Camera extrinsic parameter calibration method for transmission phase deflection wavefront measurement system

Through the small hole imaging model and the reverse direct linear transformation algorithm, combined with the Hough circular transformation detection center of mass, the complexity and accuracy of the free surface lens calibration method are solved, and efficient and accurate camera external parameter calibration is achieved.

CN118840435BActive Publication Date: 2024-11-22CHANGCHUN INST OF OPTICS FINE MECHANICS & PHYSICS CHINESE ACAD OF SCI
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411326701.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-23
Publication Date
2024-11-22
Estimated Expiration
2044-09-23

AI Technical Summary

Technical Problem

The existing free-surface lens calibration methods are complex and rely on additional equipment, making it difficult to utilize multiple feature point pairs, the calculation complexity is high and the calibration results are not accurate enough, limiting the application of the transmissive phase deflection wavefront measurement system.

Method used

Using small hole imaging model and reverse direct linear transformation algorithm, the distance between the screen and the camera's light center is calculated by taking the camera's checkerboard image at a specific position, constructing geometric relationships, and using Hough circle transformation to detect the center of mass of feature points, combining internal reference and translation The matrix constrains the initial amount, and uses singular value decomposition to solve the rotation matrix and translation vector, simplifying the calibration process and improving accuracy.

Benefits of technology

It realizes simplification and accuracy of camera external parameter calibration, avoids additional equipment dependence, makes full use of feature points to information, improves calibration accuracy, and solves the problems of cumbersome steps and high computational complexity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118840435B_ABST
    Figure CN118840435B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of camera parameter calibration, and in particular to a camera extrinsic calibration method for a transmission-type phase deflection wavefront measurement system, comprising S1: building a detection optical path; S2: calibrating camera intrinsic parameters; S3: controlling a screen to display a black and white chessboard image and a feature dot matrix image, and storing the images through a camera; S4: calculating the distance from the screen to the optical center of the camera, and based on this, calculating the image distance of the screen at a specific position; S5: using Hough circle transform to detect and extract the center of feature points in the feature dot matrix image; S6: establishing a feature point area based on the center of feature points, binarizing and extracting the centroid of feature points; S7: sorting and matching the centroids of feature points; S8: converting the coordinates of the centroid of feature points to a camera coordinate system, so that the image distance is used as the third-dimensional information of the image plane in the camera coordinate system; S9: using an inverse direct linear transformation algorithm to calculate the camera extrinsic parameters. The present invention can simplify the calibration process, does not rely on additional equipment, and improves the calibration accuracy.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of camera parameter calibration, and in particular relates to a camera extrinsic parameter calibration method for a transmission type phase deflection wavefront measurement system. Background Art

[0002] Compared with traditional spherical and aspherical surfaces, free-form surfaces provide designers with more degrees of freedom, can significantly improve the optical performance indicators of optical systems, and have great advantages in correcting aberrations. For free-form optical lenses, high-precision detection is the key basis for evaluating the overall performance of the lens. The detection accuracy of free-form surfaces limits the further expansion of optical free-form surface applications, and the detection accuracy of free-form lenses depends on the calibration accuracy of the detection system, the phase shift extraction accuracy, the wavefront fitting accuracy, and the optical path structure position accuracy. Among them, the calibration accuracy has the greatest impact on the detection accuracy of free-form lenses.

[0003] The limitations of existing calibration methods for free-form lenses are mainly complexity and reliance on additional equipment. In the reverse ray tracing calibration process, a calibration plate is required to calibrate the camera parameters, and the plane equation of the screen needs to be determined for ray tracing, which not only increases the complexity of the operation, but also may introduce additional errors. The PnP (Perspective-n-Point) calibration method is sensitive to noise and errors, and when there are more than three given paired points, it is difficult to use the information of more points, resulting in a waste of feature point pairs. The LM (Levenberg-Marquardt) optimization algorithm has high computational complexity and is sensitive to initial values. Improper selection of initial values ​​may lead to slower convergence or convergence to a suboptimal solution. In addition, there are currently few calibration methods for transmission phase deflection wavefront measurement systems, and there is a lack of a unified high-precision and efficient calibration method, which limits its promotion and popularization in practical applications. Summary of the invention

[0004] In view of this, the present invention aims to provide a camera extrinsic parameter calibration method for a transmission phase deflection wavefront measurement system, so as to solve the problems of current calibration methods relying on additional equipment, complicated steps, difficulty in using multiple feature point pairs, high calculation complexity and inaccurate calibration results.

