CN107633536B - Camera calibration method and system based on two-dimensional plane template - Google Patents
Camera calibration method and system based on two-dimensional plane template Download PDFInfo
- Publication number
- CN107633536B CN107633536B CN201710674279.5A CN201710674279A CN107633536B CN 107633536 B CN107633536 B CN 107633536B CN 201710674279 A CN201710674279 A CN 201710674279A CN 107633536 B CN107633536 B CN 107633536B
- Authority
- CN
- China
- Prior art keywords
- target
- homography matrix
- obtaining
- corner points
- parameters
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- Studio Devices (AREA)
Abstract
The invention discloses a camera calibration method and a system based on a two-dimensional plane template, wherein the method comprises the following steps: obtaining a chessboard pattern calibration image, wherein the calibration image comprises a plurality of angular points; eliminating corner points with a first error value larger than a first preset value from the plurality of corner points by adopting a random sampling consistency algorithm to obtain target corner points; acquiring a homography matrix according to the coordinates of the target corner points and the corresponding world coordinates; eliminating the homography matrix with the second error value larger than the second preset value by adopting a random sampling consistency algorithm to obtain a target homography matrix; obtaining a target quadratic curve model according to the target homography matrix, and obtaining initial parameters of a camera according to the target quadratic curve model; estimating the initial parameters by adopting a maximum likelihood estimation method to obtain target parameters, and calibrating the camera by utilizing the target parameters. The invention solves the technical problem that the accuracy of camera calibration is not high in the Zhangyingyou calibration method.
Description
Technical Field
The invention relates to the technical field of vision measurement, in particular to a camera calibration method and system based on a two-dimensional plane template.
Background
In the vision measurement, the calibration of camera parameters is a very critical link, and the calibration precision and the stability of a calibration algorithm directly influence the accuracy of a result generated by the working of a camera. The purpose of camera calibration is to estimate parameters of a camera lens and an image sensor, including internal parameters, external parameters and lens distortion coefficients, which are widely applied to the field of computer vision, such as correcting lens distortion, measuring the actual size of an object, determining the position of a camera in a scene, and the like. Camera calibration is also widely used in robotics, navigation systems, three-dimensional reconstruction, and the like.
At present, methods for calibrating cameras are roughly classified into two types: conventional calibration methods and camera self-calibration methods. The traditional calibration method is that a calibration block with known size is utilized to establish the corresponding relation between the object coordinate and the image coordinate, and then the internal and external parameters of the camera are obtained through a corresponding algorithm, so that the calibration precision is high, but the processing and maintenance of the three-dimensional calibration block are difficult, and the cost is very high; the camera self-calibration mainly utilizes camera motion constraint or scene constraint to estimate parameters, so that the flexibility is strong, but the self-calibration method has too strong constraint conditions on the camera motion and the scene, the robustness is poor in practical use, the precision is low, and the factors limit the use range of the self-calibration.
In order to solve the technical problems, Zhang Zhengyou provides a flexible plane calibration method, which is between the traditional calibration method and the camera self-calibration method, does not need a specific calibration object, only needs to print a checkerboard, and can avoid the problems of high equipment requirement and low precision of the traditional calibration method. However, the calibration method for Zhangyingyou uses a plurality of checkerboard images, calculates a homography matrix corresponding to each image by using the corresponding relation of angular points, then uses a closed solution to obtain camera internal parameters and external parameters, finally considers lens distortion, uses the closed solution obtained in the prior art as an initial value, and adopts a nonlinear search algorithm to estimate all parameters including the internal parameters, the external parameters and distortion coefficients. However, the accuracy requirement of the nonlinear search algorithm on a given initial value is very strict, and since the camera parameters are coupled, if the accuracy of the provided initial value is not high, the nonlinear optimization performance is poor, and a local optimal solution is involved, so that the accuracy and robustness for solving the camera parameters and the distortion coefficient are insufficient.
Therefore, the existing Zhang Zhengyou calibration method has the technical problem that the accuracy of camera calibration is not high.
Disclosure of Invention
The embodiment of the invention provides a camera calibration method and system based on a two-dimensional plane template, which are used for solving the technical problem that the accuracy of camera calibration is not high in the existing Zhang Zhen Yong calibration method.
The invention discloses a camera calibration method based on a two-dimensional plane template, which comprises the following steps:
obtaining a chessboard pattern calibration image, wherein the calibration image comprises a plurality of angular points;
eliminating corner points with a first error value larger than a first preset value from the plurality of corner points by adopting a random sampling consistency algorithm to obtain target corner points;
acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the target corner points and the corresponding world coordinates;
eliminating the homography matrix with the second error value larger than the second preset value by adopting a random sampling consistency algorithm to obtain a target homography matrix;
obtaining a target quadratic curve model according to the target homography matrix, and obtaining initial parameters of a camera according to the target quadratic curve model, wherein the initial parameters comprise a first internal parameter, a first external parameter and a first distortion coefficient;
estimating the initial parameters by adopting a maximum likelihood estimation method to obtain target parameters, wherein the target parameters comprise second internal parameters, second external parameters and second distortion coefficients, and calibrating the camera by utilizing the target parameters.
In the method provided by the present invention, the eliminating, by using a random sampling consistency algorithm, a corner of the plurality of corners whose first error value is greater than a first preset value to obtain a target corner includes:
acquiring a reprojection coordinate of the angular point;
acquiring a first distance between the reprojection coordinate of the corner point and the initial coordinate of the corner point;
and eliminating the corner corresponding to the first distance larger than the first preset value by adopting a random sampling consistency algorithm so as to obtain the target corner.
In the method provided by the present invention, the obtaining of the homography matrix from the world coordinate system where the checkerboard calibration image is located to the pixel coordinate system according to the coordinates of the target corner point and the corresponding world coordinates includes:
randomly selecting four corner points from the chessboard pattern calibration image, and acquiring coordinates of the four corner points;
and acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the four corner points and the world coordinates corresponding to the four corner points.
In the method provided by the invention, the eliminating the homography matrix with the second error value larger than the second preset value by adopting the random sampling consistency algorithm to obtain the target homography matrix comprises the following steps:
obtaining a preset quadratic curve model according to the homography matrix;
obtaining a second distance between the homography matrix and the quadratic curve model, and taking the second distance as a difference value;
if the difference value is larger than the second preset value, the homography matrix is not in accordance with the condition;
and removing the homography matrix which does not meet the condition from the original homography matrix to obtain a target homography matrix.
