A light source calibration method and device based on a combined target and related medium

Abstract

The application discloses a light source calibration method and device based on a combined target and related media, and the method comprises the following steps: acquiring mirror sphere image information under high exposure and uniform illumination, and calculating the initial coordinates of the sphere center corresponding to the image information by using a perspective imaging model; setting the sphere center corresponding to the initial coordinates of the sphere center as a sphere center reference point, combining the sphere center reference point and a plane reference point as a target reference point for re-projection, and performing error averaging on the error of the re-projection by using a light beam adjustment algorithm to obtain optimized coordinates of the sphere center; under low exposure, for each target posture, the sphere high light point of each light source is acquired by combining the sphere equation and the optimized coordinates of the sphere center; the incident light direction of the sphere high light point is acquired to obtain an incident light line, and the light source is calibrated according to the incident light line. The embodiment realizes accurate calibration of the light source position under the condition of non-uniform illumination, ensures the accuracy of the light source position calibration, and improves the accuracy of near-field photometric three-dimensional reconstruction.

Description

Technical Field

[0001] This invention relates to the field of light source calibration technology, and in particular to a light source calibration method, apparatus and related media based on a combined target. Background Technology

[0002] Photometric stereo vision is an image-based 3D reconstruction algorithm that can recover the surface normals and 3D shape of an object by capturing multiple frames under different lighting conditions. This allows for the dense reconstruction of the object's 3D surface normals, making it widely used in rapid manufacturing, industrial inspection, and cultural relic preservation. Its principle is to reconstruct the object's 3D shape by analyzing the changes in brightness and darkness on its surface. In near-field photometric stereo 3D reconstruction, the LED light source calibration results provide light source constraints for the photometric stereo vision model. The LED light source calibration results determine the accuracy of the object's normal recovery and the precision of the 3D reconstruction. The calibration of the light source's spatial position during the reconstruction process directly affects the accuracy of the reconstruction results.

[0003] A light source spatial positioning calibration system typically consists of multiple LED light sources and a camera. The relative positions of the LED light sources and the camera are fixed, and the system determines the spatial position of the light source by calibrating the relative positional relationship between the calibrated LED light sources and the camera. Currently, light source calibration methods mainly use a highly reflective sphere for reflection. Based on the specular reflection properties of the highlights on the surface of the highly reflective sphere, the position of the sphere's center is calculated using a specular reflection model, thus calibrating the light source position. In this process, the accuracy of the calibration result is affected by the accuracy of the highly reflective sphere edge extraction and the image plane highlight extraction. However, traditional light source calibration methods still have some drawbacks. For example, for near-field photometric stereo 3D reconstruction problems, when calibrating under single illumination conditions, the illumination received by the highly reflective sphere surface is uneven, inevitably resulting in shadows at the edges of the image contour. This makes it difficult to distinguish the sphere's contour from the background, leading to errors in image contour edge extraction, which in turn results in errors in the extracted sphere center position, thus affecting the accuracy of the light source spatial positioning calibration. Therefore, traditional light source calibration methods, when using a single reflective sphere for calibration, require adjusting the camera's exposure parameters to strike a balance between sharp sphere edges and high-quality highlights. However, shadowing is unavoidable under single illumination. Furthermore, while increasing camera exposure can alleviate shadows along the sphere's outline, it also reduces the quality of the highlights on the highly reflective sphere, decreasing highlight extraction accuracy and consequently affecting the accuracy of light source spatial positioning calibration, thus reducing the precision of near-field photometric 3D reconstruction.

[0004] Therefore, how to achieve high-precision calibration of the spatial position of the light source, thereby improving the accuracy of near-field photometric three-dimensional reconstruction, is a problem to be solved by those skilled in the art. Summary of the Invention

[0005] This invention provides a light source calibration method, apparatus, computer equipment, and storage medium based on a combined target, aiming to achieve high-precision calibration of the spatial position of the light source and improve the accuracy of near-field photometric three-dimensional reconstruction.

[0006] In a first aspect, embodiments of the present invention provide a light source calibration method based on a combined target, comprising:

[0007] Under high exposure and uniform lighting conditions, the mirror sphere image information in the combined target image is acquired, and the mirror sphere image information is calculated using the camera's perspective imaging model to obtain the initial coordinates of the sphere's center corresponding to the mirror sphere image information; wherein, the combined target image is an image of the combined target with multiple target postures captured by the camera, the combined target includes a mirror sphere and a planar target, and the planar target contains multiple planar reference points;

[0008] The sphere center corresponding to the initial coordinates of the sphere center is set as the sphere center reference point, and the sphere center reference point and the plane reference point are combined to form the target reference point. Then, the target reference point is reprojected, and the error of the reprojection is averaged using the bundle adjustment algorithm to obtain the optimized coordinates of the sphere center.

[0009] Under low exposure conditions of the camera, for each target posture, the spherical highlight point of each light source on the mirror sphere is obtained by combining the spherical equation and the optimized coordinates of the sphere center.

[0010] Obtain the incident light direction of each light source at the corresponding spherical highlight point under each target posture, and obtain the corresponding incident light ray. Then, calibrate the light source based on the incident light ray.

[0011] Secondly, embodiments of the present invention provide a light source calibration device based on a combined target, comprising:

[0012] The initial coordinate acquisition unit is used to acquire the mirror sphere image information in the combined target image under high exposure and uniform lighting conditions of the camera, and to calculate the mirror sphere image information using the camera's perspective imaging model to obtain the initial coordinates of the sphere center corresponding to the mirror sphere image information; wherein, the combined target image is an image of the combined target with multiple target postures captured by the camera, the combined target includes a mirror sphere and a planar target, and the planar target contains multiple planar reference points;

[0013] The coordinate optimization unit is used to set the sphere center corresponding to the initial coordinates of the sphere center as the sphere center reference point, and combine the sphere center reference point with the plane reference point to form a target reference point. Then, the target reference point is reprojected, and the error of the reprojection is averaged using the bundle adjustment algorithm to obtain the optimized coordinates of the sphere center.