[0005] To achieve the above object, the technical solution created by the present invention is implemented as follows:

[0006] A camera extrinsic parameter calibration method for a transmission phase deflection wavefront measurement system comprises the following steps:

[0007] S1: Build a transmissive phase deflection wavefront measurement system consisting of a screen, a lens to be tested, and a camera to detect the optical path, and adjust the aperture and focal length of the camera so that the camera can form a clear image within the predetermined axial range;

[0008] S2: Use Zhang’s calibration method to calibrate the camera’s intrinsic parameters;

[0009] S3: Adjust the optical axis of the camera to make it perpendicular to the screen, control the screen to display the black and white chessboard image and the characteristic dot matrix image, and shoot and store them through the camera;

[0010] S4: Calculate the distance from the screen to the optical center of the camera using the black and white checkerboard diagram, and construct a geometric relationship based on the distance to calculate the image distance of the screen at a specific position;

[0011] S5: Detect and extract the centers of feature points in the feature point array image using Hough circle transform;

[0012] S6: establishing a feature point region based on the center of the feature point, performing binarization processing on the feature point region, and extracting the centroid of the feature point;

[0013] S7: sorting and matching the extracted feature point centroids;

[0014] S8: based on the image distance of the screen at a specific position and the camera internal parameters, the centroid coordinates of the feature point are converted into the camera coordinate system, so that the image distance is used as the third-dimensional information of the image plane in the camera coordinate system;

[0015] S9. In the camera coordinate system, the two-dimensional plane where the screen is located is used as the projection of the camera image plane. The coordinates of the feature point matrix A with known position coordinates on the screen in the camera coordinate system and the coordinates of the feature point matrix B on the camera image plane extracted in the camera coordinate system are combined to obtain the camera extrinsic parameters using the inverse direct linear transformation algorithm.

[0016] Furthermore, in step S2, the screen is controlled to display a black and white checkerboard image, the screen posture is changed, a group of black and white checkerboard images are collected by the camera, the group of black and white checkerboard images are grayed and then the corner points are extracted, and the camera intrinsic parameters are calculated based on the extracted corner points.

[0017] Furthermore, in step S4, the process of using the black and white checkerboard image to calculate the distance from the screen to the optical center of the camera is as follows: the black and white checkerboard image displayed on the screen is combined with the black and white checkerboard image taken when calibrating the camera intrinsic parameters, and the Zhang Zhengyou calibration method of the Camera calibrator toolbox in Matlab is used to calibrate the translation matrix from the screen to the camera. The component of the translation matrix in the z direction can be regarded as the distance from the screen to the optical center of the camera.

[0018] Furthermore, in step S4, a geometric relationship is constructed based on the distance, and the formula for calculating the image distance of the screen at a specific position is:

[0019] ;

[0020] In the formula, Indicates the image distance of the screen at a specific position, l Indicates the distance from the screen to the optical center of the camera. Indicates the pixel size of the camera target surface, Represents the pixel size of the screen pixel, m represents the characteristic point matrix A The number of pixels on the screen between the two points in the center ,n Represents feature points B The number of pixels between the two points on the camera image plane.

[0021] Furthermore, in step S5, the feature point matrix image captured and saved by the camera is converted into a grayscale image, the radius range of the feature points is set, the detection target is set as a dark circle under a bright background, and Hough circle transform detection is performed according to the radius range of the feature points to obtain the center coordinates and radius of all feature points.

