A method and system for automatic monitoring of full-area deformation of reinforced concrete beams

Abstract

This invention provides an automatic monitoring method and system for the full-domain deformation of reinforced concrete beams, relating to the field of intelligent monitoring technology for engineering structures. The method includes: acquiring an original image of the reinforced concrete beam; inputting the original image into an image segmentation model to output a binarized mask image; determining the beam center point based on the foreground region in the binarized mask image; identifying the edges of the foreground region and extracting the beam's outer contour point set using an edge detection algorithm and a contour extraction algorithm; removing redundant points from the beam's outer contour point set to obtain a candidate beam corner point set; determining multiple target beam corner points from the candidate corner point set using a heuristic optimization algorithm; dividing the binarized mask image into multiple edge regions using a center radial division method; assigning each outer contour point in the beam's outer contour point set to its corresponding edge region; constructing a linear contour curve; and calculating the vertical full-domain deformation of the reinforced concrete beam based on the linear contour curve.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring technology for engineering structures, and in particular to an automatic monitoring method and system for the full-area deformation of reinforced concrete beams. Background Technology

[0002] With the continuous development of infrastructure projects such as buildings and bridges, reinforced concrete beams, as key load-bearing components, inevitably undergo plastic deformation during long-term service. These deformation characteristics directly reflect the degradation of structural performance and become important indicators for assessing structural health. Timely and accurate acquisition of full-domain deformation information of reinforced concrete beams provides an important basis for structural health status assessment and safety risk early warning.

[0003] In existing technologies, deformation monitoring of reinforced concrete beams typically relies on traditional corner point extraction methods, which identify structural boundaries by using fixed thresholds or local geometric features (such as curvature abrupt changes).

[0004] However, in actual engineering projects, due to factors such as stress deformation, image shooting angle, distortion effects, and apparent damage, the edges of reinforced concrete beams often no longer maintain a standard rectangular shape. Slight bending, corner blunting, or edge discontinuity may occur at the boundaries. Therefore, traditional corner extraction methods struggle to accurately extract the four true structural corners when faced with complex and irregular boundaries, thus affecting the accuracy of deformation measurements.

[0005] Furthermore, with the increasing complexity and deformation of building structures, the accuracy and robustness of traditional automatic deformation monitoring methods for reinforced concrete beams are gradually failing to meet actual monitoring needs. They are also unable to adapt to various non-standard contour shapes and have failed to effectively handle deformation monitoring tasks of reinforced concrete beams in complex environments. Summary of the Invention

[0006] To address the challenges of traditional corner point extraction methods in accurately extracting four true structural corner points when facing complex and irregular boundaries, thus affecting the accuracy of deformation measurement, and the fact that the accuracy and robustness of traditional automatic full-area deformation monitoring methods for reinforced concrete beams are gradually failing to meet actual monitoring needs, are difficult to adapt to various non-standard contour shapes, and have not effectively handled the technical problems of deformation monitoring tasks for reinforced concrete beams under complex backgrounds, this invention provides an automatic full-area deformation monitoring method and system for reinforced concrete beams.

[0007] The technical solutions provided by the embodiments of the present invention are as follows:

[0008] First aspect:

[0009] This invention provides an automatic monitoring method for the full-area deformation of reinforced concrete beams, comprising:

[0010] S1: Obtain the original image of the reinforced concrete beam;

[0011] S2: Input the original image into the image segmentation model and output a binarized mask image including the reinforced concrete beam;

[0012] S3: Determine the center point of the beam based on the foreground region in the binarized mask image;

[0013] S4: By using edge detection and contour extraction algorithms, the edges of the foreground region are identified, and the set of contour points outside the beam is extracted;

[0014] S5: By using the Douglas-Peucker polygon approximation algorithm, redundant points in the beam's outer contour point set are removed to obtain the candidate corner point set of the beam.

[0015] S6: With the objectives of minimizing the contour reconstruction error and minimizing the number of beam corner points, a heuristic optimization algorithm is used to determine multiple target beam corner points in the candidate beam corner point set;

[0016] S7: Based on the center point of the beam and the corner points of each target beam, the binarized mask image is divided into multiple edge regions using the center radial line division method;

[0017] S8: Divide each outer contour point in the beam's outer contour point set into the corresponding edge region;

[0018] S9: Extract all outer contour points from multiple edge regions and construct a linear contour curve;

[0019] S10: Calculate the vertical deformation of the reinforced concrete beam based on the linear profile curve and the reference curve from historical monitoring processes, so as to automatically monitor the overall deformation of the reinforced concrete beam.

[0020] The second aspect:

[0021] An automatic monitoring system for the full-area deformation of reinforced concrete beams provided in this embodiment of the invention includes:

[0022] processor;

[0023] A memory storing computer-readable instructions, which, when executed by the processor, implement the automatic monitoring method for full-area deformation of reinforced concrete beams as described in the first aspect.

[0024] Third aspect:

[0025] The present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the automatic monitoring method for full-area deformation of reinforced concrete beams as described in the first aspect.