[0014] The highlight acquisition unit is used to acquire the spherical highlight of each light source on the mirror sphere by combining the spherical equation and the optimized coordinates of the sphere center for each target posture under low exposure conditions of the camera.

[0015] The light source calibration unit is used to obtain the incident light direction of each light source at the corresponding spherical highlight point under each target posture, and obtain the corresponding incident light ray, and then calibrate the light source according to the incident light ray.

[0016] Thirdly, embodiments of the present invention provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the light source calibration method based on a combined target as described in the first aspect.

[0017] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the light source calibration method based on a combined target as described in the first aspect.

[0018] This invention first acquires image information of a mirror sphere under high exposure and uniform illumination conditions, and calculates the initial coordinates of the sphere center corresponding to the image information using a camera perspective imaging model. The sphere center corresponding to the initial coordinates is set as the sphere center reference point. Then, the sphere center reference point on the mirror sphere is combined with multiple planar reference points on the planar target to form the target reference point. Next, the target reference point is reprojected, and the error is averaged using a beam adjustment algorithm to obtain the optimized coordinates of the sphere center. Then, by combining the spherical equation and the optimized coordinates of the sphere center, the spherical highlight point of each light source in each target posture under low exposure conditions is obtained. Then, the incident light direction corresponding to the spherical highlight point is obtained, thereby obtaining the incident light ray corresponding to the highlight point. Finally, the light source is calibrated based on the incident light ray. This embodiment uses a combined target to calibrate the position of the light source, and obtains the positional relationship between the mirror sphere and the planar target in the combined target through a beam adjustment algorithm. This avoids the impact of uneven illumination or insufficient brightness on the extraction of the positional relationship. The position of the light source is calibrated based on the positional relationship between the mirror sphere and the planar target, realizing accurate calibration of the light source position under non-uniform illumination conditions, ensuring the accuracy of the light source position calibration, and improving the accuracy of near-field photometric stereoscopic reconstruction. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A schematic flowchart of a light source calibration method based on a combined target provided in an embodiment of the present invention;

[0021] Figure 2 This is an example diagram of the combined target in a light source calibration method based on a combined target provided in an embodiment of the present invention;

[0022] Figure 3 A schematic diagram of a sub-process of a light source calibration method based on a combined target provided in an embodiment of the present invention;

[0023] Figure 4 An example diagram of a camera perspective imaging model in a light source calibration method based on a combined target provided in an embodiment of the present invention;

[0024] Figure 5 An example diagram of the beam adjustment algorithm in a light source calibration method based on a combined target provided in an embodiment of the present invention;

[0025] Figure 6 An example diagram illustrating the specular reflection principle in a light source calibration method based on a combined target provided in an embodiment of the present invention;

[0026] Figure 7 A schematic block diagram of a light source calibration device based on a combined target provided in an embodiment of the present invention;

[0027] Figure 8 This is a schematic block diagram of a light source calibration device based on a combined target, provided as an embodiment of the present invention. Detailed Implementation

[0028] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0030] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0031] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0032] Please see below. Figure 1 , Figure 1 The flowchart of a light source calibration method based on a combined target provided in an embodiment of the present invention specifically includes steps S101 to S104.

[0033] S101. Under high exposure and uniform lighting conditions, acquire the mirror sphere image information in the combined target image, and use the camera perspective imaging model to calculate the mirror sphere image information to obtain the initial coordinates of the sphere center corresponding to the mirror sphere image information; wherein, the combined target image is an image of the combined target with multiple target postures taken by the camera, the combined target includes a mirror sphere and a planar target, and the planar target contains multiple planar reference points;

[0034] S102. Set the sphere center corresponding to the initial coordinates of the sphere center as the sphere center reference point, and combine the sphere center reference point with the plane reference point to form the target reference point. Then, reproject the target reference point and use the bundle adjustment algorithm to average the error of the reprojection to obtain the optimized coordinates of the sphere center.

[0035] S103. Under low exposure conditions of the camera, for each target posture, combine the spherical equation and the optimized coordinates of the sphere center to obtain the spherical highlight point of each light source on the mirror sphere.

[0036] S104. Obtain the incident light direction of each light source at the corresponding spherical highlight point under each target posture, and obtain the corresponding incident light ray. Then, calibrate the light source according to the incident light ray.

[0037] In this embodiment, the combined target is first calibrated. Specifically, the image information of the mirror sphere in the combined target image is obtained when the camera is in a high-exposure state and multiple light sources to be calibrated are in a uniform illumination state. The image information of the mirror sphere is then input into the camera perspective imaging model for calculation to obtain the initial coordinates of the center of the mirror sphere corresponding to the image information of the mirror sphere. Then, the center of the mirror sphere corresponding to the initial coordinates of the center of the mirror sphere is set as the reference point of the center of the sphere. The reference point of the center of the mirror sphere is combined with multiple reference points on the planar target as the target reference point. Next, the target reference point is reprojected, and the error generated by the reprojection is averaged using the bundle adjustment algorithm to obtain the optimized coordinates of the center of the sphere, thereby completing the calibration of the combined target. After obtaining the calibration results of the combined target, the light source is calibrated based on these results. Specifically, for each target pose under low camera exposure, the spherical highlight point of each light source on the mirror sphere is first obtained by combining the spherical equation and the optimized coordinates of the sphere center. Then, the incident light direction corresponding to each spherical highlight point of each light source under each target pose is obtained, thus obtaining the incident ray corresponding to the highlight point. Finally, the light source is calibrated based on the incident ray. This embodiment uses a combined target to calibrate the light source position and obtains the positional relationship between the mirror sphere and the planar target in the combined target through a beam adjustment algorithm. This avoids the impact of uneven illumination or insufficient brightness on the extraction of the positional relationship. The light source position is calibrated based on the positional relationship between the mirror sphere and the planar target, thereby achieving accurate calibration of the light source position under non-uniform illumination conditions, ensuring the accuracy of the light source position calibration, and improving the accuracy of near-field photometric stereoscopic reconstruction.

