Method for high-speed camera optical axis pointing calibration based on straight line vector projection slope matching

Abstract

This application discloses a method for calibrating the optical axis pointing of a high-speed camera based on linear vector projection slope matching. The method simulates the projection of a calibration plate's linear vector onto the image plane and matches it with the actual extracted linear slope to directly calculate the azimuth and elevation angles of the optical axis pointing. First, a calibration plate with at least two non-parallel linear vectors distributed on its surface is fabricated, and a coordinate system is established using the calibration plate as a reference. Then, the calibration plate is photographed, and the image plane slope of the selected linear vector is extracted. The selected linear vector is simulated and projected onto the image plane to obtain a two-dimensional projection vector containing the azimuth angle A and the elevation angle E. Finally, the optical axis pointing parameters of the high-speed camera are calculated by matching the simulated projection slope with the actual extracted slope. This application also provides a method for verifying the correctness of the calibration results. This method decouples pose parameters, simplifies calibration plate fabrication and data processing, and provides accurate external parameter support for high-precision spatial attitude testing of high-speed recording equipment.

Description

Technical Field

[0001] This application belongs to the field of camera calibration technology, specifically relating to a method for calibrating the optical axis of a high-speed camera based on linear vector projection slope matching. Background Technology

[0002] Optical measurement, as a crucial external measurement method, plays an irreplaceable role in aircraft attitude measurement, event reproduction, and fault diagnosis. High-precision measurement by optical equipment relies on accurate intrinsic and extrinsic parameters. Currently, commonly used optical quantitative equipment mainly includes two categories: theodolites and high-speed video recording equipment. Theodolites possess precise intrinsic and extrinsic parameters, enabling high-precision orientation; while high-speed video recording equipment typically only possesses intrinsic parameters and lacks precise optical axis pointing information. To achieve high-precision positioning and attitude determination with high-speed video recording equipment, its extrinsic parameters, namely the azimuth and pitch angles of the optical axis, must be obtained through calibration.

[0003] Existing calibration methods are mostly based on feature points on a calibration board, calculating camera parameters by establishing object-image mapping relationships between these feature points. These methods primarily aim at spatial positioning, requiring simultaneous calculation of the camera's position and attitude during calibration. This leads to coupling of pose parameters, complex calibration board fabrication, and cumbersome data processing. However, in attitude testing applications, the core focus is on optical axis pointing rather than absolute position. Feature point-based calibration methods not only introduce redundant parameters but also struggle to achieve accurate decoupling of the pointing parameter alone. Although the calibration principle based on linear vectors theoretically simplifies pose relationships, no linear vector methods for calibrating the optical axis pointing of high-speed cameras have been documented in the literature, both domestically and internationally. Summary of the Invention

[0004] To achieve accurate calibration of the optical axis pointing of a high-speed camera, this application aims to provide a calibration method based on the slope matching of linear vector projection. By simulating the projection of the linear vector of the calibration plate onto the image plane and matching it with the actual extracted image plane slope, the azimuth and pitch angles of the optical axis pointing can be directly decoupled and obtained.

[0005] To achieve the above technical objectives, this application specifically adopts the following technical solution: In one aspect of this application, a method for optical axis pointing calibration of a high-speed camera based on linear vector projection slope matching is provided, comprising the following steps: S1. Make a calibration plate, wherein the calibration plate is a plane and at least two non-parallel straight line vectors are distributed on the surface, and the thickness of the straight line vectors is determined according to the imaging field of view of the high-speed camera being calibrated and the extractability of the imaging straight line vectors. S2. Establish a coordinate system based on the calibration plate; S3. Take a picture of the calibration board, select at least two valid straight line vectors from the acquired image, and extract their image plane slope. ; S4. Project the selected straight line vector onto the image plane of the high-speed camera to obtain a two-dimensional projection vector containing the optical axis pointing parameters to be calibrated, wherein the optical axis pointing parameters include azimuth angle A and pitch angle E; S5. Compare the simulated projection result with the extracted actual image plane slope. Matching is performed, and the optical axis pointing parameters of the high-speed camera are obtained through slope consistency calculation.

[0006] In one implementation, the coordinate system in step S2 is defined as follows: with the lower left corner of the calibration plate as the origin, the horizontal rightward axis parallel to the ground as the X-axis, and the vertical upward axis as the Y-axis, the XYZ axes form a right-handed system; the calibration plate is placed perpendicular to the ground, with its front facing the lens in the center, and the straight line vector perpendicular to the ground is an invalid straight line vector.

[0007] In one implementation, the coordinate system in step S2 is defined as follows: the calibration plate is placed facing the lens and directly below the line of sight of the high-speed camera, with the two mutually perpendicular sides of the calibration plate being the X-axis and Y-axis, and the three axes X, Y, and Z forming a right-handed coordinate system.

[0008] In one embodiment, the calibration plate contains at least two linear vectors that are not parallel to each other.

[0009] In one implementation, the correctness of the optical axis pointing calibration result is verified by one of the following methods: For a single high-speed camera, the known straight line vector used for non-calibration is used as the target. The azimuth angle A and elevation angle E after calibration are substituted for projection. The projection slope is compared with the actual image plane slope to verify consistency. For two high-speed cameras, the calibrated azimuth angle A and elevation angle E are substituted into each camera to perform vector synthesis. The synthesized result is then compared with the actual projection of the known straight line vector to verify consistency.

[0010] The beneficial effects of this application are as follows: This application achieves optical axis pointing calibration based on linear vector projection slope matching. Compared to traditional feature point-based calibration methods, it decouples pose parameters, focusing only on the core parameter of optical axis pointing. This method reduces the complexity of calibration board fabrication, simplifies data processing, and improves the flexibility and reliability of calibration operations while maintaining calibration accuracy. This application provides theoretical support for high-speed video recording equipment aimed at attitude testing, enabling high-precision measurement of spatial attitude without precise initial external parameters. Attached Figure Description

[0011] Figure 1This is a schematic flowchart of the calibration method according to an embodiment of this application; Figure 2 This is the linear vector foundation setting of the ground high-speed camera calibration board in the embodiments of this application; Figure 3 This is the definition of the calibration coordinate system for the ground high-speed camera in the embodiments of this application; Figure 4 This is the basic linear vector setting of the airborne high-speed camera calibration board in the embodiments of this application; Figure 5 This is the definition of the calibration coordinate system for the airborne high-speed camera in this application embodiment. Detailed Implementation

[0012] The technical solution of this application will be clearly and completely described below with reference to specific embodiments. However, those skilled in the art will understand that the embodiments described below are only some embodiments of this application, not all embodiments, and are only used to illustrate this application, and should not be regarded as limiting the scope of this application. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0013] Existing high-speed video recording equipment optical axis pointing calibration relies on the object-image mapping relationship of feature points. The solution process requires simultaneous calculation of position and attitude parameters, leading to pose coupling, complex calibration board fabrication, and redundant data processing. For practical applications where only optical axis pointing is needed for attitude testing, this application utilizes the projection characteristics of straight line vectors onto the image plane. By simulating the projection of the calibration board's straight line vectors onto the high-speed camera's image plane, it matches this projection with the slope of the actual captured straight line image plane to directly calculate the azimuth and pitch angles of the optical axis pointing. This method eliminates the limitations of the number and distribution of feature points, decouples pose parameters, and requires only two or more non-parallel straight line vectors to complete the calibration. The application specifies the setting of the calibration board's straight line vectors, the definition of the coordinate system, and the projection model for both ground-based and airborne testing modes, and provides a means to verify the calibration results based on known straight line vectors, enabling high-speed video recording equipment to obtain accurate optical axis pointing without precise initial external parameters.

[0014] In one implementation, a method for calibrating the optical axis of a high-speed camera based on linear vector projection slope matching includes the following steps: S1. Fabrication of a calibration plate. The calibration plate has a planar structure, and at least two straight line vectors are evenly distributed on its surface. These straight line vectors are not parallel to each other. The thickness of the straight line vectors is determined based on the imaging field of view of the high-speed camera being calibrated and whether the straight line vectors can be accurately extracted from the image, ensuring that clear straight line images can be obtained under different imaging conditions.

[0015] S2. Establish a coordinate system based on the calibration plate. The lower left corner of the calibration plate is used as the origin of the coordinate system, serving as the starting point for spatial measurement. Starting from the origin, the direction parallel to the ground and horizontally to the right is defined as the positive X-axis; the vertically upward direction perpendicular to the ground is defined as the positive Y-axis. After determining the X and Y axes, the Z-axis direction is determined according to the right-hand rule: when the right hand grasps the Z-axis, the direction the thumb points when the four fingers move from the positive X-axis to the positive Y-axis, thus establishing a three-dimensional rectangular coordinate system.

[0016] The calibration plate is positioned perpendicular to the ground, facing the high-speed camera lens and centered in the field of view. In this placement, straight line vectors perpendicular to the ground become invalid during imaging due to projection; therefore, they should be excluded from the selection of calibration line vectors and not used in the calculations. There are at least two non-parallel straight line vectors; this design ensures a unique correspondence between the projection slopes of different straight line vectors on the image plane.

[0017] S3. Take a picture of the calibration board, select at least two valid straight line vectors from the acquired image, and extract their image plane slope. .

[0018] Fix the high-speed camera in a predetermined position, align it with the calibration board, and take a picture to obtain a digital image containing the calibration board image. During the shooting, ensure that the calibration board image is clear and that straight lines and vectors are identifiable in the image.

[0019] The acquired image is processed to select at least two valid line vectors for calibration. The determination of valid line vectors depends on the placement of the calibration plate: when the calibration plate is placed perpendicular to the ground, line vectors perpendicular to the ground appear as lines in the image plane that are aligned with the Y-axis of the image coordinate system. These lines cannot provide effective constraint information in subsequent projection calculations and are therefore not selected as valid line vectors; line vectors in other directions that are clearly imaged can be used as valid line vectors.

[0020] After selecting valid line vectors, their image plane slopes are extracted using image processing methods. Specifically, this involves: extracting line edges from the image using edge detection algorithms; then using line detection algorithms (such as Hough transform) to obtain the pixel coordinates of each line vector; and finally calculating the slope value of each line vector in the image plane coordinate system. Image plane slope Defined as the degree of inclination of a straight line on the image plane, expressed as the ratio of the change in the vertical coordinate to the change in the horizontal coordinate, in pixels.

[0021] S4. Project the selected straight line vector onto the image plane of the high-speed camera to obtain a two-dimensional projection vector containing the optical axis pointing parameters to be calibrated, wherein the optical axis pointing parameters include azimuth angle A and pitch angle E.

[0022] The simulated projection is based on a pinhole imaging model. Specifically, let the direction vector of a certain straight line vector in space be denoted as in the calibration plate coordinate system (world coordinate system). The coordinate system is defined as follows: the origin is located at the lower left corner of the calibration plate, the X-axis is horizontal to the right, the Y-axis is vertically upward, and the Z-axis is determined using a right-hand rule. The optical axis of the high-speed camera is described by the azimuth angle A and the pitch angle E, where the azimuth angle A is the angle between the projection of the optical axis onto the horizontal plane and the positive direction of the X-axis, and the pitch angle E is the angle between the optical axis and the horizontal plane. The camera has no roll angle setting (i.e., the image plane coordinate system remains parallel to the horizontal plane). In this case, the rotation matrix from the world coordinate system to the camera coordinate system... It can be represented as:

[0023] Rotation matrix Transform vectors in the world coordinate system to the camera coordinate system (the camera coordinate system is defined as: The axis extends horizontally to the right along the image plane. The axis points vertically downwards along the image plane. (Axis points along the optical axis towards the scene).

[0024] Direction vector of the line Transform to the camera coordinate system to obtain = In the camera coordinate system, the projection direction of a straight line onto the image plane is determined by... In the image plane ( The projection onto a plane determines the two-dimensional vector. The specific expression for the direction of the line at infinity on the image plane depends on... The component of . For a straight line passing through the optical center (ignoring translation), the direction of the straight line on the image plane is and . The projection directions on the image plane are consistent, that is, the two-dimensional direction vector can be written as: ,in and They are respectively In the camera coordinate system and Components on the axis (usually in the camera coordinate system) down, Forward, but the proportional relationship of the direction vectors does not affect the slope. In actual calculations, the image plane slope... for / However, attention must be paid to the relationship between the symbol conventions and the image coordinate system. To ensure consistency with subsequent slope matching, a two-dimensional projection vector is directly used. It represents the direction of a line on the image plane, and its components are determined by the spatial line direction and the rotation matrix.

[0025] Taking a high-speed camera deployed on the ground as an example, with the calibration plate placed perpendicular to the ground, any straight line vector perpendicular to the ground in the world coordinate system (such as along the Y-axis) is invalid and will not be selected. Valid straight line vectors are selected, such as a horizontal straight line vector pointing to the right. After the rotation matrix transformation, the direction in the camera coordinate system is obtained as follows: Its two-dimensional projection vector on the image plane is If a straight line vector is selected at a 45° angle to the horizontal plane... After transformation, we get Its two-dimensional projection vector is The two-dimensional vector components mentioned above are all functions of the azimuth angle A and the elevation angle E, and contain the optical axis pointing information to be calibrated.

[0026] For airborne high-speed cameras, the calibration plate is placed horizontally directly below the camera, and the coordinate system is defined differently: the X and Y axes lie in the plane of the calibration plate (perpendicular to each other), and the Z axis points vertically upwards. The straight line vector selected in the world coordinate system at this time is as follows: and After undergoing the same rotation matrix transformation, the corresponding two-dimensional projection vector expression can be obtained. For example... The projection result is , The projection result is .

[0027] After the simulated projection calculation is completed, the two-dimensional projection vector corresponding to each selected line vector is obtained. ,in The number represents the line number.

[0028] S5. Compare the simulated projection result with the actual image plane slope extracted in step three. Matching is performed, and the optical axis pointing parameters (A, E) of the high-speed camera are obtained through slope consistency calculation.

[0029] After completing the simulated projection of the line vectors, the two-dimensional projection vector corresponding to each selected line vector is obtained. This vector is a function of the azimuth angle A and the elevation angle E. The direction of the simulated projection vector is represented on the image plane as the inclination of a straight line, and its theoretical slope can be determined by the ratio of the components of the projection vector. Considering the conventions of the image plane coordinate system (usually horizontal to the right...), The positive direction of the axis is vertically downward. (positive direction of the axis) the slope of a straight line on the image plane Defined as the ratio of the change in the vertical axis to the change in the horizontal axis, i.e. The theoretical slope is different from the actual image plane slope extracted from the image in step S3. They should be consistent and form the basis for calculating the optical axis pointing parameters.

[0030] For each valid straight line vector involved in the calculation, establish an equation in which the theoretical slope equals the actual slope:

[0031] in, and The specific expression is derived in step S4 based on the direction of the line vector and the camera rotation matrix. For example, in the case of a high-speed camera deployed on the ground, if a line vector is selected... ,but , The corresponding equation is:

[0032] If both straight lines and vectors are selected ,but , The corresponding equation is:

[0033] Solving the two equations simultaneously yields solutions for A and E. For the case of an airborne high-speed camera, the form of the equations changes depending on the chosen straight-line vector, but the principle remains the same.

[0034] In practical calculations, due to measurement noise in image plane slope extraction and the potential nonlinearity of the equations, numerical optimization methods are typically used for solving the problem. The error function is constructed as follows:

[0035] Iterative solutions obtained through nonlinear least squares algorithms (such as the Gauss-Newton method or the Levenberg-Marquardt method) Minimum parameter This serves as the calibration result for the optical axis direction. If there are only two straight lines and the equation can be solved analytically, a closed-form solution can also be directly derived.

[0036] In another implementation, a method for calibrating the optical axis of an airborne high-speed camera based on linear vector projection slope matching involves placing a calibration plate, after its fabrication, directly below the camera's line of sight, with the calibration plate's plane parallel to the ground and the camera's optical axis approximately perpendicular to the center of the calibration plate. This placement method is suitable for typical airborne testing conditions where the camera is shooting downwards, ensuring that the calibration plate's plane and the camera's image plane form a conjugate relationship.

[0037] The coordinate system is defined with the calibration plate itself as the reference: two adjacent and mutually perpendicular edges of the calibration plate are taken as the coordinate axis directions, one of which is designated as the X-axis, and the adjacent perpendicular edge is designated as the Y-axis. Both the X-axis and Y-axis lie in the horizontal plane, and their positive directions can be arbitrarily assigned according to the actual orientation of the calibration plate. According to the rules of the right-hand rectangular coordinate system, the Z-axis direction is determined by the direction of the right thumb when turning from the X-axis to the Y-axis, and this direction is perpendicular to the plane of the calibration plate and upwards. There may be a rotational relationship between the established coordinate system and the world geographic coordinate system.

[0038] In this coordinate system, all linear vectors within the calibration plate plane are represented by X-axis and Y-axis directional components. Since the calibration plate is placed horizontally, there are no linear vectors perpendicular to the ground within its plane (the direction perpendicular to the ground corresponds to the Z-axis, which is not within the plane). Therefore, all linear vectors can provide effective constraint information when projected onto the image plane, and there is no situation where vertical linear vectors need to be excluded in ground calibration.

[0039] When photographing the calibration board, the high-speed camera is fixed to the airborne platform, with the lens vertically downwards and aimed directly below the line of sight, ensuring the calibration board is completely and clearly within the field of view. After acquiring the image, at least two valid line vectors are selected from it. Since the calibration board is placed horizontally, all line vectors in its plane are parallel to the X-axis or Y-axis, and there are no invalid line vectors perpendicular to the ground. Therefore, any two non-parallel line vectors can be selected as valid lines. The slope of the selected line vectors on the image plane is extracted using an image processing algorithm. The actual measured value is recorded.

[0040] Then, a simulated projection calculation of the straight line vector is performed. The calibration plate coordinate system is used as the world coordinate system, in which the X and Y axes lie in the horizontal plane, and the Z axis is perpendicular to the calibration plate and points upwards. Let the direction vector of the selected straight line vector be... Since the straight line lies in the calibration plate plane, the Z-direction component is zero. The optical axis of the high-speed camera is described by the azimuth angle A and the pitch angle E. The camera has no roll angle. The rotation matrix from the world coordinate system to the camera coordinate system is... The process is the same as for ground calibration. Transform the line vector to the camera coordinate system to obtain... = The two-dimensional projection vector of a straight line onto the image plane is determined by... In the camera coordinate system shaft and The components on the axis are denoted as The components of this vector are expressed as functions of A and E. For example, consider a straight vector along the X-axis. The projection result is Select a straight line vector along the Y-axis. The projection result is Select a straight line vector at a 45° angle to the X-axis. The projection result is .

[0041] The simulated projection results are matched with the measured slopes. The theoretical slope corresponds to each line vector. , and the measured slope Equal equations:

[0042] In the case of two straight lines, solving the simultaneous equations yields A and E. Let's choose a straight line along the X-axis. and the straight line in the Y-axis direction For example, the first equation gives In the second equation The projection vector is Its slope Therefore, Therefore, we can first determine that A must satisfy... Then, use the first equation to solve for E. If other combinations of lines are chosen, the form of the equation will change accordingly, and all can be solved through numerical optimization or analytical methods.

[0043] Construct an error function and iteratively solve it using a nonlinear least squares algorithm to minimize the parameters of the error function. This is the calibration result of the optical axis pointing of the airborne high-speed camera.

[0044] In some embodiments, after the optical axis pointing calibration is completed and the azimuth angle A and elevation angle E are obtained, the correctness of the calibration results is verified in the following manner.

[0045] For the calibration of a single high-speed camera, one or more known straight line vectors not involved in the calibration calculation are selected as verification targets. These straight line vectors also lie within the calibration plate plane, and their spatial orientation is known. The azimuth angle A and elevation angle E obtained from the calibration are substituted into the projection model to calculate the theoretical projection slope of the straight line vector on the image plane. Simultaneously, the image plane slope of the straight line vector is extracted from the actual captured image. The theoretical projection slope is compared with the extracted slope; if they are consistent within the allowable error range, the calibration result is considered correct and reliable.

[0046] For a measurement system consisting of two high-speed cameras, after each camera completes optical axis pointing calibration, it obtains a set of azimuth angles A and elevation angles E. A known straight vector in the same space is selected as the verification target. The calibration parameters of the two cameras are substituted into the projection model to obtain the theoretical projections of this straight vector onto the image planes of the two cameras. Based on the principle of binocular vision, the theoretical projections of the two cameras can be further synthesized to determine the direction or position of the straight vector in space. The synthesized result is then compared with the actual known direction of the straight vector. If the synthesized result matches the actual direction, it proves that the calibration parameters of the two cameras have internal consistency and can meet the requirements of joint measurement.

[0047] Example Reference Figure 1 As shown, the optical axis pointing calibration of the ground high-speed camera is carried out according to the following steps: Step 1: Calibration Plate Fabrication and Linear Design Requirements. The calibration plate adopts a planar structure, with at least two non-parallel linear vectors distributed within the plane. The linear vector settings are as follows: Figure 2 As shown. The calibration line vector must be a valid line vector. The calibration plate should be placed perpendicular to the ground, with its front facing the lens and centered. Line vectors perpendicular to the ground are invalid and will not be used for calibration. The object distance is set to 50m, the target line length is 1.2m, and the width is 0.2m. The high-speed camera resolution is 1536×1024, the individual pixel size is 10μm×10μm, and the focal length is f mm.

[0048] Step 2: Define the coordinate system reference. Use the direction defined by the linear vector of the calibration plate as the reference direction, such as... Figure 3 As shown. With the lower left corner of the calibration plate as the origin, the X-axis extends horizontally to the right parallel to the ground, and the Y-axis extends vertically upwards. The X, Y, and Z axes form a right-handed system.

[0049] Step 3: Photograph the calibration board and select and extract the line vectors. Select at least two valid line vectors from the calibration board and obtain the slope of the line vector image plane. . (Using horizontal straight-line vectors) and For example, projection and processing are performed.

[0050] Step 4: Linear vector simulation projection and object-image matching processing. The horizontal linear vector... and The images of high-speed cameras are projected onto the ground, and the projection results are as follows: and The combined target line vector simulation projection results and the actual slope in the third step. The optical axis pointing (azimuth A, elevation E) of the high-speed camera deployed on the ground is obtained through slope matching calculation.

[0051] Reference Figure 1As shown, the optical axis alignment calibration of the airborne high-speed camera is performed according to the following steps: Step 1: Calibration plate fabrication and straight line design requirements. The calibration plate is planar, and at least two non-parallel straight line vectors are distributed within the plane. The straight line vectors are set as follows: Figure 4 As shown. The calibration plate is placed directly below the line of sight of the high-speed camera, facing the lens. All line vectors in its plane are valid line vectors. The object distance is set at 50m, the target line length is 1.2m, and the width is 0.2m. The high-speed camera resolution is 1536×1024, the individual pixel is 10μm×10μm, and the focal length is f mm.

[0052] Step 2: Define the coordinate system reference. Use the direction defined by the linear vector of the calibration plate as the reference direction, such as... Figure 5 As shown. The two mutually perpendicular sides of the calibration plate are the X-axis and Y-axis, respectively. The XYZ axes form a right-handed system, with the Z-axis perpendicular to the plane of the calibration plate and pointing upwards.

[0053] Step 3: Photograph the calibration board and select and extract the line vectors. Select at least two valid line vectors from the calibration board and obtain the slope of the line vector image plane. . (Using horizontal straight-line vectors) and For example, projection and processing are performed.

[0054] Step 4: Linear vector simulation projection and object-image matching processing. The horizontal linear vector... and The projections onto the image plane of the airborne high-speed camera are as follows: and The combined target line vector simulation projection results and the actual slope in the third step. The optical axis pointing (azimuth A and pitch E) of the airborne high-speed camera is obtained through slope matching calculation.

[0055] After calibration, the correctness of the optical axis pointing calibration results is verified. For a single high-speed camera, a known straight line vector (not used for calibration) is used as the target. The azimuth angle A and elevation angle E after calibration are substituted into the vector vector for projection. The slope obtained from the projection is compared with the actual image plane slope to verify consistency. For two high-speed cameras, the azimuth angle A and elevation angle E after calibration are substituted into the vector vectors respectively for vector synthesis. The synthesized result is compared with the actual projection of the known straight line vector to verify consistency.

[0056] Although the embodiments of this application have been described above in conjunction with the accompanying drawings, this application is not limited to the specific embodiments and application fields described above. The specific embodiments described above are merely illustrative and instructive, not restrictive. Those skilled in the art can make many other forms based on the guidance of this specification and without departing from the scope of protection of the claims of this application, and these are all within the scope of protection of this application.

Claims

1. A method for optical axis pointing calibration of a high-speed camera based on linear vector projection slope matching, characterized in that, Includes the following steps: S1. Make a calibration plate, wherein the calibration plate is a plane and at least two non-parallel straight line vectors are distributed on the surface, and the thickness of the straight line vectors is determined according to the imaging field of view of the high-speed camera being calibrated and the extractability of the imaging straight line vectors. S2. Establish a coordinate system based on the calibration plate; S3. Take a picture of the calibration board, select at least two valid straight line vectors from the acquired image, and extract their image plane slope. ; S4. Project the selected straight line vector onto the image plane of the high-speed camera to obtain a two-dimensional projection vector containing the optical axis pointing parameters to be calibrated, wherein the optical axis pointing parameters include azimuth angle A and pitch angle E; S5. Compare the simulated projection result with the extracted actual image plane slope. Matching is performed, and the optical axis pointing parameters of the high-speed camera are obtained through slope consistency calculation.

2. The method for high-speed camera optical axis pointing calibration based on linear vector projection slope matching according to claim 1, characterized in that, In step S2, the coordinate system is defined as follows: with the lower left corner of the calibration plate as the origin, the horizontal rightward axis parallel to the ground as the X-axis, and the vertical upward axis as the Y-axis, the XYZ axes form a right-handed system; the calibration plate is placed perpendicular to the ground, with its front facing the lens in the center, and the straight line vector perpendicular to the ground is an invalid straight line vector.

3. The method for high-speed camera optical axis pointing calibration based on linear vector projection slope matching according to claim 1, characterized in that, In step S2, the coordinate system is defined as follows: the calibration plate is placed directly below the line of sight of the high-speed camera with the lens facing it, and the two mutually perpendicular sides of the calibration plate are the X-axis and Y-axis, respectively. The XYZ axes form a right-handed coordinate system.

4. The method for high-speed camera optical axis pointing calibration based on linear vector projection slope matching according to claim 1, characterized in that, The calibration plate contains at least two linear vectors that are not parallel to each other.

5. The method for high-speed camera optical axis pointing calibration based on linear vector projection slope matching according to claim 1, characterized in that, The correctness of the optical axis pointing calibration result shall be verified by one of the following methods: For a single high-speed camera, the known straight line vector used for non-calibration is used as the target. The azimuth angle A and elevation angle E after calibration are substituted for projection. The projection slope is compared with the actual image plane slope to verify consistency. For two high-speed cameras, the calibrated azimuth angle A and elevation angle E are substituted into each camera to perform vector synthesis. The synthesized result is then compared with the actual projection of the known straight line vector to verify consistency.

Images