[0022] Furthermore, the value range of the sensitivity factor for identifying dark circles is set to 0-1.

[0023] Furthermore, in step S6, according to the central coordinates and radius of the feature points, a rectangular area with a radius three times that of the center of each feature point is set as a reference, a grayscale threshold is set for each rectangular area for binarization, and then the centroid extraction algorithm is used to extract the centroid of each rectangular area as the centroid of each feature point.

[0024] Further, in step S7, the extracted feature point centroid coordinates are stored as The matrix is ​​a 2-column matrix. The first column of the matrix represents the horizontal coordinate of the centroid of the feature point, and the second column of the matrix represents the vertical coordinate of the centroid of the feature point. The matrix is ​​sorted according to the size of the horizontal coordinate of the centroid of the feature point. The data increases in turn as cycles. Set each The centroids of feature points are used as matrix units, and each matrix unit is sorted according to the size of the ordinate.

[0025] Furthermore, in step S9, a linear equation group is constructed using each pair of matching features in the feature point matrix A and the feature point matrix B, and the projection matrix is ​​obtained by solving the equation group using singular value decomposition. The rotation matrix R and the translation matrix T are extracted from the projection matrix to obtain the camera extrinsics.

[0026] Compared with the prior art, the invention can achieve the following beneficial effects:

[0027] The present invention is based on the pinhole imaging model. In the camera coordinate system, the two-dimensional plane where the screen is located is regarded as the projection of the camera image plane, and the image distance of the screen at a specific position is introduced on the camera image plane as the third dimension component. The distance from the screen to the optical center of the camera can be calculated by shooting a chessboard picture displayed on the screen at a specific position by the camera, and then a geometric relationship is constructed to solve the image distance. This not only simplifies the calibration process, but also improves the calibration accuracy. The initial quantity is jointly constrained by the internal parameter and the translation matrix to ensure the solution accuracy of the R and T matrices. The extracted feature points are subjected to Hough transformation to identify the circular domain, and the centroid of each circular domain is extracted to achieve automatic extraction. Combined with the extracted lattice centroid, it is substituted into the proposed inverse direct transformation model, and a linear equation group is constructed based on mutually matching feature point pairs. The rotation matrix R and the translation vector T are solved using singular value decomposition. This calibration method does not rely on additional equipment, and the calibration steps are simple. The centroid coordinates of the feature points are accurately extracted, and a large number of feature point pairs are used to construct an overdetermined equation group, and the information of each feature point pair is used as much as possible, which improves the calibration accuracy. Therefore, the present invention solves the problems of current calibration methods that they rely on additional equipment, have complicated steps, are difficult to use multiple feature point pairs, have high calculation complexity and produce inaccurate calibration results. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The drawings constituting part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments and descriptions of the present invention are used to explain the present invention and do not constitute an improper limitation on the present invention. In the drawings:

[0029] Figure 1 It is a calibration schematic diagram for a transmission-type phase deflection wavefront measurement system according to an embodiment of the invention;

[0030] Figure 2 It is a flow chart of a camera extrinsic parameter calibration method for a transmission-type phase deflection wavefront measurement system according to an embodiment of the invention;

[0031] Figure 3 Schematic diagram of the principle of pinhole imaging according to an embodiment of the present invention;

[0032] Figure 4 It is a schematic diagram of the principle of the geometric relationship of the pinhole imaging in the x direction according to the embodiment of the invention;

[0033] Figure 5 is a schematic diagram of a feature dot matrix captured by a camera according to an embodiment of the invention;

[0034] Figure 6 It is a schematic diagram of the principle of Hough circle transform detection according to an embodiment of the invention;

[0035] Figure 7It is a schematic diagram of a feature point array detected by using Hough circle transform according to an embodiment of the invention;

[0036] Figure 8 Schematic diagram of the detection optical path of the transmission-type phase deflection wavefront measurement system according to an embodiment of the invention;

[0037] Fig. 9 is a schematic diagram of spherical degree distribution according to an embodiment of the invention;

[0038] Fig.10 It is a schematic diagram of cylindrical lens degree distribution according to an embodiment of the invention.