In the method provided by the invention, the checkerboard calibration images comprise a plurality of images, and the angles of different calibration images and calibration cameras are different.
Based on the same inventive concept, the second aspect of the present invention provides a camera calibration system based on a two-dimensional planar template, the system comprising: the system comprises an acquisition module, a processing module and a display module, wherein the acquisition module is used for acquiring a checkerboard calibration image, and the calibration image comprises a plurality of angular points;
the first obtaining module is used for eliminating the corner points with the first error value larger than a first preset value from the plurality of corner points by adopting a random sampling consistency algorithm to obtain target corner points;
the second obtaining module is used for obtaining a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the target corner points and the corresponding world coordinates;
the third obtaining module is used for eliminating the homography matrix with the second error value larger than the second preset value by adopting a random sampling consistency algorithm to obtain a target homography matrix;
a fourth obtaining module, configured to obtain a target quadratic curve model according to the target homography matrix, and obtain initial parameters of the camera according to the target quadratic curve model, where the initial parameters include a first internal parameter, a first external parameter, and a first distortion coefficient;
and the calibration module is used for estimating the initial parameters by adopting a maximum likelihood estimation method to obtain target parameters, wherein the target parameters comprise a second internal parameter, a second external parameter and a second distortion coefficient, and the target parameters are used for calibrating the camera.
In the system provided by the present invention, the first obtaining module is further configured to:
acquiring a reprojection coordinate of the angular point;
acquiring a first distance between the reprojection coordinate of the corner point and the initial coordinate of the corner point;
and eliminating the corner corresponding to the first distance larger than the first preset value by adopting a random sampling consistency algorithm so as to obtain the target corner.
In the system provided by the present invention, the second obtaining module is further configured to:
randomly selecting four corner points from the chessboard pattern calibration image, and acquiring coordinates of the four corner points;
and acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the four corner points and the world coordinates corresponding to the four corner points.
In the system provided by the present invention, the third obtaining module is further configured to:
obtaining a preset quadratic curve model according to the homography matrix;
obtaining a second distance between the homography matrix and the quadratic curve model, and taking the second distance as a difference value;
if the difference value is larger than the second preset value, the homography matrix is not in accordance with the condition;
removing the homography matrix which does not meet the condition from the original homography matrix to obtain a target homography matrix
In the system provided by the invention, the checkerboard calibration images comprise a plurality of images, and the angles of different calibration images and calibration cameras are different.
Based on the same inventive concept, the third aspect of the present invention also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of:
obtaining a chessboard pattern calibration image, wherein the calibration image comprises a plurality of angular points;
eliminating corner points with a first error value larger than a first preset value from the plurality of corner points by adopting a random sampling consistency algorithm to obtain target corner points;
acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the target corner points and the corresponding world coordinates;
eliminating the homography matrix with the second error value larger than the second preset value by adopting a random sampling consistency algorithm to obtain a target homography matrix;
obtaining a target quadratic curve model according to the target homography matrix, and obtaining initial parameters of a camera according to the target quadratic curve model, wherein the initial parameters comprise a first internal parameter, a first external parameter and a first distortion coefficient;
estimating the initial parameters by adopting a maximum likelihood estimation method to obtain target parameters, wherein the target parameters comprise second internal parameters, second external parameters and second distortion coefficients, and calibrating the camera by utilizing the target parameters.
Based on the same inventive concept, the fourth aspect of the present invention also provides a computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the following steps when executing the program:
obtaining a chessboard pattern calibration image, wherein the calibration image comprises a plurality of angular points;
eliminating corner points with a first error value larger than a first preset value from the plurality of corner points by adopting a random sampling consistency algorithm to obtain target corner points;
acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the target corner points and the corresponding world coordinates;
eliminating the homography matrix with the second error value larger than the second preset value by adopting a random sampling consistency algorithm to obtain a target homography matrix;
obtaining a target quadratic curve model according to the target homography matrix, and obtaining initial parameters of a camera according to the target quadratic curve model, wherein the initial parameters comprise a first internal parameter, a first external parameter and a first distortion coefficient;
estimating the initial parameters by adopting a maximum likelihood estimation method to obtain target parameters, wherein the target parameters comprise second internal parameters, second external parameters and second distortion coefficients, and calibrating the camera by utilizing the target parameters.
One or more technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages:
the embodiment of the application provides a camera calibration method based on a two-dimensional plane template, which comprises the following steps: obtaining a chessboard pattern calibration image, wherein the calibration image comprises a plurality of angular points; eliminating corner points with a first error value larger than a first preset value from the plurality of corner points by adopting a random sampling consistency algorithm to obtain target corner points; acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the target corner points and the corresponding world coordinates; eliminating the homography matrix with the second error value larger than the second preset value by adopting a random sampling consistency algorithm to obtain a target homography matrix; obtaining a target quadratic curve model according to the target homography matrix, and obtaining initial parameters of a camera according to the target quadratic curve model, wherein the initial parameters comprise a first internal parameter, a first external parameter and a first distortion coefficient; estimating the initial parameters by adopting a maximum likelihood estimation method to obtain target parameters, wherein the target parameters comprise second internal parameters, second external parameters and second distortion coefficients, and calibrating the camera by utilizing the target parameters. In the above method provided by the present invention, on one hand, when obtaining the initial value, the random sampling consistency algorithm is adopted to eliminate the corner points with the corner point error value larger than the first preset value from the plurality of corner points, so as to obtain the target corner point, then the homography matrix is calculated according to the target corner point, because the angular points with lower precision are removed, the calculation accuracy of the homography matrix can be improved, thereby improving the calibration precision, on the other hand, eliminating the homography matrix with the error value larger than the second preset value by adopting a random sampling consistency algorithm, thereby obtaining a target homography matrix, then solving the calibration parameters according to the target homography matrix, the calibration image with low quality is eliminated, so that the calibration precision can be further improved, and the technical problem that the existing Zhang Zhen Yong calibration method is low in camera calibration accuracy and robustness is solved.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
Fig. 1 is a flowchart of a camera calibration method based on a two-dimensional planar template according to an embodiment of the present invention;
FIG. 2 is a structural diagram of a camera calibration system based on a two-dimensional planar template according to an embodiment of the present invention;
FIG. 3 is a block diagram of a computer-readable medium in an embodiment of the invention;
fig. 4 is a block diagram of a computer device according to an embodiment of the present invention.
