A method for damage scan trajectory positioning and characterization of large size composite repair parts

Abstract

The application discloses a kind of large-size composite material repair piece damage scanning track positioning and characterization method, it is related to nondestructive testing technical field, comprising: in the surface of large-size composite material repair piece and the probe of damage detection equipment, paste reflective marker ball, in detection scene, multiple motion capture cameras are arranged, and the whole motion capture system is calibrated using calibration block;Hand-held probe scans damage on the surface of repair piece, and motion capture camera synchronously acquires the track of marker ball, saves damage image after completing current scanning, and system synchronously saves the track coordinate set of marker ball;Marker ball track is fitted into curved surface and is gridded, and color filling is carried out according to the damage image pixel of corresponding position, and the fusion characterization of all damage data is completed by repeating scanning;Probe scanning track curved surface fitting filled grid is spatially aligned with repair piece three-dimensional model.The application aims at realizing accurate positioning and visual characterization to large-size composite material repair piece damage.

Description

Technical Field

[0001] This invention relates to the field of nondestructive testing technology, specifically to a method for locating and characterizing damage scanning trajectories in large-size composite material repair parts. Background Technology

[0002] The use of composite materials in domestically produced large aircraft continues to increase, and they are now widely used in main load-bearing components such as wings and fuselages. Throughout their manufacturing, assembly, testing, and service life, these composite materials are susceptible to impact, fatigue, and environmental corrosion, leading to hidden damage such as delamination, debonding, and fiber breakage, which seriously threaten structural integrity and flight safety. Damage assessment of these repair components is a core aspect of ensuring airworthiness and operational safety. Accurately obtaining the spatial location, geometric distribution, and physical dimensions of the damage, and correlating this information with the structural digital model, is crucial for conducting repair quality evaluation, strength assessment, and life prediction.

[0003] Ultrasonic testing is the main method for detecting hidden damage in composite materials. Currently, automated testing is widely used in the manufacturing stage. However, due to the limited space, complex structure, and variable field environment of large-sized repair parts in service, automated equipment is difficult to deploy, and manual handheld probe testing is still the main method. Existing manual testing has three bottlenecks: (1) The scanning trajectory is free and uncontrollable, making it difficult to ensure full coverage of the repair area and resulting in a high risk of missed detection; (2) The detection images lack precise spatial coordinates, and damage localization relies on manual experience annotation, resulting in low accuracy and untraceable data; (3) The detection data is disconnected from the three-dimensional model of the structure, making it impossible to support the quantitative assessment of repair quality and structural strength analysis. Therefore, there is an urgent need for a method for damage scanning trajectory localization and fusion characterization that is adaptable to the conditions of the in-service field, takes into account both detection accuracy and operational efficiency, and is compatible with existing ultrasonic equipment.

[0004] Existing research attempts to introduce motion capture into nondestructive testing (NDT). Riise J et al., in their paper "Using motion capture systems for NDE" published at NDT 2019, proposed a six-DOF probe tracking scheme based on motion capture. However, this is only used as a substitute for position encoding. The technical approach remains at the level of mapping probe coordinates to scan indices to generate 2D C-scan images, without addressing surface fitting and meshing of the scan trajectory, and further without disclosing a fusion representation method for filling the damage image into a surface mesh according to pixel indices. Therefore, it cannot achieve the organic fusion and precise positioning of the scan trajectory, damage image, and 3D model. Patent CN114372317A, applied for by the China Aircraft Strength Research Institute, acquires damage images through photography and NDT equipment. After extracting the damage contour through image recognition and matching and calibrating it on a 2D structural diagram, the coordinates of the corner points of the rectangular damage location frame are input into a 3D model to achieve positioning. However, this method relies on post-processing photography and offline image processing, and does not achieve real-time tracking of the scanning process, nor does it fit the trajectory into a surface and mesh it to carry pixel-level mapping. Essentially, it is an offline two-dimensional image coordinate transfer, which is difficult to meet the needs of on-site inspection of large-sized repair parts for real-time damage localization and high-precision fusion characterization.