[0038] It should be noted that "high exposure" and "low exposure" in this embodiment do not have strict numerical definitions. In specific application scenarios, the camera's exposure parameters need to be determined according to the requirements of the actual scenario in order to obtain the data or information required for specific processing steps. "High exposure" and "low exposure" are only relative to each other in terms of parameters.

[0039] Combination Figure 2As shown, in a practical application scenario, the mirror sphere and the planar target in the combined target are respectively a ceramic mirror standard sphere 1 and a Lambertian-type planar target 2. The positions of the ceramic mirror standard sphere 1 and the Lambertian-type planar target 2 are fixed. The sizes of the ceramic mirror standard sphere 1 and the Lambertian-type planar target 2 can be adjusted according to technology and requirements. For example, the diameter of the ceramic mirror standard sphere 1 can be set to 38.1mm, and the size of the Lambertian-type planar target 2 can be 100×100mm. An array of black background white dots with 11×9 equally spaced intervals is set on the Lambertian-type planar target 2. The center of the white dots is the planar reference point, and the interval between the centers is 8.25mm. Among them, four white dots are set as white dots for the identification and sorting of the planar reference point. It should be noted that, provided that the pose relationship between the mirror sphere and the camera can be provided, the planar target is not limited to the Lambertian planar target. Its shape can be selected according to different practical applications. For example, it can also be a triangular planar target. In other application scenarios, the planar target can also be replaced by other reference objects that can provide the pose relationship between the mirror sphere and the camera.

[0040] Combination Figure 3 As shown, in one embodiment, step S101 includes steps S301 to S304.

[0041] S301. Obtain the sphere outline of the mirror sphere in the image coordinate system through an edge extraction algorithm, and fit the center of the mirror sphere according to the sphere outline to obtain the image coordinates of the center of the sphere and the radius of the sphere in the image coordinate system.

[0042] S302. Obtain the initial camera coordinates of the sphere's center in the camera coordinate system using the camera perspective imaging model transformation according to the following formula:

[0043]

[0044] in, The initial camera coordinates of the sphere's center in the camera coordinate system are given; r' represents the true radius of the mirror sphere; r represents the radius of the sphere's center in the image coordinate system; u0 and v0 are the principal point coordinate components of the camera; u and v are the coordinate components of the image coordinates of the sphere's center in the image coordinate system; f represents the camera focal length; and d represents the camera pixel size.

[0045] S303. Perform camera calibration on the planar target to obtain the pose relationship between the planar target and the camera in high-exposure states and camera parameters under multiple target postures; the camera parameters include camera intrinsic parameters and distortion coefficients.

[0046] S304. The initial coordinates of the sphere's center in the world coordinate system are calculated according to the following formula:

[0047]

[0048]

[0049] in, Represents the initial world coordinates of the sphere's center in the world coordinate system under the i-th target's attitude, [R (i) |T (i) ] represents the high-exposure state pose relation under the i-th target pose, R and T represent the rotation matrix and translation matrix in the high-exposure state pose relation, respectively, and i represents the i-th target pose. This represents the initial camera coordinates of the ball's center in the camera coordinate system under the i-th target's pose. The initial coordinates of the sphere's center are represented by N, and N represents the total number of target postures.

[0050] In this embodiment, the outline of the mirror sphere in the image coordinate system is first obtained by an edge extraction algorithm. The center of the mirror sphere is then obtained by fitting the extracted outline. The image coordinates and radius of the sphere in the image coordinate system are obtained from this. Then, the image coordinates and radius of the sphere in the image coordinate system are transformed by the camera perspective imaging model to obtain the initial camera coordinates of the sphere in the camera coordinate system. Furthermore, by calibrating the planar target, the pose relationship between the planar target and the camera in multiple target poses and the camera parameters are obtained. The initial coordinates of the sphere center are then calculated based on the pose relationship in the high exposure state.

[0051] This embodiment extracts the spherical outline of a mirror sphere under high exposure and uniform lighting conditions, avoiding the problem of the mirror sphere's outline blending with the background in the image. By transforming the image coordinates and the sphere radius using a camera perspective imaging model, more accurate initial camera coordinates of the sphere's center in the camera coordinate system are obtained. Furthermore, by combining images of multiple target poses, the initial coordinates of the sphere's center are processed to obtain more robust initial world coordinates. The edge extraction algorithm used in this embodiment is a commonly used image information extraction technique in the field of image processing. Edge extraction algorithms are widely used in tasks such as target detection and image segmentation, and can extract the outline of objects in an image by detecting changes in brightness. In specific embodiments, different edge extraction algorithms can be selected for image processing according to specific needs; for example, the Canny edge detection algorithm can be used.

[0052] It should be noted that in the above formula for the initial camera coordinates obtained in the camera coordinate system, the left side of the equal sign... It is a three-dimensional coordinate system, and the matrix on the right side of the equals sign is... It is a 3×1 vector, and the three sub-vectors in the matrix are... They represent The coordinate components in three dimensions are (x, y, z).

[0053] Combination Figure 4 As shown, in a specific embodiment, X is represented as a point in space. When the subscripts of X are w, c, and p, they represent the coordinates of X in the world coordinate system, camera coordinate system, and image coordinate system, respectively. The camera perspective imaging model can complete the transformation between coordinate systems. For example... Figure 4 In the world coordinate system (X), w Y w Z w Coordinates under ) and After passing through the camera coordinate system (X) c Y c Z c The origin O c Finally, the image is formed in the image coordinate system, and the coordinates in the image coordinate system are obtained. and [R|T] represents the pose relationship between the planar target and the camera, while u and v represent the coordinate components of the image coordinates of the sphere's center in the image coordinate system.