Detailed Description
The embodiment of the invention provides a camera calibration method and system based on a two-dimensional plane template, which are used for solving the technical problem that the accuracy of camera calibration is not high in the existing Zhang Zhengyou calibration method.
The technical scheme in the embodiment of the application has the following general idea:
a camera calibration method based on a two-dimensional plane template, the method comprising: obtaining a chessboard pattern calibration image, wherein the calibration image comprises a plurality of angular points; eliminating corner points with a first error value larger than a first preset value from the plurality of corner points by adopting a random sampling consistency algorithm to obtain target corner points; acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the target corner points and the corresponding world coordinates; eliminating the homography matrix with the second error value larger than the second preset value by adopting a random sampling consistency algorithm to obtain a target homography matrix; obtaining a target quadratic curve model according to the target homography matrix, and obtaining initial parameters of a camera according to the target quadratic curve model, wherein the initial parameters comprise a first internal parameter, a first external parameter and a first distortion coefficient; estimating the initial parameters by adopting a maximum likelihood estimation method to obtain target parameters, wherein the target parameters comprise second internal parameters, second external parameters and second distortion coefficients, and calibrating the camera by utilizing the target parameters.
In the method, on one hand, when an initial value is obtained, angular points of a plurality of angular points, the angular point error values of which are greater than a first preset value, are removed by adopting a random sampling consistency algorithm so as to obtain target angular points, then homography matrixes are calculated according to the target angular points, the accuracy of calculation of the homography matrixes can be improved due to the removal of the angular points with lower accuracy, so that the calibration accuracy is improved, on the other hand, homography matrixes with error values of which are greater than a second preset value are removed by adopting the random sampling consistency algorithm so as to obtain the target homography matrixes, then calibration parameters are solved according to the target homography matrixes, and due to the removal of low-quality calibration images, the calibration accuracy can be further improved, and the technical problems that the accuracy and the robustness of camera calibration are not high in the existing Zhang Zhengyou calibration method are solved.
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example one
The embodiment provides a camera calibration method based on a two-dimensional plane template, please refer to fig. 1, the method includes:
step S101: obtaining a chessboard pattern calibration image, wherein the calibration image comprises a plurality of angular points;
step S102: and eliminating the corner points with the first error value of the corner points larger than a first preset value from the plurality of corner points by adopting a random sampling consistency algorithm to obtain the target corner point.
Step S103: and acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the target corner points and the corresponding world coordinates.
Step S104: and eliminating the homography matrix with the second error value larger than the second preset value by adopting a random sampling consistency algorithm to obtain a target homography matrix.
Step S105: obtaining a target quadratic curve model according to the target homography matrix, and obtaining initial parameters of a camera according to the target quadratic curve model, wherein the initial parameters comprise a first internal parameter, a first external parameter and a first distortion coefficient;
step S106: estimating the initial parameters by adopting a maximum likelihood estimation method to obtain target parameters, wherein the target parameters comprise second internal parameters, second external parameters and second distortion coefficients, and calibrating the camera by utilizing the target parameters.
It should be noted that, the applicant of the present invention finds, through long-term practice, that in the planar calibration-based method for Zhangyingyou, the accuracy of solving the initial value is mainly related to the following two aspects, i.e., the accuracy of extracting the corner points of the checkerboard, and second, the quality of the acquired calibration image. In practical application, the camera calibration process is affected by noise, image blur, illumination change, checkerboard view angle and the like, so that the extraction accuracy of the checkerboard corner points is low. In addition, the camera parameters can be accurately calibrated only by the high-quality calibration image which meets the specification, for example: the calibration precision can be reduced when the included angle between the checkerboard plane and the camera image plane is larger than 45 degrees, and the proportion of the checkerboard in the image and the distance between the checkerboard and the camera optical center can influence the precision of solving the initial value. The method provided by the invention is based on the above recognition, and provides a camera calibration method based on a two-dimensional plane template. The method specifically comprises the following steps: and (3) eliminating the angular points with larger errors by utilizing the reprojection error of each checkerboard angular point and applying a random sampling consistency algorithm, and then recalculating the homography matrix of each image. And based on the theory that the preset quadratic curve model is not changed in a single camera, the homography matrix corresponding to each image is subjected to random sampling consistency algorithm again, and low-quality calibration images which do not meet the standard are removed. And finally, obtaining a more accurate initial value, namely initial internal and external parameters (namely a first internal parameter and a second internal parameter) of the camera by adopting a closed solution, combining the initial internal and external parameters with a lens distortion factor, obtaining an accurate second internal parameter, a second external parameter and a second distortion coefficient of the camera by adopting a nonlinear search algorithm, and calibrating the camera, thereby improving the calibration accuracy and precision.
The following describes in detail a camera calibration method based on a two-dimensional planar template provided by the present application with reference to fig. 1:
step S101 is first executed: obtaining a chessboard pattern calibration image, wherein the calibration image comprises a plurality of angular points;
in a specific implementation, a plurality of checkerboard calibration images may be captured by a camera, where each checkerboard calibration image includes a plurality of corner points. Preferably, the angles of the different calibration images and the calibration camera are different, for example, the calibration images can be shot at multiple angles, so that the calibration images have enough angle change relative to the camera, thereby ensuring the calibration precision, and the number of the calibration images can be shot according to actual conditions, and can be, for example, 20, 25, 30, and the like.
Then, step S102 is executed: and eliminating the corner points with the first error value of the corner points larger than a first preset value from the plurality of corner points by adopting a random sampling consistency algorithm to obtain the target corner point.
In a specific implementation process, since the precision of the checkered corner points affects the final calibration precision, and corner points with low precision need to be removed, in this embodiment, a random sampling consistency algorithm is used to remove corner points in the plurality of corner points, in which a first error value of the corner points is greater than a first preset value, so as to obtain a target corner point.
The method for eliminating the low-precision corner points can be realized by the following steps:
acquiring a reprojection coordinate of the angular point;
acquiring a first distance between the reprojection coordinate of the corner point and the initial coordinate of the corner point;
and eliminating the corner corresponding to the first distance larger than the first preset value by adopting a random sampling consistency algorithm so as to obtain the target corner.
In a specific implementation process, a camera model is introduced first, and the camera model is divided into a linear model and a nonlinear model, wherein the linear model based on a two-dimensional plane template is as follows:
In the above formula, s represents a scale factor, (u, v) is a pixel coordinate (i.e. corner coordinate) of a camera imaging point corresponding to a corner point of the checkerboard image, (X, Y) is a world coordinate of the corner point of the checkerboard image, K is an internal parameter of the camera, and r is1,r2Rotating matrix R for camera extrinsic parameters3×3T represents the camera extrinsic parameter translation matrix, R3×3And t represents the rotation and translation matrix of the world coordinate system to the camera coordinate system where the checkerboard calibration plate is located.