[0026] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0027] In this embodiment of the invention, redundant points in the beam's outer contour point set are eliminated using the Douglas-Peucker polygon approximation algorithm to obtain a candidate corner point set for the beam. With the objectives of minimizing contour reconstruction error and minimizing the number of beam corner points, a heuristic optimization algorithm is used to determine multiple target beam corner points within the candidate corner point set. When facing complex and irregular boundaries, four true structural corner points can be accurately extracted, thereby improving the accuracy of deformation measurement. By extracting all outer contour points from multiple edge regions, a linear contour curve is constructed. Based on the linear contour curve and the reference curve from historical monitoring processes, the vertical global deformation of the reinforced concrete beam is calculated for automatic monitoring of the global deformation of the reinforced concrete beam. The accuracy and robustness meet actual monitoring requirements, can adapt to various non-standard contour shapes, and can effectively handle deformation monitoring tasks of reinforced concrete beams in complex backgrounds. Attached Figure Description

[0028] 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.

[0029] Figure 1 A flowchart illustrating an automatic monitoring method for the full-area deformation of a reinforced concrete beam provided in an embodiment of the present invention;

[0030] Figure 2 This is a structural schematic diagram of an automatic monitoring system for the full-area deformation of reinforced concrete beams provided in an embodiment of the present invention. Detailed Implementation

[0031] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0032] 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.

[0033] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent.

[0034] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0035] 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.

[0036] Reference manual attached Figure 1 The diagram shows a flowchart of an automatic monitoring method for the full-area deformation of reinforced concrete beams provided by an embodiment of the present invention.

[0037] This invention provides a method for automatic monitoring of the full-area deformation of reinforced concrete beams. This method can be implemented using an automatic monitoring device for the full-area deformation of reinforced concrete beams, which can be a terminal or a server. The processing flow of the automatic monitoring method for the full-area deformation of reinforced concrete beams may include the following steps:

[0038] S1: Obtain the original image of the reinforced concrete beam.

[0039] Alternatively, an industrial camera or other image acquisition device can be used to acquire raw images of the reinforced concrete beam.

[0040] In this embodiment of the invention, by using an industrial camera or other image acquisition device to acquire the original image of the reinforced concrete beam, the accuracy, efficiency, and adaptability of image acquisition can be improved, while reducing errors from manual operation, ensuring that subsequent image processing and deformation monitoring processes are more accurate and efficient. This provides more reliable technical support for structural health monitoring, intelligent construction, and preventative maintenance.

[0041] S2: Input the original image into the image segmentation model and output a binarized mask image including the reinforced concrete beam.

[0042] Specifically, the image segmentation model is a deep neural network-based image segmentation model.

[0043] Alternatively, the deep neural network may specifically be: U-Net, Mask R-CNN, or SAM.

[0044] U-Net is a deep neural network architecture for image segmentation, originally designed for medical image segmentation. It employs a symmetric encoder-decoder structure, where the encoder extracts image features, and the decoder restores the image's spatial resolution through upsampling operations. U-Net's unique feature is that it directly passes the feature maps from the encoder to the decoder via skip connections, thus preserving high-resolution spatial information while recovering image details. This allows U-Net to effectively handle pixel-level segmentation tasks, especially when data is limited.

[0045] Mask R-CNN is an extension of Region Convolutional Neural Network (R-CNN) for simultaneous object detection and image segmentation. It adds a branch to Faster R-CNN to generate a pixel-level mask for each detected object. Mask R-CNN first performs Region Proposal Network (RPN) generation, then performs classification, bounding box regression, and mask generation on each proposal. It can accurately generate a pixel-level segmentation mask for each object while detecting it, making it suitable for complex instance segmentation tasks.

[0046] SAM (Segment Anything Model) is a general-purpose image segmentation model based on deep learning, launched by Meta (formerly Facebook). SAM is designed to handle various types of image segmentation tasks, including semantic segmentation, instance segmentation, and panoptic segmentation. Combining large-scale pre-trained data with powerful image segmentation capabilities, SAM can automatically identify and segment object boundaries in any input image. This model can perform high-quality segmentation with minimal user interaction and has demonstrated excellent performance in applications across different domains.

[0047] It should be noted that U-Net, Mask R-CNN, and SAM are all existing technologies, and will not be described in detail here.

[0048] In this embodiment of the invention, inputting the original image into a deep neural network image segmentation model can automatically, accurately, and efficiently extract the structure of reinforced concrete beams, reducing human intervention, improving monitoring accuracy, enhancing model adaptability, and supporting real-time monitoring. Deep learning segmentation models (such as U-Net, Mask R-CNN, and SAM) not only improve processing efficiency but also provide robust segmentation performance in various complex environments, providing a reliable foundation for subsequent deformation monitoring, structural evaluation, and other tasks.

[0049] S3: Determine the center point of the beam based on the foreground region in the binarized mask image.

