coordinate.
For example, computer obtains the first coordinate of the first known point and the second known point in world coordinate system respectively is (Xw1, Yw1, 0), the second coordinate be (Xw2, Yw2, 0).
Step 206: computer determines the second position relationship between camera coordinate system and image pixel coordinates system.
This step is realized by following steps 2061-2062.
Step 2061: computer establishes image physical coordinates system, determines image pixel coordinates system and image physical coordinates system Between the third place relationship.
It should be noted that in the corresponding image of default working face that computer obtains, due to image pixel coordinates It is uo0The u axis and v axis of v as unit of the pixel of the image, i.e., the coordinate at any point in image pixel coordinates system (u, V) only indicate that the columns and line number that the point is located in array of pixels, computer cannot describe the picture by the image pixel coordinates system The physical location of vegetarian refreshments in the picture.It, should therefore, it is necessary to establish the image physical coordinates system indicated with physical unit in the picture The unit of reference axis is millimeter in image physical coordinates system.
The unit of reference axis, which can according to need, in the image physical coordinates system is configured and changes, and implements in the present invention The unit of reference axis in the image physical coordinates system is not especially limited in example;For example, coordinate in the image physical coordinates system The unit of axis can also be centimetre, micron etc..
Wherein, computer establish the step of image physical coordinates system can be with are as follows:
Referring to Fig. 3, computer obtains the optical axis center of video camera and the intersection point o of the image1, with o1For image physical coordinates The origin of system is sat using the u axis positive direction of image pixel coordinates system as the positive direction of image physical coordinates system x-axis with image pixel The v axis positive direction of mark system is the positive direction of image physical coordinates system y-axis, establishes image physical coordinates system xo1y。
It therefore, can be by determining coordinate (u, v) of any the third place point in image in image pixel coordinates system With the relationship of the coordinate (x, y) of the third place point in image physical coordinates system, and then image pixel coordinates system and figure are determined As the third place relationship between physical coordinates system is as follows:
Wherein, αxAnd αyRespectively image from image pixel coordinates system into image physical coordinates system in image physical coordinates The sampling frequency of the x-axis of system and the amplification factor in y-axis direction, i.e. video camera in the x-axis and y-axis direction of image physical coordinates system Rate, (u0, v0) it is coordinate of the coordinate origin of image physical coordinates system in image pixel coordinates system, meanwhile, (u0, v0) It is coordinate of the camera light axis center in image pixel coordinates system.
Step 2062: computer is according to the third place relationship, by following formula two, determines any second in image Set the relationship of the coordinate and coordinate of the second position point in camera coordinate system a little in image pixel coordinates system:
Formula two:
Wherein, (uc, vc) it is coordinate of any second position point in image pixel coordinates system in image, (Xc, Yc, Zc) For coordinate of the second position point in camera coordinate system, kxFor the first amplification coefficient, kyFor the second amplification coefficient, (u0, v0) be Coordinate of the coordinate origin of image physical coordinates system in image pixel coordinates system.
It should be noted that computer obtains the imaging geometry model of video camera, i.e. pin-point model, according to the small casement Type determines that camera coordinate system and the proportionate relationship of image physical coordinates system are as follows:Wherein, f For the focal length of video camera.
It willIt substitutes intoIn, by determining in image The pass of any second position point coordinate in image pixel coordinates system and coordinate of the second position point in camera coordinate system System, in turn, determines that the second position relationship between camera coordinate system and image pixel coordinates system is as follows:
Formula two:
Wherein, (uc, vc) it is coordinate of any second position point in image pixel coordinates system in image, (Xc, Yc, Zc) For coordinate of the second position point in camera coordinate system, kxFor the first amplification coefficient, kyFor the second amplification coefficient, (u0, v0) be Coordinate of the coordinate origin of image physical coordinates system in image pixel coordinates system.
207: computer determines the intrinsic parameter of video camera according to the first coordinate, the second coordinate and second position relationship.
It should be noted that computer obtains the first known point and the second known point respectively and sits in the world by step 205 The first coordinate (X in mark systemw1, Yw1, 0), the second coordinate (Xw2, Yw2, 0), by step 206, computer obtains camera coordinates Second position relationship between system and image pixel coordinates system, then, computer is according to the first coordinate, the second coordinate, second Relationship is set, determines camera intrinsic parameter.Specifically, this step can be realized by following steps 2071-2072:
Step 2071: computer obtains the first known point according to the first coordinate, the second coordinate and second position relationship respectively With third coordinate and 4-coordinate of second known point in image pixel coordinates system;
Computer is according to the first coordinate (Xw1, Yw1, 0) and the second coordinate (Xw2, Yw2, 0), respectively determine the first known point and The first position and the second position of second known point in the picture;According to first position and the second position, the first known point is determined With third coordinate (u of second known point in image pixel coordinates system1, v1) and 4-coordinate (u2, v2)。
Step 2072: computer passes through following formula according to the first coordinate, the second coordinate, third coordinate and 4-coordinate Three, determine the intrinsic parameter of video camera:
Formula three:
Wherein, (Xw1, Yw1, 0) and it is first coordinate of first known point in world coordinate system, (Xw2, Yw2, 0) and it is second Know the second coordinate a little in world coordinate system, (u1, v1) it is third coordinate of first known point in image pixel coordinates system, (u2, v2) it is the second known point 4-coordinate in image pixel coordinates system.
It should be noted that since computer has obtained the first known point and the second known point in world coordinate system respectively In the first coordinate (Xw1, Yw1, 0), the second coordinate (Xw2, Yw2, 0), the first known point and the second known point are in image pixel coordinates Third coordinate (u in system1, v1), 4-coordinate (u2, v2), therefore, by the first coordinate (Xw1, Yw1, 0), the second coordinate (Xw2, Yw2, 0), third coordinate (u1, v1) and 4-coordinate (u2, v2) substitute into formula three respectively, that is, it can determine the intrinsic parameter of video camera.It takes the photograph The intrinsic parameter of camera includes: the first amplification coefficient kx, the second amplification coefficient ky, camera light axis center is in image pixel coordinates system In coordinate (u0, v0)。