Now, we will introduce homography matrix, for example, knowing the four mapping points (u, v) and (X, Y) in each calibration image and camera image, we can find the homography matrix H from the world coordinates where the checkerboard calibration plate is located to the pixel coordinates by equation (2), and we can get a homography matrix for each calibration image as follows:
Wherein the homography matrix H has 8 degrees of freedom.
The corner point coordinates (u, v) of the obtained checkerboard image can be extracted, then the homography matrix H of the checkerboard image is calculated by using the corresponding world coordinates (X, Y) and the formula (2), and the extracted corner point coordinates are not accurate due to the influences of noise, image blurring, illumination change and the like, so that the obtained calculated matrix is inaccurate
Is also inaccurate, so the coordinates of the corner re-projection can be obtained by equation (2)
And comparing the reference image with the initial coordinates (u, v) of the angular point, wherein the Euclidean distance between the reference image and the angular point is the reprojection error
d1The smaller the value of (A) is, the higher the extraction precision of the corner point is, if the error value is greater than the first preset value, the lower the precision of the corner point is, and thus the corner point is removed. For example, the first preset value is set to δ1If d is1<δ1If the corner point meets the model requirement, it is called inner corner point, and the number t of corner points meeting the calibration requirement is recorded1(ii) a Otherwise, the corner point does not meet the requirements of the model and is called an outer corner point. To ensure accuracy, the above process is repeated a number of times, which is determined by the following formula:
N＝log(1-p)/log(1-ws)(4)
the above formula shows that in N iterations, the minimum sample number is randomly selected every time s, then the probability that no outlier exists at least once is p, p is 0.99, and w represents the probability that the arbitrarily selected sample is an inlier. Then reserve t1All the corner points of the maximum time corresponding fitting model are then used to re-estimate the homography matrix by least square method
Thereby obtaining an optimal model.
Step S103 is performed next: and acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the target corner points and the corresponding world coordinates.
In a specific implementation process, a homography matrix from a world coordinate system to a pixel coordinate system where the checkerboard calibration image is located can be obtained through formula (2).
Specifically, the obtaining a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the target corner point and the corresponding world coordinates includes:
randomly selecting four corner points from the chessboard pattern calibration image, and acquiring coordinates of the four corner points;
and acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the four corner points and the world coordinates corresponding to the four corner points.
In a specific implementation process, 4 angular point coordinates (u, v) are randomly selected in each calibration image, and a homography matrix corresponding to the calibration image is obtained through a formula (2) and world coordinates (X, Y) corresponding to the formula (2)
Then, step S104 is performed: and eliminating the homography matrix with the second error value larger than the second preset value by adopting a random sampling consistency algorithm to obtain a target homography matrix.
Specifically, the removing the homography matrix with the second error value larger than the second preset value by using the random sampling consistency algorithm to obtain the target homography matrix includes:
obtaining a preset quadratic curve model according to the homography matrix;
obtaining a second distance between the homography matrix and the quadratic curve model, and taking the second distance as a difference value;
if the difference value is larger than the second preset value, the homography matrix is not in accordance with the condition;
and removing the homography matrix which does not meet the condition from the original homography matrix to obtain a target homography matrix.
In the specific implementation process, firstly, an internal reference matrix K and a preset quadratic curve model B are introduced, and r is utilized in a Zhang Zhengyou calibration method1,r2The known condition of unit orthogonality can result in the following two constraints:
h1 TK-TK-1h2＝0
h1 TK-TK-1h1＝h2 TK-TK-1h2(5)
wherein h is1,h2Is the column vector of the homography matrix H, K is the camera intrinsic parameter matrix, B is K-TK-1For presetting a quadratic curve model (ICA for short), B is only related to an internal reference matrix K of the camera and is not related to the direction and the position of the camera according to an expression of B.
The method for obtaining the target homography matrix can be achieved by first randomly selecting two homography matrices
Calculate ICA by equation (5), i.e. B ═ K-TK-1Then calculating each homography matrix respectively
Distance d from ICA2Wherein d is2＝(h1 TK-TK-1h2)2+(h1 TK-TK-1h1-1)2+(h2 TK-TK-1h2-1)2(6) (ii) a The second preset value is delta2If d is2<δ2If the homography matrix meets the model requirement, the homography matrix is called as an inner homography matrix, and the number t of the homography matrices meeting the model is recorded2(ii) a Otherwise, the homography matrix does not conform to the model, called the outer homography matrix. To ensure accuracy, the above process N is repeated2Second (N)2Determined by equation 4), reserve t2All homography matrices corresponding to the conforming model at maximum
These homography matrices are taken as target homography matrices.
Step S105 is performed next: and obtaining a target quadratic curve model according to the target homography matrix, and obtaining initial parameters of the camera according to the target quadratic curve model, wherein the initial parameters comprise a first internal parameter and a first external parameter.
In a specific implementation process, the target homography matrix obtained in the previous step is utilized, a least square method is adopted to estimate the model again to obtain an accurate ICA model, and then the initial parameters of the camera are obtained through linear calculation.
Finally, step S106 is executed: estimating the initial parameters by adopting a maximum likelihood estimation method to obtain target parameters, wherein the target parameters comprise second internal parameters, second external parameters and second distortion coefficients, and calibrating the camera by utilizing the target parameters.
In a specific implementation process, the initial parameters obtained in step S105 are used as closed solutions, and these closed solutions are used as initial values, and then all the initial parameters are estimated by a maximum likelihood estimation method, including the first internal parameter, the first external parameter, and the first distortion coefficient. And the first distortion coefficient k1,k2The initial value is set to zero. Optimized by the formula (7)Precise internal reference A, external reference k1,k2Second distortion coefficient Ri,ti
Wherein n represents the number of images finally used for calibration after eliminating low-quality calibration images, and miRepresenting the number m of homography matrix angular points after eliminating low-precision angular points in the ith calibration imageijCoordinates representing the jth corner point in the ith image, MjRepresents mijThe corresponding known world coordinates are then used to,
represents MjThe reprojection coordinates of (a). The closed solution A, R obtained by the previous stepsiAnd k1＝0,k2And (3) solving the formula (7) by using a Levenberg-Marquarat iterative optimization algorithm as an initial value to obtain accurate target parameters, wherein the target parameters comprise A and Ri,ti,k1,k2Namely, the second internal parameter, the second external parameter and the second distortion coefficient of the camera, and then the camera is calibrated through the target parameters to finish the high-precision calibration of the camera.