[0039] The reference numerals include: 1~screen, 2~lens to be tested, 3~camera lens, 4~CCD camera. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solution and advantages of the invention more clear, the invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described here are only used to explain the invention and do not constitute a limitation of the invention.

[0041] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0042] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "first", "second", etc. are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first", "second", etc. may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0043] In the description of the invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installation", "connection" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the invention can be understood according to specific circumstances.

[0044] The following will refer to Figure 1-Figure 10 The invention is described in detail with reference to the embodiments.

[0045] The present invention provides a camera extrinsic calibration method for a transmission-type phase deflection wavefront measurement system, and calibrates the camera-to-screen extrinsic parameter R and T matrices based on an inverse direct linear transformation (DLT) calibration method of a pinhole imaging model, such as Figure 1 As shown, the R matrix and T matrix are used for the screen physical coordinate system To World Coordinate System The conversion between the camera coordinate system (defined in the system as the world coordinate system) is specifically expressed as:

[0046] (1).

[0047] like Figure 2 As shown, the camera extrinsic parameter calibration method for a transmission-type phase deflection wavefront measurement system provided by the present invention includes the following steps:

[0048] S1: Build a transmissive phase deflection wavefront measurement system consisting of a screen, a lens to be tested, and a camera to detect the optical path, and adjust the aperture and focal length of the camera so that the camera can form a clear image within the predetermined axial range.

[0049] The screen 1, the lens to be tested 2, and the camera constitute a detection optical path of a transmission-type phase deflection wavefront measurement system, and the camera includes a camera lens 3 and a CCD camera 4.

[0050] S2: Use Zhang’s calibration method to calibrate the camera’s intrinsic parameters.

[0051] In the transmission phase deflection wavefront measurement system, the calibration accuracy of the camera's internal and external parameters directly affects the system's measurement accuracy. At present, the calibration of the camera's internal parameters mostly adopts the Zhang Zhengyou calibration method. The specific steps are to control the screen to display a black and white chessboard image, constantly change the screen's posture, call the camera to take n black and white chessboard pictures, grayscale the black and white chessboard pictures and extract the corner points, and further calculate the camera's internal parameters: camera focal length , , center point coordinates and distortion coefficient .

[0052] S3: Adjust the camera to be perpendicular to the screen, control the screen to display the black and white checkerboard image and the characteristic dot matrix image, and shoot and store them through the camera.

[0053] When the optical axis of the camera is perpendicular to the screen, the optical axis of the camera coincides with the optical axis of the screen. When the screen is stabilized at a specific position in the optical path, the control program is used to automatically control the screen to display a black and white chessboard image and a characteristic dot matrix image (circular dot matrix, OK Column Total The black and white checkerboard is used to calculate the distance from the screen to the camera optical center. The feature dot matrix image is used to solve the external parameters from the screen to the camera: R matrix and T matrix.

[0054] S4: Use the black and white checkerboard diagram to calculate the distance from the screen to the optical center of the camera, and construct a geometric relationship based on the distance to calculate the image distance of the screen at a specific position.

[0055] The inverse direct linear transformation (DLT) calibration method adopted by the present invention is implemented based on the pinhole imaging model. In order to reduce the additional error introduced by the measurement and simplify the calibration process, the present invention innovatively proposes that under the pinhole imaging model, the image plane in the camera coordinate system is regarded as the object, the screen in the camera coordinate system is regarded as the image, and the image plane in the camera coordinate system is regarded as the object. As the third-dimensional information of the camera image plane, it corresponds to different , Available through Figure 3 The geometric relationship shown is solved as follows:

[0056] (2);

[0057] In the formula, Indicates the pixel size of the camera target surface (in millimeters), represents the pixel size of the screen in millimeters, n Represents feature points B The number of pixels (in pixels) at the center of the two points on the camera image plane, m represents the characteristic point matrix A The number of pixels on the screen between the two points in the center (in pixels).