Wherein, the first amplification coefficient kx, the second amplification coefficient kyIt is respectively as follows: kxx× f, kyy× f, f are video camera Focal length, therefore, kx、kyFor describing the focal length in external parameters of cameras.
In the embodiment of the present invention, the image that computer is shot by obtaining video camera establishes camera coordinate system, figure As pixel coordinate system and image physical coordinates system, the of the first known point and the second known point in world coordinate system is obtained respectively The third coordinate and the 4th of one coordinate and the second coordinate and the first known point and the second known point in image pixel coordinates system Coordinate will substitute into the public affairs determined according to the positional relationship between camera coordinate system, world coordinate system and image pixel coordinates system In formula, the outer parameter and intrinsic parameter of the video camera can be found out, cumbersome calculating process is omitted, it is simple to operation, it improves and takes the photograph The working efficiency of camera calibration.
The embodiment of the invention provides a kind of device of camera calibration for pilotless automobile, which can be answered With on computers, computer can be car-mounted terminal.Method for executing above-mentioned camera calibration.
Referring to Fig. 5, which includes:
Acquisition module 301 obtains image, image includes known to first for acquiring default working face by video camera Point and the second known point;
First establishes module 302, in the picture, establishing image pixel coordinates system;
Second establishes module 303, for the optical axis and image according to video camera, establishes camera coordinate system;
First determining module 304, for determining the first position relationship between camera coordinate system and world coordinate system, root According to first position relationship, the outer parameter of video camera is determined;
Module 305 is obtained, for obtaining the first seat of the first known point and the second known point respectively in world coordinate system Mark and the second coordinate;
Second determining module 306, for determining that the second position between camera coordinate system and image pixel coordinates system is closed System;
Third determining module 307, for determining video camera according to the first coordinate, the second coordinate and second position relationship Intrinsic parameter.
Optionally, second module 303 is established, is also used to obtain the optical axis center of video camera, using optical axis center as origin, with The direction horizontally to the right of image is X-axis positive direction, to be parallel to the optical axis of video camera, and is directed toward default working face by video camera Direction be Z axis positive direction, with perpendicular to image, and direction is Y-axis positive direction straight up, establishes camera coordinate system.
Optionally, the first determining module 304 is also used to, by following formula one, determine institute according to the first position relationship Coordinate of any first position point of image in the world coordinate system and the first position point is stated to sit in the video camera The relationship of coordinate in mark system:
Formula one:
Wherein, MoutFor the outer parameter of video camera, (Xw, Yw, Zw) it is any first position point of image in world coordinate system In coordinate, (Xc, Yc, Zc) it is coordinate of the first position point in camera coordinate system, R is spin matrix, and p is translation matrix.
Optionally, the second determining module 306, comprising:
First determination unit determines image pixel coordinates system and image physical coordinates for establishing image physical coordinates system The third place relationship between system;
Second determination unit, for by following formula two, determining any second in image according to the third place relationship The relationship of coordinate of the location point in image pixel coordinates system and coordinate of the second position point in camera coordinate system:
Formula two:
Wherein, (uc, vc) it is coordinate of any second position point in image pixel coordinates system in image, (Xc, Yc, Zc) For coordinate of the second position point in camera coordinate system, kxFor the first amplification coefficient, kyFor the second amplification coefficient, (u0, v0) be Coordinate of the coordinate origin of image physical coordinates system in image pixel coordinates system.
Optionally, third determining module 307, comprising:
Acquiring unit, for according to the first coordinate, the second coordinate and second position relationship, obtain respectively the first known point and Third coordinate and 4-coordinate of second known point in image pixel coordinates system;
Third determination unit, for passing through following public affairs according to the first coordinate, the second coordinate, third coordinate and 4-coordinate Formula three determines the intrinsic parameter of video camera:
Formula three:
Wherein, (Xw1, Yw1, 0) and it is first coordinate of first known point in world coordinate system, (Xw2, Yw2, 0) and it is second Know the second coordinate a little in world coordinate system, (u1, v1) it is third coordinate of first known point in image pixel coordinates system, (u2, v2) it is the second known point 4-coordinate in image pixel coordinates system.
In the embodiment of the present invention, the image that computer is shot by obtaining video camera establishes camera coordinate system, figure As pixel coordinate system and image physical coordinates system, the of the first known point and the second known point in world coordinate system is obtained respectively The third coordinate and the 4th of one coordinate and the second coordinate and the first known point and the second known point in image pixel coordinates system Coordinate will substitute into the public affairs determined according to the positional relationship between camera coordinate system, world coordinate system and image pixel coordinates system In formula, the outer parameter and intrinsic parameter of the video camera can be found out, cumbersome calculating process is omitted, it is simple to operation, it improves and takes the photograph The working efficiency of camera calibration.
It should be understood that the device of the camera calibration provided by the above embodiment for pilotless automobile is being used for It, only the example of the division of the above functional modules, can in practical application when the camera calibration of pilotless automobile To be as needed completed by different functional modules above-mentioned function distribution, i.e., the internal structure of device is divided into different Functional module, to complete all or part of the functions described above.In addition, provided by the above embodiment be used for unmanned vapour The device of the camera calibration of vehicle and the embodiment of the method for the camera calibration for pilotless automobile belong to same design, Specific implementation process is detailed in embodiment of the method, and which is not described herein again.
Those of ordinary skill in the art will appreciate that realizing that all or part of the steps of above-described embodiment can pass through hardware It completes, relevant hardware can also be instructed to complete by program, the program can store in a kind of computer-readable In storage medium, storage medium mentioned above can be read-only memory, disk or CD etc..
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all in spirit of the invention and Within principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.