[0005] In summary, existing technologies have not yet solved the problem of accurately integrating the three elements of "scanning trajectory - damage image - structural model" in the spatial domain during manual handheld inspection. In particular, in the scenario of scanning curved surfaces of large-size composite repair parts, there is a lack of complete technical solutions for fitting the free manual scanning trajectory into a continuous curved surface, mapping the two-dimensional damage image to three-dimensional space without distortion, and accurately aligning it with the model. Summary of the Invention

[0006] To address the aforementioned shortcomings in existing technologies, this invention provides a method for locating and characterizing damage scanning trajectories in large-size composite material repair parts. The aim is to enable real-time capture of scanning trajectories during manual inspection of large-size composite material repair parts for precise damage localization. While ensuring no scanning area is missed, the method associates damage images from the inspection process with the scanning trajectory, accurately mapping the damage based on a gridded trajectory surface, thereby achieving precise damage localization and visual characterization.

[0007] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows: a method for locating and characterizing damage scanning trajectories of large-size composite material repair parts, comprising the following steps:

[0008] S1: Reflective marker balls are attached to the surface of large composite material repair parts and the probe of the damage detection equipment. Based on the size of the repair parts, their placement, and the occlusion of the reflective marker balls when the handheld probe is scanning, multiple motion capture cameras are arranged in the detection scene, and the entire motion capture system is calibrated using calibration blocks.

[0009] S2: The handheld probe scans the surface of the repair part for damage. The motion capture camera simultaneously captures the trajectory of the reflective marker ball on the probe. After the current scan is completed, the detection equipment saves the damage image, and the motion capture system simultaneously saves the trajectory coordinate set of the reflective marker ball on the probe.

[0010] S3: Fit the trajectory of the probe's reflective marker ball into a curved surface and mesh it. Fill the corresponding damage image pixels with color and repeat the scan to complete the fusion characterization of all damage data.

[0011] S4: Obtain the 3D model of the repair part. Based on the position coordinates of the reflective marking ball on the surface of the repair part, align the mesh after fitting and filling the probe scanning trajectory surface with the 3D model in space to achieve the fusion representation and positioning of the scanning trajectory, damage image and 3D model.

[0012] Further, step S1 includes the following sub-steps:

[0013] S11: Arrange three or more reflective marker balls on the surface of large-size composite material repair parts. The reflective marker balls are placed at structural feature points and the spacing between them is maximized. Arrange two reflective marker balls on both sides of the probe scanning direction, and the distance between them is not greater than the probe scanning width.

[0014] S12: In the inspection scenario of large-size composite material repair parts, multiple motion capture cameras are deployed. The motion capture cameras emit and receive infrared light. The camera deployment covers the largest repair part to be inspected, and ensures that the probe marker ball is not excessively obstructed throughout the scanning process.

[0015] S13: Establish the coordinate system of the motion capture system through the calibration block and complete the calibration. Set the marker balls on both sides of the probe as rigid bodies and the marker balls on the surface of the repair part as model reference points. Obtain the real-time spatial coordinates of the reflective marker balls and the probe trajectory.

[0016] Furthermore, in S2, when the handheld probe scans the surface of the repair part for damage, the motion capture system records the time, speed and attitude of the corresponding position while locating the coordinates of the reflective marker spherical coordinates, and calculates the speed, speed difference and attitude angle change rate on both sides of the probe in real time. When the speed, speed difference and attitude angle change rate of the probe movement exceed the preset quality threshold, the motion capture system issues a warning to the operator through audio-visual signals, including reducing the scanning speed, maintaining a straight scanning line and stabilizing the probe attitude.

[0017] Furthermore, step S3 includes the following sub-steps:

[0018] S31: Connect the centers of the reflective marker balls at the start and end of the scan to form two line segments. Fit the two scan trajectories of the two reflective marker balls into curves. Use the two line segments and two curves to perform surface fitting on the area scanned by the probe.

[0019] S32: Determine the effective scanning width based on the distance between the reflective marker balls on both sides of the probe, crop the edge of the damaged image to obtain the effective damaged image, and parameterize and mesh the surface according to the image resolution;

[0020] S33: Index the effective damaged image by row and column resolution, discretize the parameterized curve to generate quadrilateral grid patches, use the image pixel index as the quadrilateral grid patch identifier, establish a one-to-one correspondence between image pixels and grid patches, and extract the RGB color values ​​of image pixels to fill the corresponding grid patches.