[0058] To avoid measuring When introducing errors, a black and white chessboard image is displayed on the screen, and combined with the black and white chessboard image taken when calibrating the camera intrinsic parameters, the Zhang Zhengyou calibration method of the Camera calibrator toolbox in Matlab is used to calibrate the translation matrix from the screen to the camera , since the camera optical axis is perpendicular to the screen surface, The component of the matrix in the z direction can be regarded as the distance from the screen to the optical center of the camera , Used to construct geometric relationships and calculate the image distance of the screen at a specific position , which can simplify the step of measuring the distance from the camera's optical center to the screen and also improve the accuracy.

[0059] Because the distance from the screen to the camera's optical center After that, combined with the pinhole imaging model, we can accurately solve Therefore, we only need to shoot one more black and white chessboard picture based on the n black and white chessboard pictures taken by the camera internal parameters to calculate the image distance , which simplifies the process of measuring the distance from the screen to the camera's optical center and improves the measurement accuracy.

[0060] The pinhole imaging model is a type of camera model. Its model is simple and universal. The principle is as follows: Figure 3 As shown:

[0061] Three coordinate systems are defined in the pinhole imaging model: world coordinate system (In the pinhole imaging model, the camera coordinate system is defined as the world coordinate system), the image coordinate system , pixel coordinate system . A point in space The point projected onto the camera image plane by the pinhole imaging model is .

[0062] Figure 4 shows the projection of the pinhole imaging model in the x direction. Figure 4 The geometric relationship in can be obtained:

[0063] (3);

[0064] The same is true in The directions are:

[0065] (4);

[0066] The image coordinate system can be , pixel coordinate system The transformation relationship is expressed as homogeneous coordinates:

[0067] (5);

[0068] In the formula, , That is, the focal length of the camera in the x and y directions. These two parameters will be solved when calibrating the camera's intrinsic parameters.

[0069] Since the units of the image coordinate system and the pixel coordinate system in the pinhole model are different and the positions of the two coordinate systems are also different, translation and scaling are required, which can be expressed in homogeneous coordinates as follows:

[0070] (6);

[0071] In the formula, , Respectively represent the pixel size on the CCD camera, in millimeters, , Represents the image coordinate system In pixel coordinate system The zero point of coordinates in .

[0072] Combining formulas (1), (5), and (6), we can Image point on the camera image plane The conversion relationship is expressed as:

[0073] (7);

[0074] In the formula, the first and second items are the camera intrinsic parameter matrix A:

[0075] (8);

[0076] A is the camera intrinsic parameter, and R and T are the camera extrinsic parameters.

[0077] S5: Use Hough circle transform to detect and extract the centers of feature points in the feature point array image.

[0078] Convert the feature point image captured by the camera into a grayscale image, set the radius range of the feature points, and set the detection target to a dark circle under a bright background. Since there is a certain systematic error in the experiment, the intercepted feature points are not strictly circles. Set the sensitivity factor for identifying circles to (The value range is 0-1. The larger the sensitivity factor, the more circular targets can be identified). Hough circle transform detection is performed according to the radius range of the feature points to obtain the center coordinates and radius of all feature points.

[0079] In the specific experiment, the circular feature point array captured by the camera is as follows Figure 5As shown in the figure, the circular feature points in the captured pictures are difficult to identify and the center of the circular feature points is difficult to accurately locate due to factors such as uneven light intensity and dirty lens / screen. To solve this problem, Hough circle transform detection is used in combination with the centroid extraction algorithm to identify the circular feature points and extract the centroid of the circular feature points.

[0080] The principle of Hough circle transform detection is as follows Figure 6 As shown, the coordinates of the circular feature points in the camera coordinate system can be expressed as:

[0081] (9);

[0082] For a point on the image plane , whose value is certain, assuming that , transform the point of the camera coordinate system space to the Hough space, and the Hough space will be is the coordinate, that is, a point in the camera coordinate system space corresponds to a circle in the Hough space. At least three points can determine a circle. By setting the radius range of the circle to be identified, the circle can be accurately identified without being affected by uneven light intensity. After the corresponding circular feature points are identified, they are binarized according to the grayscale transformation. The non-circular feature point area is set to a black background. The feature point array image after identification is as follows Figure 7 shown.

