A self-repairing recalibration method, device and system for a DIC system
By using pure vision to calculate the drift transformation matrix and a modular architecture, the problem of data discontinuity caused by pose drift in the DIC system during long-term monitoring was solved, realizing unified data correction and efficient management, and improving the reliability and accuracy of measurement.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- YANSUO INSTR TECH (SHANGHAI) CO LTD
- Filing Date
- 2026-04-27
- Publication Date
- 2026-07-14
AI Technical Summary
In long-term monitoring, the DIC system may experience slight pose drift due to temperature cycling, wind load, settlement, or accidental collisions, resulting in "steps" or "drifts" in the measurement data, which disrupts the continuity of the time series. Traditional recalibration operations are complex and cannot maintain the absolute physical meaning of the data.
The system drift transformation matrix is calculated using a purely visual method. A permanent benchmark is established through initial calibration, and a modular architecture is constructed, including benchmark establishment, field data acquisition, transformation calculation, and data correction modules, to achieve unified data correction and closed-loop management.
Ensuring all data is seamlessly integrated into the same physical coordinate system improves data continuity and reliability, simplifies recalibration operations, reduces costs, and achieves full lifecycle data traceability and accuracy assurance.
Smart Images
Figure CN122391377A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of computer vision and optical precision measurement technology, specifically relating to a calibration and data maintenance method for a digital image correlation (DIC) measurement system, and more particularly to a method and system for rapid on-site self-correction and recalibration that can ensure the spatiotemporal continuity and comparability of all measurement data throughout the entire lifecycle of the system. Background Technology
[0002] Digital image correlation (DIC) technology is a non-contact, full-field optical measurement method that has been widely applied in deformation and strain measurement in fields such as materials mechanics, structural health monitoring, and precision manufacturing. The measurement accuracy of a DIC system is highly dependent on the accuracy of camera parameter calibration. Traditional camera calibration methods require the use of a high-precision calibration board to acquire images from multiple perspectives (usually requiring the capture of a dozen to several dozen images), and to solve for the camera's internal parameters (such as focal length, principal point, and distortion coefficients) and external parameters through complex nonlinear optimization.
[0003] In long-term, continuous field monitoring (such as health monitoring of large machinery and long-term experimental measurements), the core challenge facing DIC systems is not only initial accuracy, but also the reliability and consistency of long-term data. In actual experimental sites or long-term monitoring environments, minute pose drifts caused by temperature cycling, wind loads, settlement, or accidental collisions can introduce systematic measurement errors, leading to "steps" or "drifts" in monitoring data. This disrupts the continuity of time series, making trend analysis and early warning judgments unfounded.
[0004] Currently, recalibrating to solve this problem requires a complete recalibration, which is no less complex than the initial calibration and severely disrupts the continuity of monitoring. More importantly, traditional recalibration establishes a new coordinate system independent of the initial calibration. The data before and after recalibration belong to two different spatial reference systems, making it impossible to directly compare and analyze continuous physical quantities (such as displacement and strain). Although alignment can be achieved through subsequent data registration, this process is complex, introduces secondary errors, and loses the absolute physical meaning of the measurement data.
[0005] For example, Chinese patent application CN118570307A discloses a method for monitoring camera pose changes and correcting the coordinate system using a tilt sensor. This method essentially relies on external sensors to directly measure pose changes, which is fundamentally different from the pure visual self-correction scheme of this invention.
[0006] Therefore, providing a method and system for rapid on-site recalibration of a DIC system based on initial calibration self-correction, which can achieve drift correction without external sensors and based solely on visual information, and ensure that all data is always based on the same initial coordinate system, is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention
[0007] To address the aforementioned technical problems, this invention provides a method and system for rapid on-site recalibration of a DIC system based on initial calibration self-correction, thereby overcoming the technical defects of existing technologies, such as data coordinate system breakage caused by recalibration, reliance on external sensors, and complex operation.
[0008] To address the aforementioned technical problems, this invention constructs a core paradigm of "one-time calibration, permanent benchmark" and employs a purely visual method to calculate the system drift transformation matrix, thereby achieving unified data correction.