[0054] In one embodiment, step S102 includes:

[0055] Obtain the coordinates of the reference points corresponding to each plane reference point on the planar target;

[0056] The coordinates of the reference point are transformed from the world coordinate system to the image coordinate system using the camera perspective imaging model, according to the following formula:

[0057]

[0058]

[0059] in, This represents the coordinates of the reference point in the camera coordinate system. This represents the coordinates of the reference point in the world coordinate system. The coordinates of the reference point in the image coordinate system are given, n represents the total number of planar reference points on the planar target, K represents the camera intrinsic parameter matrix, and δ(X) represents the coordinates of the reference point in the image coordinate system. c ) represents the camera distortion compensation term, and z represents the scale factor;

[0060] The initial coordinates of the sphere center are transformed from the world coordinate system to the image coordinate system, and the initial coordinates of the sphere center reference point and the reference point coordinates corresponding to the plane reference point are combined into the target reference coordinates.

[0061] The objective function is constructed according to the following formula, combined with the target reference coordinates:

[0062]

[0063] Where θ represents the parameter to be optimized. m represents the total number of target reference points, where m = n + 1, m = 1, 2, ..., M, and ψ represents the camera parameters. This represents the reference coordinates of the j-th target in the i-th target pose within the reprojected image coordinate system. This represents the coordinates of the j-th reference point in the image coordinate system under the i-th target posture detected.

[0064] In this embodiment, the coordinates of all planar reference points on the planar target in the world coordinate system are first obtained. These world coordinate coordinates are then input into the camera's perspective imaging model and converted to the image coordinate system. Finally, an optimization objective function is constructed based on the image coordinate system reference point coordinates. Regarding the reference point coordinates, since the planar target cannot be perfectly flat, the planar reference points on the target may deviate due to printing errors and other factors, resulting in coordinate deviations. Therefore, a bundle adjustment algorithm can be introduced to eliminate the adverse effects of reference point coordinate errors.

[0065] Compared with existing technologies, this embodiment does not require extracting the center of the mirror sphere by extracting the image contour of the mirror sphere. Instead, it uses a planar target to provide the pose relationship between the mirror sphere and the camera, and extracts the center of the mirror sphere by identifying the image edges and extracting planar reference points that are less affected by uneven lighting. This reduces the error in extracting the center of the mirror sphere and the environmental requirements for light source position calibration, and achieves high-precision light source position calibration under non-uniform lighting conditions.

[0066] In one specific embodiment, the camera parameter ψ = {[f u ,f v ,u0,v0,s],[k1,k2,k3,p1,p2]}, which includes camera intrinsic parameters [f u ,f v The camera intrinsic parameters [u0, v0, s] and distortion coefficients [k1, k2, k3, p1, p2] are important parameters in camera calibration. Camera intrinsic parameters describe the optical properties of the camera and the physical properties of pixels on the image sensor; they mainly include the focal length f. u and f vThe parameters include principal point coordinates (u0, v0), tilt factor s, etc.; distortion coefficients mainly include radial distortion k1, k2, k3 and tangential distortion p1, p2. Radial distortion is caused by the shape of the camera lens itself, while tangential distortion is caused by the camera lens itself not being parallel to the camera sensor plane or image plane. Distortion coefficients are used to describe the deformation in the image caused by lens distortion. Camera intrinsic parameters and distortion coefficients are important parameters used to describe the optical properties of the camera and image quality. By calibrating the camera, camera intrinsic parameters and distortion coefficients can be obtained, and then used to correct distortion in the image, which can improve image quality. At the same time, the pose relationship between the planar target and the camera [R|T] is used as the camera extrinsic parameter, and the camera extrinsic parameter, camera intrinsic parameter and distortion coefficient are combined as camera calibration parameters.

[0067] In another specific embodiment, the total number of target reference points within the combined target is m, where m = n + 1 and m = 1, 2, ..., n, M. That is, the target reference points include n planar reference points and 1 sphere center reference point. For ease of representation, the sphere center reference point is taken as the Mth reference point, i.e., n = m - 1. For example, if the total number of target reference points within the combined target is 100, then the sphere center reference point is the 100th target reference point, and the number of planar reference points n is 99.

[0068] Combination Figure 5 As shown, in one embodiment, step S102 further includes:

[0069] The optimization objective function is minimized to jointly optimize and adjust the coordinates of the reference point and the parameters to be optimized, thereby obtaining the joint adjusted coordinates of the sphere's center.

[0070] The joint adjustment coordinates of the sphere center are corrected using the bundle adjustment algorithm according to the following formula to obtain the optimized coordinates of the sphere center in the world coordinate system:

[0071]

[0072] in, This represents the optimized coordinates of the sphere's center. represents the true distance between each planar reference point on the planar target, and l represents the distance between each planar reference point on the planar target after processing by the bundle adjustment algorithm. This indicates the joint adjustment coordinates.

[0073] In this embodiment, the objective function is first minimized to achieve joint optimization and adjustment of the reference point coordinates and the parameters to be optimized. Based on the result of the joint optimization and adjustment, the joint adjusted coordinates of the sphere center are obtained. Finally, the joint adjusted coordinates of the sphere center are corrected to obtain the optimized coordinates of the sphere center. In specific application scenarios, since the objective function is constructed based on the camera perspective imaging model and the principle of bundle adjustment, the reference point coordinates and camera parameters need to be included in the optimization. This will lead to a scaling of the spatial scale in the world coordinate system. Therefore, the joint adjusted coordinates of the sphere center need to be corrected to obtain more accurate optimized coordinates of the sphere center.

[0074] In bundle adjustment algorithms, a bundle refers to the set of rays originating from each spatial point and converging at the optical center of each camera, such as... Figure 5 The left side of the image shows the reference point in the combined target under multiple target postures. and The emitted light beams converge at the optical center of each camera C to form a beam. Adjustment refers to averaging errors by adjusting the pose relationships and their associated reference point coordinates and camera parameters, such as... Figure 5 The right side of the diagram represents the variables to be optimized based on error averaging. The joint optimization adjustment, the reprojection error in the beam adjustment refers to the sum of the squared distances between the image coordinates of the reference point obtained by reprojection based on the camera perspective imaging model and the image coordinates of the reference point detected by the beam adjustment.