[0083] In the DLT direct linear transformation method, the more feature points there are, the more accurate the result is. However, too many feature points will lead to a decrease in the spacing between points and an unclear grayscale transformation range of the points. How to accurately identify the centroid coordinates of a sufficient number of feature points is a key issue. In the present invention, it is proposed to use Hough circle transform detection combined with a centroid extraction algorithm to identify the centroid of a circular dot matrix, thereby improving the accuracy of calibration.

[0084] S6: Establish a feature point region based on the center of the feature point, perform binarization on the feature point region, and extract the centroid of the feature point.

[0085] Hough circle transform detection also has strong anti-noise performance for fuzzy noise points caused by non-circular dirt. Since the Hough transform extracts the center of the circle, which is somewhat different from the center of mass, in order to extract the center of the circle more accurately, the center of mass extraction algorithm is used to extract the center of mass of the identified circular feature point area to replace the center of the circle coordinates, so that the precise center of the dot matrix can be identified. Specifically, according to the center coordinates and radius of the feature point, a rectangular area with a radius of three times the center of each feature point is set as the reference, and a grayscale threshold is set for each rectangular area for binarization, and then the center of mass extraction algorithm is used to extract the center of mass of each rectangular area as the center of mass of each feature point.

[0086] S7: Sort and match the extracted feature point centroids.

[0087] The extracted feature point centroid coordinates are stored as The matrix is ​​a 2-column matrix. The first column of the matrix represents the horizontal coordinate of the centroid of the feature point, and the second column of the matrix represents the vertical coordinate of the centroid of the feature point. The matrix is ​​sorted according to the size of the horizontal coordinate of the centroid of the feature point. The data increases in turn as cycles. Set each The centroids of the feature points are used as matrix units, and each matrix unit is sorted according to the size of the ordinate. The sorted coordinates can be matched one-to-one with the coordinates of the feature points displayed on the screen.

[0088] S8: Based on the image distance of the screen at a specific position and the camera internal parameters, the centroid coordinates of the feature point are converted to the camera coordinate system, so that the image distance is used as the third-dimensional information of the image surface in the camera coordinate system.

[0089] Based on the image distance of the screen at a specific position And the camera internal parameter A, the centroid coordinates of the feature point are transformed into the camera coordinate system so that the image distance As the third-dimensional information of the image plane in the camera coordinate system.

[0090] S9. In the camera coordinate system, the two-dimensional plane where the screen is located is used as the projection of the camera image plane. The coordinates of the feature point matrix A with known position coordinates on the screen in the camera coordinate system and the coordinates of the feature point matrix B on the camera image plane extracted in the camera coordinate system are combined to obtain the camera extrinsic parameters using the inverse direct linear transformation algorithm.

[0091] Based on the pinhole imaging model, it is proposed that the two-dimensional plane where the screen is located is regarded as the projection of the camera image plane in the camera coordinate system, that is, in the camera coordinate system, the screen is the image of the camera image plane after passing through the pinhole. As the third-dimensional information of the image plane in the camera coordinate system, logically, the two-dimensional plane where the screen is located is the projection of the three-dimensional camera image plane with depth information, which is complete.

[0092] The coordinates of the feature point matrix with known position coordinates preset on the screen are matched with the coordinates of the feature point matrix extracted on the camera image plane in the camera coordinate system. Each feature point can construct two linear equations, and all feature point pairs construct an overdetermined system of equations. The singular value decomposition is used to solve the overdetermined system of equations to obtain the projection matrix, and the rotation matrix R and translation vector T are extracted from the projection matrix.