Claims (10)

1. a kind of method of the camera calibration for pilotless automobile, which is characterized in that the described method includes:
Default working face is acquired by video camera, obtains image, described image includes the first known point and the second known point;
In described image, image pixel coordinates system is established;
According to the optical axis and described image of the video camera, camera coordinate system is established;
Determine the first position relationship between the camera coordinate system and world coordinate system, according to the first position relationship, Determine the outer parameter of the video camera;
Obtain first known point and second known point the first coordinate and second in the world coordinate system respectively Coordinate;
Determine the second position relationship between the camera coordinate system and described image pixel coordinate system;
According to first coordinate, second coordinate and the second position relationship, the intrinsic parameter of the video camera is determined.
2. the method according to claim 1, wherein the optical axis and described image according to the video camera, Establish camera coordinate system, comprising:
The optical axis center for obtaining the video camera is X with the direction horizontally to the right of described image using the optical axis center as origin Axis positive direction to be parallel to the optical axis of the video camera, and is Z by the direction that the video camera is directed toward the default working face Axis positive direction establishes the camera coordinate system so that direction is Y-axis positive direction perpendicular to described image, and straight up.
3. the method according to claim 1, wherein described according to the first position relationship, determine described in take the photograph The outer parameter of camera, comprising:
Choose the optical axis of the video camera and the intersection point O of default working facewFor the origin of the world coordinate system, the camera shooting X in machine coordinate systemcAxis, YcAxis is respectively parallel to the X in the world coordinate systemwAxis, YwAxis, and the camera coordinate system In XcAxis, YcX in axis and the world coordinate systemwAxis, YwAxis direction is consistent, the Z in the camera coordinate systemcAxis and institute State the Z in world coordinate systemwAxis is on the optical axis of the video camera, and direction is consistent;
Determine any first position point of described image in the generation by following formula one according to the first position relationship The relationship of coordinate and coordinate of the first position point in the camera coordinate system in boundary's coordinate system:
Formula one:
Wherein, MoutFor the outer parameter of the video camera, (Xw, Yw, Zw) it is any first position point of described image in the generation Coordinate in boundary's coordinate system, (Xc, Yc, Zc) it is coordinate of the first position point in the camera coordinate system, R is rotation Matrix, p are translation matrix,P=[0 0 d], D be the video camera optical axis center to the default working face distance.
4. the method according to claim 1, wherein the determination camera coordinate system and described image picture Second position relationship between plain coordinate system, comprising:
Image physical coordinates system is established, determines the third position between described image pixel coordinate system and described image physical coordinates system Set relationship;
According to the third place relationship, by following formula two, determine any second position point in described image described The relationship of coordinate and coordinate of the second position point in the camera coordinate system in image pixel coordinates system:
Formula two:
Wherein, (uc, vc) it is coordinate of any second position point in described image pixel coordinate system in described image, (Xc, Yc, Zc) it is coordinate of the second position point in the camera coordinate system, kxFor the first amplification coefficient, kyFor the second amplification Coefficient, (u0, v0) it is coordinate of the coordinate origin of described image physical coordinates system in described image pixel coordinate system.
5. the method according to claim 1, wherein it is described according to first coordinate, second coordinate and The second position relationship, determines the intrinsic parameter of the video camera, comprising:
According to first coordinate, second coordinate and the second position relationship, obtain respectively first known point and Third coordinate and 4-coordinate of second known point in described image pixel coordinate system;
According to first coordinate, second coordinate, the third coordinate and the 4-coordinate, by following formula three, Determine the intrinsic parameter of the video camera:
Formula three:
Wherein, (Xw1, Yw1, 0) and it is first coordinate of first known point in the world coordinate system, (Xw2, Yw2, 0) and for institute State second coordinate of second known point in the world coordinate system, (u1, v1) it is first known point in described image pixel Third coordinate in coordinate system, (u2, v2) it is second known point 4-coordinate in described image pixel coordinate system.
6. a kind of device of the camera calibration for pilotless automobile, which is characterized in that described device includes:
Acquisition module obtains image for acquiring default working face by video camera, described image include the first known point with Second known point;
First establishes module, for establishing image pixel coordinates system in described image;
Second establishes module, for the optical axis and described image according to the video camera, establishes camera coordinate system;
First determining module, for determining the first position relationship between the camera coordinate system and world coordinate system, according to The first position relationship determines the outer parameter of the video camera;
Module is obtained, for obtaining first known point and second known point the respectively in the world coordinate system One coordinate and the second coordinate;
Second determining module, for determining that the second position between the camera coordinate system and described image pixel coordinate system is closed System;
Third determining module, described in determining according to first coordinate, second coordinate and the second position relationship The intrinsic parameter of video camera.
7. device according to claim 6, which is characterized in that
Described second establishes module, is also used to obtain the optical axis center of the video camera, using the optical axis center as origin, with institute The direction horizontally to the right for stating image is X-axis positive direction, to be parallel to the optical axis of the video camera, and is directed toward institute by the video camera The direction for stating default working face is Z axis positive direction, with perpendicular to described image, and straight up, direction is Y-axis positive direction, is built Found the camera coordinate system.
8. device according to claim 6, which is characterized in that
First determining module is also used to choose the optical axis of the video camera and the intersection point O of default working facewFor the generation The origin of boundary's coordinate system, Xc axis, Y in the camera coordinate systemcAxis is respectively parallel to the X in the world coordinate systemwAxis, YwAxis, and Xc axis, Y in the camera coordinate systemcX in axis and the world coordinate systemwAxis, YwAxis direction is consistent, described Z in camera coordinate systemcZ in axis and the world coordinate systemwAxis is on the camera optical axis, and direction is consistent;Root Determine any first position point of described image in the world coordinates by following formula one according to the first position relationship The relationship of coordinate and coordinate of the first position point in the camera coordinate system in system:
Formula one:
Wherein, MoutFor the outer parameter of the video camera, (Xw, Yw, Zw) it is any first position point of described image in the generation Coordinate in boundary's coordinate system, (Xc, Yc, Zc) it is coordinate of the first position point in the camera coordinate system, R is rotation Matrix, p are translation matrix,P=[0 0 d], D be the video camera optical axis center to the default working face distance.
9. device according to claim 6, which is characterized in that second determining module, comprising:
First determination unit determines described image pixel coordinate system and described image physics for establishing image physical coordinates system The third place relationship between coordinate system;
Second determination unit, for by following formula two, determining any in described image according to the third place relationship Second position point in described image pixel coordinate system coordinate and the second position point in the camera coordinate system The relationship of coordinate:
Formula two:
Wherein, (uc, vc) it is coordinate of any second position point in described image pixel coordinate system in described image, (Xc, Yc, Zc) it is coordinate of the second position point in the camera coordinate system, kxFor the first amplification coefficient, kyFor the second amplification Coefficient, (u0, v0) it is coordinate of the coordinate origin of described image physical coordinates system in described image pixel coordinate system.
10. device according to claim 6, which is characterized in that the third determining module, comprising:
Acquiring unit, described in being obtained respectively according to first coordinate, second coordinate and the second position relationship The third coordinate and 4-coordinate of first known point and second known point in described image pixel coordinate system;
Third determination unit, for being sat according to first coordinate, second coordinate, the third coordinate and the described 4th Mark, by following formula three, determines the intrinsic parameter of the video camera:
Formula three:
Wherein, (Xw1, Yw1, 0) and it is first coordinate of first known point in the world coordinate system, (Xw2, Yw2, 0) and for institute State second coordinate of second known point in the world coordinate system, (u1, v1) it is first known point in described image pixel Third coordinate in coordinate system, (u2, v2) it is second known point 4-coordinate in described image pixel coordinate system.
CN201610811688.0A 2016-09-09 2016-09-09 The method and apparatus of camera calibration for pilotless automobile Active CN106485753B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610811688.0A CN106485753B (en) 2016-09-09 2016-09-09 The method and apparatus of camera calibration for pilotless automobile

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610811688.0A CN106485753B (en) 2016-09-09 2016-09-09 The method and apparatus of camera calibration for pilotless automobile

Publications (2)

Publication Number Publication Date
CN106485753A CN106485753A (en) 2017-03-08
CN106485753B true CN106485753B (en) 2019-09-10

Family

ID=58273585

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610811688.0A Active CN106485753B (en) 2016-09-09 2016-09-09 The method and apparatus of camera calibration for pilotless automobile

Country Status (1)

Country Link
CN (1) CN106485753B (en)

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107364393A (en) * 2017-05-25 2017-11-21 纵目科技(上海)股份有限公司 Display methods, device, storage medium and the electronic equipment of vehicle rear view image
CN113822939A (en) 2017-07-06 2021-12-21 华为技术有限公司 Method and device for calibrating external parameters of vehicle-mounted sensor
CN108416810A (en) * 2018-02-05 2018-08-17 北京邦天信息技术有限公司 Conversion method and device of the camera coordinates system to world coordinate system in camera calibration
CN109242779B (en) * 2018-07-25 2023-07-18 北京中科慧眼科技有限公司 Method and device for constructing camera imaging model and automobile automatic driving system
CN110163914B (en) * 2018-08-01 2021-05-25 京东方科技集团股份有限公司 Vision-based positioning
CN109472829B (en) * 2018-09-04 2022-10-21 顺丰科技有限公司 Object positioning method, device, equipment and storage medium
CN111243029B (en) * 2018-11-28 2023-06-23 驭势(上海)汽车科技有限公司 Calibration method and device of vision sensor
CN109540144A (en) * 2018-11-29 2019-03-29 北京久其软件股份有限公司 A kind of indoor orientation method and device
CN110378965B (en) * 2019-05-21 2022-04-15 北京百度网讯科技有限公司 Method, device and equipment for determining coordinate system conversion parameters of road side imaging equipment
CN113192123A (en) * 2020-01-14 2021-07-30 杭州海康威视数字技术股份有限公司 Image processing method, device and equipment
CN112132908B (en) * 2020-09-23 2024-02-20 西安邮电大学 Camera external parameter calibration method and device based on intelligent detection technology
CN113821107B (en) * 2021-11-23 2022-03-04 成都索贝数码科技股份有限公司 Indoor and outdoor naked eye 3D system with real-time and free viewpoint