[0021] S34: Repeat steps S31-S33 to complete the scanning and image mapping segment by segment, and realize the fusion of damage data in the whole area;

[0022] S35: When overlapping trajectories are found during scanning, the subsequent scanning surface is given a higher display priority based on the timestamp.

[0023] Furthermore, S31 also includes: after downsampling the coordinate point set of the scanning trajectory of the two reflective marker balls, the coordinate point set is fitted into a curve using the cubic spline interpolation method, the two fitted spline curves are used as trajectory boundaries, the two line segments are used as the starting boundaries of the surface, the four boundaries are connected end to end to form a closed curve, and the surface is fitted using the Königs surface method.

[0024] Furthermore, step S4 includes the following sub-steps:

[0025] S41: Obtain a 3D model of a large-size composite material repair part. The 3D model of the repair part can be a design model or a 3D scanned point cloud model.

[0026] S42: Using the three reflective marker balls on the surface of the repair part as a reference, the three-dimensional model of the repair part is aligned with the mesh after fitting and filling the probe scanning trajectory surface to achieve accurate fusion and representation of the damage image, scanning surface and three-dimensional model.

[0027] S43: Save the fused damage image, scanned surface, and 3D model data to achieve data control and visualization representation of the damage detection process.

[0028] Furthermore, S43 displays the location and distribution of damage based on the fused damage image, scanned surface, and 3D model data. At the same time, it classifies the severity according to the damage area and depth, and automatically generates a 3D interactive inspection report containing 3D location, size, grade, and heat map.

[0029] The beneficial effects of this invention are:

[0030] (1) Spatial digitization and coverage verification of manual scanning trajectory were realized. The probe pose was recorded in real time by the motion capture system and the scanning surface was generated. Combined with the three-dimensional model, the integrity of the scanning area of ​​the damage was assessed, and the missed areas were eliminated from the technical mechanism.

[0031] (2) A full data representation and management system for the detection process was constructed, realizing the fusion storage of scanning trajectory, damage image and three-dimensional model, supporting structure-based damage query, historical data tracing and repair quality visualization assessment, and meeting the requirements of full life cycle data management of aerospace structures.

[0032] (3) Real-time evaluation and optimization control of scanning quality is realized. By monitoring the scanning speed and attitude angle change rate of the probe in real time, dynamic quality evaluation of the scanning process is carried out. When the scanning speed is too fast or the attitude changes suddenly, an early warning is issued in time, which effectively avoids image stretching distortion and poor coupling caused by excessive scanning speed, and improves the reliability and repeatability of detection data.

[0033] (4) A damage severity grading and three-dimensional interactive report generation mechanism was established. Based on the fused three-dimensional model and the results of quantitative damage analysis, the damage severity was graded in multiple dimensions, and a three-dimensional interactive inspection report containing the damage location, size, grade and spatial distribution was automatically generated, providing intuitive, accurate and complete data support for strength assessment of repair parts, service decision and repair scheme optimization. Attached Figure Description

[0034] Figure 1 This is a flowchart of a method for locating and characterizing damage scanning trajectories in large-sized composite material repair parts.

[0035] Figure 2 This is a schematic diagram of the damage scanning trajectory location for large-size composite material repair parts.

[0036] Figure 3 This is a schematic diagram of image mapping and three-dimensional characterization of damage to composite repair parts. Detailed Implementation

[0037] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0038] like Figure 1 As shown, a method for locating and characterizing damage scanning trajectories in large-size composite material repair parts includes the following steps:

[0039] S1: Reflective marker balls are attached to the surface of large composite material repair parts and the probe of the damage detection equipment. Based on the size of the repair parts, their placement, and the occlusion of the reflective marker balls when the handheld probe is scanning, multiple motion capture cameras are arranged in the detection scene, and the entire motion capture system is calibrated using calibration blocks.

[0040] In an optional embodiment of the present invention, a large-size composite material wall panel repair part is used as the detection object. Multiple motion capture cameras are arranged in the detection scene. Reflective marker balls are pasted on the surface of the repair part. The motion capture system is calibrated and optimized using the calibration block built into the motion capture system.

