Automatic detection method and system for shear deformation of bridge rubber support

Abstract

The invention provides an automatic detection method and system for shear deformation of a bridge rubber support, and belongs to the technical field of engineering structure detection. According to the method, image correction based on geometric rule cushion blocks, deep learning image segmentation and a multi-stage angular point extraction strategy of'angular point coarse detection-random sampling consistency algorithm refined positioning-contour adsorption correction 'are introduced, so that the method can be used for extracting angular points under the engineering field conditions of limited shooting space, complex illumination, existence of stains or shielding and the like; high-robustness and high-precision automatic positioning can still be carried out on the corner points of the bridge rubber support; secondly, in combination with a shear deformation calculation method based on a left and right edge contour linear fitting inclination angle, the problems of subjective errors and poor repeatability caused by traditional manual measurement are avoided, the precision and stability of shear deformation detection are remarkably improved, linkage with monitoring equipment can be realized under an unattended condition, and the detection efficiency is improved. The automatic, long-term and intelligent monitoring of the shear deformation of the bridge rubber support is realized.

Description

Technical Field

[0001] This invention relates to the field of engineering structure testing technology, and in particular to an automated method and system for detecting shear deformation of bridge rubber bearings. Background Technology

[0002] Bridge bearings, as a crucial component of bridge structures, primarily bear the loads of vertical and lateral loads, as well as bridge deformation caused by temperature changes. During the long-term use of a bridge, bearings are subjected to continuous load, environmental, and temperature variations, with rubber bearings particularly susceptible to shear deformation. Shear deformation of rubber bearings not only affects the normal use of the bridge but can also lead to further structural damage and even bridge safety accidents. Therefore, regular inspection of bridge bearings, especially the detection of shear deformation, is of paramount importance for ensuring the safety and stability of bridges.

[0003] Traditional methods for detecting shear deformation in bridge bearings largely rely on manual visual inspection and conventional measurement techniques. While manual inspection is intuitive, it is not only labor-intensive and inefficient when dealing with large-scale bridges, but also susceptible to human error, leading to inaccurate results. Traditional measurement methods typically rely on manually operated tools such as angle measuring instruments and displacement sensors. While these devices offer a degree of accuracy, they require complex setup and lengthy operation, and cannot be quickly applied in various environments. Furthermore, traditional methods are often limited to localized inspections of the bridge, making it difficult to comprehensively and rapidly inspect the entire bridge bearing system, thus failing to meet the demands of modern bridge structural health monitoring for efficient, accurate, and comprehensive testing. Summary of the Invention

[0004] To address the problems in the existing technology, this invention provides an automated detection method and system for shear deformation of bridge rubber bearings. Firstly, by introducing image correction based on geometrically regular pads, deep learning image segmentation, and a multi-stage corner extraction strategy of "coarse corner detection—refined positioning using random sampling consistency algorithm—contour adsorption correction," this invention achieves highly robust and accurate automatic positioning of bridge rubber bearing corners even under engineering site conditions such as limited shooting space, complex lighting, and the presence of stains or obstructions. Secondly, by combining a shear deformation calculation method based on linear fitting of the inclination angle of the left and right edge contours, it avoids the subjective errors and poor repeatability problems caused by traditional manual measurement, significantly improving the accuracy and stability of shear deformation detection. To achieve the above objectives, the technical solution is as follows:

[0005] On one hand, the present invention provides an automated detection method for shear deformation of bridge rubber bearings, the method comprising:

[0006] S1. Obtain the front view image of the bridge rubber bearing based on the bridge rubber bearing;

[0007] S2. Based on the front view image of the bridge rubber bearing, obtain the binarized mask of the rubber bearing through the image segmentation model;

[0008] S3. Based on the binary mask of the rubber support, the outer edge contour of the rubber support is obtained through a contour detection algorithm.

[0009] S4. Based on the outer edge contour of the rubber bearing, a quadrilateral approximation algorithm is used for geometric fitting to obtain the four corner points of the rubber bearing.

[0010] S5. Based on the four corner points of the rubber bearing, a robust straight line fitting is performed using a random sampling consensus algorithm to obtain the four corner points of the rubber bearing after refinement.

[0011] S6. Based on the four corner points of the rubber bearing after fine processing, extract the pixel coordinates of the outer contours on both sides of the binary mask of the rubber bearing and map them onto the front view image of the bridge rubber bearing to obtain the rubber bearing image with pixel coordinates.

[0012] S7. Based on the image of the rubber support with pixel coordinates, perform linear fitting on the pixel coordinates of the outer contours on both sides of the binary mask of the rubber support to obtain the pixel fitting line.

[0013] S8. Based on the fitted straight line of the pixel, the shear deformation on both sides of the rubber support is obtained by calculating the tilt angle.

[0014] Optionally, in step S1, obtaining a front view image of the bridge rubber bearing includes:

[0015] S11. Based on the bridge rubber bearing, the geometric reference of the image is obtained through the geometrically regular pad structure below the rubber bearing.

[0016] S12. Based on the geometric reference of the bridge rubber bearing and the image, take a frontal view of the bridge rubber bearing and use a perspective correction algorithm to obtain the frontal view image of the bridge rubber bearing.

[0017] Optionally, the image segmentation model includes: a deep learning-based semantic segmentation network model.

[0018] Optionally, in step S3, the outer edge contour of the rubber support is obtained using a contour detection algorithm based on the binary mask of the rubber support, including:

[0019] S31. Based on the binary mask of the rubber bearing, through connected component analysis, the target connected component is retained and small noise blocks are deleted to obtain the purified rubber bearing mask.

[0020] S32. Based on the purified rubber support mask, the outer edge contour of the rubber support is obtained by processing it through the findContours function.

[0021] Optionally, the quadrilateral approximation algorithm includes the Douglas–Peucker algorithm.

[0022] Optionally, in step S5, a robust straight-line fitting algorithm is used based on the four corner points of the rubber bearing to obtain the four corner points of the rubber bearing after refinement, including:

[0023] S51. Based on the four corner points of the rubber support, obtain the local neighborhood of the four corner points;

[0024] S52. Based on the local neighborhood of the four corner points, a robust straight line fitting is performed using a random sampling consensus algorithm to obtain four sets of two adjacent fitted straight lines.

[0025] S53. Based on the four sets of two adjacent fitted straight lines, the four corner points of the rubber bearing after fine processing are obtained.

[0026] Optionally, in S8, based on the fitted straight line of the pixel, the shear deformation on both sides of the rubber support is calculated by tilt angle, including:

[0027] S81. Based on the fitted line of the pixel, obtain the slope of the fitted line of the pixel.

[0028] S82. Based on the slope of the fitted line of the pixel, obtain the tilt angle of the fitted line of the pixel;

[0029] S83. Based on the inclination angle of the fitted line of the pixel, the shear deformation on both sides of the rubber support is obtained through formula (1).

[0030] (1)

[0031] In the formula: This refers to the shear deformation on both sides of the rubber bearing. The vertical height of the rubber bearing. The angle of inclination of the line fitted to the pixel.

[0032] On the other hand, the present invention provides an automated detection system for shear deformation of bridge rubber bearings. This system is applied to an automated detection method for shear deformation of bridge rubber bearings. The system includes:

[0033] The image acquisition module is used to acquire a front view image of the bridge rubber bearing.

[0034] The binarization module is used to obtain a binarization mask of the rubber bearing based on the front view image of the bridge rubber bearing through an image segmentation model.

[0035] The contour acquisition module is used to obtain the outer edge contour of the rubber support based on the binary mask of the rubber support and through a contour detection algorithm.

[0036] The corner point acquisition module is used to obtain the four corner points of the rubber support by performing geometric fitting using a quadrilateral approximation algorithm based on the outer edge contour of the rubber support.

[0037] The corner point processing module is used to perform robust line fitting based on the four corner points of the rubber bearing using a random sampling consensus algorithm, so as to obtain the four corner points of the rubber bearing after fine processing.

[0038] The pixel extraction module is used to extract the pixel coordinates of the outer contours on both sides of the binary mask of the rubber bearing based on the four corner points after the rubber bearing has been refined, and then map them onto the front view image of the bridge rubber bearing to obtain a rubber bearing image with pixel coordinates.

[0039] The linear fitting module is used to perform linear fitting on the pixel coordinates of the outer contours of both sides of the binary mask of the rubber support based on the image of the rubber support with pixel coordinates, so as to obtain the pixel fitting line.

[0040] The deformation calculation module is used to fit a straight line based on the pixel and calculate the shear deformation on both sides of the rubber support by tilt angle.

[0041] Compared with the prior art, the technical solution of the present invention has at least the following beneficial effects:

[0042] The above-mentioned scheme, on the one hand, introduces a multi-stage corner extraction strategy based on geometric regular pad blocks, deep learning image segmentation, and "coarse corner detection - fine positioning using random sampling consistency algorithm - contour adsorption correction," which enables highly robust and accurate automatic positioning of bridge rubber bearing corners even under engineering site conditions such as limited shooting space, complex lighting, and the presence of stains or occlusions. On the other hand, by combining a shear deformation calculation method based on linear fitting of the tilt angle of the left and right edge contours, it avoids the subjective errors and poor repeatability problems caused by traditional manual measurement, significantly improving the accuracy and stability of shear deformation detection. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 This is a flowchart of an embodiment of the automated detection method for shear deformation of bridge rubber bearings according to the present invention;

[0045] Figure 2 This is a flowchart illustrating the process of obtaining a front view image of a bridge rubber bearing in an embodiment of the automated detection method for shear deformation of bridge rubber bearings of the present invention.

[0046] Figure 3 This is a flowchart illustrating the process of obtaining the outer edge contour of a rubber bearing in an embodiment of the automated detection method for shear deformation of bridge rubber bearings of the present invention.

[0047] Figure 4 This is a flowchart of the four corner points of the rubber bearing after fine processing, obtained in an embodiment of the automated detection method for shear deformation of bridge rubber bearings of the present invention.

[0048] Figure 5 This is a flowchart illustrating the automated detection method for shear deformation of bridge rubber bearings according to the present invention, showing the shear deformation on both sides of the rubber bearing.

[0049] Figure 6 This is a schematic diagram illustrating the extraction of the outer edge contour of the rubber bearing in an embodiment of the automated detection method for shear deformation of bridge rubber bearings of the present invention;

[0050] Figure 7 This is a system block diagram of an embodiment of the automated detection system for shear deformation of bridge rubber bearings of the present invention. Detailed Implementation

[0051] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0052] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0053] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0054] like Figure 1 The flowchart shown is an embodiment of the automated detection method for shear deformation of bridge rubber bearings according to the present invention. The present invention provides an automated detection method for shear deformation of bridge rubber bearings, which is implemented by an automated detection system for shear deformation of bridge rubber bearings. The method includes:

[0055] S1. Obtain the front view image of the bridge rubber bearing based on the bridge rubber bearing;

[0056] Specifically, such as Figure 2 The flowchart shown in the embodiment of the automated detection method for shear deformation of bridge rubber bearings of the present invention illustrates the process of obtaining a front view image of a bridge rubber bearing. Step S1, obtaining the front view image of the bridge rubber bearing, includes:

[0057] S11. Based on the bridge rubber bearing, the geometric reference of the image is obtained through the geometrically regular pad structure below the rubber bearing.

[0058] S12. Based on the geometric reference of the bridge rubber bearing and the image, take a frontal view of the bridge rubber bearing and use a perspective correction algorithm to obtain the frontal view image of the bridge rubber bearing.

[0059] Furthermore, image acquisition equipment (such as high-resolution industrial cameras, drone photography systems, etc.) is used to photograph the bridge rubber bearings, ensuring the camera lens maintains a frontal view as much as possible. In cases where space constraints at the engineering site prevent frontal photography, the rectangular pad structure beneath the rubber bearing is used as a geometric reference. A perspective correction algorithm is then employed to reconstruct the frontal view of the rubber bearing, improving the accuracy and stability of subsequent corner detection.

[0060] S2. Based on the front view image of the bridge rubber bearing, obtain the binarized mask of the rubber bearing through the image segmentation model;

[0061] Specifically, the image segmentation model includes: a semantic segmentation network model based on deep learning.

[0062] S3. Based on the binary mask of the rubber support, the outer edge contour of the rubber support is obtained through a contour detection algorithm.

[0063] Specifically, such as Figure 3 The flowchart shown in the embodiment of the automated detection method for shear deformation of bridge rubber bearings of the present invention, which obtains the outer edge contour of the rubber bearing, is as follows: Figure 6 The schematic diagram shown in the embodiment of the automated detection method for shear deformation of bridge rubber bearings of the present invention illustrates the extraction of the outer edge contour of the rubber bearing. In step S3, based on the binary mask of the rubber bearing, the outer edge contour of the rubber bearing is obtained through a contour detection algorithm, including:

[0064] S31. Based on the binary mask of the rubber bearing, through connected component analysis, the target connected component is retained and small noise blocks are deleted to obtain the purified rubber bearing mask.

[0065] S32. Based on the purified rubber support mask, the outer edge contour of the rubber support is obtained by processing it through the findContours function.

[0066] S4. Based on the outer edge contour of the rubber bearing, a quadrilateral approximation algorithm is used for geometric fitting to obtain the four corner points of the rubber bearing.

[0067] Specifically, the quadrilateral approximation algorithm includes the Douglas–Peucker algorithm.

[0068] S5. Based on the four corner points of the rubber bearing, a robust straight line fitting is performed using a random sampling consensus algorithm to obtain the four corner points of the rubber bearing after refinement.

[0069] Specifically, such as Figure 4 The flowchart shown in the embodiment of the automated detection method for shear deformation of bridge rubber bearings of the present invention obtains the four corner points of the rubber bearing after refined processing. In step S5, based on the four corner points of the rubber bearing, a robust straight line fitting algorithm is used to obtain the four corner points of the rubber bearing after refined processing, including:

[0070] S51. Based on the four corner points of the rubber support, obtain the local neighborhood of the four corner points;

[0071] S52. Based on the local neighborhood of the four corner points, a robust straight line fitting is performed using a random sampling consensus algorithm to obtain four sets of two adjacent fitted straight lines.

[0072] S53. Based on the four sets of two adjacent fitted straight lines, the four corner points of the rubber bearing after fine processing are obtained.

[0073] Furthermore, let the local neighborhood pixels of the corner point be ( The fitted straight line is in least squares form:

[0074] (2)

[0075] in,( , , ) is the optimal parameter set obtained by the random sampling consensus algorithm after multiple random sampling and consensus verifications.

[0076] For two adjacent fitted lines:

[0077] (3)

[0078] (4)

[0079] in, , , The coefficient of the first straight line; , , is the coefficient of the second line.

[0080] Intersection of two lines ( As the precise corner position, the calculation formula is:

[0081] (5)

[0082] (6)

[0083] This refinement method can obtain corner coordinates with high consistency and robustness under conditions of noise, stains, shadows and other interference.

[0084] S6. Based on the four corner points of the rubber bearing after fine processing, extract the pixel coordinates of the outer contours on both sides of the binary mask of the rubber bearing and map them onto the front view image of the bridge rubber bearing to obtain the rubber bearing image with pixel coordinates.

[0085] Specifically, for the refined corner point P=( ), which is the set of pixels at the contour edge { ( The distance calculation formula for} is as follows:

[0086] (7)

[0087] In the formula, Let be the Euclidean distance between the corner point and the j-th edge pixel.

[0088] Find the edge point with the smallest distance:

[0089] (8)

[0090] In the formula, The precise corner point position is defined after adsorption, ensuring strict alignment between the corner point and the actual contour of the rubber support. This process precisely constrains the corner point to the actual boundary of the rubber support, making subsequent contour extraction and deformation calculations more accurate.

[0091] Based on the corner coordinates of the rubber support, the pixel coordinates of the outer contours of the left and right sides of the binarized mask are extracted and mapped onto the original image for visualization. For example... Figure 6 The schematic diagram shown in the embodiment of the automated detection method for shear deformation of bridge rubber bearings of the present invention is a schematic diagram of extracting the outer edge contour of the rubber bearing. The four corner points are defined as follows: upper left corner point A (… Point B in the upper right corner Point C in the lower right corner ), lower left corner point D ( The outer contour pixels are arranged in clockwise (or counterclockwise) order to obtain an ordered contour sequence. Based on the index positions of the four corner points (top left A, top right B, bottom right C, and bottom left D) obtained from the previous detection within this contour sequence, the contour is divided into four segments. The contour segment connecting the top left and bottom left corners is defined as the left outer contour AB, and the contour segment connecting the top right and bottom right corners is defined as the right outer contour CD. The pixels of the left and right outer contours are mapped back to the original image for visualization, which is used for subsequent deformation calculations and manual verification.

[0092] S7. Based on the image of the rubber support with pixel coordinates, perform linear fitting on the pixel coordinates of the outer contours on both sides of the binary mask of the rubber support to obtain the pixel fitting line.

[0093] S8. Based on the fitted straight line of the pixel, the shear deformation on both sides of the rubber support is obtained by calculating the tilt angle.

[0094] Specifically, such as Figure 5 The flowchart shown in the embodiment of the automated detection method for shear deformation of bridge rubber bearings of the present invention obtains the shear deformation on both sides of the rubber bearing. In step S8, the shear deformation on both sides of the rubber bearing is obtained by fitting a straight line based on the pixel and calculating the inclination angle, including:

[0095] S81. Based on the fitted line of the pixel, obtain the slope of the fitted line of the pixel.

[0096] S82. Based on the slope of the fitted line of the pixel, obtain the tilt angle of the fitted line of the pixel;

[0097] S83. Based on the inclination angle of the fitted line of the pixel, the shear deformation on both sides of the rubber support is obtained through formula (1).

[0098] (1)

[0099] In the formula: This refers to the shear deformation on both sides of the rubber bearing. The vertical height of the rubber bearing. The angle of inclination of the line fitted to the pixel.

[0100] Furthermore, linear least squares fitting is performed on the left and right contour pixels respectively to obtain the slopes of the left and right edge lines. and Calculate the corresponding inclination angle based on the slope. and They are respectively:

[0101] (9)

[0102] (10)

[0103] By combining the actual height dimensions of the rubber bearing and calculating its shear deformation based on the inclination angle difference or corresponding geometric relationships, the automatic and accurate measurement of the shear deformation of bridge rubber bearings is achieved, exhibiting high repeatability and stability. Let the vertical height of the rubber bearing be... Shear deformation on both sides and They are respectively:

[0104] (11)

[0105] (12)

[0106] like Figure 7 The diagram shown is a system block diagram of an embodiment of the automated detection system for shear deformation of bridge rubber bearings according to the present invention. The present invention provides an automated detection system for shear deformation of bridge rubber bearings, which is applied to an automated detection method for shear deformation of bridge rubber bearings. The system includes: an image acquisition module, a binarization module, a contour acquisition module, a corner point acquisition module, a corner point processing module, a pixel extraction module, a line fitting module, and a deformation calculation module. Specifically,

[0107] The image acquisition module is used to acquire a front view image of the bridge rubber bearing.

[0108] The binarization module is used to obtain a binarization mask of the rubber bearing based on the front view image of the bridge rubber bearing through an image segmentation model.

[0109] The contour acquisition module is used to obtain the outer edge contour of the rubber support based on the binary mask of the rubber support and through a contour detection algorithm.

[0110] The corner point acquisition module is used to obtain the four corner points of the rubber support by performing geometric fitting using a quadrilateral approximation algorithm based on the outer edge contour of the rubber support.

[0111] The corner point processing module is used to perform robust line fitting based on the four corner points of the rubber bearing using a random sampling consensus algorithm, so as to obtain the four corner points of the rubber bearing after fine processing.

[0112] The pixel extraction module is used to extract the pixel coordinates of the outer contours on both sides of the binary mask of the rubber bearing based on the four corner points after the rubber bearing has been refined, and then map them onto the front view image of the bridge rubber bearing to obtain a rubber bearing image with pixel coordinates.

[0113] The linear fitting module is used to perform linear fitting on the pixel coordinates of the outer contours of both sides of the binary mask of the rubber support based on the image of the rubber support with pixel coordinates, so as to obtain the pixel fitting line.

[0114] The deformation calculation module is used to fit a straight line based on the pixel and calculate the shear deformation on both sides of the rubber support by tilt angle.

[0115] This invention provides an automated detection method and system for shear deformation of bridge rubber bearings. Firstly, by introducing a multi-stage corner extraction strategy—based on pad-based image correction, deep learning image segmentation, and a "coarse corner detection—refined positioning using random sampling consistency algorithm—contour adsorption correction" approach—this invention achieves highly robust and accurate automatic positioning of bridge rubber bearing corners even under engineering site conditions such as limited shooting space, complex lighting, and the presence of stains or obstructions. Secondly, by combining a shear deformation calculation method based on linear fitting of the inclination angle of the left and right edge contours, it avoids the subjective errors and poor repeatability problems caused by traditional manual measurement, significantly improving the accuracy and stability of shear deformation detection. It can also be linked with monitoring equipment under unattended conditions to achieve automated, long-term, and intelligent monitoring of shear deformation of bridge rubber bearings.

[0116] It is understood that the present invention has been described through the above embodiments and should not be construed as limiting the implementation and scope of the present invention. Those skilled in the art will recognize that various changes or equivalent substitutions can be made to these features and embodiments without departing from the spirit and scope of the present invention. Furthermore, under the teachings of the present invention, these features and embodiments can be modified to adapt to specific situations and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the protection scope of the present invention.

Claims

1. An automated detection method for shear deformation of bridge rubber bearings, characterized in that, The method includes: S1. Obtain the front view image of the bridge rubber bearing based on the bridge rubber bearing; S2. Based on the front view image of the bridge rubber bearing, obtain the binarized mask of the rubber bearing through the image segmentation model; S3. Based on the binary mask of the rubber support, the outer edge contour of the rubber support is obtained through a contour detection algorithm. S4. Based on the outer edge contour of the rubber support, a quadrilateral approximation algorithm is used for geometric fitting to obtain the four corner points of the rubber support. S5. Based on the four corner points of the rubber bearing, a robust straight line fitting is performed using a random sampling consensus algorithm to obtain the four corner points of the rubber bearing after refinement. S6. Based on the four corner points of the rubber bearing after fine processing, extract the pixel coordinates of the outer contours on both sides of the binary mask of the rubber bearing and map them onto the front view image of the bridge rubber bearing to obtain a rubber bearing image with pixel coordinates. S7. Based on the rubber support image with pixel coordinates, perform linear fitting on the pixel coordinates of the outer contours on both sides of the binary mask of the rubber support to obtain a pixel fitting line. S8. Based on the pixel-fitted straight line, the shear deformation on both sides of the rubber support is obtained by calculating the tilt angle.

2. The automated detection method for shear deformation of bridge rubber bearings according to claim 1, characterized in that, In step S1, obtaining a front view image of the bridge rubber bearing includes: S11. Based on the bridge rubber bearing, the geometric reference of the image is obtained through the geometrically regular pad structure below the rubber bearing; S12. Based on the geometric reference of the bridge rubber bearing and the image, take a frontal view image of the bridge rubber bearing and use a perspective correction algorithm to obtain the frontal view image of the bridge rubber bearing.

3. The automated detection method for shear deformation of bridge rubber bearings according to claim 1, characterized in that, The image segmentation model includes: a semantic segmentation network model based on deep learning.

4. The automated detection method for shear deformation of bridge rubber bearings according to claim 1, characterized in that, In step S3, the outer edge contour of the rubber support is obtained using a contour detection algorithm based on the binary mask of the rubber support, including: S31. Based on the binary mask of the rubber support, through connected component analysis, the target connected component is retained and small noise blocks are deleted to obtain the purified rubber support mask. S32. Based on the purified rubber support mask, the outer edge contour of the rubber support is obtained by processing it through the findContours function.

5. The automated detection method for shear deformation of bridge rubber bearings according to claim 1, characterized in that, The quadrilateral approximation algorithm includes the Douglas–Peucker algorithm.

6. The automated detection method for shear deformation of bridge rubber bearings according to claim 1, characterized in that, In step S5, a robust straight-line fitting algorithm is used based on the four corner points of the rubber bearing to obtain the four corner points of the rubber bearing after refinement, including: S51. Based on the four corner points of the rubber support, obtain the local neighborhood of the four corner points; S52. Based on the local neighborhood of the four corner points, a robust straight line fitting is performed using a random sampling consensus algorithm to obtain four sets of two adjacent fitted straight lines. S53. Based on the four sets of two adjacent fitted straight lines, obtain the four corner points of the rubber bearing after fine processing.

7. The automated detection method for shear deformation of bridge rubber bearings according to claim 1, characterized in that, In step S8, based on the pixel-fitted straight line, the shear deformation on both sides of the rubber support is calculated through the tilt angle, including: S81. Obtain the slope of the pixel-fitted line based on the pixel-fitted line; S82. Obtain the tilt angle of the pixel fitting line based on the slope of the pixel fitting line; S83. Based on the inclination angle of the pixel-fitted straight line, the shear deformation on both sides of the rubber support is obtained using formula (1). （1） In the formula: This refers to the shear deformation on both sides of the rubber bearing. The vertical height of the rubber bearing. The angle of inclination of the line fitted to the pixel.

8. An automated detection system for shear deformation of bridge rubber bearings, used to implement the automated detection method for shear deformation of bridge rubber bearings as described in any one of claims 1-7, characterized in that, The system includes: The image acquisition module is used to acquire a front view image of the bridge rubber bearing. The binarization module is used to obtain a binarization mask of the rubber bearing based on the front view image of the bridge rubber bearing through an image segmentation model. The contour acquisition module is used to obtain the outer edge contour of the rubber support based on the binary mask of the rubber support and through a contour detection algorithm. The corner point acquisition module is used to obtain the four corner points of the rubber support by performing geometric fitting using a quadrilateral approximation algorithm based on the outer edge contour of the rubber support. The corner point processing module is used to perform robust line fitting using a random sampling consensus algorithm based on the four corner points of the rubber bearing, so as to obtain the four corner points of the rubber bearing after fine processing. The pixel extraction module is used to extract the pixel coordinates of the outer contours on both sides of the binary mask of the rubber bearing based on the four corner points after the fine processing of the rubber bearing, and map them onto the front view image of the bridge rubber bearing to obtain a rubber bearing image with pixel coordinates. The linear fitting module is used to perform linear fitting on the pixel coordinates of the outer contours of both sides of the binary mask of the rubber support based on the rubber support image with pixel coordinates, so as to obtain the pixel fitting line. The deformation calculation module is used to obtain the shear deformation on both sides of the rubber support by fitting a straight line based on the pixel and calculating the tilt angle.

Images