[0050] It should be noted that the foreground region refers to the set of pixels corresponding to the reinforced concrete beam in the binarized mask image; these pixels typically represent the actual structure of the beam. The background region, on the other hand, refers to areas unrelated to the beam, and its pixel values ​​are set to 0.

[0051] Specifically, the foreground area is:

[0052] F={(x j ,y j )|M(x j ,y j )=1,j=1,2,…,M}

[0053] Where F represents the foreground region, x j The x-coordinate of the j-th pixel in the foreground region is represented by y. j Let represent the ordinate of the j-th pixel in the foreground region, and M represent the total number of pixels in the foreground region.

[0054] In an embodiment of the present invention,

[0055] In one possible implementation, S3 specifically refers to:

[0056] The center point of the beam is determined using the following formula:

[0057]

[0058] Where x0 represents the abscissa of the center point of the beam and y0 represents the ordinate of the center point of the beam.

[0059] In this embodiment of the invention, the center point of the beam is calculated by a weighted average of pixels in the foreground region, providing a high-precision, fast, and automated method for center point localization. This not only avoids errors caused by manual intervention but also adapts to beams of various shapes and deformations, thus providing a reliable foundation for subsequent deformation monitoring. This method significantly improves the accuracy, automation, and robustness of the monitoring system, making it an indispensable part of structural health monitoring.

[0060] S4: By using edge detection and contour extraction algorithms, the edges of the foreground region are identified, and the set of contour points outside the beam is extracted.

[0061] Edge detection algorithms are image processing techniques designed to identify regions in an image where brightness changes significantly, often corresponding to object boundaries. These algorithms analyze pixel intensity variations to pinpoint the areas of fastest change in the image, thus marking the edges. Common edge detection algorithms include Canny edge detection, the Sobel operator, and the Prewitt operator. Canny edge detection is a classic algorithm that effectively extracts edge information from images through multi-stage processing, including noise removal, gradient calculation, non-maximum suppression, and edge connectivity.

[0062] Contour extraction algorithms are used to identify and extract the contours of boundaries or objects from images, typically applied after edge detection. These algorithms obtain the geometric shape information of objects by connecting edge points in an image to form continuous closed curves. Common contour extraction algorithms include the `findContours` function in OpenCV, which extracts a set of contour points from an image and organizes them into a hierarchical structure, supporting further analysis and processing of the contours, such as shape analysis, area calculation, and contour matching. These algorithms are widely used in object recognition, shape analysis, and image segmentation.

[0063] Specifically, the binary mask image is first processed using edge detection algorithms (such as Canny edge detection or the Sobel operator) to identify the edges of the foreground region. Edge detection algorithms analyze pixel intensity changes in the image to extract significant transition points from the background to the foreground, i.e., edges. Next, contour extraction algorithms (such as the findContours function in OpenCV) are used to further process the edge-detected image to extract the boundaries of the foreground region. Contour extraction algorithms group connected edge pixels in the image into closed contour segments and return the coordinate set of all contours. Finally, the extracted beam body contour point set contains the boundary information of the foreground region and can serve as the basis for subsequent analysis (such as corner extraction and deformation calculation).

[0064] Specifically, the set of points on the outer contour of the beam is as follows:

[0065] C={(x i ,y i )|i=1,2,…,N}

[0066] Where C represents the set of points on the outer contour of the beam, x i The x-coordinate of the i-th point on the outer contour of the beam is represented by y. i Let N represent the ordinate of the i-th external contour point of the beam, and N represent the total number of external contour points of the beam.

[0067] In this embodiment of the invention, the outer contour point set of the reinforced concrete beam is extracted using edge detection and contour extraction algorithms. This enables accurate identification of the beam's geometry, removal of irrelevant information, reduction of noise interference, and provides reliable data support for subsequent deformation analysis and corner point extraction. This process not only improves the accuracy of deformation monitoring but also ensures the automation and efficiency of the monitoring system, enabling it to adapt to complex real-world environments and structural forms, thus providing strong technical support for structural health monitoring.

[0068] S5: By using the Douglas-Peucker polygon approximation algorithm, redundant points in the beam's outer contour point set are removed to obtain the candidate corner point set of the beam.

[0069] The Douglas-Peucker polygon approximation algorithm is used to simplify polygons or curves. Its main goal is to reduce the number of points in the polygon or curve by minimizing geometric errors, while preserving its original shape as much as possible. The basic idea of ​​the algorithm is to recursively select the farthest point from each end of the curve and determine whether that point significantly affects the overall shape. If the vertical distance of the point is greater than a given tolerance error, the point is retained, and the remaining part is processed recursively. Otherwise, the point is removed, and the two endpoints are directly connected. By iterating this process, a simplified polygon or curve is finally obtained, reducing the number of points while maintaining the main features of the curve. It is commonly used in map simplification, path optimization, and other applications.