Based on the same inventive concept, the embodiment of the invention also provides a system corresponding to the camera calibration method based on the two-dimensional plane template, which is specifically referred to as embodiment two.
Example two
The second embodiment of the present invention provides a camera calibration system based on a two-dimensional plane template, please refer to fig. 2, the system includes:
an obtaining module 201, configured to obtain a checkerboard calibration image, where the calibration image includes a plurality of corner points;
a first obtaining module 202, configured to remove a corner, of the multiple corners, with a first error value greater than a first preset value by using a random sampling consistency algorithm, to obtain a target corner;
a second obtaining module 203, configured to obtain a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the target corner point and the corresponding world coordinates;
a third obtaining module 204, configured to eliminate the homography matrix with the second error value larger than the second preset value by using a random sampling consistency algorithm, and obtain a target homography matrix;
a fourth obtaining module 205, configured to obtain a target quadratic curve model according to the target homography matrix, and obtain initial parameters of the camera according to the target quadratic curve model, where the initial parameters include a first internal parameter, a first external parameter, and a first distortion coefficient;
a calibration module 206, configured to estimate the initial parameter by using a maximum likelihood estimation method to obtain a target parameter, where the target parameter includes a second internal parameter, a second external parameter, and a second distortion coefficient, and calibrate the camera by using the target parameter.
In the system provided in this embodiment, the first obtaining module 202 is further configured to:
acquiring a reprojection coordinate of the angular point;
acquiring a first distance between the reprojection coordinate of the corner point and the initial coordinate of the corner point;
and eliminating the corner corresponding to the first distance larger than the first preset value by adopting a random sampling consistency algorithm so as to obtain the target corner.
In the system provided in this embodiment, the second obtaining module 203 is further configured to:
randomly selecting four corner points from the chessboard pattern calibration image, and acquiring coordinates of the four corner points;
and acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the four corner points and the world coordinates corresponding to the four corner points.
In the system provided in this embodiment, the third obtaining module 204 is further configured to:
obtaining a preset quadratic curve model according to the homography matrix;
obtaining a second distance between the homography matrix and the quadratic curve model, and taking the second distance as a difference value;
if the difference value is larger than the second preset value, the homography matrix is not in accordance with the condition;
and removing the homography matrix which does not meet the condition from the original homography matrix to obtain a target homography matrix.
In the system provided in this embodiment, the checkerboard calibration images include a plurality of images, and the angles of the different calibration images and the calibration camera are different.
Various modifications and embodiments based on the first embodiment are also applicable to the system of the present embodiment, and the detailed description of the embodiments is clear to those skilled in the art from the foregoing detailed description, so that the detailed description is omitted here for the sake of brevity.
Based on the same inventive concept, the embodiment of the present invention further provides a computer readable medium corresponding to the camera calibration method based on the two-dimensional plane template, which is specifically referred to as embodiment three.
EXAMPLE III
A third embodiment of the present invention provides a computer-readable storage medium 300, please refer to fig. 3, on which a computer program 301 is stored, and the program, when executed by a processor, implements the following steps:
obtaining a chessboard pattern calibration image, wherein the calibration image comprises a plurality of angular points;
eliminating corner points with a first error value larger than a first preset value from the plurality of corner points by adopting a random sampling consistency algorithm to obtain target corner points;
acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the target corner points and the corresponding world coordinates;
eliminating the homography matrix with the second error value larger than the second preset value by adopting a random sampling consistency algorithm to obtain a target homography matrix;
obtaining a target quadratic curve model according to the target homography matrix, and obtaining initial parameters of a camera according to the target quadratic curve model, wherein the initial parameters comprise a first internal parameter, a first external parameter and a first distortion coefficient;
estimating the initial parameters by adopting a maximum likelihood estimation method to obtain target parameters, wherein the target parameters comprise second internal parameters, second external parameters and second distortion coefficients, and calibrating the camera by utilizing the target parameters.
Various modifications and specific examples of the calibration method in the first embodiment are also applicable to the computer readable medium in the present embodiment, and the computer readable medium in the present embodiment is clear to those skilled in the art from the foregoing detailed description of the camera calibration method, so that the detailed description is omitted here for brevity of the description.
Based on the same inventive concept, the embodiment of the present invention further provides a computer device corresponding to the camera calibration method based on the two-dimensional plane template, which is specifically referred to as embodiment four.
Example four
A third embodiment of the present invention provides a computer device, please refer to fig. 4, including a memory 401, a processor, and a computer program stored in the memory and capable of running on the processor, where the processor implements the following steps when executing the computer program:
obtaining a chessboard pattern calibration image, wherein the calibration image comprises a plurality of angular points;
eliminating corner points with a first error value larger than a first preset value from the plurality of corner points by adopting a random sampling consistency algorithm to obtain target corner points;
acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the target corner points and the corresponding world coordinates;
eliminating the homography matrix with the second error value larger than the second preset value by adopting a random sampling consistency algorithm to obtain a target homography matrix;
obtaining a target quadratic curve model according to the target homography matrix, and obtaining initial parameters of a camera according to the target quadratic curve model, wherein the initial parameters comprise a first internal parameter, a first external parameter and a first distortion coefficient;
estimating the initial parameters by adopting a maximum likelihood estimation method to obtain target parameters, wherein the target parameters comprise second internal parameters, second external parameters and second distortion coefficients, and calibrating the camera by utilizing the target parameters.
For convenience of explanation, fig. 4 only shows a part related to the embodiment of the present invention, and details of the technology are not disclosed, please refer to the method part of the embodiment of the present invention. The memory 401 may be used for storing software programs and modules, and the processor 402 executes the software programs and modules stored in the memory 401, so as to execute various functional applications and data processing of the mobile terminal.
The memory 401 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function, and the like; the storage data area may store data created according to use of the computer device, and the like. The control center of the processor 402 of the mobile communication terminal connects various parts of the entire mobile communication terminal by using various interfaces and lines, and performs various functions of the mobile terminal and processes data by operating or executing software programs and/or modules stored in the memory 401 and calling data stored in the memory 401, thereby integrally monitoring the mobile terminal. Optionally, processor 402 may include one or more processing units.
