Part deformation determination method and device, computer equipment and storage medium

Abstract

The invention relates to a component deformation determination method and device, computer equipment and a storage medium. Comprising the steps of obtaining point cloud data of a part and image data, collected by at least two image collection devices, of the part; aiming at each piece of image data, performing parallax correction on the image data according to the point cloud data, the first collinearity equation and external parameters of image acquisition equipment for acquiring the image data to obtain a correction graph corresponding to the image data; performing pixel matching on the correction images corresponding to different image data to obtain a correction pixel coordinate group corresponding to each position point of the component; therefore, parallax correction of the image data acquired by different acquisition devices is realized, the actual three-dimensional coordinate of each position point is determined according to each correction pixel coordinate group, the determined actual three-dimensional coordinate is more accurate, and according to the difference value between the actual three-dimensional coordinate of each position point and the initial three-dimensional coordinate, the actual three-dimensional coordinate of each position point can be corrected according to the difference value between the actual three-dimensional coordinate of each position point and the initial three-dimensional coordinate. And the deformation occurrence condition of the component is determined, so that the determined deformation occurrence condition is more reliable.

Description

Technical Field

[0001] This application relates to the field of visual measurement technology, and in particular to a method, apparatus, computer device, and storage medium for determining component deformation. Background Technology

[0002] Substations, as crucial nodes in power transmission and distribution systems, house various high-voltage electrical equipment and supporting structural components. However, these components are constantly subjected to various stresses, including wind loads, temperature variations, and electromagnetic forces, making them prone to minor deformations such as component tilting, loose connections, and weld cracks. In severe cases, these deformations can lead to electrical faults or even safety accidents. Therefore, deformation measurement of these components allows for timely intervention upon detection of deformation, preventing electrical faults and ensuring electrical safety.

[0003] Traditional techniques typically use theodolites or total stations to measure the deformation of structural components, which suffers from low measurement accuracy. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, computer equipment, and storage medium for determining component deformation to address the aforementioned technical problems, thereby improving the accuracy of component deformation measurement.

[0005] In a first aspect, this application provides a method for determining component deformation, including:

[0006] Acquire point cloud data of the component, as well as image data of the component acquired by at least two image acquisition devices;

[0007] For each image data, based on the point cloud data, the first collinearity equation, and the external parameters of the image acquisition device that acquired the image data, parallax correction is performed on the image data to obtain a corrected image corresponding to the image data; wherein, the first collinearity equation is used to describe the correspondence between the three-dimensional coordinates of the component, the external parameters of the image acquisition device, and the pixel coordinates of the component;

[0008] Pixel matching is performed on the correction images corresponding to different image data to obtain the correction pixel coordinate group corresponding to each position point of the component; wherein, the correction pixel coordinate group corresponding to each position point includes the pixel coordinates of the position point in each image data;

[0009] Based on each set of corrected pixel coordinates, the actual three-dimensional coordinates of each position point are determined, and the deformation of the component is determined based on the difference between the actual three-dimensional coordinates and the initial three-dimensional coordinates of each position point.

[0010] In one embodiment, the step of performing parallax correction on the image data based on the point cloud data, the first collinearity equation, and the external parameters of the image acquisition device that acquires the image data, to obtain a corrected image corresponding to the image data, includes:

[0011] Feature extraction is performed on the image data to obtain the image geometric features;

[0012] Based on the point cloud geometric features and the image geometric features, a correspondence is established between the pixel coordinates of each pixel in the image data and the vertical coordinates of the point cloud data; wherein, the point cloud geometric features are obtained by feature extraction from the point cloud data;

[0013] For each pixel in the image data, the theoretical three-dimensional coordinates of the pixel are determined based on the first collinearity equation, the external parameters of the image acquisition device that acquires the image data, and the vertical coordinates corresponding to the pixel.

[0014] The theoretical three-dimensional coordinates of each pixel are projected onto a preset plane to obtain a corrected image of the image data.

[0015] In one embodiment, the step of performing pixel matching on the correction maps corresponding to different image data to obtain a set of corrected pixel coordinates corresponding to each location point of the component includes:

[0016] Pixel matching is performed on the correction images corresponding to different image data to obtain the matching pixel coordinate group corresponding to each position point of the component; wherein, the matching pixel coordinate group corresponding to each position point includes the pixel coordinates of the position point in each correction image;

[0017] Based on the theoretical three-dimensional coordinates corresponding to the matched pixel coordinate group, the external parameters of the image acquisition device, and the first collinearity equation, the corrected pixel coordinate group corresponding to each position point of the component is obtained.

[0018] In one embodiment, each corrected pixel coordinate group determines the actual three-dimensional coordinates of each location point, including:

[0019] Based on each set of corrected pixel coordinates, determine the orientation vector of each image acquisition device;

[0020] The ray equations of each image acquisition device are determined based on the internal parameters and direction vectors of each image acquisition device.

[0021] Solve the determined ray equation to obtain the actual three-dimensional coordinates of each position point.

[0022] In one embodiment, the image data includes a preset number of reference points, and the external parameters of each image acquisition device are determined in the following way:

[0023] Obtain a template image with the same shooting conditions as the image data acquired by the image acquisition device; wherein, the shooting conditions include at least the shooting angle and the image size;

[0024] Pixel matching is performed on the image data and the template image to obtain the reference pixel coordinates corresponding to each reference point;

[0025] The external parameters of the image acquisition device are determined based on the second collinearity equation, the reference pixel coordinates and actual three-dimensional coordinates corresponding to each reference point, and the internal parameters and distortion parameters of the image acquisition device; wherein, the second collinearity equation is used to describe the relationship between the pixel coordinates, three-dimensional coordinates of the reference point, the internal parameters, distortion parameters and external parameters of the image acquisition device.

[0026] In one embodiment, the step of performing pixel matching between the image data and the template image to obtain the reference pixel coordinates corresponding to each reference point includes:

[0027] Based on the image data and the template image, determine the displacement deviation of each reference point;

[0028] Based on the initial pixel coordinates of each reference point and the displacement deviation of the corresponding reference point, the reference pixel coordinates of each reference point in the image data are determined, wherein the initial pixel coordinates of each reference point are the pixel coordinates of each reference point in the template image.

[0029] Secondly, this application also provides a component deformation determining device, comprising:

[0030] An acquisition module is used to acquire point cloud data of the component, as well as image data of the component acquired by at least two image acquisition devices;

[0031] The correction module is used to perform parallax correction on each image data according to the point cloud data, the first collinearity equation, and the external parameters of the image acquisition device that acquires the image data, to obtain a corrected image corresponding to the image data; wherein, the first collinearity equation is used to describe the correspondence between the three-dimensional coordinates of the component, the external parameters of the image acquisition device, and the pixel coordinates of the component;

[0032] The matching module is used to perform pixel matching on the correction images corresponding to different image data to obtain the correction pixel coordinate group corresponding to each position point of the component; wherein, the correction pixel coordinate group corresponding to each position point includes the pixel coordinates of the position point in each image data.

[0033] The determination module is used to determine the actual three-dimensional coordinates of each position point according to each set of corrected pixel coordinates, and to determine the deformation of the component according to the difference between the actual three-dimensional coordinates and the initial three-dimensional coordinates of each position point.

[0034] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0035] Acquire point cloud data of the component, as well as image data of the component acquired by at least two image acquisition devices;

[0036] For each image data, based on the point cloud data, the first collinearity equation, and the external parameters of the image acquisition device that acquired the image data, parallax correction is performed on the image data to obtain a corrected image corresponding to the image data; wherein, the first collinearity equation is used to describe the correspondence between the three-dimensional coordinates of the component, the external parameters of the image acquisition device, and the pixel coordinates of the component;

[0037] Pixel matching is performed on the correction images corresponding to different image data to obtain the correction pixel coordinate group corresponding to each position point of the component; wherein, the correction pixel coordinate group corresponding to each position point includes the pixel coordinates of the position point in each image data;

[0038] Based on each set of corrected pixel coordinates, the actual three-dimensional coordinates of each position point are determined, and the deformation of the component is determined based on the difference between the actual three-dimensional coordinates and the initial three-dimensional coordinates of each position point.

[0039] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0040] Acquire point cloud data of the component, as well as image data of the component acquired by at least two image acquisition devices;

[0041] For each image data, based on the point cloud data, the first collinearity equation, and the external parameters of the image acquisition device that acquired the image data, parallax correction is performed on the image data to obtain a corrected image corresponding to the image data; wherein, the first collinearity equation is used to describe the correspondence between the three-dimensional coordinates of the component, the external parameters of the image acquisition device, and the pixel coordinates of the component;

[0042] Pixel matching is performed on the correction images corresponding to different image data to obtain the correction pixel coordinate group corresponding to each position point of the component; wherein, the correction pixel coordinate group corresponding to each position point includes the pixel coordinates of the position point in each image data;

[0043] Based on each set of corrected pixel coordinates, the actual three-dimensional coordinates of each position point are determined, and the deformation of the component is determined based on the difference between the actual three-dimensional coordinates and the initial three-dimensional coordinates of each position point.

[0044] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0045] Acquire point cloud data of the component, as well as image data of the component acquired by at least two image acquisition devices;

[0046] For each image data, based on the point cloud data, the first collinearity equation, and the external parameters of the image acquisition device that acquired the image data, parallax correction is performed on the image data to obtain a corrected image corresponding to the image data; wherein, the first collinearity equation is used to describe the correspondence between the three-dimensional coordinates of the component, the external parameters of the image acquisition device, and the pixel coordinates of the component;

[0047] Pixel matching is performed on the correction images corresponding to different image data to obtain the correction pixel coordinate group corresponding to each position point of the component; wherein, the correction pixel coordinate group corresponding to each position point includes the pixel coordinates of the position point in each image data;

[0048] Based on each set of corrected pixel coordinates, the actual three-dimensional coordinates of each position point are determined, and the deformation of the component is determined based on the difference between the actual three-dimensional coordinates and the initial three-dimensional coordinates of each position point.

[0049] The aforementioned component deformation determination method, apparatus, computer equipment, and storage medium acquire point cloud data of the component and image data of the component acquired by at least two image acquisition devices. For each image data, disparity correction is performed on the image data based on the point cloud data, a first collinearity equation, and the external parameters of the image acquisition devices, resulting in a corrected image corresponding to the image data. The first collinearity equation describes the correspondence between the component's three-dimensional coordinates, the external parameters of the image acquisition devices, and the component's pixel coordinates. Pixel matching is performed on the corrected images corresponding to different image data to obtain a corrected pixel coordinate set corresponding to each position point of the component. Each corrected pixel coordinate set includes the pixel coordinates of the position point in each image data. This achieves disparity correction of image data acquired by different acquisition devices, and further determines the actual three-dimensional coordinates of each position point based on each corrected pixel coordinate set, making the determined actual three-dimensional coordinates more accurate. Furthermore, the component deformation is determined based on the difference between the actual three-dimensional coordinates and the initial three-dimensional coordinates of each position point, making the determined deformation more reliable. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is an application environment diagram of the component deformation determination method in one embodiment;

[0052] Figure 2 This is a flowchart illustrating a method for determining component deformation in one embodiment;

[0053] Figure 3 This is a flowchart illustrating the process of obtaining the corrected image corresponding to the image data in one embodiment;

[0054] Figure 4 This is a flowchart illustrating the process of obtaining the corrected pixel coordinate group corresponding to each position point of the component in one embodiment;

[0055] Figure 5 This is a flowchart illustrating the process of determining the actual three-dimensional coordinates of each location point in one embodiment.

[0056] Figure 6 This is a flowchart illustrating the process of determining the external parameters of an image acquisition device in one embodiment;

[0057] Figure 7 This is a structural block diagram of a component deformation determining device in one embodiment;

[0058] Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0060] The component deformation determination method provided in this application embodiment can be applied to, for example, Figure 1In the application environment shown, at least two image acquisition devices 101 are used to simultaneously acquire image data of various components in the substation 102 and transmit the acquired image data to a terminal, so that the terminal can use the parallax-corrected image data for deformation detection. The image acquisition device 101 can be a camera, and the terminal can be various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can be smart speakers, smart TVs, smart air conditioners, smart vehicle devices, projection devices, etc. Portable wearable devices can be smartwatches, smart bracelets, head-mounted devices, etc. Head-mounted devices can be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc.

[0061] In one exemplary embodiment, such as Figure 2 As shown, a method for determining component deformation is provided. Taking the application of this method to a terminal as an example, the method includes the following steps:

[0062] S201, acquire point cloud data of the component, as well as image data of the component acquired by at least two image acquisition devices.

[0063] For example, point cloud data of a component can be acquired using a LiDAR acquisition device. Image data of the component can also be acquired using camera image acquisition devices positioned at different locations.

[0064] Before using image acquisition devices to acquire image data of components, it is necessary to determine the location of each image acquisition device. Assuming that two image acquisition devices are used to acquire image data of components, the following describes how to determine the location of each image acquisition device.

[0065] In the observation area of ​​the substation components, two fixed cameras are installed to form a stereo photogrammetry system, creating a close-range binocular observation baseline. The system deployment design should fully consider the following factors to achieve the optimal balance between measurement accuracy and monitoring coverage:

[0066] Baseline length design: can be based on the formula , where δ z For the desired accuracy, z is the distance to the target object; f is the camera focal length; B is the baseline length; δ d This represents the parallax measurement error.

[0067] For example, when the distance z = 10 meters and the desired accuracy δ z When z is less than 2 mm, the calculated baseline length must be no less than 2 meters; when z = 20 meters, the baseline must be no less than 40 meters.

[0068] Camera installation specifications:

[0069] The camera is parallel to the front face of the target component to reduce the interference of projection distortion on subsequent matching; the height is slightly higher than the center of the component to ensure complete imaging of the upper and lower edges of the component; the line-of-sight intersection angle is less than 20° to avoid insufficient overlap or parallax during stereo matching; a rigid baseline and a rotating platform are used to fix the dual cameras through a rigid structure, and the rotating platform enables multi-angle and large-area measurements to cover more components in the substation.

[0070] S202, for each image data, based on the point cloud data, the first collinearity equation, and the external parameters of the image acquisition device that acquired the image data, parallax correction is performed on the image data to obtain the corrected image data.

[0071] For example, since the image acquisition devices are arranged in different locations, the acquired image data may have visual biases. For each frame of image data, parallax correction can be performed on the image data based on point cloud data, the first collinearity equation, and the external parameters of the image acquisition device that acquired the image data, to obtain the corrected image data.

[0072] The first collinearity equation describes the correspondence between the three-dimensional coordinates of the component, the external parameters of the image acquisition device, and the pixel coordinates of the component. The expression for the first collinearity equation is as follows:

[0073] (1)

[0074] Where x0, y0, and f are the internal parameters of the image acquisition device, X s Y s Z s a1~c3 are the external parameters of the image acquisition device; Z is the vertical coordinate in the point cloud data.

[0075] For example, based on the first collinearity equation and point cloud data, each pixel in the image data acquired by each image acquisition device can be projected onto a preset plane to obtain a corrected image of the image data.

[0076] S203, perform pixel matching on the correction maps corresponding to different image data to obtain the correction pixel coordinate group corresponding to each position point of the component.

[0077] For example, suppose there are two image acquisition devices, and the corrected image corresponding to the different image data acquired by the two image acquisition devices is I. Lortho and I Rortho Correction of Figure I Lortho and I RorthoPixel matching is performed to obtain the pixel coordinate sets (X1, Y1) and (X2, Y2) corresponding to each location point of the component in each correction image. Since the correction images are not the actual acquired image data, the pixel coordinate sets (X1, Y1) and (X2, Y2) need to be mapped back to the image data acquired by the image acquisition device to obtain the corrected pixel coordinate sets (x1, y1) and (x2, y2) corresponding to each location point of the component. The corrected pixel coordinate set corresponding to each location point includes the pixel coordinates of the location point in each image data.

[0078] For example, pixel matching of the correction images corresponding to different image data can be performed by using an image matching algorithm to perform element-by-element matching of the correction images corresponding to different image data. The image matching algorithm can be Normalized Cross-Correlation (NCC), Scale-Invariant Feature Transform (SIFT), etc.

[0079] S204. Based on each set of corrected pixel coordinates, determine the actual three-dimensional coordinates of each position point, and based on the difference between the actual three-dimensional coordinates and the initial three-dimensional coordinates of each position point, determine the deformation of the component.

[0080] Furthermore, the actual three-dimensional coordinates of each location point can be determined based on the coordinate groups of each corrected pixel and the internal parameters of the image acquisition device. Then, the difference between the actual three-dimensional coordinates and the initial three-dimensional coordinates of each location point can be compared with a preset threshold to determine the deformation status of the component.

[0081] For example, the difference between the actual 3D coordinates and the initial 3D coordinates of each location point is determined according to the following formula:

[0082] (2)

[0083] Among them, (X) i Y i Z i Let (x0, y0, z0) be the actual 3D coordinates of a position point of the component calculated based on the i-th frame image, and let (x0, y0, z0) be the initial 3D coordinates of a position point of the component. The deformation along the x-axis is... This represents the deformation along the y-axis. This represents the deformation along the z-axis.

[0084] For example, if the difference between the actual 3D coordinates and the initial 3D coordinates along any axis is greater than or equal to a preset threshold, then the component is determined to have deformed; if the difference between the actual 3D coordinates and the initial 3D coordinates along any axis is less than the preset threshold, then the component is determined not to have deformed. The preset threshold can be set according to actual needs, for example, it could be 1mm or 1cm.

[0085] The aforementioned method for determining component deformation involves acquiring point cloud data of the component and image data of the component acquired by at least two image acquisition devices. For each image data, disparity correction is performed on the image data based on the point cloud data, a first collinearity equation, and the external parameters of the image acquisition devices, resulting in a corrected image map. The first collinearity equation describes the correspondence between the component's three-dimensional coordinates, the external parameters of the image acquisition devices, and the component's pixel coordinates. Pixel matching is performed on the corrected images corresponding to different image data to obtain a corrected pixel coordinate set for each position point of the component. Each corrected pixel coordinate set includes the pixel coordinates of the position point in each image data. This achieves disparity correction of image data acquired by different acquisition devices, and then determines the actual three-dimensional coordinates of each position point based on each corrected pixel coordinate set, making the determined actual three-dimensional coordinates more accurate. Furthermore, the component deformation is determined based on the difference between the actual three-dimensional coordinates and the initial three-dimensional coordinates of each position point, making the determined deformation more reliable.

[0086] In some alternative implementations, see [link to relevant documentation]. Figure 3 , Figure 3 A flowchart for obtaining a corrected image corresponding to image data is provided, which specifically includes the following steps:

[0087] S301, extract features from the image data to obtain the image geometric features.

[0088] For example, edge detection algorithms or structure tensor algorithms can be used to extract features from image data to obtain image geometric features. These geometric features include corner features, edge features, and planar features of components. Corner features include the corners of a transformer casing and the intersections of equipment supports; edge features include the horizontal or vertical edges of equipment; and planar features include the top and side surfaces of a transformer.

[0089] S302, Based on the geometric features of the point cloud and the geometric features of the image, establish the correspondence between the pixel coordinates of each pixel in the image data and the vertical coordinates of the point cloud data.

[0090] Point cloud geometric features are obtained by extracting features from point cloud data; point cloud geometric features also include corner features, edge features, and planar features of components.

[0091] For example, a correspondence between the pixel coordinates of each pixel in the image data and the vertical coordinates of the point cloud data can be established through type matching and geometric constraints. Type matching can be understood as point-to-point, line-to-line, or face-to-face matching; for example, corner points in the point cloud data correspond to corner points in the image data, and edge lines in the point cloud data correspond to edge lines in the image data. Geometric constraints can be understood as distance constraints, structural constraints, and angular constraints, such as the distance from a point to a line needing to be less than 0.5 mm, and the angular deviation of line features needing to be less than 1°.

[0092] S303, for each pixel in the image data, determine the theoretical three-dimensional coordinates of the pixel based on the first collinearity equation, the external parameters of the image acquisition device that acquires the image data, and the vertical coordinates corresponding to the pixel.

[0093] For example, by using the back projection of the first collinearity equation, i.e., the three-dimensional coordinates of the point and the external parameters of the image acquisition device, the pixel coordinates of the point in the image data can be calculated; conversely, if the ordinate (depth) and the external parameters of the image acquisition device are known, the pixel coordinates in the image data can be projected onto a uniform plane (parallax-free plane) to eliminate the difference in viewing angle.

[0094] Based on this, for each pixel in the image data, the theoretical three-dimensional coordinates of the pixel can be determined according to the first collinearity equation, the external parameters of the image acquisition device that acquires the image data, and the vertical coordinates corresponding to the pixel.

[0095] In the case where the image acquisition device is a camera, the external parameters of the image acquisition device that acquires the image data can be camera parameters, which are known quantities. The vertical coordinates corresponding to the pixels are determined according to the correspondence between the pixel coordinates of each pixel in the image data and the vertical coordinates of the point cloud data. By substituting the camera parameters and the vertical coordinates corresponding to the pixels into the first collinearity equation, the theoretical three-dimensional coordinates (X, Y, Z) of the pixels can be obtained.

[0096] S304 projects the theoretical three-dimensional coordinates of each pixel onto a preset plane to obtain a corrected image of the image data.

[0097] For example, the preset plane can be a plane parallel to one of the planes of the component. The theoretical three-dimensional coordinates of each pixel can be orthographically projected onto the preset plane to obtain a corrected image of the image data.

[0098] In the above embodiments, parallax correction is performed on the image data using point cloud data, the first collinearity equation, and the external parameters of the image acquisition device to obtain the corrected image data, thereby eliminating the parallax of the image data without the acquisition device and improving the reliability of the actual three-dimensional coordinates of the subsequently determined component position points.

[0099] In some alternative implementations, see [link to relevant documentation]. Figure 4 , Figure 4 A flowchart is provided to obtain the corrected pixel coordinate group corresponding to each position point of a component, specifically including the following steps:

[0100] S401, perform pixel matching on the correction images corresponding to different image data to obtain the matching pixel coordinate group corresponding to each position point of the component.

[0101] For example, let I be the corrected image corresponding to different image data. Lortho and I Rortho Taking this as an example, an image matching algorithm can be used to perform pixel matching on the correction images corresponding to different image data, obtaining the matching pixel coordinate sets (X1, Y1) and (X2, Y2) corresponding to each position point of the component. The matching pixel coordinate set corresponding to each position point includes the pixel coordinates of the position point in each correction image.

[0102] S402, based on the theoretical three-dimensional coordinates corresponding to the matching pixel coordinate group, the external parameters of the image acquisition device, and the first collinearity equation, obtain the corrected pixel coordinate group corresponding to each position point of the component.

[0103] For example, the theoretical three-dimensional coordinates corresponding to the matching pixel coordinate group can be determined based on S303. Then, based on the theoretical three-dimensional coordinates corresponding to the matching pixel coordinate group, the external parameters of the image acquisition device, and the first collinearity equation, the corrected pixel coordinate group corresponding to each position point of the component can be obtained. Wherein, the theoretical three-dimensional coordinates corresponding to the matching pixel coordinate group are already determined, and the external parameters of the image acquisition device are known. The theoretical three-dimensional coordinates corresponding to the matching pixel coordinate group and the external parameters of the image acquisition device are substituted into the first equation for solving to obtain the corrected pixel coordinate group corresponding to each position point of the component.

[0104] In the above embodiments, by performing pixel matching on the correction images corresponding to different image data and mapping the matching pixel coordinate groups back into the image data, the parallax of the non-acquisition device is eliminated, thereby improving the reliability of the actual three-dimensional coordinates of the subsequently determined component position points.

[0105] In some alternative implementations, see [link to relevant documentation]. Figure 5 , Figure 5 A flowchart illustrating the process of determining the actual three-dimensional coordinates of each location point is provided, specifically including the following steps:

[0106] S501, determine the direction vector of each image acquisition device based on each set of corrected pixel coordinates.

[0107] For example, the direction vector of each image acquisition device is a unit vector pointing from the center of the image to the image point, describing the "spatial direction when viewing the image point" of the image acquisition device, and can be determined based on the internal and external parameters of the image acquisition device.

[0108] For example, taking a camera as the image acquisition device and (x1, y1) and (x2, y2) as the corrected pixel coordinate sets, let's assume (x1, y1) are the corrected pixel coordinates in the image data acquired by the left camera, and (x2, y2) are the corrected pixel coordinates in the image data acquired by the right camera. (x1, y1) and (x2, y2) can be converted into "image space coordinates" (coordinates with the camera's center of image capture as the origin).

[0109] Taking the determination of the orientation vector of the left camera as an example, the image space coordinates corresponding to the corrected pixel coordinates in the image data acquired by the left camera are x. 像1 =(x1-x0) / f,y 像1 =(y1-y0) / f,z 像1 =-1. Then, using the direction cosine matrix of the left camera's extrinsic parameters, the image space coordinates are converted into an "object space direction vector." The direction cosine matrix of the left camera's extrinsic parameters is determined by the attitude parameters in the image acquisition device's extrinsic parameters. The object space direction vector of the left camera is as follows:

[0110] (3)

[0111] Finally, V1 is normalized to a unit vector to obtain the direction vector v1 of the left camera.

[0112] The direction vector of the right camera is obtained in the same way, resulting in the direction vector v2 of the right camera. The specific acquisition process will not be described in detail here.

[0113] S502, determine the ray equation of each image acquisition device based on the internal parameters and direction vector of each image acquisition device.

[0114] For example, the ray equations of each image acquisition device can be determined by taking the shooting center of the left and right cameras as the starting point and combining the direction vectors.

[0115] For example, the resulting ray equation is as follows:

[0116] (4)

[0117] Where s is the scaling factor on the left camera ray, representing the length of travel from the camera center along the v1 direction, t is the scaling factor on the right camera ray, representing the length of travel from the camera center along the v2 direction, and D is the vector difference between the two camera centers.

[0118] S503, solve the determined ray equation to obtain the actual three-dimensional coordinates corresponding to each position point.

[0119] For example, to solve the determined ray equation, we can substitute s into the left ray equation to obtain P(s), and substitute t into the right ray equation to obtain Q(t). Ideally, the two rays should intersect at a single point, but due to measurement errors, they often do not intersect. Therefore, we can take the midpoint of the shortest line connecting the two rays as the coordinates of the spatial point, that is, find s* and t* that minimize the following expression:

[0120] (5)

[0121] Where d is the distance between the two rays. Therefore, the actual three-dimensional coordinates corresponding to each position point are:

[0122] (6)

[0123] in, For each location point, the actual three-dimensional coordinates are... Let d be the coordinate point on the equation of the left ray when d reaches its minimum value. The coordinates of the right ray equation when d reaches its minimum value.

[0124] In the above embodiments, by determining the direction vector of each image acquisition device and determining the actual three-dimensional coordinates of each position point based on the direction vector and the internal parameters of the image acquisition device, the determined actual three-dimensional coordinates are more consistent with the spatial position of each position point of the component, thereby improving the accuracy of the deformation results.

[0125] In some alternative implementations, see [link to relevant documentation]. Figure 6 , Figure 6 A flowchart illustrating the process of determining the external parameters of an image acquisition device is provided, specifically including the following steps:

[0126] S601, Obtain a template image with the same shooting conditions as the image data acquired by the image acquisition device.

[0127] The shooting conditions include at least the shooting angle and image size. For example, a preset number of reference points can be selected in advance, and raw image data containing all reference points can be acquired at different shooting angles and sizes. A template image can then be constructed based on the raw image data. It should be noted that the number of reference points is no less than the number of external parameters of the image acquisition device; the external parameters of the image acquisition device are the exterior orientation elements.

[0128] For example, the selection of reference points does not rely on traditional manual layout methods, but fully utilizes existing, stable, and distinctive features at the substation site as natural reference points to simplify workflows and improve operational efficiency. Features with the following characteristics are given priority as reference points:

[0129] Structures with regular geometric shapes and clear edges, such as corner points of high-voltage equipment foundations, curb corners, intersections of metal fences, and corner points of fixed ground markers; easily identifiable and unobstructed feature points in the image; whose appearance is stable and does not easily change with time or viewing angle in subsequent image sequence and multi-angle image matching; distributed around the perimeter and center of the image area to ensure the stability of a single image. High-precision total stations can be used to measure the selected reference points.

[0130] To ensure the consistency and calculability of the measurement results, a local coordinate system was established on-site using a station-based measurement method. Specifically, an arbitrary point on-site was selected as the station point, and one or two visible points were chosen as orientation points. The three-dimensional coordinates of each reference point were determined using angle and distance measurements. During the measurement process, each reference point was measured at least twice to ensure stable accuracy. This method eliminates the need for manual markers, reduces the hazards posed by live equipment on-site, improves operational efficiency, and is suitable for substation environments with complex equipment and limited space.

[0131] To improve the image stability and positioning accuracy of cameras under conditions such as rotation, zoom, and changes in viewing angle, a template image library can be established in advance. Subsequently, the pixel coordinates of reference points in the currently acquired image data can be determined based on the template images in the template image library.

[0132] The following describes the process of creating a template image library.

[0133] Step 1: Acquisition of reference points and component images.

[0134] At the substation site, calibrated cameras were used to photograph the area. Image frames with uniform lighting and clear imaging were selected, and image blocks containing reference points and components were extracted as sample images. Multiple images were captured from each reference point and component under different angles and focal lengths to enhance the robustness of subsequent matching.

[0135] Step 2: Sample labeling and management.

[0136] For each image sample, the pixel positions of reference points in the image are manually or semi-automatically labeled, and their 3D coordinates and reference point numbers are bound together. The components to be monitored also need to have their pixel positions in the image labeled, and their 3D coordinates and component numbers bound together. All samples are categorized and managed according to their reference point numbers, forming an initial image sample database.

[0137] Step 3: Image enhancement and feature extraction.

[0138] To improve the matching ability of samples under different shooting conditions, image samples can be preprocessed by brightness normalization, scale normalization, edge enhancement, etc., and local features can be extracted and stored in the template image library for fast matching.

[0139] Step 4: Establish geometric constraints.

[0140] While collecting samples, the camera shooting parameters (focal length, internal parameters, and external parameters) are recorded. Combined with the three-dimensional coordinates of the reference point, the initial geometric relationship between the image and the ground features is established.

[0141] The reference point and component template image library constructed using the above method possesses rich adaptability to changes in viewpoint and scale. It can quickly and accurately identify and relocate reference points when external disturbances (such as rotation, zoom, or slight displacement) occur in the sequence of images, thus providing stable and high-precision support for subsequently determining the external parameters of the image acquisition equipment based on each image. Simultaneously, it also provides a sufficiently rich and accurate data foundation for the corresponding substation components to be monitored.

[0142] S602, perform pixel matching on the image data and the template image to obtain the reference pixel coordinates corresponding to each reference point.

[0143] For example, pixel matching can be performed on the image data and the template image to determine the reference pixel coordinates corresponding to each reference point. For instance, the pixel coordinates of each reference point in the template image can be directly used as the reference pixel coordinates corresponding to each reference point, or the pixel coordinates of each reference point in the template image can be used to correct the pixel coordinates of each reference point in the image data to obtain the reference pixel coordinates corresponding to each reference point.

[0144] For example, the displacement deviation of each reference point can be determined based on the image data and the template image. For instance, the first derivative can be calculated for both the image data and the template image, reflecting the pixel grayscale change rate, to obtain gradient maps G1 (template image) and G2 (image data). A gradient threshold m (e.g., m=20) is set, retaining only "high-net-value derivative pixels" with gradient values ​​greater than m (these pixels are areas with obvious features such as edges and corners, excluding interference from low-texture and noisy areas), forming a set of pixels to be matched. .

[0145] In point set A matching error function is established to calculate the difference between G1 and G2, and the pixel offset (p1, p2) that minimizes the error is found. Then, based on the initial pixel coordinates of each reference point and the displacement deviation of the corresponding reference point, the reference pixel coordinates of each reference point in the image data are determined, where the initial pixel coordinates of each reference point are the pixel coordinates of each reference point in the template image.

[0146] For example, the initial pixel coordinates of the reference point in the template image are (x 10 y 10 Then the reference pixel coordinates of the reference point in the image data are (x=x 10 +p1, y=y 10 +p2).

[0147] It should be noted that in S201, image data of the component acquired by at least two image acquisition devices can be obtained. Similarly, template images corresponding to the image data of the component acquired by at least two image acquisition devices can be selected according to the shooting conditions. Then, the first derivative can be calculated on the image data and the template image to obtain a gradient map. Based on the gradient map, the displacement deviation of each pixel of the component in the image data can be determined. Then, based on the initial pixel coordinates and displacement deviation of each pixel of the component in the template image, the reference pixel coordinates of each pixel of the component in the image data can be determined. Then, the subsequent steps are performed based on the reference pixel coordinates of the component in the image data.

[0148] S603, based on the second collinearity equation, the reference pixel coordinates and actual three-dimensional coordinates corresponding to each reference point, and the internal parameters and distortion parameters of the image acquisition device, determine the external parameters of the image acquisition device.

[0149] The second collinearity equation describes the relationship between the pixel coordinates and 3D coordinates of the reference point, the internal parameters, distortion parameters, and external parameters of the image acquisition device. The expression for the second collinearity equation is as follows:

[0150] (7)

[0151] in, / The distortion correction amount is determined by the distortion parameter, (X) c Y c Z c (x) represents the three-dimensional coordinates of the reference point. c y c () are the pixel coordinates of the reference point.

[0152] For example, the reference pixel coordinates and actual 3D coordinates corresponding to each reference point, as well as the internal parameters and distortion parameters of the image acquisition device, can be substituted into the second collinearity equation to solve for the external parameters of the image acquisition device. For instance, the nonlinear equation can be linearized and solved using the least squares method.

[0153] In the above embodiments, the external parameters of the image acquisition device are determined by the reference pixel coordinates and the actual three-dimensional coordinates of the reference point, which facilitates parallax correction of the image data acquired by the image acquisition device in the future.

[0154] In this embodiment, based on reference point image library sequence image registration and exterior orientation adaptive calculation, an image library is constructed using reference points and substation component library. Combined with the least squares matching algorithm with "high net value" first derivative constraint, a fast resection and dynamic exterior orientation re-estimation of single images within the shooting area is achieved, effectively improving the system robustness and environmental adaptability.

[0155] Furthermore, a parallax-free correction and matching algorithm based on the exterior orientation elements and Z-constraints of a single image is proposed: using the distance information provided by the lidar as the Z-constraint, the image points are mapped to a unified parallax-free surface through the geometric projection model of the single image, thereby improving the accuracy and reliability of stereo matching.

[0156] The embodiments of this application achieve high-precision, high-efficiency, and fully automated detection of minute deformations of substation components with minimal manual setup of reference points. They are applicable to the structural safety assessment and early warning of key components in substation operation and maintenance monitoring.

[0157] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0158] Based on the same inventive concept, this application also provides a component deformation determining apparatus for implementing the component deformation determining method described above. The solution provided by this apparatus is similar to the solution described in the above method; therefore, the specific limitations in one or more component deformation determining apparatus embodiments provided below can be found in the limitations of the component deformation determining method described above, and will not be repeated here.

[0159] In one exemplary embodiment, such as Figure 7 As shown, a component deformation determining device is provided, comprising:

[0160] The acquisition module 10 is used to acquire point cloud data of the component and image data of the component acquired by at least two image acquisition devices;

[0161] The correction module 20 is used to perform parallax correction on each image data according to the point cloud data, the first collinearity equation, and the external parameters of the image acquisition device that acquired the image data, to obtain the correction map corresponding to the image data; wherein, the first collinearity equation is used to describe the correspondence between the three-dimensional coordinates of the component, the external parameters of the image acquisition device, and the pixel coordinates of the component.

[0162] The matching module 30 is used to perform pixel matching on the correction images corresponding to different image data to obtain the correction pixel coordinate group corresponding to each position point of the component; wherein, the correction pixel coordinate group corresponding to each position point includes the pixel coordinates of the position point in each image data.

[0163] The determination module 40 is used to determine the actual three-dimensional coordinates of each position point according to each set of corrected pixel coordinates, and to determine the deformation of the component according to the difference between the actual three-dimensional coordinates of each position point and the initial three-dimensional coordinates.

[0164] The aforementioned method for determining component deformation involves acquiring point cloud data of the component and image data of the component acquired by at least two image acquisition devices. For each image data, disparity correction is performed on the image data based on the point cloud data, a first collinearity equation, and the external parameters of the image acquisition devices, resulting in a corrected image map. The first collinearity equation describes the correspondence between the component's three-dimensional coordinates, the external parameters of the image acquisition devices, and the component's pixel coordinates. Pixel matching is performed on the corrected images corresponding to different image data to obtain a corrected pixel coordinate set for each position point of the component. Each corrected pixel coordinate set includes the pixel coordinates of the position point in each image data. This achieves disparity correction of image data acquired by different acquisition devices, and then determines the actual three-dimensional coordinates of each position point based on each corrected pixel coordinate set, making the determined actual three-dimensional coordinates more accurate. Furthermore, the component deformation is determined based on the difference between the actual three-dimensional coordinates and the initial three-dimensional coordinates of each position point, making the determined deformation more reliable.

[0165] In one embodiment, the correction module 20 is specifically used for:

[0166] Feature extraction is performed on the image data to obtain image geometric features. Based on the point cloud geometric features and the image geometric features, a correspondence is established between the pixel coordinates of each pixel in the image data and the vertical coordinates of the point cloud data. The point cloud geometric features are obtained by feature extraction from the point cloud data. For each pixel in the image data, the theoretical three-dimensional coordinates of the pixel are determined according to the first collinearity equation, the external parameters of the image acquisition device, and the vertical coordinates of the pixel. The theoretical three-dimensional coordinates of each pixel are projected onto a preset plane to obtain a corrected image of the image data.

[0167] In one embodiment, the matching module 30 is specifically used for:

[0168] Pixel matching is performed on the correction images corresponding to different image data to obtain the matching pixel coordinate group corresponding to each position point of the component; wherein, the matching pixel coordinate group corresponding to each position point includes the pixel coordinates of the position point in each correction image; based on the theoretical three-dimensional coordinates corresponding to the matching pixel coordinate group, the external parameters of the image acquisition device and the first collinearity equation, the correction pixel coordinate group corresponding to each position point of the component is obtained.

[0169] In one embodiment, the determining module 40 is further configured to:

[0170] Based on the coordinates of each corrected pixel, determine the direction vector of each image acquisition device; based on the internal parameters and direction vector of each image acquisition device, determine the ray equation of each image acquisition device; solve the determined ray equation to obtain the actual three-dimensional coordinates corresponding to each position point.

[0171] In one embodiment, the determining module 40 is further configured to:

[0172] A template image is acquired under the same shooting conditions as the image data acquired by the image acquisition device; wherein the shooting conditions include at least the shooting angle and image size; pixel matching is performed on the image data and the template image to obtain the reference pixel coordinates corresponding to each reference point; based on the second collinearity equation, the reference pixel coordinates and actual three-dimensional coordinates corresponding to each reference point, and the internal parameters and distortion parameters of the image acquisition device, the external parameters of the image acquisition device are determined; wherein the second collinearity equation is used to describe the relationship between the pixel coordinates, three-dimensional coordinates, internal parameters, distortion parameters and external parameters of the reference point.

[0173] In one embodiment, the determining module 40 is further configured to:

[0174] Based on the image data and the template image, determine the displacement deviation of each reference point; based on the initial pixel coordinates of each reference point and the corresponding displacement deviation of the reference point, determine the reference pixel coordinates of each reference point in the image data, wherein the initial pixel coordinates of each reference point are the pixel coordinates of each reference point in the template image.

[0175] Each module in the aforementioned component deformation determining device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0176] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 8As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements a component deformation determination method. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0177] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0178] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the component deformation determination method described in any of the above embodiments.

[0179] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the component deformation determination method described in any of the above embodiments.

[0180] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the component deformation determination method described in any of the above embodiments.

[0181] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0182] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0183] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for determining component deformation, characterized in that, The method includes: Acquire point cloud data of the component, as well as image data of the component acquired by at least two image acquisition devices; For each image data, based on the point cloud data, the first collinearity equation, and the external parameters of the image acquisition device that acquired the image data, parallax correction is performed on the image data to obtain a corrected image corresponding to the image data; wherein, the first collinearity equation is used to describe the correspondence between the three-dimensional coordinates of the component, the external parameters of the image acquisition device, and the pixel coordinates of the component; Pixel matching is performed on the correction images corresponding to different image data to obtain the correction pixel coordinate group corresponding to each position point of the component; wherein, the correction pixel coordinate group corresponding to each position point includes the pixel coordinates of the position point in each image data; Based on each set of corrected pixel coordinates, the actual three-dimensional coordinates of each position point are determined, and the deformation of the component is determined based on the difference between the actual three-dimensional coordinates and the initial three-dimensional coordinates of each position point.

2. The method according to claim 1, characterized in that, The step of performing disparity correction on the image data based on the point cloud data, the first collinearity equation, and the external parameters of the image acquisition device that acquired the image data, to obtain a corrected image corresponding to the image data, includes: Feature extraction is performed on the image data to obtain the image geometric features; Based on the point cloud geometric features and the image geometric features, a correspondence is established between the pixel coordinates of each pixel in the image data and the vertical coordinates of the point cloud data; wherein, the point cloud geometric features are obtained by feature extraction from the point cloud data; For each pixel in the image data, the theoretical three-dimensional coordinates of the pixel are determined based on the first collinearity equation, the external parameters of the image acquisition device that acquires the image data, and the vertical coordinates corresponding to the pixel. The theoretical three-dimensional coordinates of each pixel are projected onto a preset plane to obtain a corrected image of the image data.

3. The method according to claim 2, characterized in that, The step of performing pixel matching on the correction maps corresponding to different image data to obtain the correction pixel coordinate group corresponding to each position point of the component includes: Pixel matching is performed on the correction images corresponding to different image data to obtain the matching pixel coordinate group corresponding to each position point of the component; wherein, the matching pixel coordinate group corresponding to each position point includes the pixel coordinates of the position point in each correction image; Based on the theoretical three-dimensional coordinates corresponding to the matched pixel coordinate group, the external parameters of the image acquisition device, and the first collinearity equation, the corrected pixel coordinate group corresponding to each position point of the component is obtained.

4. The method according to claim 1, characterized in that, Each corrected pixel coordinate group determines the actual three-dimensional coordinates of each position point, including: Based on each set of corrected pixel coordinates, determine the orientation vector of each image acquisition device; The ray equations of each image acquisition device are determined based on the internal parameters and direction vectors of each image acquisition device. Solve the determined ray equation to obtain the actual three-dimensional coordinates of each position point.

5. The method according to claim 1, characterized in that, The image data includes a preset number of reference points, and the external parameters of each image acquisition device are determined in the following way: Obtain a template image with the same shooting conditions as the image data acquired by the image acquisition device; wherein, the shooting conditions include at least the shooting angle and the image size; Pixel matching is performed on the image data and the template image to obtain the reference pixel coordinates corresponding to each reference point; The external parameters of the image acquisition device are determined based on the second collinearity equation, the reference pixel coordinates and actual three-dimensional coordinates corresponding to each reference point, and the internal parameters and distortion parameters of the image acquisition device; wherein, the second collinearity equation is used to describe the relationship between the pixel coordinates, three-dimensional coordinates of the reference point, the internal parameters, distortion parameters and external parameters of the image acquisition device.

6. The method according to claim 5, characterized in that, The step of performing pixel matching between the image data and the template image to obtain the reference pixel coordinates corresponding to each reference point includes: Based on the image data and the template image, determine the displacement deviation of each reference point; Based on the initial pixel coordinates of each reference point and the displacement deviation of the corresponding reference point, the reference pixel coordinates of each reference point in the image data are determined, wherein the initial pixel coordinates of each reference point are the pixel coordinates of each reference point in the template image.

7. A component deformation determining device, characterized in that, The device includes: An acquisition module is used to acquire point cloud data of the component, as well as image data of the component acquired by at least two image acquisition devices; The correction module is used to perform parallax correction on each image data according to the point cloud data, the first collinearity equation, and the external parameters of the image acquisition device that acquires the image data, to obtain a corrected image corresponding to the image data; wherein, the first collinearity equation is used to describe the correspondence between the three-dimensional coordinates of the component, the external parameters of the image acquisition device, and the pixel coordinates of the component; The matching module is used to perform pixel matching on the correction images corresponding to different image data to obtain the correction pixel coordinate group corresponding to each position point of the component; wherein, the correction pixel coordinate group corresponding to each position point includes the pixel coordinates of the position point in each image data. The determination module is used to determine the actual three-dimensional coordinates of each position point according to each set of corrected pixel coordinates, and to determine the deformation of the component according to the difference between the actual three-dimensional coordinates and the initial three-dimensional coordinates of each position point.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images