[0070] Specifically, the Douglas-Peucker polygon approximation algorithm simplifies the outer contour point set recursively, eliminating redundant points to obtain a simpler contour. The core idea of ​​the algorithm is to approximate the contour curve with the fewest possible polyline segments while preserving the overall shape of the contour. First, starting from the first and last points of the outer contour point set, the algorithm calculates the perpendicular distance from all points to this straight line and finds the point farthest from the line based on a preset error tolerance. If the distance to this point exceeds the tolerance, it is retained and used as the endpoint of a new polyline segment. If the distance is less than the tolerance, the point is considered to have little impact on the contour shape and can be eliminated. This process is iterated recursively until all points are simplified to the fewest possible polyline segments. Finally, the resulting simplified point set is the candidate corner point set for the beam, containing the points that have the greatest impact on the beam's shape; these points accurately represent the main features of the beam.

[0071] Specifically, the candidate corner point set of the beam is as follows:

[0072] P={(x1′,y1′),(x2′,y2′),…,(x′ K ,y′ K )}, 4≤K≤N

[0073] Where P represents the set of candidate corner points of the beam, x ′ k Let y′ represent the x-coordinate of the k-th candidate corner point of the beam. k Let represent the ordinate of the k-th candidate corner point of the beam, where k = 1, 2, ..., K, and K represents the total number of candidate corner points of the beam.

[0074] In this embodiment of the invention, redundant points are eliminated and the set of points on the outer contour of the beam is simplified by using the Douglas-Peucker polygon approximation algorithm. This not only reduces the amount of data and improves computational efficiency, but also preserves the main shape features of the beam, ensuring the accuracy of subsequent corner point extraction and deformation monitoring. This algorithm is highly adaptable, capable of handling complex structural deformations, and effectively improves the automation and robustness of the entire monitoring system.

[0075] S6: With the objectives of minimizing the contour reconstruction error and minimizing the number of beam corner points, a heuristic optimization algorithm is used to determine multiple target beam corner points in the candidate beam corner point set.

[0076] Optionally, the heuristic optimization algorithm may specifically be a genetic algorithm, a particle swarm optimization algorithm, or a simulated annealing algorithm.

[0077] Genetic algorithms, or genetic algorithms, are optimization algorithms that simulate natural selection and genetic mechanisms to find optimal solutions by mimicking biological evolution. Their basic steps include selection, crossover, mutation, and evaluation. In a genetic algorithm, each solution is represented as an "individual" or "chromosome." New solutions are generated by selecting superior individuals for crossover and mutation, and the quality of each solution is evaluated based on a fitness function. After multiple generations of evolution, the fitness of the population gradually increases, eventually finding the optimal solution. Genetic algorithms are suitable for various optimization problems, especially performing well in situations with large or complex solution spaces.

[0078] Particle Swarm Optimization (PSO) is an optimization algorithm that simulates the foraging behavior of bird flocks. In PSO, each solution in the solution space is considered a "particle," and each particle has a position and velocity. Particles search for the optimal solution by continuously updating their position and velocity. Each particle adjusts its position not only based on its own historical experience but also on information about the optimal solution in the swarm. Through cooperation and information exchange among particles, PSO can efficiently search the solution space and achieve good results in various optimization problems.

[0079] Simulated annealing is a stochastic search-based optimization algorithm inspired by the physical annealing process. The algorithm simulates the atomic motion during metal cooling, gradually reducing the "temperature" of the search process and thus decreasing the range of solutions. Initially, the algorithm allows for poor solutions to avoid getting trapped in local optima. As the temperature decreases, the algorithm gradually converges to the global optimum. Simulated annealing is particularly suitable for solving large-scale or complex combinatorial optimization problems, and it is especially effective in handling optimization problems with multiple local optima.

[0080] It should be noted that genetic algorithms, particle swarm optimization algorithms, and simulated annealing algorithms are all existing technologies, and will not be elaborated upon in this invention.

[0081] In this embodiment of the invention, heuristic optimization algorithms (such as genetic algorithms, particle swarm optimization, and simulated annealing) are used to minimize the contour reconstruction error and the number of corner points. While maintaining the accuracy of the beam's geometry, the number and quality of corner points are optimized. This method improves the accuracy of corner point extraction, reduces redundant points, and enhances the robustness of the system, providing efficient and accurate data support for subsequent deformation monitoring and structural health assessment.

[0082] In one possible implementation, S6 specifically includes sub-steps S601 and S602:

[0083] S601: Construct a dual-objective fitness function with the objectives of minimizing contour reconstruction error and minimizing the number of beam corner points:

[0084] Fitness = w1·E rec +w2·N pts

[0085]

[0086] Where Fitness represents the dual-objective fitness function, w1 represents the weighting coefficient of the contour reconstruction error, and E rec The curve represents the contour reconstruction error, w2 represents the weighting coefficient for the number of corner points of the beam, and N represents the curve reconstruction error. pts The number of corner points of the beam is represented by 'l', 'L' represents the set of sides that form a closed quadrilateral, and 'dist' represents the number of corner points. 2 (x i ,l) represents the vertical distance from the i-th point on the outer contour of the beam to edge l.