Various changes and specific examples of the calibration method in the first embodiment are also applicable to the computer device in the present embodiment, and those skilled in the art can clearly know the computer device in the present embodiment through the foregoing detailed description of the camera calibration method, so that details are not described herein again for the sake of brevity of the description.
One or more technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages:
the embodiment of the application provides a camera calibration method based on a two-dimensional plane template, which comprises the following steps: obtaining a chessboard pattern calibration image, wherein the calibration image comprises a plurality of angular points; eliminating corner points with a first error value larger than a first preset value from the plurality of corner points by adopting a random sampling consistency algorithm to obtain target corner points; acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the target corner points and the corresponding world coordinates; eliminating the homography matrix with the second error value larger than the second preset value by adopting a random sampling consistency algorithm to obtain a target homography matrix; obtaining a target quadratic curve model according to the target homography matrix, and obtaining initial parameters of a camera according to the target quadratic curve model, wherein the initial parameters comprise a first internal parameter and a first external parameter; estimating the initial parameters by adopting a maximum likelihood estimation method to obtain target parameters, wherein the target parameters comprise second internal parameters, second external parameters and distortion coefficients, and calibrating the camera by utilizing the target parameters. In the above method provided by the present invention, on one hand, when obtaining the initial value, the random sampling consistency algorithm is adopted to eliminate the corner points with the corner point error value larger than the first preset value from the plurality of corner points, so as to obtain the target corner point, then the homography matrix is calculated according to the target corner point, because the angular points with lower precision are removed, the calculation accuracy of the homography matrix can be improved, thereby improving the calibration precision, on the other hand, eliminating the homography matrix with the error value larger than the second preset value by adopting a random sampling consistency algorithm, thereby obtaining a target homography matrix, then solving the calibration parameters according to the target homography matrix, the calibration image with low quality is eliminated, so that the calibration precision can be further improved, and the technical problem that the existing Zhang Zhen Yong calibration method is low in camera calibration accuracy and robustness is solved.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various modifications and variations can be made in the embodiments of the present invention without departing from the spirit or scope of the embodiments of the invention. Thus, if such modifications and variations of the embodiments of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to encompass such modifications and variations.
Claims (9)
1. A camera calibration method based on a two-dimensional plane template is characterized by comprising the following steps:
obtaining a chessboard pattern calibration image, wherein the calibration image comprises a plurality of angular points;
eliminating corner points with a first error value larger than a first preset value from the plurality of corner points by adopting a random sampling consistency algorithm to obtain target corner points;
acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the target corner points and the corresponding world coordinates;
eliminating the homography matrix with the second error value larger than the second preset value by adopting a random sampling consistency algorithm to obtain a target homography matrix;
obtaining a target quadratic curve model according to the target homography matrix, and obtaining initial parameters of a camera according to the target quadratic curve model, wherein the initial parameters comprise a first internal parameter, a first external parameter and a first distortion coefficient;
estimating the initial parameters by adopting a maximum likelihood estimation method to obtain target parameters, wherein the target parameters comprise second internal parameters, second external parameters and second distortion coefficients, and calibrating the camera by utilizing the target parameters;
the eliminating the homography matrix with the second error value larger than the second preset value by adopting the random sampling consistency algorithm to obtain the target homography matrix comprises the following steps:
obtaining a preset quadratic curve model according to the homography matrix;
obtaining a second distance between the homography matrix and the quadratic curve model, and taking the second distance as a difference value;
if the difference value is larger than the second preset value, the homography matrix is not in accordance with the condition;
and removing the homography matrix which does not meet the condition from the original homography matrix to obtain a target homography matrix.
2. The method of claim 1, wherein the removing the corner points with the first error value greater than the first predetermined value from the plurality of corner points by using a random sampling consistency algorithm to obtain the target corner point comprises:
acquiring a reprojection coordinate of the angular point;
acquiring a first distance between the reprojection coordinate of the corner point and the initial coordinate of the corner point;
and eliminating the corner corresponding to the first distance larger than the first preset value by adopting a random sampling consistency algorithm so as to obtain the target corner.
3. The method as claimed in claim 1, wherein said obtaining a homography matrix from a world coordinate system to a pixel coordinate system in which said checkerboard calibration image is located according to the coordinates of said target corner points and the corresponding world coordinates comprises:
randomly selecting four corner points from the chessboard pattern calibration image, and acquiring coordinates of the four corner points;
and acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the four corner points and the world coordinates corresponding to the four corner points.
4. The method according to any one of claims 1 to 3, wherein said checkerboard calibration images comprise a plurality of and different calibration images are at different angles from the calibration camera.
5. A camera calibration system based on a two-dimensional planar template, the system comprising:
the system comprises an acquisition module, a processing module and a display module, wherein the acquisition module is used for acquiring a checkerboard calibration image, and the calibration image comprises a plurality of angular points;
the first obtaining module is used for eliminating the corner points with the first error value larger than a first preset value from the plurality of corner points by adopting a random sampling consistency algorithm to obtain target corner points;
the second obtaining module is used for obtaining a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the target corner points and the corresponding world coordinates;
the third obtaining module is used for eliminating the homography matrix with the second error value larger than the second preset value by adopting a random sampling consistency algorithm to obtain a target homography matrix;
a fourth obtaining module, configured to obtain a target quadratic curve model according to the target homography matrix, and obtain initial parameters of the camera according to the target quadratic curve model, where the initial parameters include a first internal parameter, a first external parameter, and a first distortion coefficient;
the calibration module is used for estimating the initial parameters by adopting a maximum likelihood estimation method to obtain target parameters, wherein the target parameters comprise a second internal parameter, a second external parameter and a second distortion coefficient, and the target parameters are used for calibrating the camera;
wherein the third obtaining module is specifically configured to:
obtaining a preset quadratic curve model according to the homography matrix;
obtaining a second distance between the homography matrix and the quadratic curve model, and taking the second distance as a difference value;
if the difference value is larger than the second preset value, the homography matrix is not in accordance with the condition;
and removing the homography matrix which does not meet the condition from the original homography matrix to obtain a target homography matrix.
6. The system of claim 5, wherein the first obtaining module is further configured to:
acquiring a reprojection coordinate of the angular point;
acquiring a first distance between the reprojection coordinate of the corner point and the initial coordinate of the corner point;
and eliminating the corner corresponding to the first distance larger than the first preset value by adopting a random sampling consistency algorithm so as to obtain the target corner.
