CN114387348A - Calibration method of large view field camera with ground-based sky background - Google Patents

Calibration method of large view field camera with ground-based sky background Download PDF

Info

Publication number
CN114387348A
CN114387348A CN202111456931.9A CN202111456931A CN114387348A CN 114387348 A CN114387348 A CN 114387348A CN 202111456931 A CN202111456931 A CN 202111456931A CN 114387348 A CN114387348 A CN 114387348A
Authority
CN
China
Prior art keywords
aerial vehicle
unmanned aerial
camera
calibration
center
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111456931.9A
Other languages
Chinese (zh)
Inventor
史潇天
王康
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.)
Wuhan Huazhong Tianjing Tongshi Technology Co ltd
Original Assignee
Wuhan Huazhong Tianjing Tongshi 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 Wuhan Huazhong Tianjing Tongshi Technology Co ltd filed Critical Wuhan Huazhong Tianjing Tongshi Technology Co ltd
Priority to CN202111456931.9A priority Critical patent/CN114387348A/en
Publication of CN114387348A publication Critical patent/CN114387348A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
    • G06T7/85—Stereo camera calibration
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T19/00—Manipulating 3D models or images for computer graphics
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/20—Image enhancement or restoration using local operators
    • G06T5/30—Erosion or dilatation, e.g. thinning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2200/00—Indexing scheme for image data processing or generation, in general
    • G06T2200/04—Indexing scheme for image data processing or generation, in general involving 3D image data

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computer Graphics (AREA)
  • Computer Hardware Design (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Image Processing (AREA)

Abstract

The invention discloses a calibration method of a large-view-field camera of a ground-based sky background.A bulb is arranged under an RTK antenna of an unmanned aerial vehicle, so that the unmanned aerial vehicle is ensured to be always lighted up in the flying process, and the distance from the center of the RTK antenna to the center of the bulb is measured; measuring the three-dimensional coordinates of the fixed camera by using an RTK antenna; dividing a fixed camera picture into grids, operating an unmanned aerial vehicle to fly through each grid, identifying the center of the unmanned aerial vehicle by using a morphological algorithm, resolving the position of the unmanned aerial vehicle in the picture, and extracting feature points; calculating pointing information corresponding to the center of the unmanned aerial vehicle based on a time angle method; and resolving the pointing information of each pixel by utilizing a Newton-Gaussian interpolation algorithm and finishing calibration. The method utilizes the unmanned aerial vehicle to simulate the fixed star, calibrates the pointing information of each pixel of the camera by identifying the unmanned aerial vehicle in the field of view in real time, is suitable for the calibration problem of the large-field-of-view camera of the foundation sky background, and can improve the calibration precision and the calibration speed.