[0087] S602: Based on the bi-objective fitness function, a heuristic optimization algorithm is used to determine multiple target beam corner points in the candidate beam corner point set. The target beam corner points include the upper left corner point A, the upper right corner point B, the lower right corner point C, and the lower left corner point D.

[0088] In this embodiment of the invention, by constructing a dual-objective fitness function and using a heuristic optimization algorithm to simultaneously minimize the contour reconstruction error and the number of beam corner points, the accuracy of the structural shape can be guaranteed while reducing redundant corner points, thereby improving subsequent computational efficiency and the accuracy of deformation monitoring. The heuristic optimization algorithm ensures the robustness and accuracy of corner point extraction through global search. Overall, this optimization method not only improves the level of automation and intelligence but also enhances the adaptability and stability of the system, making it particularly suitable for large-scale, complex structural health monitoring tasks.

[0089] S7: Based on the center point of the beam and the corner points of each target beam, the binarized mask image is divided into multiple edge regions using the center radial line division method.

[0090] The central radial line segmentation method is an image region segmentation method based on the geometric center and radial rays. In this method, the geometric center of the image is first determined as a reference point. Then, rays are emitted from the center point to the four corner points of the image, dividing the entire image into multiple fan-shaped regions. These rays divide the image edges into different regions, each corresponding to a specific boundary or structural part. This method is often used to segment complex structures, effectively dividing the image into regions based on the radial distribution at the center point, facilitating subsequent edge recognition, analysis, and processing.

[0091] In this embodiment of the invention, by using the central radial line segmentation method, complex reinforced concrete beam structure images can be efficiently and accurately divided into multiple edge regions, thereby providing more accurate basic data for subsequent deformation analysis, corner point extraction, etc. This method is not only automated and robust with strong adaptability, but also improves the accuracy and processing efficiency of monitoring. Especially in large-scale structural monitoring and real-time monitoring scenarios, it can significantly improve the performance and reliability of the system.

[0092] Optionally, the edge regions include the upper edge region AOB, the right edge region BOC, the lower edge region COD, and the left edge region AOD.

[0093] In one possible implementation, S7 specifically includes sub-steps S701 to S705:

[0094] S701: Using the center point of the beam as the origin O of the binary mask image, four rays are emitted sequentially to the upper left corner A, the upper right corner B, the lower right corner C, and the lower left corner D of the beam to obtain the first ray OA, the second ray OB, the third ray OC, and the fourth ray OD.

[0095] Optionally, the first ray OA is specifically:

[0096]

[0097] Among them, y OA,z Let x represent the ordinate of the z-th point on the first ray OA. A The x-coordinate of point A, the upper left corner of the beam, is represented by y. A Let x0 represent the ordinate of the upper left corner point A of the beam, x0 represent the abscissa of the center point of the beam, and y0 represent the ordinate of the center point of the beam. OA,z Let x represent the x-coordinate of the z-th point on the first ray OA.

[0098] Optionally, the second ray OB is specifically:

[0099]

[0100] Among them, y OB,z′ Let x represent the ordinate of the z′-th point on the second ray OB. B The x-coordinate of point B at the upper right corner of the beam is represented by y. B This represents the ordinate of point B at the upper right corner of the beam, X. OB,z′ This represents the x-coordinate of the z′-th point on the second ray OB.

[0101] Optionally, the third ray OC is specifically:

[0102]

[0103] Among them, y OC,z″ Let x represent the ordinate of the z″-th point on the third ray OC. C The x-coordinate of point C at the lower right corner of the beam is represented by y. C This represents the ordinate of the lower right corner point C of the beam, x. OC,z″ Let x represent the x-coordinate of the z″-th point on the third ray OC.

[0104] Optionally, the fourth ray OD is specifically:

[0105]

[0106] Among them, y OD,z″′ Let x represent the ordinate of the z″′-th point on the fourth ray OD. D The x-coordinate of point D at the lower left corner of the beam is represented by y. D This represents the ordinate of the lower left corner point D of the beam, x. OD,z″′ This represents the x-coordinate of the z″′-th point on the fourth ray OD.

[0107] S702: The region between the first ray OA and the second ray OB is taken as the upper edge region AOB.

[0108] S703: The region between the second ray OB and the third ray OC is taken as the right edge region BOC.

[0109] S704: The region between the third ray OC and the fourth ray OD is taken as the lower edge region COD.

[0110] S705: The region between the first ray OA and the fourth ray OD is used as the left edge region AOD to complete the division of the binarized mask image.

[0111] In this embodiment of the invention, by using the center-radial line segmentation method, the binarized mask image of a reinforced concrete beam can be effectively divided into multiple edge regions, thereby simplifying the image's geometry and improving edge recognition accuracy. This method not only improves the accuracy and automation of deformation analysis but also enhances the system's adaptability to different beam shapes, environmental disturbances, and deformations. Overall, the center-radial line segmentation method provides an efficient, accurate, and robust structural monitoring solution, demonstrating significant advantages, especially in large-scale and complex environments.