7. The system of claim 5, wherein the second obtaining module is further configured to:
randomly selecting four corner points from the chessboard pattern calibration image, and acquiring coordinates of the four corner points;
and acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the four corner points and the world coordinates corresponding to the four corner points.
8. A computer-readable storage medium, on which a computer program is stored, which program, when executed by a processor, carries out the steps of:
obtaining a chessboard pattern calibration image, wherein the calibration image comprises a plurality of angular points;
eliminating corner points with a first error value larger than a first preset value from the plurality of corner points by adopting a random sampling consistency algorithm to obtain target corner points;
acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the target corner points and the corresponding world coordinates;
eliminating the homography matrix with the second error value larger than the second preset value by adopting a random sampling consistency algorithm to obtain a target homography matrix;
obtaining a target quadratic curve model according to the target homography matrix, and obtaining initial parameters of a camera according to the target quadratic curve model, wherein the initial parameters comprise a first internal parameter, a first external parameter and a first distortion coefficient;
estimating the initial parameters by adopting a maximum likelihood estimation method to obtain target parameters, wherein the target parameters comprise second internal parameters, second external parameters and second distortion coefficients, and calibrating the camera by utilizing the target parameters;
the eliminating the homography matrix with the second error value larger than the second preset value by adopting the random sampling consistency algorithm to obtain the target homography matrix comprises the following steps:
obtaining a preset quadratic curve model according to the homography matrix;
obtaining a second distance between the homography matrix and the quadratic curve model, and taking the second distance as a difference value;
if the difference value is larger than the second preset value, the homography matrix is not in accordance with the condition;
and removing the homography matrix which does not meet the condition from the original homography matrix to obtain a target homography matrix.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program performs the steps of:
obtaining a chessboard pattern calibration image, wherein the calibration image comprises a plurality of angular points;
eliminating corner points with a first error value larger than a first preset value from the plurality of corner points by adopting a random sampling consistency algorithm to obtain target corner points;
acquiring a homography matrix from a world coordinate system where the checkerboard calibration image is located to a pixel coordinate system according to the coordinates of the target corner points and the corresponding world coordinates;
eliminating the homography matrix with the second error value larger than the second preset value by adopting a random sampling consistency algorithm to obtain a target homography matrix;
obtaining a target quadratic curve model according to the target homography matrix, and obtaining initial parameters of a camera according to the target quadratic curve model, wherein the initial parameters comprise a first internal parameter, a first external parameter and a first distortion coefficient;
estimating the initial parameters by adopting a maximum likelihood estimation method to obtain target parameters, wherein the target parameters comprise second internal parameters, second external parameters and second distortion coefficients, and calibrating the camera by utilizing the target parameters;
the eliminating the homography matrix with the second error value larger than the second preset value by adopting the random sampling consistency algorithm to obtain the target homography matrix comprises the following steps:
obtaining a preset quadratic curve model according to the homography matrix;
obtaining a second distance between the homography matrix and the quadratic curve model, and taking the second distance as a difference value;
if the difference value is larger than the second preset value, the homography matrix is not in accordance with the condition;
and removing the homography matrix which does not meet the condition from the original homography matrix to obtain a target homography matrix.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710674279.5A CN107633536B (en) | 2017-08-09 | 2017-08-09 | Camera calibration method and system based on two-dimensional plane template |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710674279.5A CN107633536B (en) | 2017-08-09 | 2017-08-09 | Camera calibration method and system based on two-dimensional plane template |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107633536A CN107633536A (en) | 2018-01-26 |
| CN107633536B true CN107633536B (en) | 2020-04-17 |
Family
ID=61099131
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710674279.5A Expired - Fee Related CN107633536B (en) | 2017-08-09 | 2017-08-09 | Camera calibration method and system based on two-dimensional plane template |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107633536B (en) |
Families Citing this family (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108447095A (en) * | 2018-01-31 | 2018-08-24 | 潍坊歌尔电子有限公司 | A kind of fisheye camera scaling method and device |
| CN110345875B (en) * | 2018-04-04 | 2021-04-27 | 灵动科技（北京）有限公司 | Calibration and ranging method, device, electronic equipment and computer readable storage medium |
| CN108510551B (en) * | 2018-04-25 | 2020-06-02 | 上海大学 | Method and system for calibrating camera parameters under long-distance large-field-of-view condition |
| CN108682037B (en) * | 2018-04-27 | 2022-03-08 | 腾讯科技（深圳）有限公司 | Method, device and equipment for relocating camera attitude tracking process and storage medium |
| CN108895959B (en) * | 2018-04-27 | 2020-03-27 | 电子科技大学 | Camera calibration plate corner point calculation method based on sub-pixels |
| CN108876854B (en) | 2018-04-27 | 2022-03-08 | 腾讯科技（深圳）有限公司 | Method, device and equipment for relocating camera attitude tracking process and storage medium |
| CN108734744B (en) * | 2018-04-28 | 2022-02-18 | 国网山西省电力公司电力科学研究院 | Long-distance large-view-field binocular calibration method based on total station |
| CN110728714B (en) * | 2018-07-16 | 2023-06-20 | Oppo广东移动通信有限公司 | Image processing method and device, storage medium and electronic equipment |
| CN109146932B (en) * | 2018-07-17 | 2021-08-24 | 北京旷视科技有限公司 | Method, device and system for determining world coordinates of target point in image |
| CN109242910B (en) * | 2018-08-21 | 2021-10-08 | 电子科技大学 | Monocular camera self-calibration method based on any known plane shape |
| CN109003312B (en) * | 2018-08-24 | 2022-01-28 | 重庆邮电大学 | Camera calibration method based on nonlinear optimization |