Description

Calibration method of large view field camera with ground-based sky background
Technical Field
The invention belongs to the technical field of calibration of measuring cameras, and particularly relates to a calibration method of a large view field camera with a ground-based sky background.
Background
The camera is an important sensor for identifying and measuring a target object, does not transmit any signal to the target, and is widely used for unmanned aerial vehicle identification and tracking in military and civil fields. At present, the real-time high-precision positioning of the unmanned aerial vehicle by combining a multiphase unit and adopting a multi-view vision method becomes an important means for obtaining the three-dimensional coordinate of the unmanned aerial vehicle. The method can realize positioning only by acquiring images of the unmanned aerial vehicle at different cameras at the same time. However, before multi-view vision measurement, calibration is required after the camera is installed. Therefore, calibration also becomes an important guarantee for high-precision intersection measurement.
In the complex arrangement of camera groups, cameras need to be calibrated respectively. However, calibration requires the deployment of a certain number of control points. In the layout, the distribution and the precision of the control points in the view field have certain requirements. In the sky background, a plurality of fixedly installed control points must be arranged in the air, and high-precision measurement of the points is realized. In practical engineering application, the mode has high cost and long time period, and cannot meet the requirements of production and practice.
Disclosure of Invention
Aiming at the problem of camera calibration under the sky background, the method provides a novel camera calibration method, and the method divides the calibration into camera position calibration and pixel-by-pixel attitude calibration, wherein the camera position is firstly acquired by using RTK in the calibration process, and then the pixel-by-pixel attitude calibration is realized by adopting a dense time angle method.
The technical scheme adopted by the invention for solving the technical problems is as follows: a calibration method of a large view field camera with a foundation sky background comprises the following steps:
step 1, reconstructing an unmanned aerial vehicle, installing a circular bulb under an RTK (Real-time kinematic) antenna of the unmanned aerial vehicle, connecting the circular bulb with a power supply of the unmanned aerial vehicle, always keeping on during the flight process of the unmanned aerial vehicle, and measuring the distance from the center of the RTK antenna to the center of the bulb;
step 2, measuring the three-dimensional coordinate of the fixed camera by using an RTK antenna;
step 3, dividing the fixed camera picture into 20 multiplied by 20 grids, operating the unmanned aerial vehicle to fly at a constant speed of less than 1.5m/s, flying each grid within a range of 20 m-50 m away from the fixed camera, identifying the center of the unmanned aerial vehicle by using a morphological algorithm, resolving the position of the unmanned aerial vehicle in the picture, and extracting feature points;
step 4, calculating the pointing information corresponding to the center of the unmanned aerial vehicle based on a time angle method;
and 5, resolving the pointing information of each pixel by utilizing a Newton-Gaussian interpolation algorithm and finishing calibration.
The calibration method of the large view field camera with the background of the ground sky comprises the following steps of 3:
step 3.1, gray level conversion: judging whether the image is color or black and white, and if the image is color, uniformly converting the image into a black and white image;
and 3.2, carrying out binarization processing on the moving target image by using a frame difference method, and calculating a frame difference result D (x, y, k, k-1) by using the following formula:
Figure BDA0003387998040000021
wherein Ik(x, y) is the ith frame image;
3.3, reserving an unmanned aerial vehicle region by using a morphological algorithm, eliminating other noises, and carrying out open operation calculation on the binary image by using structural elements according to the following formula:
Figure BDA0003387998040000022
wherein g is a structural element,
Figure BDA0003387998040000024
the expression carries out on operation on the binary image by using structural elements, theta represents corrosion operation,
Figure BDA0003387998040000023
indicating the dilation operation.
The calibration method of the large view field camera with the background of the ground sky comprises the following steps of:
step 5.1, taking a function model of an Error In Variance (EIV) model as a limiting condition:
L-eL＝(A-EA)X
wherein L is an observation vector having a value of m × 1; a is a coefficient matrix of dimension and rank (A) ═ n<m；eLA random error vector that is L; eAA random error matrix of A; x is an n multiplied by 1 dimensional unknown parameter vector;
step 5.2, determining a random model:
Figure BDA0003387998040000031
wherein eA＝vec(EA) Vec denotes the column vectorization operator of the matrix, stacking each column of the matrix in left-to-right order, QLAnd QAAre each eLAnd eAThe dimensionalities of the co-factor matrix are mxm and mn × mn respectively; sigma0Is the error in the unit weight.
Step 5.3, constructing a WTLS Lagrange objective function, utilizing Euler-Lagrange to conduct derivation on each variable through necessary conditions, enabling the derivative to be zero, and determining an interpolation iterative formula as follows:
Figure BDA0003387998040000032
wherein i is the corresponding iteration number.
The invention has the beneficial effects that:
the method and the device adopt the unmanned aerial vehicle to calibrate the camera, avoid the problem of arrangement of control points under the sky background, and do not need to arrange high-precision control points during installation.
2, realizing high-precision real-time identification of the unmanned aerial vehicle through a morphological algorithm;
and 3, interpolation calculation is carried out by utilizing a Newton-Gaussian interpolation algorithm, so that the influence of inconsistent identification precision on the calibration result is effectively solved.
Drawings
FIG. 1 is a flow chart of a calibration method of the present invention;
fig. 2 is a schematic diagram of the unmanned aerial vehicle identification effect using the calibration method of the present invention.
Detailed Description
The invention is described in further detail below with reference to the figures and specific examples.
The invention researches a calibration method of a large view field camera with a foundation sky background. The invention aims at the problem that control points are difficult to distribute under the conditions of a foundation and a large view field, and realizes the high-precision calibration of the pointing information of the camera by using the unmanned aerial vehicle.
As shown in fig. 1 and fig. 2, the technical solution for calibrating a single-scene large-view-field camera according to the present invention is as follows:
step 1, reforming an unmanned aerial vehicle, installing a circular bulb under an RTK (Real-time kinematic) antenna of the unmanned aerial vehicle, connecting the circular bulb with a power supply of the unmanned aerial vehicle, ensuring that the unmanned aerial vehicle always keeps on during the flight process, and measuring the distance from the center of the antenna to the center of the bulb.
And 2, measuring the three-dimensional coordinate of the fixed camera by using the RTK antenna.
And 3, dividing the fixed camera picture into 20 multiplied by 20 grids, operating the unmanned aerial vehicle to fly at a constant speed of less than 1.5m/s, flying each grid within a range of 20 m-50 m away from the fixed camera, identifying the center of the unmanned aerial vehicle by using a morphological algorithm, resolving the position of the unmanned aerial vehicle in the picture, and extracting feature points. The steps of extracting the feature points are as follows:
step 3.1, gray level conversion: the images are judged to be color or black and white, and are unified into a black and white image.
And 3.2, carrying out binarization processing on the moving target image by using a frame difference method:
Figure BDA0003387998040000041
Ikand (x, y) is the ith frame image and D (x, y, k, k-1) frame difference result.
And 3.3, reserving the unmanned aerial vehicle area by adopting a morphological algorithm, and eliminating other noises.
Figure BDA0003387998040000042
Wherein g is a structural element,
Figure BDA0003387998040000043
the expression carries out on operation on the binary image by using structural elements, theta represents corrosion operation,
Figure BDA0003387998040000044
indicating the dilation operation.
And 4, calculating the pointing information corresponding to the center of the unmanned aerial vehicle based on a time angle method.
And 5, resolving the pointing information of each pixel by utilizing a Newton-Gaussian interpolation algorithm and finishing calibration.
And 5.1, taking a function model of the variable Error (EIV) model as a limiting condition.
L-eL＝(A-EA)X
Wherein L is an observation vector having a value of m × 1; a is a coefficient matrix of dimension and rank (A) ═ n<m；eLA random error vector that is L; eAA random error matrix of A; x is an n X1 dimensional unknown parameter vector.
Step 5.2, determining a random model:
Figure BDA0003387998040000051
wherein e isA＝vec(EA) Vec represents a column vectorization operator of the matrix, and each column of the matrix is stacked in the order from left to right; qLAnd QAAre each eLAnd eAThe dimensionalities of the co-factor matrix are mxm and mn × mn respectively; sigma0Is the error in the unit weight.
And 5.3, constructing a WTLS Lagrange objective function, deriving all variables by using Euler-Lagrange through necessary conditions, and determining an interpolation iterative formula by making the derivative be zero:
Figure BDA0003387998040000052
wherein i is the corresponding iteration number.
The above-described embodiments are merely illustrative of the principles and effects of the present invention, and some embodiments may be applied, and it will be apparent to those skilled in the art that various changes and modifications may be made without departing from the inventive concept of the present invention, and these embodiments are within the scope of the present invention.

Claims (3)

1. A calibration method for a large view field camera with a foundation sky background is characterized by comprising the following steps: comprises the following steps
Step 1, installing a bulb below an RTK antenna of an unmanned aerial vehicle, ensuring that the bulb is always on in the flight process of the unmanned aerial vehicle, and measuring the distance from the center of the RTK antenna to the center of the bulb;
step 2, measuring the three-dimensional coordinate of the fixed camera by using an RTK antenna;
step 3, dividing the fixed camera picture into 20 multiplied by 20 grids, operating the unmanned aerial vehicle to fly at a constant speed of less than 1.5m/s, flying each grid within a range of 20 m-50 m away from the fixed camera, identifying the center of the unmanned aerial vehicle by using a morphological algorithm, resolving the position of the unmanned aerial vehicle in the picture, and extracting feature points;
step 4, calculating the pointing information corresponding to the center of the unmanned aerial vehicle based on a time angle method;
and 5, resolving the pointing information of each pixel by utilizing a Newton-Gaussian interpolation algorithm and finishing calibration.
2. The method for calibrating a large field of view camera with a background of the sky as claimed in claim 1, wherein the step 3 is specifically as follows:
step 3.1, gray level conversion: judging whether the image is color or black and white, and if the image is color, converting the image into a black and white image;
and 3.2, carrying out binarization processing on the moving target image by using a frame difference method, and calculating a frame difference result D (x, y, k, k-1) by using the following formula:
Figure FDA0003387998030000011
wherein Ik(x, y) is the ith frame image;
3.3, reserving an unmanned aerial vehicle region by using a morphological algorithm, eliminating other noises, and carrying out open operation calculation on the binary image by using structural elements according to the following formula:
Figure FDA0003387998030000012
wherein g is a structural element, theta represents corrosion operation,
Figure FDA0003387998030000013
indicating the dilation operation.
3. The method for calibrating a large field of view camera with a background of the sky as claimed in claim 1, wherein the step 5 is specifically as follows:
step 5.1, taking a function model of the variable error model as a limiting condition:
L-eL＝(A-EA)X
wherein L is an observation vector having a value of m × 1; a is a coefficient matrix of dimension and rank (A) ═ n<m；eLA random error vector that is L; eAA random error matrix of A; x is an n multiplied by 1 dimensional unknown parameter vector;
step 5.2, determining a random model:
Figure FDA0003387998030000021
wherein eA＝vec(EA) Vec denotes the column of the matrixThe vectorization operator is to stack each column of the matrix in the order from left to right, QLAnd QAAre each eLAnd eAThe dimensionalities of the co-factor matrix are mxm and mn × mn respectively; sigma0Is the error in the unit weight.
Step 5.3, constructing a WTLS Lagrange objective function, deriving each variable through necessary conditions by using an Euler-Lagrange theorem, enabling a derivative to be zero, and determining an interpolation iterative formula as follows:
Figure FDA0003387998030000022
wherein i is the corresponding iteration number.
CN202111456931.9A 2021-12-01 2021-12-01 Calibration method of large view field camera with ground-based sky background Pending CN114387348A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111456931.9A CN114387348A (en) 2021-12-01 2021-12-01 Calibration method of large view field camera with ground-based sky background