[0041] S1 includes the following steps:

[0042] S11: Arrange three or more reflective marker balls on the surface of large-size composite material repair parts. The reflective marker balls are placed at structural feature points and the spacing between them is maximized. Arrange two reflective marker balls on both sides of the probe scanning direction, and the distance between them is not greater than the probe scanning width.

[0043] The reflective marker spheres are coated with reflective material, and the spheres should be positioned as far apart as possible. They should ideally be located at structural features on the surface of the repaired part, such as corners or sharp points. Different diameter marker spheres can be selected based on the size of the object being inspected and the inspection scene to ensure better capture by the motion capture camera.

[0044] Specifically, this invention selects three corner points of a large-size composite material wall panel as feature points, such as... Figure 2 As shown. Select an 8mm diameter reflective marker ball on the probe and a 12mm diameter reflective marker ball on the composite material panel.

[0045] The Euclidean distance between the marker balls on both sides of the probe should not exceed the probe scanning width. The Euclidean distance is defined as the effective scanning area of ​​the probe, in order to avoid missed areas and eliminate unstable signals at the edges.

[0046] S12: In the inspection scenario of large-size composite material repair parts, multiple motion capture cameras are deployed. The motion capture cameras emit and receive infrared light. The camera deployment covers the largest repair part to be inspected, and ensures that the probe marker ball is not excessively obstructed throughout the scanning process.

[0047] Based on the principles of motion capture camera placement, as many motion capture cameras as possible should be placed on one side of the probe's scanning direction to ensure that the motion capture cameras are not excessively obstructed during manual scanning. Simultaneously, the motion capture cameras should be placed on the side of the repair part that needs to be inspected, so that the placement requirements can still be met when facing different large-sized repair parts.

[0048] The manual handheld probe is used to scan blind spots on the surface of large composite materials. During the scanning process of the entire repair part, all kinds of obstructions must be removed. It is necessary to ensure that the two reflective marker balls of the probe in the forward direction can be captured by two or more motion capture cameras at the same time. If they cannot be captured, motion capture cameras need to be added at the corresponding positions.

[0049] Specifically, this invention targets large-size composite material wall panels, arranging eight motion capture cameras above them to cover various scanning directions, such as... Figure 2 As shown.

[0050] S13: Establish the coordinate system of the motion capture system through the calibration block and complete the calibration. Set the marker balls on both sides of the probe as rigid bodies and the marker balls on the surface of the repair part as model reference points. Obtain the real-time spatial coordinates of the reflective marker balls and the probe trajectory.

[0051] The motion capture system is calibrated using a calibration block. The location of the calibration block is the origin and direction of the entire motion capture system's coordinates, denoted as . After calibration, all reflective marker balls can be obtained. The coordinates of the center of the sphere in this coordinate system.

[0052] Based on the origin The spatial coordinates of the centers of the three reflective marker spheres on the surface of the large composite material repair part are: , and The coordinates of the centers of the two reflective marker spheres on either side of the probe are: , .

[0053] When the position of the motion capture camera in the detection scene does not change, the entire system only needs to be calibrated once. Subsequent testing after replacing or repairing parts does not require recalibration. The coordinates of the center of the reflective marker ball and the scanning trajectory of the probe can be obtained directly.

[0054] S2: The handheld probe scans the surface of the repair part for damage. The motion capture camera simultaneously captures the trajectory of the reflective marker ball on the probe. After the current scan is completed, the detection equipment saves the damage image, and the motion capture system simultaneously saves the trajectory coordinate set of the reflective marker ball on the probe.

[0055] When using a handheld probe to scan the surface of a repair part for damage, the probe must remain still for at least 2 seconds at the start and end of the scan to allow the motion capture system to start and stop collecting data synchronously. The collected trajectory data does not include any extra data from the probe other than what is being detected, thus enabling the synchronous acquisition and storage of the damage image and the trajectory of the reflective marker ball.

[0056] When a handheld probe is used to scan for damage on the surface of a repair part, the motion capture system records the time at the corresponding position while locating the coordinates of the reflective marker sphere. ,speed and posture And calculate the velocity on both sides of the probe in real time. Speed ​​difference and attitude angle change rate , The motion capture system is responsible for the change in attitude angle over a certain period of time. When the speed, speed difference, and attitude angle change rate of the probe movement exceed the preset quality threshold, the system will issue an early warning to the operator through audio-visual signals. This warning includes reducing the scanning speed, maintaining a straight-line scanning, and stabilizing the probe attitude to ensure the quality of damage image acquisition and the accuracy of scanning trajectory data.

[0057] When the resolution of the acquired effective damaged image is too high or too low, the image is downsampled or upsampled to ensure a better display effect. In this embodiment, the image width is selected to be greater than 400 pixels. Then, the fitted surface is parameterized and meshed according to the sampled resolution.

[0058] S3: Fit the trajectory of the probe's reflective marker ball into a curved surface and mesh it. Fill the corresponding damage image pixels with color and repeat the scan to complete the fusion characterization of all damage data.

[0059] S3 includes the following steps:

[0060] S31: Connect the centers of the reflective marker balls at the start and end of the scan to form two line segments. Fit the two scan trajectories of the two reflective marker balls into curves. Use the two line segments and two curves to perform surface fitting on the area scanned by the probe.

[0061] Coordinates of the center of the probe's reflective marker sphere and Two tracks will be created during the scanning process, namely and Two sets of coordinate points, with the acquisition time of each point synchronized, are used to fit the coordinate point sets into a curve after downsampling based on the number of points in the set.

[0062] Two straight line segments are used as the initial boundaries of the surface, and two fitted spline curves are used as the trajectory boundaries. The four boundaries are joined end-to-end to form a closed curve, and the surface is fitted using the Coons surface method. The surface is then mapped to a two-dimensional parameter domain, with the scanning direction as the reference point. Axis parameter direction, value range ,in Corresponding to the starting position of the scan, Corresponding to the end position of the scan; with the probe width direction perpendicular to the scan direction as... Axis parameter direction, value range ,in Corresponding to the first trajectory curve, Corresponding to the second trajectory curve. The three-dimensional coordinates of the marker sphere trajectory are established through this parameterization. Two-dimensional parametric coordinates of the surface The mapping relationship, that is, any point on the surface can be represented as This provides a unified parameter space benchmark for subsequent mesh generation and image mapping.

[0063] Damage images obtained after probe scanning can be exported in real time via the device's network port / USB, facilitating subsequent fusion with motion capture trajectory data.

[0064] S32: Determine the effective scanning width based on the distance between the reflective marker balls on both sides of the probe, crop the edge of the damaged image to obtain the effective damaged image, and parameterize and mesh the surface according to the image resolution;

[0065] Based on the defined effective scanning area of ​​the probe, the coordinates of the center of the probe's reflective sphere are... and The Euclidean distance between them should be less than the width of a single probe scan. This is to ensure image quality. (The image is taken here.) Width of the effective scanning area of ​​the probe The specific formula is as follows:

[0066]

[0067] Secondly, the obtained damaged image is cropped and deleted proportionally on both sides in the width direction. To obtain the middle The width of the image is taken as the effective damage image, and the resolution W×L (width×length) of the image is calculated. The width resolution W of the effective damage image corresponds to the straight line direction of the surface fitting in step S31, and the length resolution L of the effective damage image corresponds to the curve direction of the surface fitting in step S31.

[0068] S33: Index the effective damaged image by row and column resolution, discretize the parameterized curve to generate quadrilateral grid patches, use the image pixel index as the quadrilateral grid patch identifier, establish a one-to-one correspondence between image pixels and grid patches, and extract the RGB color values ​​of image pixels to fill the corresponding grid patches.

[0069] The effective damage image obtained in step S32 is indexed by pixel according to the row and column resolution W×L, where The row index corresponds to the probe width direction. The column index corresponds to the scan direction; the parameterized surface is in The direction is discretized into W equal parts. The orientation is discretized into L equal parts, generating W×L quadrilateral mesh patches. Indexed by image pixels. As a grid patch identifier, making the first Line 1 The grid patch of the row and the image number Line 1 The columns of pixels form a one-to-one correspondence; the coverage area of ​​the grid patch in the parameter domain is... , This parameter region is the mapping range of pixels on the curved surface, used to extract image pixels. The RGB color values ​​are filled into the corresponding mesh patches, thereby achieving distortion-free patch-by-patch color filling and ensuring that the texture color of the curved surface remains consistent with the original damaged image, such as... Figure 3 As shown.

[0070] S34: Repeat steps S31-S33 to complete the scanning and image mapping segment by segment, and realize the fusion of damage data in the whole area; the colored surface is in the same coordinate system, and the damage image corresponding to each scanned surface is fused, which completely records the entire damage detection process and damage distribution.

[0071] S35: When overlapping trajectories are detected during scanning, the subsequent scanned surface is assigned a higher display priority based on the timestamp. During visualization, only the surface with the higher priority is displayed in the overlapping area, while the surface with the lower priority is occluded.

[0072] S4: Obtain the 3D model of the repair part. Based on the position coordinates of the reflective marking ball on the surface of the repair part, align the mesh after fitting and filling the probe scanning trajectory surface with the 3D model in space to achieve the fusion representation and positioning of the scanning trajectory, damage image and 3D model.

[0073] S4 includes the following steps:

[0074] S41: Obtain a 3D model of a large-size composite material repair part. The 3D model of the repair part can be a design model or a 3D scanned point cloud model. The model must include the area of ​​the three reflective marker spheres on the surface of the large-size composite material repair part.

[0075] When selecting a point cloud model, the corresponding two-dimensional image of the point cloud should be acquired simultaneously, and the image pixel color should be assigned to the point cloud based on the spatial position mapping to achieve the fusion coloring of the image and the point cloud, thereby reflecting the structural information of the repair part. Then, the colored surface and the point cloud model are aligned by the position of the marker sphere to provide position information for strength assessment and repair evaluation.

[0076] S42: Using the three reflective marker spheres on the surface of the repair part as a reference, the 3D model of the repair part is 3D aligned with the mesh after fitting and filling the probe scanning trajectory surface, achieving accurate fusion and representation of the damage image, scanning surface, and 3D model, such as... Figure 3 As shown;

[0077] S43: The fused damage image, scanned surface, and 3D model data are saved to achieve data control and visualization of the damage detection process. The fused model accurately displays the location and distribution of damage, and the point cloud / model can be used to measure the location and size of damage, with the measurement results accurately reflecting the true size of the damage.

[0078] Based on the fused damage images, scanned surfaces, and 3D model data, S43 displays the location and distribution of damage. It also classifies the severity according to the damage area and depth, and automatically generates a 3D interactive inspection report containing 3D location, size, grade, and heat map. The report also supports 3D browsing, slice analysis, and data export, providing intuitive and comprehensive data support for strength assessment of repair parts, service decisions, and repair scheme optimization.

[0079] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of the invention.

Claims

1. A method for locating and characterizing damage scanning trajectories in large-size composite material repair parts, characterized in that, Includes the following steps: S1: Reflective marker balls are attached to the surface of large composite material repair parts and the probe of the damage detection equipment. Based on the size of the repair parts, their placement, and the occlusion of the reflective marker balls when the handheld probe is scanning, multiple motion capture cameras are arranged in the detection scene, and the entire motion capture system is calibrated using calibration blocks. S2: The handheld probe scans the surface of the repair part for damage. The motion capture camera simultaneously captures the trajectory of the reflective marker ball on the probe. After the current scan is completed, the detection equipment saves the damage image, and the motion capture system simultaneously saves the trajectory coordinate set of the reflective marker ball on the probe. S3: Fit the trajectory of the probe's reflective marker ball into a curved surface and mesh it. Fill the corresponding damage image pixels with color and repeat the scan to complete the fusion characterization of all damage data. S4: Obtain the 3D model of the repair part. Based on the position coordinates of the reflective marking ball on the surface of the repair part, align the mesh after fitting and filling the probe scanning trajectory surface with the 3D model in space to achieve the fusion representation and positioning of the scanning trajectory, damage image and 3D model.