[0009] The system adopts a modular architecture, consisting of a benchmark establishment module, a field acquisition module, a transformation calculation module, a data correction module, and a storage unit. The modules work together to form a complete closed-loop system.
[0010] Specifically, this invention establishes an initial calibration coordinate system through initial calibration, obtains and stores the camera's intrinsic parameters and the first extrinsic parameters between the calibration board and the camera. When system drift correction is needed, at least one image of the calibration board is acquired, and the second extrinsic parameters between the current camera coordinate system and the calibration board are calculated using the intrinsic parameters. Based on the difference between the first and second extrinsic parameters, a rigid body transformation matrix is calculated to transform the current camera coordinate system back to the initial calibration coordinate system, and this rigid body transformation matrix is applied to the 3D coordinate data to achieve unified data transformation. This invention also provides two parallel strategies for calculating the rigid body transformation matrix: a single-point anchoring simplified strategy (Strategy B) and a global constraint optimization strategy (Strategy A), to adapt to scenarios with different accuracy and convenience requirements.
[0011] Thus, this invention realizes a complete closed loop from initial calibration to drift correction and data traceability, significantly improving the data continuity, ease of operation and measurement reliability of the DIC system for long-term monitoring.
[0012] According to one aspect of the present invention, a method for rapid on-site recalibration of a DIC system based on initial calibration self-correction is provided, comprising the following steps: An initial calibration coordinate system is established through initial calibration, and the camera's internal parameters and the camera's first external parameters relative to the calibration plate in the initial calibration coordinate system are obtained. Acquire at least one image of the field calibration board, obtain the correspondence between the two-dimensional image coordinates and the three-dimensional spatial coordinates of the feature points of the calibration board based on the image of the field calibration board, and use the internal parameters to calculate the second external parameters of the current camera coordinate system relative to the calibration board; Based on the first external parameter and the second external parameter, calculate the rigid body transformation matrix used to transform the current camera coordinate system back to the initial calibration coordinate system; The rigid body transformation matrix is applied to the three-dimensional coordinate data obtained by DIC measurement, and the three-dimensional coordinate data is uniformly transformed to the initial calibration coordinate system.
[0013] Preferably, the calibration plate is placed within the camera's field of view at any position and orientation.
[0014] Preferably, each feature point of the calibration board has a unique coded identifier.
[0015] Preferably, the rigid body transformation matrix is calculated in one of the following two ways: Method 1: Solve the current pose of a single field calibration board image using the PnP algorithm, compare it with the first external parameter, calculate the relative rigid body transformation, and then take the inverse to obtain the rigid body transformation matrix; Method 2: Jointly optimize multiple field calibration board images and multiple initial calibration images. Under the constraint of keeping the internal parameters unchanged, with the goal of minimizing the overall reprojection error of the feature points of the calibration board on the multiple field calibration board images and the multiple initial calibration images, solve for the rigid body transformation matrix.
[0016] Preferably, the three-dimensional coordinate data is converted to the initial calibration coordinate system using the following formula: X_corrected=T_cam×X_original; Where X_corrected is the transformed 3D coordinate data, T_cam is the rigid body transformation matrix, and X_original is the original 3D coordinate data.
[0017] Preferably, the method is executed based on the assumption that the internal parameters remain stable, and a manual calibration prompt is triggered when the internal parameters change beyond a preset threshold.
[0018] Preferably, the method further includes: saving the rigid body transformation matrix obtained from each correction calculation, wherein the saved rigid body transformation matrices are concatenated in the correction sequence to form a correction transformation chain, so that the three-dimensional coordinate data can be traced back to the initial calibration coordinate system by applying the correction transformation chain.
[0019] In another aspect, the present invention provides a field-based rapid recalibration device for a DIC system based on initial calibration self-correction, comprising: The reference establishment module is used to establish an initial calibration coordinate system through initial calibration, and to obtain the camera's internal parameters and the camera's first external parameters relative to the calibration plate in the initial calibration coordinate system. The field acquisition module is used to acquire at least one field calibration board image, obtain the correspondence between the two-dimensional image coordinates and the three-dimensional spatial coordinates of the feature points of the calibration board based on the field calibration board image, and calculate the second external parameter of the current camera coordinate system relative to the calibration board using the internal parameters; The transformation calculation module is used to calculate, based on the first external parameter and the second external parameter, a rigid body transformation matrix for transforming the current camera coordinate system back to the initial calibration coordinate system; The data correction module is used to apply the rigid body transformation matrix to the three-dimensional coordinate data obtained by DIC measurement, and to uniformly transform the three-dimensional coordinate data to the initial calibration coordinate system.
[0020] Preferably, the device further includes a storage unit for storing the rigid body transformation matrix obtained from each correction calculation, wherein the stored rigid body transformation matrices are concatenated in the correction sequence to form a correction transformation chain.
[0021] Another aspect of the present invention provides a rapid on-site recalibration system for a DIC system based on initial calibration self-correction, comprising: Camera, used to acquire initial calibration board images and on-site calibration board images; A calibration plate, as a known physical reference, is used to provide feature points with known geometric structures; A storage unit is used to store the internal parameters and the first external parameters; A processing unit is configured to perform the method described in any of the preceding claims, or to implement the apparatus described in any of the preceding claims.
[0022] Compared with the prior art, the present invention has the following beneficial effects: Fundamental Innovation: It proposes a brand-new DIC system data management paradigm of "one-time calibration, permanent benchmark, and unified coordinate system", which fundamentally solves the global problem of incomparability of data from multiple periods due to different reference systems in long-term monitoring, and provides a data consistency foundation for building a truly usable structural digital twin.
[0023] Data continuity assurance: Through the process of "calculating deviation - unified correction", all data before and after recalibration are seamlessly connected to the same physical coordinate system, the time series is smooth and continuous, and trend analysis, early warning and life assessment can be performed directly.
[0024] Flexible precision-efficiency balance: It offers two parallel implementation strategies: global optimization and single-point anchoring. The former prioritizes accuracy and robustness, suitable for laboratory or periodic in-depth maintenance; the latter prioritizes ease of operation, suitable for daily rapid inspection and drift correction in the field, realizing on-demand configuration of maintenance methods.
[0025] It is highly practical in the field: the single-point anchoring strategy simplifies the complex field recalibration operation to "place the board anywhere and take a picture", which takes as little as 1 minute. It has low technical requirements and greatly increases the willingness and frequency of users to perform regular verification, thus ensuring data quality in the process.
[0026] No external sensors required: Drift correction can be completed using only visual information from the calibration board image, eliminating the need for additional hardware such as tilt sensors, thus reducing system cost and deployment complexity.
[0027] Data lifecycle traceability: By saving the history of corrections and transformations, any data can be traced back to a unified initial coordinate system, realizing full lifecycle traceability management of measurement data.
[0028] Compared to solutions relying on external sensors, this invention requires no additional hardware, reducing the time for a single correction operation from several hours (traditional recalibration) to less than one minute, and lowering the projection error after correction from over 40% to a stable, low level (see [link]). Figure 10 This has resulted in significant efficiency improvements and accuracy assurance.
[0029] Terminology Explanation To facilitate understanding of the technical solution of this invention, some technical terms appearing in the specification are explained as follows: DIC (Digital Image Correlation): A non-contact, full-field optical measurement method that calculates the three-dimensional displacement and strain distribution of an object's surface by tracking changes in the image coordinates of speckle or feature points on the surface of the object.
[0030] Calibration board: A planar or three-dimensional reference object with a known precise geometric structure. Its surface usually contains regularly arranged feature points (such as checkerboard corner points, dot arrays, coded marker points) and is used for camera calibration and pose determination.
[0031] Internal parameters: Parameters that describe the camera's imaging geometry, including focal length (fx, fy), principal point coordinates (Cx, Cy), and lens distortion coefficients (radial distortion K1, K2, tangential distortion p1, p2, etc.).
[0032] External parameters: These are parameters that describe the relative pose relationship between the camera coordinate system and the world coordinate system or calibration board coordinate system. They typically include the rotation matrix R and the translation vector t.
[0033] PnP (Perspective-n-Point): A classic problem in computer vision. Given the camera's intrinsic parameters, the coordinates of a point in 3D space, and the coordinates of its 2D projection points on the image, the problem asks for the camera's pose (rotation matrix and translation vector) relative to this 3D space.
[0034] Rigid body transformation matrix: A mathematical expression describing the changes in position and orientation of a rigid body in three-dimensional space. It is composed of rotation matrices and translation vectors and is used to transform coordinates in one coordinate system to another.
[0035] Reprojection error: The Euclidean distance between the three-dimensional spatial points projected onto the image plane by the camera's intrinsic and extrinsic parameters and the actual observed two-dimensional image points is an important indicator for measuring the accuracy of calibration and pose solving.
[0036] Corrected transformation chain: A data chain composed of rigid body transformation matrices obtained from each correction calculation, concatenated in time sequence, used to trace the three-dimensional coordinate data at any time back to the initial calibration coordinate system.
[0037] The terms used above are only for explaining specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Those skilled in the art will understand that other existing technologies with the same or similar functions can be employed without departing from the concept of the present invention. Attached Figure Description
[0038] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0039] Figure 1 This is a structural diagram of the DIC system field rapid recalibration system based on initial calibration self-correction provided in an embodiment of the present invention; Figure 2 This is a flowchart illustrating the overall process of the rapid on-site recalibration method for a DIC system based on initial calibration self-correction in this embodiment of the invention. Figure 3 This is a schematic diagram of the initial calibration and three-dimensional reconstruction deformation analysis in an embodiment of the present invention; Figure 4 This is a diagram showing the change in the field of view of the object under test within the left camera before and after a collision, as described in this embodiment of the invention. Figure 5 This is a schematic diagram of a calibration plate image acquired after a collision in an embodiment of the present invention for recalibration correction; Figure 6 This is a schematic diagram illustrating the principle of Strategy A (global constraint strategy) in an embodiment of the present invention, demonstrating the joint optimization of multi-view information and on-site information; Figure 7 This is a schematic diagram illustrating the principle of Strategy B (single-point anchoring strategy) in an embodiment of the present invention, showing the process of calculating the relative transformation of "single-point comparison"; Figure 8 This is a comparative schematic diagram of the displacement-time curves of the monitoring points before and after recalibration correction in an embodiment of the present invention; Figure 9 This is a comparative schematic diagram of the displacement-time curves of the monitoring points after recalibration correction in an embodiment of the present invention; Figure 10 This is a comparison chart of projection errors before and after recalibration following a collision with the camera bracket in an embodiment of the present invention. Detailed Implementation
[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0041] The embodiments of this invention are written in a progressive manner. Example 1: System Architecture
[0042] As shown in Figure 1, this invention provides a rapid on-site recalibration system for a DIC system based on initial calibration self-correction. Its core lies in maintaining a unified spatial reference frame based on the initial calibration and calculating the system drift transformation matrix using a purely visual method to achieve unified data correction. The system adopts a modular architecture, consisting of the following core modules that work together to achieve a complete closed loop from initial calibration to drift correction and data traceability.
[0043] Specifically, the system includes the following functional modules: The benchmark establishment module is used to establish an initial calibration coordinate system through initial calibration and obtain the camera's intrinsic parameters and the camera's first extrinsic parameters relative to the calibration board in the initial calibration coordinate system. The input of this module includes multi-view calibration board images, and the camera's intrinsic parameter matrix K, distortion coefficients, and extrinsic parameters (rotation matrix R_ref_i and translation vector t_ref_i) corresponding to each image are obtained by solving the bundle adjustment or nonlinear optimization algorithm.
[0044] The on-site acquisition module is used to acquire at least one image of the on-site calibration board, obtain the correspondence between the two-dimensional image coordinates and the three-dimensional spatial coordinates of the feature points of the calibration board based on the on-site calibration board image, and calculate the second external parameter of the current camera coordinate system relative to the calibration board using the internal parameters. This module supports the acquisition of single or multiple on-site images, providing current pose information for subsequent drift calculations.
[0045] The transformation calculation module is used to calculate the rigid body transformation matrix for transforming the current camera coordinate system back to the initial calibration coordinate system based on the first external parameters and the second external parameters. This module supports two calculation strategies: Strategy B (single-point anchoring simplified strategy) and Strategy A (global constraint optimization strategy), allowing users to flexibly choose according to their accuracy and convenience requirements.
[0046] The data correction module is used to apply the rigid body transformation matrix to the three-dimensional coordinate data obtained by DIC measurement, and to uniformly transform the three-dimensional coordinate data to the initial calibration coordinate system. The correction formula is X_corrected=T_cam×X_original.
[0047] The storage unit is used to store the rigid body transformation matrix obtained from each correction calculation. The stored rigid body transformation matrices are concatenated in the correction sequence to form a correction transformation chain, enabling any three-dimensional coordinate data to be traced back to the initial calibration coordinate system by applying the correction transformation chain. It should be noted that the storage unit and its implemented correction transformation chain management function are supplementary enhancements to the core correction process, not essential modules of the system, and can be selectively configured according to actual needs.
[0048] The aforementioned functional modules work together to form a complete closed-loop system. After initial calibration, the benchmark establishment module stores internal parameters and the first external parameter. When system drift correction is needed, the field acquisition module acquires images of the field calibration board, the transformation calculation module calculates the rigid body transformation matrix according to the selected strategy, and the data correction module applies this matrix to complete the unified data transformation. Simultaneously, the storage unit saves the history of the transformation matrix for each correction, achieving traceable management of the data throughout its entire lifecycle.
[0049] All modules communicate through standardized interfaces to ensure the scalability and stability of the system. Example 2: General Process
[0050] like Figure 2 As shown, the method of the present invention is divided into three stages: initial calibration, normal measurement, and drift correction and data unification, and specifically includes the following steps: Step 1: Initial Calibration and Reference Information Establishment The camera is initially calibrated, an initial calibration coordinate system is established, and the camera's internal parameters and its first external parameters relative to the calibration plate in the initial calibration coordinate system are obtained.
[0051] Specifically, at the DIC system deployment site, a calibration board is used to acquire no fewer than a preset number of N (N≥15) calibration board images according to a standard multi-view high-precision calibration procedure. Based on these N images, the high-precision intrinsic parameter matrix K of the camera, distortion coefficients, and the extrinsic parameters (rotation matrix R_ref_i and translation vector t_ref_i) of the camera relative to the calibration board for each image are obtained through bundle adjustment or nonlinear optimization algorithms. The intrinsic parameters and the extrinsic parameters corresponding to the initial calibration are stored together to form the system's reference information.
[0052] Step 2: Drift Correction and Data Unification When system drift needs to be corrected, a rapid on-site recalibration process is executed, which essentially calculates the deviation of the current system state relative to the unified spatial reference frame.
[0053] Specifically, the calibration board is placed within the camera's field of view at any position and orientation, and at least one image of the calibration board is captured. Based on the image of the calibration board, the correspondence between the two-dimensional image coordinates and the three-dimensional spatial coordinates of the feature points of the calibration board is obtained, and the second external parameter of the current camera coordinate system relative to the calibration board is calculated using the internal parameters.
[0054] Based on the first external parameter and the second external parameter, calculate the rigid body transformation matrix T_cam used to transform the current camera coordinate system back to the initial calibration coordinate system. The calculation can be performed using one of the following two strategies: Strategy A (Global Constraint Optimization Strategy): This strategy jointly optimizes multiple field calibration board images with multiple initial calibration images. Under the constraint of fixed internal parameters, the objective is to minimize the overall reprojection error of the feature points of the calibration board on the multiple field calibration board images and the multiple initial calibration images. The rigid body transformation matrix is solved based on the following principle: Figure 6 As shown, this strategy forms strong geometric constraints by reusing initial multi-view information, resulting in high solution accuracy and robustness.
[0055] Strategy B (Single-point Anchoring Simplified Strategy): The current pose of a single field calibration board image is solved using the PnP algorithm, compared with the first external parameters, and the relative rigid body transformation is calculated and then inverted to obtain the rigid body transformation matrix. The principle is as follows: Figure 7 As shown, this strategy simplifies the complex on-site recalibration operation into 'single-point comparison, one-click correction', which takes as little as 1 minute and has low technical requirements.
[0056] The rigid body transformation matrix is applied to the three-dimensional coordinate data obtained through DIC measurement, and the three-dimensional coordinate data is uniformly transformed to the initial calibration coordinate system. The uniform correction formula is: X_corrected = T_cam × X_original.
[0057] Step 3: Data Traceability Management (Optional Supplementary Operation) This step is a supplementary management operation to the core processes in steps 1-2 above. It is not a necessary step, but it helps to achieve full lifecycle traceability management of data.
[0058] Specifically, the rigid body transformation matrix obtained from each correction calculation is saved, and the saved rigid body transformation matrices are concatenated in the correction sequence to form a correction transformation chain, so that the three-dimensional coordinate data can be traced back to the initial calibration coordinate system by applying the correction transformation chain.
[0059] This supplementary operation ensures that all historical and future measurement data can be traced back to the same initial unified spatial reference frame, further enhancing the manageability and auditability of long-term monitoring data.
[0060] Example 3: Recalibration application after camera collision drift The following example, which illustrates the application of the method of the present invention, is an accidental collision that occurred during long-term monitoring of the DIC system.
[0061] like Figure 3 As shown, the initial calibration was completed according to the standard DIC calibration procedure. After calibration, a reference data package was obtained, containing camera intrinsic and extrinsic parameters: intrinsic parameters include fx, fy, Cx, Cy and distortion coefficients, and extrinsic parameters include translation vectors Tx, Ty, Tz and rotation angles (α, β, γ). Subsequently, normal data acquisition and measurement, 3D reconstruction and deformation analysis were performed, and the data was output and saved.
[0062] like Figure 3 As shown, the camera mount was manually moved slightly to simulate the drift caused by the collision. The position of the object in the left camera's field of view before and after the collision shows that the actual displacement did not change, but the camera's field of view drifted due to the collision.
[0063] After the collision is complete, keep the object in its original position, such as Figure 4 As shown, the calibration board is placed in the camera's field of view at any position and orientation, and at least one image of the calibration board in the field is acquired.
[0064] according to Figure 2 The process shown and Figure 6 , Figure 7 The principle illustrated involves recalibrating the camera system while keeping its internal parameters unchanged. Specifically: Strategy A (global constraint optimization strategy) is adopted: M calibration board images captured on-site and Q reference images selected from the reference package are input into the optimization model. The optimization variables are the unified camera drift transformation matrix T_cam and the calibration board poses corresponding to the M on-site images. The optimization objective is to minimize the overall reprojection error of the calibration board feature points on all input images. After optimization convergence, the obtained T_cam is the system drift transformation matrix.
[0065] Alternatively, strategy B (single-point anchoring minimalist strategy) can be adopted: for a single field calibration board image, the pose (R_cur, t_cur) of the current calibration board relative to the camera is solved using the PnP algorithm with known camera parameters. The anchor pose is the external parameter (i.e., the first external parameter) corresponding to the calibration board image obtained through multi-view optimization in the initial calibration. The relative rigid body transformation T from the anchor pose (R_anchor, t_anchor) to the current pose is calculated, and the inverse transformation is taken to obtain T_cam.
[0066] Using the calculated transformation matrix and camera intrinsic parameters as references, the data after the collision is unified onto the original system coordinate system for subsequent measurements.
[0067] like Figure 8 As shown, the displacement cloud map and displacement changes over time before recalibration (the three-dimensional reconstruction analysis was performed using the initial calibration parameters) show that a significant displacement step can be seen during the collision, which differs greatly from the actual displacement reading of the measured object.
[0068] like Figure 9 As shown, the recalibrated and corrected displacement cloud map and the displacement change over time show that the corrected displacement did not produce a step change and is consistent with the actual displacement reading of the measured object.
[0069] like Figure 10 As shown, the projection error is large after a collision, exceeding 40%, which seriously affects DIC 3D reconstruction, and consequently, deformation analysis and continuous time-series monitoring. After recalibration correction, the projection error remains at a stable and low level.
[0070] One or more embodiments in this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of this application. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments in this application should be included within the protection scope of this application.
[0071] If flowcharts are used in this application, they are used to illustrate the operations performed by the system according to embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these flows, or one or more steps can be removed from these flows.
[0072] The foregoing provides a detailed description of a self-correcting recalibration method, apparatus, and system for a DIC system. The above description of the disclosed embodiments enables those skilled in the art to implement or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A self-correcting recalibration method for a DIC system, characterized in that, Includes the following steps: An initial calibration coordinate system is established through initial calibration, and the camera's internal parameters and the camera's first external parameters relative to the calibration plate in the initial calibration coordinate system are obtained. Acquire at least one image of the field calibration board, obtain the correspondence between the two-dimensional image coordinates and the three-dimensional spatial coordinates of the feature points of the calibration board based on the image of the field calibration board, and use the internal parameters to calculate the second external parameters of the current camera coordinate system relative to the calibration board; Based on the first external parameter and the second external parameter, calculate the rigid body transformation matrix used to transform the current camera coordinate system back to the initial calibration coordinate system; The rigid body transformation matrix is applied to the three-dimensional coordinate data obtained by DIC measurement, and the three-dimensional coordinate data is uniformly transformed to the initial calibration coordinate system.
2. The self-correcting recalibration method for a DIC system according to claim 1, characterized in that, The calibration plate is placed in any position and orientation within the camera's field of view.
3. The self-correcting recalibration method for a DIC system according to claim 1, characterized in that, Each feature point on the calibration board has a unique coded identifier.
4. The self-correcting recalibration method for a DIC system according to claim 1, characterized in that, The rigid body transformation matrix can be calculated in one of the following two ways: Method 1: Solve the current pose of a single image of the field calibration board using the PnP algorithm, compare it with the first external parameter, calculate the relative rigid body transformation, and then take the inverse to obtain the rigid body transformation matrix; Method 2: Jointly optimize multiple field calibration board images and multiple initial calibration images. Under the constraint of keeping the internal parameters unchanged, with the goal of minimizing the overall reprojection error of the feature points of the calibration board on the multiple field calibration board images and the multiple initial calibration images, solve for the rigid body transformation matrix.
5. The self-correcting recalibration method for a DIC system according to claim 1, characterized in that, The method is executed based on the assumption that the internal parameters remain stable, and triggers a manual calibration prompt when the internal parameters change beyond a preset threshold.
6. The self-correcting recalibration method for a DIC system according to claim 1, characterized in that, The three-dimensional coordinate data is transformed to the initial calibration coordinate system using the following formula: X_corrected=T_cam×X_original; Where X_corrected is the transformed 3D coordinate data, T_cam is the rigid body transformation matrix, and X_original is the original 3D coordinate data.
7. The self-correcting recalibration method for a DIC system according to claim 1, characterized in that, It also includes the following steps: saving the rigid body transformation matrix obtained from each correction calculation, and the saved rigid body transformation matrices are concatenated in the correction time sequence to form a correction transformation chain, so that the three-dimensional coordinate data can be traced back to the initial calibration coordinate system by applying the correction transformation chain.
8. A self-correcting recalibration device for a DIC system, characterized in that, include: The reference establishment module is used to establish an initial calibration coordinate system through initial calibration, and to obtain the camera's internal parameters and the camera's first external parameters relative to the calibration plate in the initial calibration coordinate system. The field acquisition module is used to acquire at least one field calibration board image, obtain the correspondence between the two-dimensional image coordinates and the three-dimensional spatial coordinates of the feature points of the calibration board based on the field calibration board image, and calculate the second external parameter of the current camera coordinate system relative to the calibration board using the internal parameters; The transformation calculation module is used to calculate, based on the first external parameter and the second external parameter, a rigid body transformation matrix for transforming the current camera coordinate system back to the initial calibration coordinate system; The data correction module is used to apply the rigid body transformation matrix to the three-dimensional coordinate data obtained by DIC measurement, and to uniformly transform the three-dimensional coordinate data to the initial calibration coordinate system.
9. The self-correcting recalibration device for a DIC system according to claim 8, characterized in that, It also includes a storage unit for storing the rigid body transformation matrix obtained from each correction calculation. The stored rigid body transformation matrices are concatenated in the correction time sequence to form a correction transformation chain.
10. A self-correcting recalibration system for a DIC system, characterized in that, include: Camera, used to acquire initial calibration board images and on-site calibration board images; A calibration plate, as a known physical reference, is used to provide feature points with known geometric structures; A storage unit is used to store the internal parameters and the first external parameters; A processing unit for performing the method of any one of claims 1 to 7, or for implementing the apparatus of claim 8 or 9.