A method and apparatus for ground calibration of a long focal length linear array optical camera

Abstract

This invention discloses a ground calibration method and apparatus for a long-focal-length linear array optical camera, comprising: combining a slit target with a collimator, supplemented by a beam splitter and a plane mirror to construct an infinitely far planar target, and establishing a mathematical model for high-precision ground calibration of the linear array camera; accurately extracting the center of the integral image of the linear array camera through a multi-scale image extraction method; accurately calculating the coordinates of the target feature points using the geometric constraints of the slit target's own scribe lines; virtually orthogonalizing the linear array camera into a planar array based on the linear array camera's imaging model and performing subsequent calculations; constructing a calibration system apparatus to verify the feasibility and effectiveness of ground calibration of the long-focal-length linear array optical camera. This invention is suitable for ground calibration of long-focal-length linear array optical cameras in a ground laboratory environment, and has the advantages of high precision, high speed, and low cost, which is of great significance for ensuring high-resolution imaging and high-precision positioning in aerospace, remote sensing and mapping and other fields.

Description

Technical Field

[0001] This invention relates to the technical field of optical sensor calibration, and specifically to a ground calibration method and apparatus for a telephoto linear array optical camera. Background Technology

[0002] Long-focal-length linear array optical cameras, with their wide range, long distance, high resolution, and pushbroom imaging capabilities, are widely used in aerospace remote sensing, aerial surveying, and other fields, possessing extremely important practical application value. With the increasing need to safeguard national sovereignty and territorial integrity, the rapidly developing high-resolution Earth observation technology is of paramount importance to ensuring national security. Among these, long-focal-length linear array optical cameras in various spaceborne and airborne systems are key components for acquiring high-resolution Earth images. However, due to the complexity of the observation environment and limitations in technology, the geometric quality of current high-resolution satellite images often falls short of the high-precision geometric positioning accuracy requirements in practical applications, significantly restricting their practical application and service capabilities. Therefore, high-precision calibration of the optical linear array cameras on the ground or in orbit, along with image error correction, is necessary to ensure accurate Earth observation.

[0003] Due to limitations in optical lens design and manufacturing processes, lenses cannot always maintain a linear transformation according to the ideal imaging model during practical applications. A certain degree of nonlinear distortion always exists, causing the actual imaging point to deviate from the ideal imaging point. This error is particularly pronounced for telephoto cameras, severely reducing their spatial positioning accuracy. Therefore, before deploying a telephoto optical camera in practical use, its intrinsic parameters must be calibrated to ensure high-precision measurements. Especially for aerospace optical cameras, ground calibration before launch is fundamental for achieving high-precision detection with large-aperture telephoto optical systems. It is crucial for ensuring high-precision detection after orbital insertion and provides initial parameters for subsequent on-orbit calibration. However, as camera focal lengths and sizes increase, the requirements for testing equipment and sites become increasingly stringent, rendering existing calibration modes and methods inadequate for their calibration needs.

[0004] Currently, most long-focal-length cameras use precision angle measurement for ground calibration. Wu Guodong et al. (Wu Guodong, Han Bing, He Xu. Laboratory calibration method for geometric parameters of linear CCD cameras using precision angle measurement [J]. Optics and Precision Engineering, 2007(10):1628-1632.) used a high-precision two-dimensional precision turntable and collimator to perform high-precision calibration of linear array cameras, achieving a calibration accuracy of micrometers for principal distance and principal point. Yuan Guoqin et al. (Yuan Guoqin, Ding Yalin, Hui Shouwen, Liu Liguo, Yu Chunfeng. Group progressive calibration algorithm for surveying cameras based on precision angle measurement) [J]. Acta Optica Sinica, 2012, 32(01): 134-139.) proposes a grouped asymptotic algorithm based on the precision angle measurement method. The grouped asymptotic approximation and weight theory are used to improve the precision angle measurement method, avoiding the influence of theoretical errors and sampling points on the algorithm accuracy and improving the calibration accuracy. The precision angle measurement method is intuitive and simple and easy to implement, but it has strict requirements for the calibration environment. The accuracy of the turntable directly determines the calibration accuracy, and the production of high-precision turntables is expensive and time-consuming. The precision angle measurement method also requires a large amount of manual intervention and adjustment, and cannot achieve rapid calibration.

[0005] While traditional geometric calibration methods based on ground control points can effectively correct image system errors, they suffer from drawbacks such as high cost and poor timeliness due to their over-reliance on high-precision ground calibration field data. Some satellites' cameras in other bands cannot image control points observed from the Earth, and global ground control points have not yet been established in the overseas observation areas of reconnaissance satellites, thus preventing on-orbit calibration. Furthermore, on-orbit polynomial distortion models inherently suffer from the drawback of easily getting trapped in local minima; they are only mathematically optimal and lack physical meaning. While they offer high accuracy after calibration, their accuracy cannot be guaranteed to be optimal over long-term operation. Therefore, conducting ground calibration of long-focal-length, large-aperture optical cameras is of great significance for achieving high-precision long-distance detection.

[0006] Currently, the key technical challenges restricting the ground calibration of long focal length line array optical cameras mainly include: (1) The laboratory high-precision calibration method relying on a three-axis turntable has stringent requirements for testing equipment and environment, is time-consuming to assemble and adjust, has low calibration efficiency, uncontrollable system / human error, and the angle measurement accuracy directly determines the calibration accuracy, making the overall calibration mode quite complex; (2) The target imaging quality is uneven, and the image feature extraction accuracy cannot guarantee the high-precision calibration requirements. At the same time, the longer the focal length, the more sensitive the calibration results are to image noise, meaning that even a small positioning deviation can lead to a large calibration error; (3) The calibration of long focal length, large aperture cameras requires a larger turntable and a collimator with a longer focal length and a larger field of view, meaning that the cost of long focal length line array camera calibration is high. Therefore, overcoming the above key technical challenges and researching a ground-based calibration method for long focal length line array optical cameras is of great significance for ensuring high-resolution imaging and high-precision positioning of optical payloads in aerospace and other fields. Summary of the Invention

[0007] To address the aforementioned technical problems, this invention provides a ground calibration method and apparatus for long-focal-length linear array optical cameras. This apparatus features a compact and small structure, high calibration efficiency, strong versatility, and simple system composition, enabling ground calibration of new long-focal-length linear array optical cameras in the laboratory.

[0008] To achieve the above objectives, the present invention adopts the following technical solution:

[0009] A ground calibration method for a long-focal-length linear array optical camera includes the following steps:

[0010] Step a: In a ground laboratory environment, a high-precision ground calibration hardware device for a long focal length line array optical camera is set up. Only a planar slit target is used in conjunction with a collimator, supplemented by a beam splitter and a planar reflector to achieve high-precision ground calibration of the long focal length line array optical camera.

[0011] Step b: Perform image feature modeling and centroid extraction. High-precision extraction of image feature centroids based on multi-scale models is used for high-precision calibration and solution of camera intrinsic parameters.

[0012] Step c: By designing a slit target and modeling the target feature points based on the geometric constraints of the slit target's own markings, the spatial point coordinates and the ray vector at infinity are obtained.

[0013] Step d: By performing virtual orthogonalization on the linear array camera, a virtual area array camera imaging model and an intrinsic parameter calibration mathematical model are obtained, which are used for subsequent solutions to the linear array camera problem.

[0014] Step e: The oscillating plane mirror changes the incident direction of the second set of light rays, covering the entire field of view of the telephoto linear array optical camera. Substitute these steps into the above steps and obtain the intrinsic parameters of the linear array optical camera through nonlinear optimization.

[0015] Furthermore, in step a, the infinity imaging target is designed as a miniature planar slit target mounted on the focal plane of the collimator to generate an infinity planar target. The target is split into two groups of infinity target rays by a beam splitter. The first group of rays directly enters the camera to be calibrated, and the second group enters the field of view of the camera to be calibrated synchronously after being reflected by a mirror, thus realizing a calibration mode that does not rely on a large-aperture collimator and a large high-precision turntable.

[0016] Furthermore, in step b, an adaptive multi-scale centroid extraction method is used to extract the centroid of feature points in the integral imaging of the slit target by the linear array camera. This allows centroid extraction to be completed under optimal scale conditions for different image centers, thereby improving the accuracy of image centroid extraction.

[0017] This invention also provides a ground calibration device for a long-focal-length linear array optical camera, comprising an image analysis and calculation module, a target imaging and image acquisition module, and an infinity target module. The image analysis and calculation module includes an image feature extraction and linear array camera intrinsic parameter modeling and solving module, which processes the image of the infinity target and optimizes the camera's intrinsic parameters based on the linear array camera intrinsic parameter model. The target imaging and image acquisition module includes a long-focal-length linear array camera optical lens, a high-resolution camera, and a control box, used to complete the imaging of the infinity target and provide data for the image analysis and calculation module. The infinity target module comprises a slit target, a collimator, a beam splitter, and a plane mirror, used to generate the infinity target for data acquisition by the target imaging and image acquisition module.

[0018] The advantages of this invention compared to the prior art are as follows:

[0019] This invention combines a slit target with a collimator, supplemented by a beam splitter and a plane mirror to construct an infinitely far planar target, establishing a mathematical model for high-precision ground calibration of a linear array camera. It achieves accurate extraction of the center of the integral image from the linear array camera through a multi-scale image extraction method; it uses the geometric constraints of the slit target's inscription lines to accurately calculate the coordinates of target feature points; based on the linear array camera's imaging model, it virtually orthogonals the linear array camera into a planar array for subsequent calculations; and it constructs a calibration system to verify the feasibility and effectiveness of ground calibration for long-focal-length linear array optical cameras. This invention is suitable for ground calibration of long-focal-length linear array optical cameras in a ground-based laboratory environment, offering advantages such as high precision, high speed, and low cost. It is of great significance for ensuring high-resolution imaging and high-precision positioning in aerospace, remote sensing, and mapping fields. Attached Figure Description

[0020] Figure 1 This is a flowchart of a ground calibration method for a long focal length linear array optical camera according to the present invention;

[0021] Figure 2 Schematic diagram of ground calibration principle for a telephoto linear array optical camera;

[0022] Figure 3 A schematic diagram of a slit target design for imaging with a telephoto linear array optical camera. Detailed Implementation

[0023] This invention combines a slit target with a collimator, supplemented by a beam splitter and a plane mirror to construct an infinitely far planar target, establishing a mathematical model for high-precision ground calibration of a linear array camera; it achieves accurate extraction of the center of the integral image of the linear array camera through a multi-scale image extraction method; it uses the geometric constraints of the slit target's own markings to accurately calculate the coordinates of the target's feature points; based on the linear array camera's imaging model, it virtually orthogonals the linear array camera into a planar array and performs subsequent calculations; and it constructs a calibration system device to verify the feasibility and effectiveness of ground calibration of a telephoto linear array optical camera.

[0024] like Figure 1 As shown, a ground calibration method for a long-focal-length linear array optical camera according to the present invention includes the following steps:

[0025] Step 11: Set up a high-precision ground calibration hardware device for the telephoto linear array optical camera in a ground laboratory environment.

[0026] like Figure 2 As shown, the apparatus for implementing this invention includes an image analysis and calculation module, a target imaging and image acquisition module, and an infinity target module. The image analysis and calculation module includes an image feature extraction and linear array camera intrinsic parameter modeling and solving module, which processes the image of the infinity target and optimizes the camera's intrinsic parameters based on the linear array camera intrinsic parameter model. The target imaging and image acquisition module includes a telephoto linear array camera optical lens, a high-resolution camera, and a control box, used to complete the imaging of the infinity target and provide data for the image analysis and calculation module. The infinity target module consists of a slit target, a collimator, a beam splitter, and a plane mirror, used to generate the infinity target for data acquisition by the target imaging and image acquisition module.

[0027] in Figure 2The relationships between the modules are as follows: In the infinity target module, the slit target is mounted on the focal plane of the collimator. The light rays forming the infinity target are split into two beams by a beam splitter. One beam passes through the beam splitter and enters the camera, while the other beam is deflected and enters the camera after passing through a plane mirror. The telephoto linear array camera, optical camera, and imaging target together form a high-resolution camera. The control box triggers the linear array camera to achieve imaging of the infinity target. The image analysis and calculation module processes the target image and calculates the camera's intrinsic parameters. This invention combines a slit target with a collimator. The slit target is placed at the focal plane of the collimator, and a beam splitter and a plane mirror are used to construct an infinitely far planar target, establishing a high-precision ground calibration mathematical model for a line-scan camera. A multi-scale image extraction method is used to accurately extract the center of the integral image from the line-scan camera. The geometric constraints of the slit target's markings are used to accurately calculate the coordinates of the target's feature points. Based on the line-scan camera's imaging model, the line-scan camera is virtually orthogonalized into a planar array for subsequent calculations. The beam splitter divides the beams into two groups of infinitely far target rays. The first group directly enters the camera to be calibrated, while the second group, reflected by the mirror, simultaneously enters the camera's field of view. The tilting plane mirror can change the incident direction of the second group of rays, covering the entire field of view of the large-aperture optical camera.

[0028] Step 12: Image Feature Modeling and Centroid Extraction. High-precision centroid extraction of image features based on a multi-scale model is employed, laying the foundation for high-precision calibration and solution of camera intrinsic parameters.

[0029] Step 121: Let g be obtained by convolving the two-dimensional image f(x,y) with the partial derivatives of Gaussian of each order. x g y g xx g xy g yy Let f(x,y) be the first-order derivative convolution result of the two-dimensional image with a Gaussian convolution kernel relative to x, the first-order derivative convolution result of the two-dimensional image with a Gaussian convolution kernel relative to y, the second-order derivative convolution result of the two-dimensional image with a Gaussian convolution kernel relative to x, the partial derivative convolution result of the two-dimensional image with a Gaussian convolution kernel relative to xy, and the second-order derivative convolution result of the two-dimensional image with a Gaussian convolution kernel relative to y. Then, the second-order Taylor expansion of the two-dimensional image f(x,y) in the adjacent images of the light stripe pixel (x0,y0) can be expressed as:

[0030]

[0031] Step 122: According to the CStegger algorithm, the image center point is the point where the first derivative is zero and the second derivative is at its maximum along the edge direction of the light stripe image. The Hessian matrix of the two-dimensional image is... Therefore, the edge direction corresponds to the eigenvector of the largest absolute eigenvalue of the Hessian matrix. Let the edge direction be represented by n = (n x,n y ) represents, and ||(n) x ,n y )||=1, where |||| is the modulo operation, and the image grayscale function is in (n x ,n y The second derivative in the direction of ) corresponds to the largest absolute eigenvalue of the Hessian matrix. Let the candidate scale list be S. List ={σ1,σ i ,…σ N}, where σ i Let i be the size of the Gaussian convolution kernel, i be the scale number, and N be the number of candidate scales. Then, the list of normalized curves corresponding to each pixel is as follows: Where I i Represents a light bar image. This indicates that the normalized value is obtained by solving the operation to get C. i And C List ={C1,C i ,…C N},in As can be seen from the multi-scale light spot extraction method, selecting C i The σ corresponding to the maximum value of (x,y) i The optimal scale is used for subsequent Gaussian convolution kernels. Equation (1) can be expressed along the edge direction as:

[0032]

[0033] Where f[] represents the Taylor expansion equation in the edge direction, t is the coefficient to be determined, and n x n y This is the score of the edge direction vector.

[0034] For the edges of the lines, make We can obtain:

[0035]

[0036] Step 123: Therefore, the maximum or minimum value of the image gray level is (p x p y )=((tn x +x0),(tn y +y0)). If That is, the point whose first derivative is zero lies within the current pixel, and (n x ,n y If the second derivative in the direction of ) is greater than a specified threshold, then the point (p) x ,p y () is the center point of the line.

[0037] Step 13: As Figure 3As shown, the coordinates of spatial target feature points are solved. A slit target is designed, and based on the image point coordinates, local coordinate modeling of target feature points is performed using the self-constraints of the parallel and oblique lines of the slit target, resulting in the spatial point coordinates of the target and the ray vector at infinity. The specific steps include:

[0038] Step 131: Based on the fact that the line scan camera only performs horizontal single-pixel imaging, a backlit light source and a light-transmitting design at the slit lines are designed to ensure that the line scan camera only images the single horizontal intersection when imaging the slit target. Simultaneously, to construct feature points when the line scan camera intersects with the planar target, a combination of vertical parallel lines and oblique lines is designed to obtain the planar slit target, facilitating the calculation of target feature point coordinates based on the invariant cross-ratio characteristic. The planar slit target consists of vertically equidistant parallel lines and oblique lines, with the horizontal axis as the X-axis and the vertical axis as the Y-axis. Let L0, L2, ... L 2n The lines are vertically equidistant parallel lines (numbered evenly), with spacing of a, L1, L3, ... L 2n+1 The target is represented by diagonal lines and numbers (odd numbers), where n∈{0,1,2,3,…,N}. The target markings are light-transmitting slits, while the remaining positions are opaque planes, and the target is placed at the original reticle position of the collimator.

[0039] Step 132: When the collimator illuminates the planar target, and the line-scan camera's line of sight intersects with the planar target to form an image, the line-scan camera's integral imaging can obtain the integral image of each intersecting bright spot, specifically represented by multiple vertical bright lines. Let P be the target feature point corresponding to the slit target. i Its corresponding image point is p i where it satisfies the constraint relationship

[0040] Step 133: Let the line array camera intersect with the slit of the planar target. The total change in height from the intersection of the first vertical slit to the last vertical slit is h, where a is the distance between each vertical line of the slit target. According to the geometric constraints of the planar slit target, the local coordinate components of the image formed by the intersection of the line array camera's field of view and the planar target can be expressed as follows:

[0041]

[0042] X 2n =a×n,

[0043] Where X0 = 0, Y0 = 0.

[0044] Step 14: Image the slit target using a linear array camera. After coordinate extraction, it can be considered a virtual orthogonalized area array camera calibration model. This includes the following steps:

[0045] Step 141: The integral imaging characteristics of the linear array camera are perspective projection imaging in the horizontal direction and time integral image in the vertical direction. After the coordinate extraction in step 13, the distribution of one-dimensional target points can be determined. At this time, the one-dimensional image coordinates and their corresponding planar target coordinates are obtained simultaneously.

[0046] Step 142: First, construct symmetrical coordinates for the image feature points about y = x to obtain the vertical axis image point coordinates orthogonal to the existing one-dimensional coordinates. The original one-dimensional image points can be regarded as the u-axis of the two-dimensional image, and the orthogonalized vertical axis can be regarded as the v-axis of the image, thus realizing the virtual orthogonalization of the linear array camera coordinates. Similarly, construct symmetrical coordinates for the corresponding planar target feature points about y = x. At this point, the virtual orthogonalization imaging of the slit target feature points by the linear array camera is completed. This can be virtually represented as synchronous imaging of the horizontal and vertical target points, constituting synchronous imaging of the spatial planar target under the virtual orthogonalized two-dimensional camera image plane.

[0047] Based on the perspective imaging model of a linear array camera, the infinite ray V is constructed from the feature points of the target at infinity. r The corresponding image point p can be represented as ρp = K·R t ·V r Where ρ is a non-zero coefficient. Let u0 and v0 be the camera intrinsic parameter matrix. u0 and v0 are the coordinates of the principal points on the virtual camera image plane, satisfying u0 = v0, f x f x are the equivalent focal lengths in the u and v directions, respectively, and are equal. R is the rotation matrix between the target at infinity and the camera coordinate system, establishing an imaging model of the spatial target feature points on the virtual orthogonal camera image plane.

[0048] Step 143: Based on the virtual orthogonal imaging in Step 142, the camera intrinsic parameters can be solved using a method based on a planar target. Since the total change in the height h of the vertical slit in Step 133 is an unknown scaling factor set during the single imaging of the target, the translation vector between the slit and the infinitely far target constructed by the collimator relative to the virtual orthogonal camera, as well as the subsequent intrinsic parameter solution process, does not need to be solved. Only the intrinsic parameters and rotation vector need to be solved. That is, this change in height h does not affect the accuracy of the camera intrinsic parameter solution.

[0049] Since multiple sets of vectors can be projected onto the linear array camera for imaging via a plane mirror, the two sets of light vector imaging models are as follows, with the first set of fixed imaging vectors represented as:

[0050] v1=K·R t1 ·V r (4)

[0051] Among them, V r R is a vector at infinity. t1 For the first group V rIn the camera coordinate system, the external parameters are as follows: the ideal point coordinates are v1, and after lens distortion model transformation, the image point coordinates are p1 = distFunc(v1).

[0052] Where distFunc(v1) is the distortion function, let p d =[u d ,v d ,1] T p represents the homogeneous coordinates of a point in the distorted image. n =[x n y n ,1] T Let x = u - u0, y = v - v0, where x is the homogeneous coordinate of the normalized image point. The physical model of lens distortion is expressed as:

[0053] u d =u+x(k1r 2 +k2r 4 +k3r 6 )+p1(2x 2 +y 2 )+2p2xy

[0054] v d =v+y(k1r 2 +k2r 4 +k3r 6 )+2p1xy+p2(x 2 +2y 2 (5)

[0055] in, k1, k2, and k3 are the radial distortion coefficients of the lens, and p1 and p2 are the tangential distortion coefficients. Therefore, the camera's intrinsic parameters mainly include (f x ,f y ,γ,u0,v0,k1,k2,k3,p1,p2).

[0056] Similarly, the second group V r The imaging vector constructed after the mirror rotates freely is:

[0057] v i =K·R ti ·V r ,i=2,3…,n (6)

[0058] Among them, R ti For V r In the camera coordinate system, the extrinsic parameter v i After lens distortion model transformation, the coordinates of the imaging point are p. i =distFunc(v i ).

[0059] The intrinsic parameters of the camera to be calibrated are obtained by minimizing the error between the back-projected feature points of the target at infinity and the feature points of the actual image, using a nonlinear optimization method.

[0060] Step 15: Solve for the intrinsic parameters of the optical camera. The target rays are split into two groups at infinity using a beam splitter. The first group enters the camera to be calibrated directly, while the second group is reflected by a mirror and simultaneously enters the camera's field of view. A tilting plane mirror can change the incident direction of the second group of rays, covering the entire field of view of the large-aperture optical camera. By analyzing the acquired target image data, image features are extracted and combined with camera and target parameters. Using the ground calibration model of the telephoto linear array camera, the objective function is to minimize the back-projection error of the target feature points. Nonlinear optimization is then used to derive the intrinsic parameters of the camera to be calibrated.

Claims

1. A ground calibration method for a long-focal-length linear array optical camera, characterized in that, Includes the following steps: Step a) In a ground-based laboratory environment, a high-precision ground calibration hardware device for a long-focal-length linear array optical camera is set up. Only a planar slit target and a collimator are used, supplemented by a beam splitter and a wobbly planar mirror to achieve high-precision ground calibration of the long-focal-length linear array optical camera. The infinity imaging target is designed as a miniature planar slit target installed on the focal plane of the collimator to generate an infinity planar target. The target is split into two groups of infinity target rays by the beam splitter. The first group of rays directly enters the camera to be calibrated, and the second group enters the field of view of the camera to be calibrated synchronously after being reflected by the mirror. This achieves a calibration mode that does not rely on a large-aperture collimator and a large high-precision turntable. Step b: Perform image feature modeling and centroid extraction. High-precision extraction of image feature centroids based on multi-scale models is used for high-precision calibration and solution of camera intrinsic parameters. Step c: By designing a slit target and modeling the target feature points based on the geometric constraints of the slit target's own markings, the spatial point coordinates and the ray vector at infinity are obtained. Step d: By performing virtual orthogonalization on the linear array camera, a virtual area array camera imaging model and an intrinsic parameter calibration mathematical model are obtained, which are used for subsequent solution of the linear array camera. Step e: The oscillating plane mirror changes the incident direction of the second set of light rays, covering the entire field of view of the telephoto linear array optical camera. Substitute these steps into the above steps and obtain the intrinsic parameters of the linear array optical camera through nonlinear optimization.

2. The ground calibration method for a long focal length linear array optical camera according to claim 1, characterized in that: In step b, an adaptive multi-scale centroid extraction method is used to extract the centroid of feature points in the integral imaging of the slit target by the linear array camera. This allows centroid extraction to be completed under optimal scale conditions for different image centers, thereby improving the accuracy of image centroid extraction.

3. An apparatus for implementing the ground calibration method for a long focal length linear array optical camera as described in any one of claims 1-2, characterized in that, It includes an image analysis and calculation module, a target imaging and image acquisition module, and an infinity target module. The image analysis and calculation module includes an image feature extraction and linear array camera intrinsic parameter modeling and solving module, which realizes the image processing of infinity target and optimizes the camera intrinsic parameters based on the linear array camera intrinsic parameter model. The target imaging and image acquisition module includes a telephoto linear array camera optical lens, a high-resolution camera, and a control box, which is used to complete the imaging of the infinity target and provide data for the image analysis and calculation module. The infinity target module consists of a slit target, a collimator, a beam splitter, and a plane mirror, and is used to generate an infinity target for data acquisition by the target imaging and image acquisition module.

Images