2. The method for locating and characterizing damage scanning trajectories of large-size composite material repair parts according to claim 1, characterized in that, S1 includes the following steps: S11: Arrange three or more reflective marker balls on the surface of large-size composite material repair parts. The reflective marker balls are placed at structural feature points and the spacing between them is maximized. Arrange two reflective marker balls on both sides of the probe scanning direction, and the distance between them is not greater than the probe scanning width. S12: In the inspection scenario of large-size composite material repair parts, multiple motion capture cameras are deployed. The motion capture cameras emit and receive infrared light. The camera deployment covers the largest repair part to be inspected, and ensures that the probe marker ball is not excessively obstructed throughout the scanning process. S13: Establish the coordinate system of the motion capture system through the calibration block and complete the calibration. Set the marker balls on both sides of the probe as rigid bodies and the marker balls on the surface of the repair part as model reference points. Obtain the real-time spatial coordinates of the reflective marker balls and the probe trajectory.

3. The method for locating and characterizing damage scanning trajectories of large-size composite material repair parts according to claim 1, characterized in that, When the handheld probe in S2 scans the surface of the repair part for damage, the motion capture system records the time, speed and attitude of the corresponding position while locating the coordinates of the reflective marker spherical coordinates. It also calculates the speed, speed difference and attitude angle change rate on both sides of the probe in real time. When the speed, speed difference and attitude angle change rate of the probe movement exceed the preset quality threshold, the motion capture system issues a warning to the operator through audio-visual signals, including reducing the scanning speed, maintaining a straight scanning line and stabilizing the probe attitude.

4. The method for locating and characterizing damage scanning trajectories of large-size composite material repair parts according to claim 1, characterized in that, S3 includes the following steps: S31: Connect the centers of the reflective marker balls at the start and end of the scan to form two line segments. Fit the two scan trajectories of the two reflective marker balls into curves. Use the two line segments and two curves to perform surface fitting on the area scanned by the probe. S32: Determine the effective scanning width based on the distance between the reflective marker balls on both sides of the probe, crop the edge of the damaged image to obtain the effective damaged image, and parameterize and mesh the surface according to the image resolution; S33: Index the effective damaged image by row and column resolution, discretize the parameterized curve to generate quadrilateral grid patches, use the image pixel index as the quadrilateral grid patch identifier, establish a one-to-one correspondence between image pixels and grid patches, and extract the RGB color values ​​of image pixels to fill the corresponding grid patches. S34: Repeat steps S31-S33 to complete the scanning and image mapping segment by segment, and realize the fusion of damage data in the whole area; S35: When overlapping trajectories are found during scanning, the subsequent scanning surface is given a higher display priority based on the timestamp.

5. The method for locating and characterizing damage scanning trajectories of large-size composite material repair parts according to claim 4, characterized in that, S31 further includes: after downsampling the coordinate point set of the scanning trajectory of the two reflective marker balls, the coordinate point set is fitted into a curve using the cubic spline interpolation method, the two fitted spline curves are used as trajectory boundaries, the two line segments are used as the starting boundaries of the surface, the four boundaries are connected end to end to form a closed curve, and the surface is fitted using the Coons surface method.

6. The method for locating and characterizing damage scanning trajectories of large-size composite material repair parts according to claim 1, characterized in that, S4 includes the following sub-steps: S41: Obtain a 3D model of a large-size composite material repair part. The 3D model of the repair part can be a design model or a 3D scanned point cloud model. S42: Using the three reflective marker balls on the surface of the repair part as a reference, the three-dimensional model of the repair part is aligned with the mesh after fitting and filling the probe scanning trajectory surface to achieve accurate fusion and representation of the damage image, scanning surface and three-dimensional model. S43: Save the fused damage image, scanned surface, and 3D model data to achieve data control and visualization representation of the damage detection process.

7. The method for locating and characterizing damage scanning trajectories of large-size composite material repair parts according to claim 6, characterized in that, Based on the fused damage image, scanned surface, and 3D model data, S43 displays the location and distribution of damage. It also classifies the severity according to the damage area and depth, and automatically generates a 3D interactive inspection report containing 3D location, size, grade, and heat map.

Images