[0075] In one specific embodiment, the camera parameters ψ, the high-exposure pose relationship [RT], and the planar reference point are used. and the initial coordinates of the sphere's center As the initial value of the objective function, the initial value of the parameter θ to be optimized is obtained. Setting the initial value of the objective function can improve the convergence speed of the optimization process and ensure the global optimality of the optimization result. By minimizing the optimization objective function, the joint optimization and adjustment of the reference point coordinates and camera calibration parameters can be achieved, resulting in more accurate joint adjustment coordinates.

[0076] In another specific embodiment, since in the camera perspective imaging model, when any reference point is imaged by the camera, the equations for the coordinates of that reference point in both the horizontal and vertical components can be listed, based on this and combined with the parameter θ to be optimized and the camera parameter ψ, it can be concluded that: only when 2MN>10+6N+3M can the joint optimization and adjustment of the reference point coordinates and the camera calibration parameters be achieved. Here, 2MN represents the reference point coordinates in the image coordinate system. The equations for the horizontal and vertical components are: 10 represents the ten parameters included in the camera parameter ψ; 6N represents R and T in the camera extrinsic parameters [R|T], where T is a three-dimensional vector and R is a 3×3 rotation matrix that can be obtained by converting a three-dimensional rotation vector. 6N is composed of two "three-dimensional" vectors; 3M represents the three dimensions of the reference point coordinates.

[0077] In one embodiment, step S103 includes:

[0078] Images of the mirror sphere under different target postures were acquired under low-exposure conditions of the camera.

[0079] Highlights are extracted from the mirror sphere image, and corresponding planar reference points on the planar target in the target pose are identified to obtain the pose relationship between the planar target and the camera in low-exposure state.

[0080] Transform the optimized coordinates of the sphere center in the world coordinate system to the camera coordinate system according to the following formula:

[0081]

[0082] in, This represents the optimized coordinates of the sphere's center in the camera coordinate system. R represents the optimized coordinates of the sphere's center in the world coordinate system. '(i) and T '(i) These represent the rotation and translation matrices in the pose relationship of the low-exposure state, respectively.

[0083] The intersection point of the spherical surface of the combined target and the reflected light ray is obtained according to the spherical equation, and the intersection point is taken as the highlight point of the spherical surface.

[0084] In this embodiment, firstly, with the camera in low exposure mode, images of the mirror sphere are acquired corresponding to different target poses. Highlights are extracted from these images, and planar reference points on the planar target in each target pose are identified. This yields the pose relationship between the planar target and the camera in low exposure mode. Then, to facilitate the calculation of the light source position, the optimized coordinates of the sphere center are transformed from the world coordinate system to the camera coordinate system. Finally, based on the transformed optimized coordinates, the intersection point of the sphere and the reflected light rays in the combined target is obtained according to the spherical equation, and this intersection point is used as the spherical highlight point. In specific application scenarios, the optimized coordinates of the sphere center and the true radius of the mirror sphere can be combined to establish a spherical equation representing the mirror sphere in space. Then, based on the spherical equation, the normal vector and reflected light rays corresponding to the highlight point are calculated, thereby obtaining the intersection point of the sphere and the reflected light rays in the combined target, and thus the spherical highlight point.

[0085] Combination Figure 6As shown, in one embodiment, step S104 includes:

[0086] For each light source under each target posture, obtain the normal of the corresponding spherical specular point;

[0087] The incident light rays of each light source at the highlight point on the spherical surface are calculated according to the following formula based on the principle of specular reflection:

[0088]

[0089] in, Let represent the incident light ray at the spherical highlight point of the j-th light source under the i-th target posture. The spherical specular highlight represents the reflected light from the j-th light source in the i-th target posture. The normal to the j-th light source is represented by the i-th target orientation.

[0090] In this embodiment, the normal to the spherical specular highlight point corresponding to each light source under each target posture is first obtained. Then, based on the principle of specular reflection, the incident light rays of each light source at the spherical specular highlight point are calculated using the normal. In specific application scenarios, based on the principle of specular reflection, it can be known that the incident and reflected light rays at the spherical specular highlight point are symmetrical about the normal. The intersection point between the reflected light rays under different target postures is O, and the light rays reaching the image plane from point O are equal to the reflected light rays at the spherical specular highlight point. Specifically, as shown... Figure 6 As shown, the incident light emitted by the light source L in target posture 1 is reflected to point P on the image plane after passing through the corresponding highlight point H on the sphere. Similarly, the incident light emitted by the light source L in target posture 2 is also reflected to the image plane and intersects with the reflected light formed in target posture 1 at point O. In the figure, [R|T] represents the pose relationship between the planar target and the camera, and θ1 and θ2 represent the incident angle and reflection angle corresponding to the highlight point H, respectively.

[0091] In one embodiment, step S104 further includes:

[0092] Determine whether the incident rays from each light source at the highlight point on the spherical surface intersect;

[0093] If an intersection point is determined to exist, the intersection point will be used as the final light source calibration result.

[0094] If no intersection is found, the target point that is closest to each incident ray is selected, and the target point is used as the final light source calibration result.

[0095] In this embodiment, the system determines whether there is an intersection point between the incident rays at the spherical speckle. If an intersection point exists, it is used as the final light source calibration result. If no intersection point exists, the target point with the shortest distance to each incident ray is selected as the final light source calibration result. Specifically, the intersection point of the incident rays from each light source to the spherical speckle can be obtained by observing the intersection of rays, thus obtaining the light source calibration result. However, in some application scenarios, different incident rays at the spherical speckle may not have a common intersection point. Therefore, to reduce the error of the light source calibration result, when no common intersection point exists, the target point with the shortest distance to each incident ray is selected as the final light source calibration result.

[0096] In one specific embodiment, the camera used in the light source calibration method based on the combined target provided in this embodiment is a MER-503-36U3M / C Daheng camera, employing eight LUXEON Rebel LED light sources. The camera resolution is 2448*2048. By synchronizing the camera and LED light sources, eight images can be continuously acquired. The working distance between the camera and the combined target is approximately 350mm. The diameter of the mirror sphere in the combined target is 38.1mm, and the pattern of the planar target is an 11×9 black background with white dots, with the center spacing of the white dots being 8.25mm. The calibration target size is 100×100mm. Based on this, light source calibration is performed. The optimized coordinates of the sphere center of the combined target and the light source calibration results are shown in Tables 1 and 2.