[0093] Since a large number of feature point pairs are used to construct an overdetermined set of equations, the information of each feature point pair is used as much as possible, thereby improving the calibration accuracy. This solves the problems of the current calibration method's reliance on additional equipment, cumbersome steps, difficulty in using multiple feature point pairs, high computational complexity, and inaccurate calibration results.

[0094] The coordinates of the feature points on the camera image plane are , the corresponding pixel coordinates on the screen coordinate system are , which can be expressed as:

[0095] (10);

[0096] In the formula, is the scale coefficient, R is the rotation matrix, T is the translation matrix, and equation (10) can be expanded as follows:

[0097] (11);

[0098] Expand (11) to:

[0099] (12);

[0100] (13);

[0101] (14);

[0102] In the formula, a pair of matching feature points represents two limiting equations. There are 12 variables in the R matrix and the T matrix. , the number of point pairs that theoretically match each other The equation can be solved. In practice, the more pairs of points that match each other, the more accurate the solution.

[0103] Normalize the coordinates of the feature point pairs and solve the overdetermined equation:

[0104] (15);

[0105] In the formula, , k is an intermediate variable used to represent the matrix of 12 variables. The solved R and T matrices are the camera extrinsic parameters.

[0106] The present invention uses a centroid extraction algorithm to extract the centroid of the identified circular feature point area to replace the center coordinates of the circle, so that the precise center of the circular feature point can be identified. Substitute the point matrix into (15) to obtain the exact R matrix and T matrix.

[0107] In order to verify the accuracy of the calibration method proposed in this invention, a Figure 8 The detection optical path of the transmission phase deflection wavefront measurement system shown in the figure uses the five-step phase shift method for phase shifting, uses the least squares method to solve the wrapped phase, calculates the wavefront slope through coordinate transformation, uses Zernike for wavefront reconstruction to obtain wavefront information, and solves the main curvature of the wavefront. and , substituting into the diopter calculation model, the spherical power of the measured lens can be solved and cylindrical , specifically:

[0108] (16).

[0109] In the formula and represents the principal curvature of the wavefront, Indicates the spherical power of the lens being tested, Indicates the cylindrical power of the lens being tested.

[0110] For an aspherical lens, the theoretical design value of spherical power is 400 degrees, and the theoretical design value of cylindrical power is 0 degrees. Actual measurement is carried out to verify the accuracy of the proposed calibration method. The spherical power distribution curve is shown in Fig. 9 As shown, the cylindrical lens power distribution curve is as follows Fig.10 As shown, the test results show that the refractive distribution curve and distribution value of the spherical lens are consistent with the theoretical design, further demonstrating the applicability and accuracy of the external parameter calibration method we proposed.

[0111] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the disclosure of the present invention can be performed in parallel, sequentially or in different orders, as long as the desired results of the technical solution disclosed in the present invention can be achieved, and this document does not limit this.

[0112] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A camera extrinsic calibration method for a transmission phase deflection wavefront measurement system, characterized in that: The steps include: S1: Build a transmissive phase deflection wavefront measurement system consisting of a screen, a lens to be tested, and a camera to detect the optical path, and adjust the aperture and focal length of the camera so that the camera can form a clear image within the predetermined axial range; S2: Use Zhang’s calibration method to calibrate the camera’s intrinsic parameters; S3: Adjust the optical axis of the camera to make it perpendicular to the screen, control the screen to display the black and white chessboard image and the characteristic dot matrix image, and shoot and store them through the camera; S4: Calculate the distance from the screen to the optical center of the camera using the black and white checkerboard diagram, and construct a geometric relationship based on the distance to calculate the image distance of the screen at a specific position; S5: Detect and extract the centers of feature points in the feature point array image using Hough circle transform; S6: establishing a feature point region based on the center of the feature point, performing binarization processing on the feature point region, and extracting the centroid of the feature point; S7: sorting and matching the extracted feature point centroids; S8: based on the image distance of the screen at a specific position and the camera internal parameters, the centroid coordinates of the feature point are converted into the camera coordinate system, so that the image distance is used as the third-dimensional information of the image plane in the camera coordinate system; S9. In the camera coordinate system, the two-dimensional plane where the screen is located is used as the projection of the camera image plane. The coordinates of the feature point matrix A with known position coordinates on the screen in the camera coordinate system and the coordinates of the feature point matrix B on the camera image plane extracted in the camera coordinate system are combined to obtain the camera extrinsic parameters using the inverse direct linear transformation algorithm.