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101026778A (en) * 2007-03-14 2007-08-29 北京理工大学 Distortion measurement and correction method for CCD shooting system and comprehensive test target
CN101226638A (en) * 2007-01-18 2008-07-23 中国科学院自动化研究所 Method and apparatus for standardization of multiple camera system
CN103745452A (en) * 2013-11-26 2014-04-23 理光软件研究所(北京)有限公司 Camera external parameter assessment method and device, and camera external parameter calibration method and device

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100386090B1 (en) * 2001-04-02 2003-06-02 한국과학기술원 Camera calibration system and method using planar concentric circles

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101226638A (en) * 2007-01-18 2008-07-23 中国科学院自动化研究所 Method and apparatus for standardization of multiple camera system
CN101026778A (en) * 2007-03-14 2007-08-29 北京理工大学 Distortion measurement and correction method for CCD shooting system and comprehensive test target
CN103745452A (en) * 2013-11-26 2014-04-23 理光软件研究所(北京)有限公司 Camera external parameter assessment method and device, and camera external parameter calibration method and device

Also Published As

Publication number Publication date
CN106485753A (en) 2017-03-08

Similar Documents

Publication Publication Date Title
CN106485753B (en) The method and apparatus of camera calibration for pilotless automobile
KR101666959B1 (en) Image processing apparatus having a function for automatically correcting image acquired from the camera and method therefor
TWI555379B (en) An image calibrating, composing and depth rebuilding method of a panoramic fish-eye camera and a system thereof
Wang et al. 360sd-net: 360 stereo depth estimation with learnable cost volume
TWI555378B (en) An image calibration, composing and depth rebuilding method of a panoramic fish-eye camera and a system thereof
WO2018068678A1 (en) Method and device for determining external parameter of stereoscopic camera
JP5739584B2 (en) 3D image synthesizing apparatus and method for visualizing vehicle periphery
WO2019192358A1 (en) Method and apparatus for synthesizing panoramic video, and electronic device
CN109118544B (en) Synthetic aperture imaging method based on perspective transformation
CN106570899B (en) Target object detection method and device
JP6174104B2 (en) Method, apparatus and system for generating indoor 2D plan view
CN110874854B (en) Camera binocular photogrammetry method based on small baseline condition
CN113329179B (en) Shooting alignment method, device, equipment and storage medium
CN111123962A (en) Rotor unmanned aerial vehicle repositioning photographing method for power tower inspection
CN113192646A (en) Target detection model construction method and different target distance monitoring method and device
WO2023236508A1 (en) Image stitching method and system based on billion-pixel array camera
CN111325828B (en) Three-dimensional face acquisition method and device based on three-dimensional camera
CN112307912A (en) Method and system for determining personnel track based on camera
CN108269234A (en) A kind of lens of panoramic camera Attitude estimation method and panorama camera
CN112470189B (en) Occlusion cancellation for light field systems
CN115359130A (en) Radar and camera combined calibration method and device, electronic equipment and storage medium
CN105352482B (en) 332 dimension object detection methods and system based on bionic compound eyes micro lens technology
CN111383264B (en) Positioning method, positioning device, terminal and computer storage medium
CN108537831B (en) Method and device for performing CT imaging on additive manufacturing workpiece
CN115147495A (en) Calibration method, device and system for vehicle-mounted system

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20220221

Address after: 241006 Anshan South Road, Wuhu Economic and Technological Development Zone, Anhui Province

Patentee after: Wuhu Sambalion auto technology Co.,Ltd.

Address before: 241006 Changchun Road, Wuhu economic and Technological Development Zone, Wuhu, Anhui, 8

Patentee before: CHERY AUTOMOBILE Co.,Ltd.

TR01 Transfer of patent right

Effective date of registration: 20240412

Address after: 241000 10th Floor, Block B1, Wanjiang Wealth Plaza, Guandou Street, Jiujiang District, Wuhu City, Anhui Province

Patentee after: Dazhuo Intelligent Technology Co.,Ltd.

Country or region after: China

Patentee after: Dazhuo Quxing Intelligent Technology (Shanghai) Co.,Ltd.

Address before: 241006 Anshan South Road, Wuhu Economic and Technological Development Zone, Anhui Province

Patentee before: Wuhu Sambalion auto technology Co.,Ltd.

Country or region before: China