[0097]

[0098] Table 1 (Calibration results using only a mirror sphere light source)

[0099]

[0100] Table 2 (Calibration Results of Combined Targets)

[0101] Based on the characteristic that the light sources of the camera are distributed on the same plane, the diagonal distance error can be used to effectively evaluate the light source distribution, thereby determining the accuracy of the calibrated light source position. By comparing the light source calibration results based on the diagonal distance of the LED light source in the photometric stereo system with the light source calibration results based on the combined target provided in this embodiment and the light source calibration method using only a mirror sphere under the same non-uniform illumination conditions, for example, the distance error is calculated based on the data in Tables 1 and 2, and the comparison results shown in Table 3 are obtained. It can be shown that the light source calibration method based on the combined target provided in this embodiment can effectively improve the accuracy of light source calibration compared with the light source calibration method using only a mirror sphere.

[0102]

[0103]

[0104] Table 3 (Diagonal error of light source / in mm)

[0105] Here, the diagonal distance of the LED light source is measured using a laser tracker.

[0106] Figure 7 This is a schematic block diagram of a light source calibration device 700 based on a combined target provided in this embodiment. The device 700 includes:

[0107] The initial coordinate acquisition unit 701 is used to acquire the mirror sphere image information in the combined target image under high exposure and uniform lighting conditions of the camera, and to calculate the mirror sphere image information using the camera perspective imaging model to obtain the initial coordinates of the sphere center corresponding to the mirror sphere image information; wherein, the combined target image is an image of the combined target with multiple target postures taken by the camera, the combined target includes a mirror sphere and a planar target, and the planar target contains multiple planar reference points;

[0108] The coordinate optimization unit 702 is used to set the sphere center corresponding to the initial coordinates of the sphere center as the sphere center reference point, and combine the sphere center reference point with the plane reference point to form a target reference point. Then, the target reference point is reprojected, and the error of the reprojection is averaged using the bundle adjustment algorithm to obtain the optimized coordinates of the sphere center.

[0109] The highlight acquisition unit 703 is used to acquire the spherical highlight of each light source on the mirror sphere by combining the spherical equation and the optimized coordinates of the sphere center for each target posture under low exposure conditions of the camera.

[0110] The light source calibration unit 704 is used to obtain the incident light direction of each light source at the corresponding spherical highlight point under each target posture, and obtain the corresponding incident light ray, and then calibrate the light source according to the incident light ray.

[0111] Combination Figure 8 As shown, in one embodiment, the initial coordinate acquisition unit 701 includes:

[0112] The edge extraction unit 801 is used to obtain the sphere outline of the mirror sphere in the image coordinate system through the edge extraction algorithm, and to fit the center of the mirror sphere according to the sphere outline to obtain the image coordinates of the center of the sphere and the radius of the sphere in the image coordinate system.

[0113] The initial coordinate transformation unit 802 is used to transform the initial camera coordinates of the sphere center in the camera coordinate system according to the following formula using the camera perspective imaging model:

[0114]

[0115] in, The initial camera coordinates of the sphere's center in the camera coordinate system are given; r' represents the true radius of the mirror sphere; r represents the radius of the sphere's center in the image coordinate system; u0 and v0 are the principal point coordinate components of the camera; u and v are the coordinate components of the image coordinates of the sphere's center in the image coordinate system; f represents the camera focal length; and d represents the camera pixel size.

[0116] The camera calibration unit 803 is used to perform camera calibration on the planar target to obtain the pose relationship between the planar target and the camera in high-exposure states and camera parameters under multiple target postures; the camera parameters include camera intrinsic parameters and distortion coefficients.

[0117] The initial coordinate calculation unit 804 is used to calculate the initial coordinates of the sphere center in the world coordinate system according to the following formula:

[0118]

[0119]

[0120] in, Represents the initial world coordinates of the sphere's center in the world coordinate system under the i-th target's attitude, [R (i) T (i) ] represents the high-exposure state pose relation under the i-th target pose, R and T represent the rotation matrix and translation matrix in the high-exposure state pose relation, respectively, and i represents the i-th target pose. This represents the initial camera coordinates of the ball's center in the camera coordinate system under the i-th target's pose. The initial coordinates of the sphere's center are represented by N, and N represents the total number of target postures.

[0121] In one embodiment, the coordinate optimization unit 702 includes:

[0122] The reference point acquisition unit is used to acquire the coordinates of the reference points corresponding to each plane reference point on the plane target.

[0123] The reference point transformation unit is used to transform the coordinates of the reference point from the world coordinate system to the image coordinate system using the camera perspective imaging model according to the following formula:

[0124]

[0125]

[0126] in, This represents the coordinates of the reference point in the camera coordinate system. This represents the coordinates of the reference point in the world coordinate system. The coordinates of the reference point in the image coordinate system are given, n represents the total number of planar reference points on the planar target, K represents the camera intrinsic parameter matrix, and δ(X) represents the coordinates of the reference point in the image coordinate system. c ) represents the camera distortion compensation term, and z represents the scale factor;

[0127] The reference coordinate combination unit is used to transform the initial coordinates of the sphere center from the world coordinate system to the image coordinate system, and to combine the initial coordinates of the sphere center and the reference point coordinates corresponding to the sphere center reference point and the plane reference point into the target reference coordinates.

[0128] The optimization function construction unit is used to construct the optimization objective function according to the following formula, combined with the target reference coordinates:

[0129]

[0130] Where θ represents the parameter to be optimized. m represents the total number of target reference points, where m = n + 1, m = 1, 2, ..., M, and ψ represents the camera parameters. This represents the reference coordinates of the j-th target in the i-th target pose within the reprojected image coordinate system. This represents the coordinates of the j-th reference point under the i-th target posture.