2. The camera extrinsic parameter calibration method for a transmission-type phase deflection wavefront measurement system according to claim 1, characterized in that: In step S2, the screen is controlled to display a black and white checkerboard image, the screen posture is changed, a group of black and white checkerboard images are collected by the camera, the group of black and white checkerboard images are grayed and then the corner points are extracted, and the camera intrinsic parameters are calculated according to the extracted corner points.

3. The camera extrinsic parameter calibration method for a transmission-type phase deflection wavefront measurement system according to claim 2, characterized in that: In step S4, the process of using the black and white checkerboard image to calculate the distance from the screen to the optical center of the camera is as follows: the black and white checkerboard image displayed on the screen is combined with the black and white checkerboard image taken when calibrating the camera intrinsic parameters, and the Zhang Zhengyou calibration method of the Cameracalibrator toolbox in Matlab is used to calibrate the translation matrix from the screen to the camera. The component of the translation matrix in the z direction can be regarded as the distance from the screen to the optical center of the camera.

4. The camera extrinsic parameter calibration method for a transmission-type phase deflection wavefront measurement system according to claim 3, characterized in that: In step S4, a geometric relationship is constructed based on the distance, and the formula for calculating the image distance of the screen at a specific position is: ; In the formula, Indicates the image distance of the screen at a specific position, l Indicates the distance from the screen to the optical center of the camera. Indicates the pixel size of the camera target surface, Represents the pixel size of the screen pixel, m Represents feature points A The number of pixels on the screen between the two points in the center ,n Represents feature points B The number of pixels between the two points on the camera image plane.

5. The camera extrinsic parameter calibration method for a transmission-type phase deflection wavefront measurement system according to claim 1, characterized in that: In step S5, the feature point matrix image captured and saved by the camera is converted into a grayscale image, the radius range of the feature points is set, the detection target is set as a dark circle under a bright background, and Hough circle transform detection is performed according to the radius range of the feature points to obtain the center coordinates and radius of all feature points.

6. The camera extrinsic parameter calibration method for a transmission-type phase deflection wavefront measurement system according to claim 5, characterized in that: Set the sensitivity factor for identifying dark circles to a value range of 0-1.

7. The camera extrinsic parameter calibration method for a transmission-type phase deflection wavefront measurement system according to claim 5, characterized in that: In step S6, according to the center coordinates and radius of the feature points, a rectangular area with a radius three times the center of each feature point is set as a reference, a grayscale threshold is set for each rectangular area for binarization, and then the centroid extraction algorithm is used to extract the centroid of each rectangular area as the centroid of each feature point.

8. The camera extrinsic parameter calibration method for a transmission-type phase deflection wavefront measurement system according to claim 1, characterized in that: In step S7, the extracted feature point centroid coordinates are stored as The matrix is ​​a 2-column matrix. The first column of the matrix represents the horizontal coordinate of the centroid of the feature point, and the second column of the matrix represents the vertical coordinate of the centroid of the feature point. The matrix is ​​sorted according to the size of the horizontal coordinate of the centroid of the feature point. The data increases in turn as cycles. Set each The centroids of feature points are used as matrix units, and each matrix unit is sorted according to the size of the ordinate.

9. The camera extrinsic parameter calibration method for a transmission-type phase deflection wavefront measurement system according to claim 1, characterized in that: In step S9, a linear equation group is constructed using each pair of matching features in the feature point matrix A and the feature point matrix B, and the projection matrix is ​​obtained by solving the equation group using singular value decomposition. The rotation matrix R and the translation matrix T are extracted from the projection matrix to obtain the camera extrinsics.