[0112] S8: Divide each outer contour point in the beam's outer contour point set into the corresponding edge region.

[0113] Specifically, firstly, all outer contour points in the beam's outer contour point set are traversed, and the edge region to which each point belongs is determined based on its coordinate value. According to preset conditions, the edge region of each outer contour point is determined one by one: points on the first ray OA, second ray OB, third ray OC, and fourth ray OD with the same abscissa as the outer contour point are selected, and their ordinates are compared. When the ordinate of an outer contour point is simultaneously greater than or equal to y... OC and y OD When the outer contour point is located in the lower edge region, it is determined that the outer contour point is located in the lower edge region. This occurs when the ordinate of the outer contour point is simultaneously less than or equal to y. OA and y OB When the point is located in the upper edge region, it is determined that the point is within the upper edge region. When the ordinate of the outer contour point is greater than or equal to y... OA And less than or equal to y OD When the point is located in the left edge region, it is determined that the point is located in the left edge region. When the ordinate of the outer contour point is greater than or equal to y... OB And less than or equal to y OC When the point is located, it is determined that it is in the right edge region. Through these determinations, each outer contour point is accurately assigned to its corresponding edge region.

[0114] In this embodiment of the invention, by dividing each outer contour point into a corresponding edge region, accurate edge identification and deformation analysis can be achieved, improving the accuracy and efficiency of deformation monitoring. The automated division process avoids manual intervention and errors, adapts to complex beam deformation situations, and provides high-quality data support for subsequent tasks such as health assessment and deformation calculation. This makes the monitoring system more efficient and intelligent, providing accurate data and decision support in real-time, complex structural monitoring.

[0115] S9: Extract all outer contour points in the edge region and construct a linear contour curve.

[0116] In this embodiment of the invention, by extracting all outer contour points of multiple edge regions and constructing linear contour curves, not only is the accuracy and automation of deformation analysis improved, but the integrity of the contour and the accuracy of the data are also ensured. This method can efficiently and accurately capture the deformation and boundary information of the structure, providing solid data support for subsequent deformation calculations, structural health assessments, and other tasks.

[0117] In one possible implementation, S9 specifically includes sub-steps S901 to S903:

[0118] S901: Extract all outer contour points in the upper or lower edge region.

[0119] S902: Arrange all extracted outer contour points in ascending order of their horizontal coordinates.

[0120] S903: Connect all outer contour points in sequence to construct a linear contour curve.

[0121] In this embodiment of the invention, by extracting all outer contour points of the upper and lower edge regions, arranging them in ascending order of their horizontal coordinates, and connecting them sequentially, it is possible to ensure that the constructed linear contour curve is accurate, continuous, and smooth. This process not only improves the accuracy of deformation analysis and health assessment but also enhances the automation and efficiency of the monitoring system, avoids human error, and provides strong data support for subsequent structural deformation analysis, visualization, and health assessment.

[0122] S10: Calculate the vertical deformation of the reinforced concrete beam based on the linear profile curve and the reference curve from historical monitoring processes, so as to automatically monitor the overall deformation of the reinforced concrete beam.

[0123] In this embodiment of the invention, the vertical global deformation of reinforced concrete beams is calculated by comparing the linear profile curve with a reference baseline curve, providing a high-precision, automated deformation monitoring system. This not only accurately assesses the deformation of the beam and provides early warning of potential structural risks, but also offers reliable technical support for long-term structural health monitoring, maintenance decisions, and intelligent construction.

[0124] In one possible implementation, the formula for calculating the vertical global deformation is as follows:

[0125] ΔY i =y″ i -y′ i

[0126] Where, ΔYi y″ represents the vertical global deformation of the i-th outer contour point in the upper or lower edge region. i Let y′ be the ordinate of the i-th outer contour point in the upper or lower edge region of the current frame's binarized mask image. i It represents the ordinate of the i-th point in the upper or lower edge region of the previous frame's binarized mask image.

[0127] Reference manual attached Figure 2 The diagram shows a structural schematic of an automatic monitoring system for the full-area deformation of reinforced concrete beams provided by the present invention.

[0128] The present invention also provides an automatic monitoring system 20 for the full-area deformation of reinforced concrete beams, applied to the above-mentioned automatic monitoring method for the full-area deformation of reinforced concrete beams, comprising:

[0129] Processor 201;

[0130] The memory 202 stores computer-readable instructions, which, when executed by the processor 201, implement the automatic monitoring method for full-area deformation of reinforced concrete beams as described in the method embodiment.

[0131] The automatic monitoring system 20 for full-area deformation of reinforced concrete beams provided by the present invention can perform the above-mentioned automatic monitoring method for full-area deformation of reinforced concrete beams and achieve the same or similar technical effects. To avoid repetition, the present invention will not elaborate further.

[0132] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0133] In this embodiment of the invention, redundant points in the beam's outer contour point set are eliminated using the Douglas-Peucker polygon approximation algorithm to obtain a candidate corner point set for the beam. With the objectives of minimizing contour reconstruction error and minimizing the number of beam corner points, a heuristic optimization algorithm is used to determine multiple target beam corner points within the candidate corner point set. When facing complex and irregular boundaries, four true structural corner points can be accurately extracted, thereby improving the accuracy of deformation measurement. By extracting all outer contour points from multiple edge regions, a linear contour curve is constructed. Based on the linear contour curve and the reference curve from historical monitoring processes, the vertical global deformation of the reinforced concrete beam is calculated for automatic monitoring of the global deformation of the reinforced concrete beam. The accuracy and robustness meet actual monitoring requirements, can adapt to various non-standard contour shapes, and can effectively handle deformation monitoring tasks of reinforced concrete beams in complex backgrounds.

[0134] It should be understood that the processor in the embodiments of the present invention can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0135] It should also be understood that the memory in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0136] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0137] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0138] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0139] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0140] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0141] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0142] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0143] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0144] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0145] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0146] This invention provides a computer-readable storage medium storing a computer program thereon, characterized in that, when executed by a processor, the program implements the automatic monitoring method for full-area deformation of reinforced concrete beams as described in the method embodiment.

[0147] The present invention provides a computer-readable storage medium that can realize the steps and effects of the automatic monitoring method for full-area deformation of reinforced concrete beams in the above-described method embodiments. To avoid repetition, the present invention will not repeat the details.

[0148] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0149] In this embodiment of the invention, redundant points in the beam's outer contour point set are eliminated using the Douglas-Peucker polygon approximation algorithm to obtain a candidate corner point set for the beam. With the objectives of minimizing contour reconstruction error and minimizing the number of beam corner points, a heuristic optimization algorithm is used to determine multiple target beam corner points within the candidate corner point set. When facing complex and irregular boundaries, four true structural corner points can be accurately extracted, thereby improving the accuracy of deformation measurement. By extracting all outer contour points from multiple edge regions, a linear contour curve is constructed. Based on the linear contour curve and the reference curve from historical monitoring processes, the vertical global deformation of the reinforced concrete beam is calculated for automatic monitoring of the global deformation of the reinforced concrete beam. The accuracy and robustness meet actual monitoring requirements, can adapt to various non-standard contour shapes, and can effectively handle deformation monitoring tasks of reinforced concrete beams in complex backgrounds.

[0150] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0151] The following points need to be explained:

[0152] (1) The accompanying drawings of the embodiments of the present invention only involve the structures involved in the embodiments of the present invention. Other structures can refer to the general design.

[0153] (2) For clarity, the thickness of layers or regions is enlarged or reduced in the drawings used to describe embodiments of the present invention; that is, these drawings are not drawn to actual scale. It is understood that when an element such as a layer, film, region, or substrate is referred to as being “above” or “below” another element, the element may be “directly” located “above” or “below” the other element, or there may be intermediate elements.

[0154] (3) Where there is no conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other to obtain new embodiments.

[0155] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. The scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for automatic monitoring of the full-area deformation of reinforced concrete beams, characterized in that, include: S1: Obtain the original image of the reinforced concrete beam; S2: Input the original image into the image segmentation model and output a binarized mask image including the reinforced concrete beam; S3: Determine the center point of the beam based on the foreground region in the binarized mask image; S4: Using edge detection and contour extraction algorithms, identify the edges of the foreground region and extract the set of contour points outside the beam body; S5: Using the Douglas-Peucker polygon approximation algorithm, redundant points in the beam body outline point set are removed to obtain the candidate corner point set of the beam body; S6: With the objectives of minimizing the contour reconstruction error and minimizing the number of beam corner points, a heuristic optimization algorithm is used to determine multiple target beam corner points in the candidate beam corner point set; S7: Based on the center point of the beam and each of the corner points of the target beam, the binarized mask image is divided into multiple edge regions using the center radial line division method; S8: Divide each outer contour point in the beam body outline point set into the corresponding edge region; S9: Extract all outer contour points in the edge region and construct a linear contour curve; S10: Calculate the vertical global deformation of the reinforced concrete beam based on the linear profile curve and the reference curve from the historical monitoring process, so as to automatically monitor the global deformation of the reinforced concrete beam.

2. The automatic monitoring method for full-area deformation of reinforced concrete beams according to claim 1, characterized in that, The image segmentation model is specifically a deep neural network-based image segmentation model.

3. The automatic monitoring method for full-area deformation of reinforced concrete beams according to claim 1, characterized in that, The foreground region specifically refers to: F={(x j ,y j )|M(x j ,y j )=1,j=1,2,…,M} Where F represents the foreground region, x j The x-coordinate of the j-th pixel in the foreground region is represented by y. j The ordinate of the j-th pixel in the foreground region is represented by , and M represents the total number of pixels in the foreground region. The specific set of points on the outer contour of the beam is as follows: C={(x i ,y i )|i=1,2,…,N} Where C represents the set of points on the outer contour of the beam, x i The x-coordinate of the i-th point on the outer contour of the beam is represented by y. i represents the ordinate of the i-th external contour point of the beam, and N represents the total number of external contour points of the beam. The specific set of candidate corner points for the beam is as follows: P={(x′1,y′1),(x′2,y′2),…,(x′ K ,and' K )},4≤K≤N Where P represents the set of candidate corner points of the beam, x′ k Let y′ represent the x-coordinate of the k-th candidate corner point of the beam. k Let represent the ordinate of the k-th candidate corner point of the beam, where k = 1, 2, ..., K, and K represents the total number of candidate corner points of the beam.