[0131] In one embodiment, the coordinate optimization unit 702 further includes:

[0132] The minimization processing unit is used to minimize the optimization objective function to jointly optimize and adjust the reference point coordinates and the parameters to be optimized, so as to obtain the joint adjusted coordinates of the sphere center.

[0133] The correction processing unit is used to correct the joint adjustment coordinates of the sphere center using a bundle adjustment algorithm according to the following formula, to obtain the optimized coordinates of the sphere center in the world coordinate system:

[0134]

[0135] in, This represents the optimized coordinates of the sphere's center. represents the true distance between each planar reference point on the planar target, and l represents the distance between each planar reference point on the planar target after processing by the bundle adjustment algorithm. This indicates the joint adjustment coordinates.

[0136] In one embodiment, the highlight acquisition unit 703 includes:

[0137] The image acquisition unit is used to acquire images of the mirror sphere under different target postures when the camera is under low exposure conditions.

[0138] The highlight extraction unit is used to extract highlights from the mirror sphere image and identify planar reference points on the planar target in the corresponding target pose, thereby obtaining the pose relationship between the planar target and the camera in low-exposure state.

[0139] An optimized coordinate transformation unit is used to transform the optimized coordinates of the sphere center in the world coordinate system to the camera coordinate system according to the following formula:

[0140]

[0141] in, This represents the optimized coordinates of the sphere's center in the camera coordinate system. R represents the optimized coordinates of the sphere's center in the world coordinate system. '(i) and T '(i) These represent the rotation and translation matrices in the pose relationship of the low-exposure state, respectively.

[0142] The intersection point acquisition unit is used to acquire the intersection point of the sphere of the combined target and the reflected light rays according to the spherical equation, and to use the intersection point as the highlight point of the sphere.

[0143] In one embodiment, the light source calibration unit 704 includes:

[0144] The normal acquisition unit is used to acquire the normal of the corresponding spherical specular point for each light source under each target posture.

[0145] The specular reflection unit is used to calculate the incident light rays of each of the light sources at the specular highlight point according to the following formula, based on the principle of specular reflection:

[0146]

[0147] in, Let represent the incident light ray at the spherical highlight point of the j-th light source under the i-th target posture. The spherical specular highlight represents the reflected light from the j-th light source in the i-th target posture. The normal to the j-th light source is represented by the i-th target orientation.

[0148] In one embodiment, the light source calibration unit 704 further includes:

[0149] The intersection point determination unit is used to determine whether there is an intersection point between the incident rays of each light source at the highlight point on the spherical surface.

[0150] The first determination unit is used to determine the intersection point as the final light source calibration result if an intersection point is determined to exist.

[0151] The second determination unit is used to select the target point that is closest to each incident light ray if no intersection point is determined, and to use the target point as the final light source calibration result.

[0152] Since the embodiments of the apparatus and the embodiments of the method correspond to each other, please refer to the description of the embodiments of the method for the embodiments of the apparatus, which will not be repeated here.

[0153] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed, can perform the steps provided in the above embodiments. The storage medium may include various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0154] This invention also provides a computer device, which may include a memory and a processor. The memory stores a computer program, and when the processor calls the computer program in the memory, it can implement the steps provided in the above embodiments. Of course, the computer device may also include various network interfaces, power supplies, and other components.

[0155] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to in the method section. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of the claims of this application.

[0156] It should also be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. A light source calibration method based on a combined target, characterized in that, include: Under high exposure and uniform lighting conditions, the mirror sphere image information in the combined target image is acquired, and the mirror sphere image information is calculated using the camera's perspective imaging model to obtain the initial coordinates of the sphere's center corresponding to the mirror sphere image information; wherein, the combined target image is an image of the combined target with multiple target postures captured by the camera, the combined target includes a mirror sphere and a planar target, and the planar target contains multiple planar reference points; The sphere center corresponding to the initial coordinates of the sphere center is set as the sphere center reference point, and the sphere center reference point and the plane reference point are combined to form the target reference point. Then, the target reference point is reprojected, and the error of the reprojection is averaged using the bundle adjustment algorithm to obtain the optimized coordinates of the sphere center. Under low exposure conditions of the camera, for each target posture, the spherical highlight point of each light source on the mirror sphere is obtained by combining the spherical equation and the optimized coordinates of the sphere center. Obtain the incident light direction of each light source at the corresponding spherical highlight point under each target posture, and obtain the corresponding incident light ray. Then, calibrate the light source based on the incident light ray.

2. The light source calibration method based on a combined target according to claim 1, characterized in that, The step of acquiring the mirror sphere image information in the combined target image and calculating the mirror sphere image information using a camera perspective imaging model to obtain the initial coordinates of the sphere center corresponding to the mirror sphere image information includes: The outline of the mirror sphere in the image coordinate system is obtained by an edge extraction algorithm, and the center of the mirror sphere is fitted according to the outline to obtain the image coordinates of the center of the sphere and the radius of the sphere in the image coordinate system. The initial camera coordinates of the sphere's center in the camera coordinate system are obtained using the following formula, based on the camera perspective imaging model transformation: in, The initial camera coordinates of the sphere's center in the camera coordinate system are given; r' represents the true radius of the mirror sphere; r represents the radius of the sphere's center in the image coordinate system; u0 and v0 are the principal point coordinate components of the camera; u and v are the coordinate components of the image coordinates of the sphere's center in the image coordinate system; f represents the camera focal length; and d represents the camera pixel size. Camera calibration is performed on the planar target to obtain the pose relationship between the planar target and the camera in high-exposure states and camera parameters under multiple target postures; the camera parameters include camera intrinsic parameters and distortion coefficients. The initial coordinates of the sphere's center in the world coordinate system are calculated using the following formula: in, Represents the initial world coordinates of the sphere's center in the world coordinate system under the i-th target's attitude, [R (i) T (i) ] represents the high-exposure state pose relation under the i-th target pose, R and T represent the rotation matrix and translation matrix in the high-exposure state pose relation, respectively, and i represents the i-th target pose. This represents the initial camera coordinates of the ball's center in the camera coordinate system under the i-th target's pose. The initial coordinates of the sphere's center are represented by N, and N represents the total number of target postures.