| CN109521781A (en) * | 2018-10-30 | 2019-03-26 | 普宙飞行器科技（深圳）有限公司 | Unmanned plane positioning system, unmanned plane and unmanned plane localization method |
| CN109658465B (en) * | 2018-12-07 | 2023-07-04 | 广州华端科技有限公司 | Data processing in image reconstruction process, image reconstruction method and device |
| CN109753930B (en) * | 2019-01-03 | 2021-12-24 | 京东方科技集团股份有限公司 | Face detection method and face detection system |
| CN111652937B (en) * | 2019-03-04 | 2023-11-03 | 广州汽车集团股份有限公司 | Vehicle-mounted camera calibration method and device |
| CN111833404B (en) * | 2019-04-23 | 2023-10-31 | 富联精密电子(天津)有限公司 | Camera correction system and camera correction method |
| CN110298892B (en) * | 2019-07-05 | 2022-10-11 | 河南科技大学 | Method for calibrating internal and external parameters of single-line-array camera |
| CN111047614B (en) * | 2019-10-10 | 2023-09-29 | 南昌市微轲联信息技术有限公司 | Feature extraction-based method for extracting target corner of complex scene image |
| CN110866956A (en) * | 2019-10-28 | 2020-03-06 | 中国科学院深圳先进技术研究院 | Robot calibration method and terminal |
| CN112907674B (en) * | 2019-11-19 | 2022-06-07 | 魔门塔(苏州)科技有限公司 | Camera calibration method and device for unmanned vehicle |
| CN112907677B (en) * | 2019-12-04 | 2023-07-25 | 杭州海康威视数字技术股份有限公司 | Camera calibration method and device for single-frame image and storage medium |
| CN111311491B (en) * | 2020-01-20 | 2020-12-04 | 当家移动绿色互联网技术集团有限公司 | Image processing method and device, storage medium and electronic equipment |
| CN113256725B (en) * | 2020-02-10 | 2023-06-20 | 武汉Tcl集团工业研究院有限公司 | Camera calibration method, device and storage medium |
| CN111566701B (en) * | 2020-04-02 | 2021-10-15 | 深圳市瑞立视多媒体科技有限公司 | Method, device and equipment for calibrating scanning field edge under large-space environment and storage medium |
| CN111429532B (en) * | 2020-04-30 | 2023-03-31 | 南京大学 | Method for improving camera calibration accuracy by utilizing multi-plane calibration plate |
| CN111951335A (en) * | 2020-08-13 | 2020-11-17 | 珠海格力电器股份有限公司 | Method, device, processor and image acquisition system for determining camera calibration parameters |
| CN114511640A (en) * | 2020-11-17 | 2022-05-17 | 北京四维图新科技股份有限公司 | Method, device and storage medium for calibrating camera by using map |
| CN112634375B (en) * | 2020-12-21 | 2022-08-05 | 杭州东信北邮信息技术有限公司 | Plane calibration and three-dimensional reconstruction method in AI intelligent detection |
| CN113516719B (en) * | 2021-04-21 | 2023-08-15 | 深圳臻像科技有限公司 | Camera calibration method, system and storage medium based on multiple homography matrixes |
| CN113192145B (en) * | 2021-05-08 | 2023-02-28 | 深圳市商汤科技有限公司 | Equipment calibration method and device, electronic equipment and storage medium |
| CN113284190B (en) * | 2021-06-09 | 2023-04-25 | 上海临港绝影智能科技有限公司 | Calibration method, device, equipment, storage medium and product |
| CN113706632B (en) * | 2021-08-31 | 2024-01-16 | 上海景吾智能科技有限公司 | Calibration method and system based on three-dimensional vision calibration plate |
| CN114549658A (en) * | 2022-02-18 | 2022-05-27 | 北京达佳互联信息技术有限公司 | Camera calibration method and device and electronic equipment |
| CN115147499A (en) * | 2022-07-11 | 2022-10-04 | 深圳思谋信息科技有限公司 | Calibration parameter determination method, hybrid calibration plate, device, equipment and medium |
| CN117226853B (en) * | 2023-11-13 | 2024-02-06 | 之江实验室 | Robot kinematics calibration method, device, storage medium and equipment |
| CN118238148A (en) * | 2024-05-13 | 2024-06-25 | 北京迁移科技有限公司 | Hand-eye calibration system, method, readable storage medium and computer program product |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8009930B2 (en) * | 2008-03-26 | 2011-08-30 | City University Of Hong Kong | Auto-calibration method for a projector-camera system |
| CN101673397B (en) * | 2009-09-30 | 2012-04-25 | 青岛大学 | Digital camera nonlinear calibration method based on LCDs |
| CN105100640B (en) * | 2015-01-23 | 2018-12-18 | 武汉智源泉信息科技有限公司 | A kind of local registration parallel video joining method and system |
| CN104851104B (en) * | 2015-05-29 | 2017-12-26 | 大连理工大学 | Using the flexible big view calibration method of target high speed camera close shot |
| CN105488810B (en) * | 2016-01-20 | 2018-06-29 | 东南大学 | A kind of focusing light-field camera inside and outside parameter scaling method |
-
2017
- 2017-08-09 CN CN201710674279.5A patent/CN107633536B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN107633536A (en) | 2018-01-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107633536B (en) | Camera calibration method and system based on two-dimensional plane template | |
| CN109035320B (en) | Monocular vision-based depth extraction method | |
| CN114399554B (en) | Calibration method and system of multi-camera system | |
| CN109544643B (en) | Video camera image correction method and device | |
| JP2020523703A5 (en) | ||
| CN102750697A (en) | Parameter calibration method and device | |
| CN107633533B (en) | High-precision circular mark point center positioning method and device under large-distortion lens | |
| CN104376553B (en) | A kind of vision measuring method at the single-shaft-rotation angle based on mobile camera and dual-target | |
| CN113096192B (en) | Image sensor internal parameter calibration method, device, equipment and storage medium | |
| EP3189658A1 (en) | Camera calibration | |
| CN106570907B (en) | Camera calibration method and device | |
| CN112470192A (en) | Dual-camera calibration method, electronic device and computer-readable storage medium | |
| CN110345875B (en) | Calibration and ranging method, device, electronic equipment and computer readable storage medium | |
| CN111383264B (en) | Positioning method, positioning device, terminal and computer storage medium | |
| CN116433737A (en) | Method and device for registering laser radar point cloud and image and intelligent terminal | |
| CN110599586A (en) | Semi-dense scene reconstruction method and device, electronic equipment and storage medium | |
| CN104167001B (en) | Large-visual-field camera calibration method based on orthogonal compensation | |
| CN114549652A (en) | Camera calibration method, device, equipment and computer readable medium | |
| CN109389645B (en) | Camera self-calibration method and system, camera, robot and cloud server | |
| CN116012242A (en) | Camera distortion correction effect evaluation method, device, medium and equipment | |
| CN116188591A (en) | Multi-camera global calibration method and device and electronic equipment | |
| CN111915681B (en) | External parameter calibration method, device, storage medium and equipment for multi-group 3D camera group | |
| CN114463437A (en) | Camera calibration method, device, equipment and computer readable medium | |
| CN112002016B (en) | Continuous curved surface reconstruction method, system and device based on binocular vision | |
| CN113379845A (en) | Camera calibration method and device, electronic equipment and storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200417 Termination date: 20200809 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |