Tilted field 3-d measurement method and system using a dual-prism assisted shamir double telecentric lens

Abstract

The application provides a kind of oblique field three-dimensional measurement method and system of double-prism auxiliary Sharm double-telecentric lens, belongs to optical non-contact measurement field.The method comprises: adjusting image surface inclination to establish clear imaging condition;Collect and segment reference full-field image, obtain reference sub-image set;Collect and segment deformed full-field image, obtain deformed sub-image sequence;Parameter set is obtained by calibration;Matching and reconstruction are carried out based on parameter set, and three-dimensional point cloud sequence is generated;Full-field three-dimensional displacement field is calculated.The application realizes the cooperation of large-tilt-angle oblique field under large depth of field clear imaging and constant magnification projection, effectively suppresses pseudo-strain error;Single camera structure is used, synchronous error is eliminated and the structure is compact;Meanwhile, decoupling model is provided to reduce calibration complexity.

Description

Technical Field

[0001] This invention relates to the field of optical non-contact measurement, and in particular to a method and system for three-dimensional measurement of the oblique field of view using a double-prism-assisted Sham double telecentric lens. Background Technology

[0002] Three-dimensional digital image correlation (3D-DIC) technology, as an important non-contact full-field optical measurement method in the field of experimental mechanics, can quantitatively acquire the three-dimensional displacement and strain fields of the surface of the measured object. It has been widely used in scenarios such as material mechanical property testing, structural health monitoring, and deformation analysis of aerospace components. With the increasing complexity of engineering measurement needs, traditional 3D-DIC systems typically use two or more cameras to simultaneously acquire images from different perspectives, and achieve three-dimensional reconstruction through stereo matching and triangulation principles. However, multi-camera systems have revealed the following inherent defects in practical applications: the system is bulky and difficult to deploy in narrow spaces (such as engine compartments or observation windows of high-temperature and high-pressure containers); the relative motion between cameras can destroy pre-calibrated internal and external parameters, leading to decreased measurement accuracy or even failure; and the hardware synchronization requirements for multiple cameras are high, which can easily introduce synchronization errors.

[0003] To overcome the aforementioned problems, single-camera virtual binocular 3D-DIC technology based on dual-prism beam splitting has emerged. This technology uses a dual-prism mounted in front of a single camera lens to effectively divide the camera's image sensor into two virtual viewpoints with a fixed angle of intersection through light refraction. Since the left and right sub-images are acquired synchronously by the same camera, hardware synchronization errors are fundamentally eliminated, while the physical size of the system is significantly reduced, and stability in vibrating or confined environments is improved.

[0004] However, when a single-camera dual-prism measurement system is applied to an oblique field of view (i.e., the camera optical axis is at a large angle to the object being measured), the following key technical bottlenecks still exist:

[0005] First, there is the problem of local defocusing due to insufficient depth of field. In strabismus measurement tasks, due to the limitations of the object-image conjugate law in traditional lens imaging, only a very narrow strip area in the center of the object plane can be clearly imaged. As the object distance deviation increases, the speckle pattern in the edge area quickly becomes blurred, causing the correlation coefficient of digital image correlation matching to drop sharply, which severely limits the effective measurement field of view.

[0006] Second, there is the spurious strain error caused by perspective projection distortion. Conventional imaging lenses use a central perspective projection model, which inherently exhibits the phenomenon of "objects appearing larger when closer and smaller when farther away." In an oblique field of view, the distance from different points on the surface of the object to the lens is unequal, resulting in differences in magnification across different areas of the image. When the object undergoes out-of-plane displacement or vibration along the optical axis, this non-uniform magnification change introduces severe spurious strain into the displacement field, significantly masking the true stress and deformation information of the material.

[0007] Third, existing improvement schemes struggle to balance depth-of-field compensation and geometric fidelity. Some studies attempt to address depth-of-field or distortion issues separately using single methods (such as employing Schahm lenses or telecentric lenses), but these methods fail to achieve both full-field sharp focus and constant magnification imaging while maintaining a compact single-camera structure. This is particularly true for applications requiring high-precision 3D deformation measurements at large tilt angles (such as non-in-situ mechanical testing and in-situ loading observation), where current technology lacks a systematic and collaborative optical measurement solution to address these problems. Summary of the Invention

[0008] To address the aforementioned problems, the present invention aims to provide a single-camera 3D measurement method and system that can simultaneously achieve clear focus across the entire field of view, suppress perspective distortion and spurious strain, and has a compact structure under large tilt angles of view.

[0009] The first aspect of the present invention provides a method for three-dimensional measurement of the oblique field of view using a double-prism assisted Sham double telecentric lens, comprising the following steps:

[0010] Based on the spatial tilt angle of the inclined plane to be measured, the image plane tilt angle of the photosensitive element in the image acquisition module is adjusted to establish clear imaging conditions applicable to the entire field of oblique view.

[0011] Under the clear imaging conditions, a full-field image containing two virtual perspectives formed by the beam splitting of the double prism is acquired by a single trigger as a reference image, and the image is segmented to obtain reference sub-image sets corresponding to the first and second virtual perspectives.

[0012] During the deformation or displacement of the object, multiple full-field images are continuously acquired according to a preset sampling frequency, and each image is segmented to form a sequence of deformed sub-images corresponding to the first virtual viewpoint and the second virtual viewpoint.

[0013] The measurement system is calibrated to obtain a set of calibration parameters for mapping two-dimensional pixel coordinates to three-dimensional spatial coordinates; the set of calibration parameters includes at least: a virtual binocular parallax parameter characterizing the beam splitting characteristics of the double prism and a mapping coefficient characterizing the sensitivity of image plane tilt to out-of-plane displacement;

[0014] Based on the calibration parameter set, cross-view image matching and temporal image matching are performed on the reference sub-image set and the deformed sub-image sequence. Combined with the triangulation principle and the mapping coefficient, the three-dimensional spatial coordinates of each point on the surface of the object under test at each deformation time are reconstructed to generate a three-dimensional point cloud sequence.

[0015] The three-dimensional point cloud sequence is compared with the three-dimensional point cloud at the reference time to obtain the full-field three-dimensional displacement field of the surface of the object under test.

[0016] Furthermore, it also includes: confirming the full-field strain distribution on the surface of the object under test based on the full-field three-dimensional displacement field.

[0017] Further calibration procedures include:

[0018] Within the horizontal beam-splitting plane, the virtual binocular parallax parameters and the horizontal magnification of the dual telecentric lenses are calibrated based on a refractive model using a double prism.

[0019] Within the vertical focusing plane, the mapping coefficients and vertical scale factor introduced by the image plane tilt angle are calibrated based on the Sham tilt imaging model.

[0020] Furthermore, the virtual binocular parallax parameters satisfy the following: the parallel offset generated by the light rays after passing through the double prisms is proportional to the out-of-plane coordinates of the measured point, and this proportionality coefficient is determined only by the prism angle and refractive index, thereby realizing a linear mapping from the out-of-plane coordinates to the lateral offset of the image point.

[0021] Furthermore, the mapping coefficient is used to establish a linear mapping between the surface displacement of the object under test and the image point offset: the offset of the image point in the vertical direction is linearly related to the surface displacement of the object.

[0022] Furthermore, in the reconstruction, the linear coupling relationship between the longitudinal coordinates of the image point under Sham tilt imaging, which simultaneously includes in-plane coordinates and out-of-plane coordinates, is utilized; the out-of-plane sensitivity coefficient in this linear coupling relationship is positively correlated with the image plane tilt angle, and its global consistency is given by the constant magnification of the dual telecentric lens.

[0023] Furthermore, in the reconstruction, the characteristic of constant magnification within the object-side depth of field of a dual telecentric lens is utilized, thereby suppressing image scaling caused by object displacement at the physical level and reducing pseudo-strain errors in the three-dimensional displacement field.

[0024] Furthermore, the temporal image matching involves performing correlation operations on deformed sub-images and reference sub-images at different times under the same virtual viewpoint to obtain the in-plane displacement field under that viewpoint.

[0025] Furthermore, the segmentation operation uses the vertical central axis of the photosensitive target as a reference to divide a single full-field image into two sub-images with fixed parallax, corresponding to the observation results of two virtual viewpoints.

[0026] A second aspect of the present invention provides a three-dimensional measurement system for the oblique field of view using a double-prism assisted Sham double telecentric lens, comprising:

[0027] The image plane tilt angle adjustment module is used to adjust the image plane tilt angle of the photosensitive element in the image acquisition module based on the spatial tilt angle of the inclined plane to be measured, so as to establish clear imaging conditions applicable to the entire field of view of the oblique view.

[0028] The reference image acquisition and segmentation module is used to acquire a full-field image containing two virtual viewpoints formed by the split beam of the double prism as a reference image by a single trigger under the clear imaging condition, and to segment the image to obtain reference sub-image sets corresponding to the first virtual viewpoint and the second virtual viewpoint.

[0029] The deformable image acquisition and segmentation module is used to continuously acquire multiple full-field images at a preset sampling frequency during the deformation or displacement of an object, and to segment each image to form a deformable sub-image sequence corresponding to the first virtual viewpoint and the second virtual viewpoint.

[0030] The calibration module is used to calibrate the measurement system and obtain a set of calibration parameters for mapping two-dimensional pixel coordinates to three-dimensional spatial coordinates; the set of calibration parameters includes at least: virtual binocular parallax parameters characterizing the beam splitting characteristics of the double prism and mapping coefficients characterizing the sensitivity of image plane tilt to out-of-plane displacement;

[0031] The 3D reconstruction module is used to perform cross-view image matching and temporal image matching on the reference sub-image set and the deformed sub-image sequence based on the calibration parameter set, and to reconstruct the 3D spatial coordinates of each point on the surface of the object under test at each deformation time by combining the triangulation principle and the mapping coefficient, thereby generating a 3D point cloud sequence.

[0032] The displacement field acquisition module is used to compare the three-dimensional point cloud sequence with the three-dimensional point cloud at the reference time to obtain the full-field three-dimensional displacement field of the surface of the object under test.

[0033] The present invention has the following advantages or beneficial effects:

[0034] 1. Traditional single-camera dual-prism systems suffer from limited depth of field due to ordinary lenses, resulting in only locally clear images during oblique viewing. Furthermore, they exhibit perspective distortion (objects appear larger when closer and smaller when farther away), and out-of-plane displacement introduces severe spurious strain. This invention addresses this by utilizing Scham's law to adjust the image plane tilt angle, ensuring a conjugate relationship between the sensor plane, the lens principal plane, and the oblique surface being measured. This significantly expands the effective depth of field under oblique viewing conditions, ensuring clear and discernible speckle patterns across the entire field of view and improving the stability of digital image correlation calculations. Simultaneously, it employs dual telecentric lenses to provide constant magnification, ensuring that the image proportions remain unchanged with object distance within a certain depth range. This physically suppresses perspective distortion and spurious strain errors caused by out-of-plane movement. The synergistic effect of depth-of-field compensation and geometric fidelity enables the system to obtain clear images across the entire field under large-angle oblique viewing conditions while guaranteeing the authenticity and reliability of three-dimensional displacement and strain measurement results.

[0035] 2. Compared to traditional binocular or multi-camera 3D-DIC systems, this invention requires only a single industrial camera. Two virtual viewpoints with a fixed angle of intersection are formed by beam splitting through a front-end dual-prism. The left and right sub-images are acquired synchronously by the same photosensitive chip, eliminating the need for additional external triggering synchronization devices. This fundamentally avoids synchronization errors between the two cameras and mechanical drift issues related to the relative pose of the cameras. Simultaneously, the overall optical structure is compact, facilitating deployment in confined spaces (such as engine compartments or observation windows of high-temperature, high-pressure containers) or environments with strong vibrations, significantly improving the system's engineering adaptability and long-term operational reliability.

[0036] 3. This invention decomposes the complex three-dimensional optical path into two independent planes for analysis: a horizontal beam-splitting plane and a vertical focusing plane. Within the horizontal beam-splitting plane, a linear mapping relationship is established from the out-of-plane coordinates to the lateral offset of the image point based on a biprism refraction model. Within the vertical focusing plane, a linear coupling relationship is established between the longitudinal coordinates of the image point and its in-plane and out-of-plane coordinates based on a Schahm tilt imaging model. This decoupled modeling approach clearly reveals the functional division of each optical element, providing a theoretical basis for subsequent step-by-step calibration and rapid three-dimensional reconstruction, reducing the complexity of parameter calibration, and improving calibration accuracy and efficiency. Attached Figure Description

[0037] The above and other features and advantages of the present invention will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0038] Figure 1 This is a flowchart illustrating a method for three-dimensional measurement of the oblique field of view using a double-prism-assisted Sham double telecentric lens, according to an exemplary embodiment.

[0039] Figure 2 This is a schematic diagram of the overall structure of the system in the XZ direction;

[0040] Figure 3 This is a schematic diagram of the overall structure of the system in the YZ direction;

[0041] Figure 4 Schematic diagram of the principle of dual-prism dual-telecentric virtual binocular imaging;

[0042] Figure 5 Schematic diagram of the principle of Scham's law for bi-telecentric tilt imaging;

[0043] Figure 6 This is a structural block diagram of a three-dimensional measurement system for a slant field of view using a dual-prism assisted Sham dual telecentric lens, according to an exemplary embodiment.

[0044] Figure 7 This is a schematic diagram of the structure of an electronic device according to the present invention;

[0045] Figure 8This is a schematic diagram of the structure of a computer-readable storage medium according to the present invention. Detailed Implementation

[0046] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, they are provided so that the invention will be thorough and complete, and the concept of the exemplary embodiments will be fully conveyed to those skilled in the art. The same reference numerals in the drawings denote the same or similar structures, and therefore their detailed description will be omitted.

[0047] The terms “a,” “one,” “the,” and “the” are used to indicate the existence of one or more elements / components / etc.; the terms “including” and “having” are used to indicate an open-ended meaning of inclusion and that other elements / components / etc. may exist in addition to the listed elements / components / etc.

[0048] Example 1

[0049] like Figures 1 to 5 As shown, this embodiment discloses a single-camera oblique field three-dimensional measurement method using a double-prism-assisted Sham double telecentric lens, which is particularly suitable for the three-dimensional deformation and morphology measurement of objects under large tilt angle oblique field.

[0050] Step S100: System construction and focusing steps.

[0051] Based on the spatial tilt angle of the inclined plane to be measured, the image plane tilt angle of the photosensitive element in the image acquisition module is adjusted to establish clear imaging conditions applicable to the entire field of oblique view.

[0052] In this embodiment, the light beam undergoes refraction via a double prism, projection via a double telecentric lens, and image plane tilt adjustment, involving complex three-dimensional spatial geometric relationships. To establish an accurate imaging mapping model and simplify parameter calculations, this invention projects the three-dimensional physical optical path onto two mutually perpendicular feature reference planes for analysis: the horizontal beam-splitting plane (XZ plane) and the vertical focusing plane (YZ plane). The necessity of this decomposition modeling lies in:

[0053] Functional decoupling: The beam splitting action of the biprism in the system mainly occurs in the horizontal direction, determining the parallax baseline of the virtual binoculars; while the depth-of-field compensation action of Scham's law mainly occurs in the vertical tilt direction. By decomposing these two core physical processes, "stereo vision construction" and "large depth-of-field focusing compensation" can be characterized independently.

[0054] Geometric simplification: Complex spatial oblique ray tracing is transformed into paraxial optical path analysis in two two-dimensional planes, which facilitates the direct derivation of explicit analytical relationships between image plane position, magnification, and deflection angle using geometric optics formulas.

[0055] Calibration support: This coordinate system division method corresponds completely to the subsequent image segmentation processing logic, providing a physical benchmark for decoupling the internal and external parameters of the virtual viewpoint.

[0056] Based on the tilt angle β of the surface to be measured, the Schahm formula is used. (Where M is the lateral magnification of the lens) Calculate and adjust the camera's image plane tilt angle α. By adjusting the object distance, ensure that the inclined plane under test can be clearly imaged across the entire field of view in both the left and right sub-images of the single camera, and that the overlapping area of ​​the two sub-images completely covers the target under test. Activate the image acquisition module, observe the left and right sub-images formed by the photosensitive target surface in real time, fine-tune the image plane tilt angle and object distance, and utilize the large depth-of-field characteristics of the dual telecentric lenses and the focal plane rotation characteristics of Scherm's law to ensure that the inclined plane under test can achieve conjugate focusing across a wide range of depth directions in the YZ plane, so that the contrast of the speckle pattern across the entire field of view meets the requirements for 3D-DIC calculation, thereby establishing clear imaging conditions suitable for the entire field of view of the oblique view.

[0057] Step S200: Reference image acquisition step.

[0058] Under the aforementioned clear imaging conditions, a full-field image containing two virtual perspectives formed by the split beam of the double prism is acquired by a single trigger as a reference image, and the image is segmented to obtain reference sub-image sets corresponding to the first and second virtual perspectives.

[0059] Specifically, a random speckle field is prepared on the surface of the inclined plane to be tested by manual spraying or transfer printing. The speckle particle size is optimized according to the spatial resolution of the system to ensure that a high-contrast, gray-level gradient-rich feature pattern is obtained in the image acquisition module for subsequent digital image correlation (DIC) analysis. Before the object under test deforms, the camera is triggered to acquire a single full-field reference image. Using image segmentation technology, the full-field image is split into two sub-images, left and right, based on the vertical central axis of the target surface, corresponding to the initial reference state of the object surface under two virtual viewpoints, respectively, thus obtaining reference sub-image sets corresponding to the first and second virtual viewpoints.

[0060] Step S300: Deformed image sequence acquisition step.

[0061] During the deformation or displacement of the object, multiple full-field images are continuously acquired according to a preset sampling frequency, and each image is segmented to form a sequence of deformed sub-images corresponding to the first and second virtual perspectives.

[0062] During experimental loading or object deformation (e.g., starting the testing machine for quasi-static tensile or compressive experiments, or driving the translation stage to generate a predetermined rigid body displacement), single deformation images are synchronously acquired according to the set sampling frequency. Left and right deformation sub-image sequences at corresponding moments are extracted through image segmentation, ensuring that image acquisition before and after deformation is completed under completely consistent optical paths, forming deformation sub-image sequences corresponding to the first and second virtual viewpoints. During acquisition, the dual telecentric optical paths and the Schahm compensation angle are kept locked to ensure geometric consistency throughout the experiment.

[0063] Step S400: System calibration procedure.

[0064] The measurement system is calibrated to obtain a set of calibration parameters for mapping two-dimensional pixel coordinates to three-dimensional spatial coordinates; the set of calibration parameters includes at least: virtual binocular parallax parameters characterizing the spectral dispersion characteristics of the double prism and mapping coefficients characterizing the sensitivity of image plane tilt to out-of-plane displacement.

[0065] This step employs a virtual binocular camera calibration model incorporating refraction correction. By acquiring multiple sets of calibration images in different poses, the intrinsic and extrinsic parameters of the system's left and right virtual viewpoints, as well as the geometric transformation parameters introduced by image plane tilt, are precisely determined. Specifically, based on the decoupled modeling approach, after the system's optical path is fully locked, a high-precision calibration board is placed within the field of view, and system parameters are acquired step by step. The complete imaging model and formula derivation are presented below from the horizontal beam splitting plane (XZ plane) and the vertical focusing plane (YZ plane), respectively.

[0066] I. XZ Plane: Double Prism, Double Telecentric Virtual Binocular Imaging Model

[0067] Since the light rays passing through the double telecentric lens are parallel to its optical axis, the biprism placed in front of the double telecentric lens will not change the direction of the imaging light rays, but only cause them to be parallelly shifted. Assuming that the optical axis of the double telecentric lens is perpendicular to the rear surface of the biprism, the above-mentioned light ray shift can be obtained through geometric optical analysis within the cross-section of the biprism. If the object point is observed through the biprism, the light rays emitted from the object point P are first deflected at the front surface of the wedge prism, and then pass through the rear surface without deflection to form imaging rays parallel to the OZ coordinate axis.

[0068] exist Figure 4 In the diagram, let the included angle of the biprism be θ, and the thickness along the OZ axis be... The relative refractive index is N, the distance between object point P and the origin of the XOZ coordinate system is Z, and the distance the light ray is deflected by the upper wedge prism is... The light rays are deflected by a distance of the lower wedge prism. Since the imaging rays passing through the wedge prism and the refracted rays from the front surface are collinear, the angles between both rays and the normal to the front surface are θ. The angles between the incident ray and the normals to the front surfaces of the upper and lower wedge prisms are respectively... and According to Snell's law, the angle i between the incident ray and the normal to the front surface satisfies the following relationship:

[0069] (1)

[0070] Based on angle i, the angle between the incident ray and the optical axis of the lens can be calculated. This leads to the establishment of information about the distance of light deflection. The equation:

[0071] (2)

[0072] (3)

[0073] After deriving and simplifying the expression for h from the above equation, we can obtain: and The mathematical expressions are the same:

[0074] (4)

[0075] In the formula, K is a parameter determined by the wedge prism's included angle α and relative refractive index N. It can be seen that the ray deflection produced by the wedge prism with these parameters depends only on the out-of-plane coordinate Z of the object being measured. From the deflection... and The intersection of the prime minister's pipeline and the rear surface of the biprism can be determined. and Spatial coordinates:

[0076] (5)

[0077] (6)

[0078] (7)

[0079] The maximum thickness of the biprism And the X coordinate of object point P, can be used to... and Replace with:

[0080] (8)

[0081] (9)

[0082] Substitute it and The spatial coordinates can be obtained as follows:

[0083] (10)

[0084] (11)

[0085] Since the orthographic projection process is independent of the object's position relative to the surface, the distance from the biprism to the dual telecentric lenses does not affect the image position of object point P in the image. Furthermore, by utilizing the constant magnification of the dual telecentric lenses, the image can be projected onto the target surface. and After scaling the in-plane coordinates by a fixed ratio η, the position of the image point formed by the object point P on the camera's CMOS sensor can be directly obtained. Based on this, combined with the pixel physical size of the CMOS sensor... And the X-axis coordinates of the pixel projection point formed by the spatial coordinate axis OZ on the CMOS sensing element are... This allows establishing an analytical relationship between the 3D coordinates of an object point and the pixel coordinates of an image point. Figure 4 midpoint , For example, its pixel coordinates are:

[0086] (12)

[0087] (13)

[0088] A coordinate system is established with the center of the camera's CMOS sensor as the origin. The mathematical model result for imaging is as follows:

[0089] (14)

[0090] (15)

[0091] II. YZ Plane: Scham's Law-based Bipolar Tilt Imaging Model

[0092] Because the object-side principal ray of a dual telecentric lens is always parallel to its optical axis, the imaging rays at different distances from the surface of the measured object all propagate in a direction parallel to the OZ coordinate axis, and the system imaging satisfies the orthographic projection relationship. When the CMOS sensing element is tilted relative to the lens image plane by Scheimpflug, the imaging rays propagating parallel to the optical axis will produce an additional positional offset on the tilted image plane, thereby establishing a linear relationship between the object point's distance from the surface coordinates and the image point position.

[0093] exist Figure 5 In this case, the optical axis of the dual telecentric lens is along the OZ direction, the CMOS sensor is rotated by an angle α around the OX axis, and the object point P is located in the YOZ section. Its spatial coordinates are expressed as:

[0094]

[0095] Let β be the angle by which the CMOS sensor rotates about the OX axis, and let α be the angle of tilt of the object surface relative to the reference plane perpendicular to the optical axis. According to the Scheimpflug imaging principle, the object surface, the lens image plane, and the CMOS sensor plane intersect on the same straight line, thus ensuring that a clear image can still be obtained even with an tilted object surface.

[0096] Without introducing Scheimpflug tilt, the CMOS sensor element is perpendicular to the lens optical axis. In this case, the imaging light rays emitted from the object point P propagate parallel to the OZ direction and form an image point on the CMOS sensor element. Let its Y-axis coordinate be Because a dual telecentric lens has a constant magnification, the in-plane coordinates of the object point and the image point satisfy a linear proportional relationship:

[0097] (16)

[0098] Where η is the magnification of the dual telecentric lens imaging. When the CMOS sensor element is tilted by an angle β around the OX axis, since the CMOS plane is no longer perpendicular to the optical axis, the actual imaging distance on the tilted image plane will be geometrically stretched compared to the perpendicular image plane. Therefore, the imaging scale of the object point's in-plane coordinate Y on the CMOS sensor element will be corrected, and its corresponding longitudinal coordinate can be expressed as:

[0099] (17)

[0100] As shown in the above equation, the CMOS tilt angle β will cause the vertical scale of the image plane to be magnified, and the magnification factor is:

[0101]

[0102] Further consider the influence of the object point's out-of-plane coordinate Z on the imaging position. Since the imaging rays from a dual telecentric lens propagate parallel to the optical axis, the imaging rays at different out-of-plane positions will produce an additional longitudinal offset on the tilted CMOS plane. Let this additional offset be h, which can be established based on the Scheimpflug tilt geometry:

[0103] (18)

[0104] It can be seen that the additional offset h is linearly related to the object point's coordinate Z above the surface, and its scaling factor is determined by the object surface tilt angle α. Therefore, the final longitudinal coordinate of the object point P on the tilted CMOS sensing element satisfies:

[0105] (19)

[0106] Substituting h, we get:

[0107] (20)

[0108] Expanding on this further:

[0109] (twenty one)

[0110] make:

[0111] (twenty two)

[0112] (twenty three)

[0113] The above imaging relationship can then be further expressed as:

[0114] (twenty four)

[0115] Therefore, it can be seen that the longitudinal coordinates of the image point in the Scheimpflug dual telecentric imaging system are not only related to the in-plane coordinates (Y) of the object point, but also have a linear mapping relationship with the out-of-plane coordinates (Z) of the object point. Wherein:

[0116] parameter Characterizes the image plane scale changes caused by CMOS tilt;

[0117] parameter It characterizes the sensitivity of out-of-plane coordinates to the position of the image point.

[0118] Further combining the physical pixel size PSZ of the CMOS sensor element, and the vertical coordinate of the pixel projection point of the spatial coordinate axis OZ on the CMOS sensor element. It can establish an analytical relationship between the spatial coordinates of object points and the pixel coordinates of image points.

[0119] Let the ordinate of pixel N on the CMOS sensor be y, then we have:

[0120] (25)

[0121] Furthermore, the relative coordinates of the image point are established with the optical axis projection point as a reference:

[0122] (26)

[0123] The final analytical model of the Scheimpflug dual telecentric imaging system can be expressed as:

[0124] (27)

[0125] Right now:

[0126] (28)

[0127] As can be seen from the above formula, the longitudinal coordinates of the image point in the Scheimpflug dual telecentric imaging system simultaneously contain both the in-plane coordinates and out-of-plane coordinates of the object point. The out-of-plane coordinates Z are linearly mapped to the longitudinal pixel coordinates of the CMOS sensor element through the Scheimpflug tilt relationship, and its mapping sensitivity is jointly determined by the object tilt angle α and the CMOS tilt angle β.

[0128] III. Implementation of Decoupled System Parameter Calibration

[0129] Based on the above models of the XZ and YZ planes, the specific calibration operations in this step are as follows:

[0130] Horizontal parameter calibration: Based on the XZ plane model (Equations 1-15), the double prism coefficient K and the effective magnification η of the double telecentric lens in the XZ plane are calibrated.

[0131] Vertical parameter calibration: Based on the YZ plane model (Equations 16-28), the image plane scale factor A(β) introduced by the physical tilt angles α and β and the out-of-plane mapping sensitivity B(α,β) are calibrated.

[0132] Comprehensive distortion correction: Combine the decoupled analytical model to establish a unified mapping matrix from pixel coordinates (x,y) to spatial coordinates (X,Y,Z).

[0133] This yields a set of calibration parameters for mapping two-dimensional pixel coordinates to three-dimensional spatial coordinates. This set of parameters includes at least: virtual binocular parallax parameters (K, η, etc.) characterizing the beam splitting characteristics of the double prism, a scale factor (η) characterizing the constant magnification of the double telecentric lens, and mapping coefficients (A, B, etc.) characterizing the sensitivity of image plane tilt to out-of-plane displacement.

[0134] Step S5: 3D reconstruction step.

[0135] Based on the calibration parameter set, cross-view image matching and temporal image matching are performed on the reference sub-image set and the deformed sub-image sequence. Combined with the triangulation principle and the mapping coefficient, the three-dimensional spatial coordinates of each point on the surface of the object under test at each deformation time are reconstructed to generate a three-dimensional point cloud sequence.

[0136] Specifically, the acquired reference images and deformed image sequences undergo digital preprocessing:

[0137] Sub-image segmentation: Using the center line of the photosensitive target as the boundary, each frame of the original full-field image is segmented into two sub-images with parallax information, and feature information from the two virtual viewpoints is extracted.

[0138] Cross-view matching: A digital image correlation algorithm is used to perform correlation operations on the left and right sub-images at the same time to obtain pixel-level disparity.

[0139] Temporal image matching: Correlation calculations are performed on deformed sub-images and reference sub-images at different times under the same virtual viewpoint to obtain the in-plane displacement field under that viewpoint.

[0140] Coordinate analysis solution: Substitute the matched pixel coordinates (x, y) into the joint analytical equation system established in Section (I) and Section (II) above. Calculate the out-of-plane coordinates Z using the XZ plane model (Equations 14 and 15), and correct the in-plane coordinates Y using the YZ plane model (Equations 27 and 28). Finally, reconstruct the three-dimensional spatial coordinates (X, Y, Z) of the measured point at each time point, generating a three-dimensional point cloud sequence.

[0141] In this reconstruction process, the linear coupling relationship between the longitudinal coordinates of the image point under Sham tilt imaging, which simultaneously includes in-plane and out-of-plane coordinates, is utilized (see Equations 24 and 28). The out-of-plane sensitivity coefficient in this linear coupling relationship is positively correlated with the image plane tilt angle, and its global consistency is given by the constant magnification of the dual telecentric lens. At the same time, the characteristic of the dual telecentric lens with constant magnification within the object-side depth of field is utilized to suppress image scaling caused by object out-of-plane displacement at the physical level and reduce pseudo-strain error in the three-dimensional displacement field.

[0142] Step S600: Deformation field calculation steps.

[0143] The three-dimensional point cloud sequence is compared with the three-dimensional point cloud at the reference time to calculate the full-field three-dimensional displacement field of the surface of the object under test. Specifically, the full-field displacement vector of the inclined plane under test in the oblique view field is obtained by calculating the difference between the three-dimensional point cloud coordinates at the reference time and the deformation time.

[0144] Step S700: Full-field strain analysis step.

[0145] After obtaining the full-field three-dimensional displacement field, the full-field strain distribution is further obtained through numerical differentiation of the displacement field. That is, the full-field three-dimensional displacement field is numerically differentiated to calculate the full-field strain distribution on the surface of the object under test. Because the system adopts dual telecentric lenses and Schahm's law compensation, the output results can effectively eliminate the spurious strain error introduced by the off-surface movement of the specimen (such as thinning during the stretching process or the axial displacement of the translation stage), realizing high-precision mechanical property evaluation or displacement vector verification.

[0146] Example 2

[0147] like Figure 6 As shown, this embodiment discloses a three-dimensional measurement system for the oblique field of view using a dual-prism assisted Schamm double telecentric lens, comprising:

[0148] The image plane tilt adjustment module is used to adjust the image plane tilt angle of the photosensitive element in the image acquisition module based on the spatial tilt angle of the inclined plane to be measured, so as to establish clear imaging conditions applicable to the entire field of view of the oblique view.

[0149] The reference image acquisition and segmentation module is used to acquire a full-field image containing two virtual viewpoints formed by the split beam of the double prism as a reference image by a single trigger under the clear imaging condition, and to segment the image to obtain reference sub-image sets corresponding to the first virtual viewpoint and the second virtual viewpoint.

[0150] The deformable image acquisition and segmentation module is used to continuously acquire multiple full-field images at a preset sampling frequency during the deformation or displacement of an object, and to segment each image to form a deformable sub-image sequence corresponding to the first virtual viewpoint and the second virtual viewpoint.

[0151] The calibration module is used to calibrate the measurement system and obtain a set of calibration parameters for mapping two-dimensional pixel coordinates to three-dimensional spatial coordinates. The set of calibration parameters includes at least: virtual binocular parallax parameters characterizing the spectral dispersion characteristics of the double prism and mapping coefficients characterizing the sensitivity of image plane tilt to out-of-plane displacement.

[0152] The 3D reconstruction module is used to perform cross-view image matching and temporal image matching on the reference sub-image set and the deformed sub-image sequence based on the calibration parameter set, and to reconstruct the 3D spatial coordinates of each point on the surface of the object under test at each deformation time by combining the triangulation principle and the mapping coefficient, thereby generating a 3D point cloud sequence.

[0153] The displacement field acquisition module is used to compare the three-dimensional point cloud sequence with the three-dimensional point cloud at the reference time to obtain the full-field three-dimensional displacement field of the surface of the object under test.

[0154] The strain field acquisition module is used to determine the full-field strain distribution on the surface of the object under test based on the full-field three-dimensional displacement field.

[0155] The specific methods by which the modules in the above embodiments perform their operations have been described in detail in Embodiment 1, and will not be elaborated upon here.

[0156] Example 3

[0157] like Figure 7 As shown, the present invention also provides an electronic device, including a processor 101, a communication interface 102, a memory 103, and a communication bus 104. The processor 101, communication interface 102, and memory 103 communicate with each other via the communication bus 104. The memory 103 is used to store computer programs. In this embodiment, when the processor 101 executes the program stored in the memory 103, it implements the steps of the method in Embodiment 1.

[0158] The electronic device provided in this embodiment of the invention has a similar implementation principle and technical effect to the above embodiments, and will not be described again here.

[0159] The aforementioned memory 103 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. Memory 103 has storage space for program code used to perform any of the method steps described above. For example, the storage space for program code may include individual program codes for implementing the various steps in the methods described above. This program code can be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, optical discs (CDs), memory cards, or floppy disks. Such computer program products are typically portable or fixed storage units. The storage unit may have storage segments or storage spaces arranged similarly to memory 103 in the aforementioned electronic device. The program code may be compressed, for example, in a suitable form. Typically, the storage unit includes programs for performing the method steps according to embodiments of the invention, i.e., code that can be read by a processor such as 101, which, when run by the electronic device, causes the electronic device to perform the various steps in the methods described above.

[0160] Example 4

[0161] like Figure 8 As shown, the present invention also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the method described above.

[0162] The computer-readable storage medium may be included in the device / apparatus described in the above embodiments; or it may exist independently and not assembled into the device / apparatus. The computer-readable storage medium carries one or more programs that, when executed, implement the method according to the embodiments of the present invention.

[0163] According to embodiments of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium, such as including, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In the present invention, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0164] In this embodiment of the invention, the term "multiple" refers to two or more, unless otherwise explicitly defined. The terms "install," "connect," and "fix" should be interpreted broadly. For example, "connect" can mean a fixed connection, a detachable connection, or an integral connection. Those skilled in the art can understand the specific meaning of the above terms in this embodiment of the invention based on the specific circumstances.

[0165] In the description of the embodiments of the present invention, it should be understood that the terms "upper" and "lower" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the embodiments of the present invention and simplifying the description, and do not indicate or imply that the device or unit referred to must have a specific orientation or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the embodiments of the present invention.

[0166] In the description of this specification, the terms "an embodiment," "a preferred embodiment," etc., refer to a specific feature, structure, material, or characteristic described in connection with that embodiment or example, which is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0167] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. For those skilled in the art, the embodiments of the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of the present invention should be included within the protection scope of the embodiments of the present invention.

Claims

1. A method for three-dimensional measurement of the oblique field of view using a double-prism assisted Schahm telecentric lens, characterized in that, Includes the following steps: Based on the spatial tilt angle of the inclined plane to be measured, the image plane tilt angle of the photosensitive element in the image acquisition module is adjusted to establish clear imaging conditions applicable to the entire field of oblique view. Under the clear imaging conditions, a full-field image containing two virtual perspectives formed by the beam splitting of the double prism is acquired by a single trigger as a reference image, and the image is segmented to obtain reference sub-image sets corresponding to the first and second virtual perspectives. During the deformation or displacement of the object, multiple full-field images are continuously acquired according to a preset sampling frequency, and each image is segmented to form a sequence of deformed sub-images corresponding to the first virtual viewpoint and the second virtual viewpoint. The measurement system is calibrated to obtain a set of calibration parameters for mapping two-dimensional pixel coordinates to three-dimensional spatial coordinates; The calibration parameter set includes at least: a virtual binocular parallax parameter characterizing the beam splitting characteristics of the double prism and a scaling factor characterizing the linear change of the image plane tilt angle with respect to the object tilt direction. Based on the calibration parameter set, cross-view image matching and temporal image matching are performed on the reference sub-image set and the deformed sub-image sequence. Combined with the triangulation principle and the mapping coefficient, the three-dimensional spatial coordinates of each point on the surface of the object under test at each deformation time are reconstructed to generate a three-dimensional point cloud sequence. The three-dimensional point cloud sequence is compared with the three-dimensional point cloud at the reference time to obtain the full-field three-dimensional displacement field of the surface of the object under test.

2. The method according to claim 1, characterized in that, Also includes: Based on the full-field three-dimensional displacement field, the full-field strain distribution on the surface of the object under test is determined.

3. The method according to claim 1, characterized in that, The calibration process includes: Within the horizontal beam-splitting plane, the virtual binocular parallax parameters and the horizontal magnification of the dual telecentric lenses are calibrated based on a refractive model using a double prism. Within the vertical focusing plane, the mapping coefficients and vertical scale factor introduced by the image plane tilt angle are calibrated based on the Sham tilt imaging model.

4. The method according to claim 1, characterized in that, The virtual binocular parallax parameters satisfy the following: the parallel offset generated by the light after passing through the double prism is proportional to the out-of-plane coordinates of the measured point, and this proportionality coefficient is determined only by the prism angle and refractive index, thereby realizing a linear mapping from the out-of-plane coordinates to the lateral offset of the image point.

5. The method according to claim 1, characterized in that, The mapping coefficient is used to establish a linear mapping between the surface displacement of the object under test and the image point offset: the offset of the image point in the vertical direction is linearly related to the surface displacement of the object.

6. The method according to claim 1, characterized in that, In the reconstruction, the linear coupling relationship between the longitudinal coordinates of the image point under Sham tilt imaging, which simultaneously includes in-plane coordinates and out-of-plane coordinates, is utilized. The out-of-plane sensitivity coefficient in this linear coupling relationship is positively correlated with the image plane tilt angle, and its global consistency is given by the constant magnification of the dual telecentric lens.

7. The method according to claim 1, characterized in that, In the reconstruction, the characteristic of constant magnification within the object-side depth of field of a dual telecentric lens is utilized, thereby suppressing image scaling caused by object displacement at the physical level and reducing pseudo-strain errors in the three-dimensional displacement field.

8. The method according to claim 1, characterized in that, The temporal image matching involves performing correlation operations on deformed sub-images and reference sub-images at different times under the same virtual viewpoint to obtain the in-plane displacement field under that viewpoint.

9. The method according to claim 1, characterized in that, The segmentation operation uses the vertical central axis of the photosensitive target as a reference to divide a single full-field image into two sub-images with fixed parallax, corresponding to the observation results of two virtual viewpoints.

10. A three-dimensional measurement system for the oblique field of view using a double-prism assisted Schahm telecentric lens, characterized in that, include: The image plane tilt angle adjustment module is used to adjust the image plane tilt angle of the photosensitive element in the image acquisition module based on the spatial tilt angle of the inclined plane to be measured, so as to establish clear imaging conditions applicable to the entire field of view of the oblique view. The reference image acquisition and segmentation module is used to acquire a full-field image containing two virtual viewpoints formed by the split beam of the double prism as a reference image by a single trigger under the clear imaging condition, and to segment the image to obtain reference sub-image sets corresponding to the first virtual viewpoint and the second virtual viewpoint. The deformable image acquisition and segmentation module is used to continuously acquire multiple full-field images at a preset sampling frequency during the deformation or displacement of an object, and to segment each image to form a deformable sub-image sequence corresponding to the first virtual viewpoint and the second virtual viewpoint. The calibration module is used to calibrate the measurement system and obtain a set of calibration parameters for mapping two-dimensional pixel coordinates to three-dimensional spatial coordinates. The calibration parameter set includes at least: a virtual binocular parallax parameter characterizing the spectral splitting characteristics of the double prism, and a scaling factor characterizing the linear change of the image plane tilt angle with respect to the object tilt direction; The 3D reconstruction module is used to perform cross-view image matching and temporal image matching on the reference sub-image set and the deformed sub-image sequence based on the calibration parameter set, and to reconstruct the 3D spatial coordinates of each point on the surface of the object under test at each deformation time by combining the triangulation principle and the mapping coefficient, thereby generating a 3D point cloud sequence. The displacement field acquisition module is used to compare the three-dimensional point cloud sequence with the three-dimensional point cloud at the reference time to obtain the full-field three-dimensional displacement field of the surface of the object under test.

Images