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111456931.9A CN114387348A (en) 2021-12-01 2021-12-01 Calibration method of large view field camera with ground-based sky background

Publications (1)

Publication Number Publication Date
CN114387348A true CN114387348A (en) 2022-04-22

Family

ID=81196345

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111456931.9A Pending CN114387348A (en) 2021-12-01 2021-12-01 Calibration method of large view field camera with ground-based sky background

Country Status (1)

Country Link
CN (1) CN114387348A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115345943A (en) * 2022-08-08 2022-11-15 恩纳基智能科技无锡有限公司 Calibration method based on differential mode concept

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115345943A (en) * 2022-08-08 2022-11-15 恩纳基智能科技无锡有限公司 Calibration method based on differential mode concept
CN115345943B (en) * 2022-08-08 2024-04-16 恩纳基智能装备(无锡)股份有限公司 Calibration method based on differential mode concept

Similar Documents

Publication Publication Date Title
CN110853075B (en) Visual tracking positioning method based on dense point cloud and synthetic view
CN108369743B (en) Mapping a space using a multi-directional camera
CN110517325B (en) Coordinate transformation and method and system for positioning objects around vehicle body through coordinate transformation
CN112946591B (en) External parameter calibration method and device, electronic equipment and storage medium
CN113748357A (en) Attitude correction method, device and system of laser radar
CN111207762B (en) Map generation method and device, computer equipment and storage medium
CN110046563B (en) Power transmission line section elevation correction method based on unmanned aerial vehicle point cloud
CN104019829A (en) Vehicle-mounted panorama camera based on POS (position and orientation system) and external parameter calibrating method of linear array laser scanner
CN114217665B (en) Method and device for synchronizing time of camera and laser radar and storage medium
CN111913169B (en) Laser radar internal reference and point cloud data correction method, device and storage medium
CN112154394A (en) Terrain detection method, movable platform, control device, system and storage medium
CN105046704A (en) Vision measurement calibration method based on circular target array and observation equation
CN103822582A (en) Relative movement measuring system applicable to rendezvousing and docking experimental platform
CN114659523B (en) Large-range high-precision attitude measurement method and device
CN114387348A (en) Calibration method of large view field camera with ground-based sky background
CN118172442A (en) Method for mapping sea land topography of airborne coastal zone
CN117058209B (en) Method for calculating depth information of visual image of aerocar based on three-dimensional map
CN116740681B (en) Target detection method, device, vehicle and storage medium
CN112525145A (en) Aircraft landing relative attitude dynamic vision measurement method and system
CN117197241A (en) Robot tail end absolute pose high-precision tracking method based on multi-eye vision
CN115236643B (en) Sensor calibration method, system, device, electronic equipment and medium
CN114972524A (en) External reference calibration method for large-scale common-view-free camera
Wan et al. Ortho-rectification of high-resolution SAR image in mountain area by DEM
CN109919998B (en) Satellite attitude determination method and device and terminal equipment
CN113124821A (en) Structure measurement method based on curved mirror and plane mirror

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