3. The light source calibration method based on a combined target according to claim 2, characterized in that, The process involves setting the sphere center corresponding to the initial coordinates of the sphere center as the sphere center reference point, combining the sphere center reference point with the plane reference point to form the target reference point, then reprojecting the target reference point, and using a bundle adjustment algorithm to average the reprojection error to obtain the optimized coordinates of the sphere center. This includes: Obtain the coordinates of the reference points corresponding to each plane reference point on the planar target; The coordinates of the reference point are transformed from the world coordinate system to the image coordinate system using the camera perspective imaging model, according to the following formula: in, This represents the coordinates of the reference point in the camera coordinate system. This represents the coordinates of the reference point in the world coordinate system. The coordinates of the reference point in the image coordinate system are given, n represents the total number of planar reference points on the planar target, K represents the camera intrinsic parameter matrix, and δ(X) represents the coordinates of the reference point in the image coordinate system. c ) represents the camera distortion compensation term, and z represents the scale factor; The initial coordinates of the sphere center are transformed from the world coordinate system to the image coordinate system, and the initial coordinates of the sphere center reference point and the reference point coordinates corresponding to the plane reference point are combined into the target reference coordinates. The objective function is constructed according to the following formula, combined with the target reference coordinates: Where θ represents the parameter to be optimized. m represents the total number of target reference points, where m = n + 1, m = 1, 2, ..., M, and ψ represents the camera parameters. This represents the reference coordinates of the j-th target in the i-th target pose within the reprojected image coordinate system. This represents the coordinates of the j-th reference point in the i-th target pose within the image coordinate system.

4. The light source calibration method based on a combined target according to claim 3, characterized in that, The process of setting the sphere center reference point as the sphere center reference point corresponding to the initial coordinates of the sphere center, combining the sphere center reference point with the plane reference point to form a target reference point, then reprojecting the target reference point, and using a bundle adjustment algorithm to average the error of the reprojection to obtain the optimized coordinates of the sphere center, further includes: The objective function is minimized to jointly optimize and adjust the parameters to be optimized, thereby obtaining the joint adjusted coordinates of the sphere's center. The joint adjustment coordinates of the sphere center are corrected using the bundle adjustment algorithm according to the following formula to obtain the optimized coordinates of the sphere center in the world coordinate system: in, This represents the optimized coordinates of the sphere's center. represents the true distance between each planar reference point on the planar target, and l represents the distance between each planar reference point on the planar target after processing by the bundle adjustment algorithm. This indicates the joint adjustment coordinates.

5. The light source calibration method based on a combined target according to claim 4, characterized in that, For each target posture, combining the spherical equation and the optimized coordinates of the sphere's center, the spherical highlight point of each light source on the mirror sphere is obtained, including: Images of the mirror sphere under different target postures were acquired under low-exposure conditions of the camera. Highlights are extracted from the mirror sphere image, and corresponding planar reference points on the planar target in the target pose are identified to obtain the pose relationship between the planar target and the camera in low-exposure state. Transform the optimized coordinates of the sphere center in the world coordinate system to the camera coordinate system according to the following formula: in, This represents the optimized coordinates of the sphere's center in the camera coordinate system. R' represents the optimized coordinates of the sphere's center in the world coordinate system. (i) and T' (i) These represent the rotation and translation matrices in the pose relationship of the low-exposure state, respectively. The intersection point of the spherical surface of the combined target and the reflected light ray is obtained according to the spherical equation, and the intersection point is taken as the highlight point of the spherical surface.

6. The light source calibration method based on a combined target according to claim 5, characterized in that, The process of obtaining the incident light direction of each light source at the corresponding spherical highlight point under each target posture, obtaining the corresponding incident ray, and then calibrating the light source based on the incident ray includes: For each light source under each target posture, obtain the normal of the corresponding spherical specular point; The incident light rays of each light source at the highlight point on the spherical surface are calculated according to the following formula based on the principle of specular reflection: in, Let represent the incident light ray at the spherical highlight point of the j-th light source under the i-th target posture. The spherical specular highlight represents the reflected light from the j-th light source in the i-th target posture. The normal to the j-th light source is represented by the i-th target orientation.

7. The light source calibration method based on a combined target according to claim 6, characterized in that, The step of obtaining the incident light direction of each light source at the corresponding spherical highlight point under each target posture, obtaining the corresponding incident ray, and then calibrating the light source based on the incident ray, further includes: Determine whether the incident rays from each light source at the highlight point on the spherical surface intersect; If an intersection point is determined to exist, the intersection point will be used as the final light source calibration result. If no intersection is found, the target point that is closest to each incident ray is selected, and the target point is used as the final light source calibration result.

8. A light source calibration device based on a combined target, characterized in that, include: The initial coordinate acquisition unit is used to acquire the mirror sphere image information in the combined target image under high exposure and uniform lighting conditions of the camera, and to calculate the mirror sphere image information using the camera's perspective imaging model to obtain the initial coordinates of the sphere center corresponding to the mirror sphere image information; wherein, the combined target image is an image of the combined target with multiple target postures captured by the camera, the combined target includes a mirror sphere and a planar target, and the planar target contains multiple planar reference points; The coordinate optimization unit is used to set the sphere center corresponding to the initial coordinates of the sphere center as the sphere center reference point, and combine the sphere center reference point with the plane reference point to form a target reference point. Then, the target reference point is reprojected, and the error of the reprojection is averaged using the bundle adjustment algorithm to obtain the optimized coordinates of the sphere center. The highlight acquisition unit is used to acquire the spherical highlight of each light source on the mirror sphere by combining the spherical equation and the optimized coordinates of the sphere center for each target posture under low exposure conditions of the camera. The light source calibration unit is used to obtain the incident light direction of each light source at the corresponding spherical highlight point under each target posture, and obtain the corresponding incident light ray, and then calibrate the light source according to the incident light ray.

9. A computer device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the light source calibration method based on a combined target as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the light source calibration method based on a combined target as described in any one of claims 1 to 7.

Images