4. The automatic monitoring method for full-area deformation of reinforced concrete beams according to claim 3, characterized in that, Specifically, S3 is: The center point of the beam is determined using the following formula: Where x0 represents the abscissa of the center point of the beam and y0 represents the ordinate of the center point of the beam.

5. The automatic monitoring method for full-area deformation of reinforced concrete beams according to claim 1, characterized in that, S6 specifically includes: S601: Construct a dual-objective fitness function with the objectives of minimizing the contour reconstruction error and minimizing the number of beam corner points: Fitness=w1·E rec +w2·N pts Where Fitness represents the dual-objective fitness function, w1 represents the weighting coefficient of the contour reconstruction error, and E rec The curve represents the contour reconstruction error, w2 represents the weighting coefficient for the number of corner points of the beam, and N represents the curve reconstruction error. pts The number of corner points of the beam is represented by 'l', 'L' represents the set of sides that form a closed quadrilateral, and 'dist' represents the number of corner points. 2 (x i ,l) represents the perpendicular distance from the i-th point on the outer contour of the beam to edge l; S602: Based on the bi-objective fitness function, multiple target beam corner points in the candidate beam corner point set are determined through the heuristic optimization algorithm, wherein the target beam corner points include the upper left corner point A, the upper right corner point B, the lower right corner point C, and the lower left corner point D.

6. The automatic monitoring method for full-area deformation of reinforced concrete beams according to claim 5, characterized in that, The edge regions include the upper edge region AOB, the right edge region BOC, the lower edge region COD, and the left edge region AOD; Specifically, S7 includes: S701: Using the center point of the beam as the origin O of the binary mask image, four rays are emitted sequentially to the upper left corner A, the upper right corner B, the lower right corner C, and the lower left corner D of the beam to obtain the first ray OA, the second ray OB, the third ray OC, and the fourth ray OD. S702: The region between the first ray OA and the second ray OB is taken as the upper edge region AOB; S703: The region between the second ray OB and the third ray OC is taken as the right edge region BOC; S704: The region between the third ray OC and the fourth ray OD is taken as the lower edge region COD; S705: The region between the first ray OA and the fourth ray OD is used as the left edge region AOD to complete the division of the binarized mask image.

7. The automatic monitoring method for full-area deformation of reinforced concrete beams according to claim 6, characterized in that, The first ray OA is specifically: Among them, y OA,z Let x represent the ordinate of the z-th point on the first ray OA. A The x-coordinate of point A, the upper left corner of the beam, is represented by y. A Let x0 represent the ordinate of the upper left corner point A of the beam, x0 represent the abscissa of the center point of the beam, and y0 represent the ordinate of the center point of the beam. OA,z This represents the x-coordinate of the z-th point on the first ray OA; The second ray OB is specifically: Among them, y OB,z′ Let x represent the ordinate of the z′-th point on the second ray OB. B The x-coordinate of point B at the upper right corner of the beam is represented by y. B This represents the ordinate of point B at the upper right corner of the beam, x. OB,z′ This represents the x-coordinate of the z′-th point on the second ray OB; The third ray OC is specifically: Among them, y OC,z″ Let x represent the ordinate of the z″-th point on the third ray OC. C The x-coordinate of point C, the lower right corner of the beam, is represented by y. C This represents the ordinate of the lower right corner point C of the beam, x. OC,z″ This represents the x-coordinate of the z″-th point on the third ray OC; The fourth ray OD is specifically: Among them, y OD,z″′ Let x represent the ordinate of the z″′-th point on the fourth ray OD. D The x-coordinate of point D at the lower left corner of the beam is represented by y. D This represents the ordinate of the lower left corner point D of the beam, x. OD,z″′ This represents the x-coordinate of the z″′-th point on the fourth ray OD.

8. The automatic monitoring method for full-area deformation of reinforced concrete beams according to claim 6, characterized in that, S9 specifically includes: S901: Extract all outer contour points in the upper edge region or the lower edge region; S902: Arrange all extracted outer contour points in ascending order of their horizontal coordinates; S903: Connect all outer contour points sequentially according to their arrangement order to construct the linear contour curve.

9. The automatic monitoring method for full-area deformation of reinforced concrete beams according to claim 1, characterized in that, The specific formula for calculating the vertical global deformation is as follows: ΔY i =y″ i -y′ i Where, ΔY i y″ represents the vertical global deformation of the i-th outer contour point in the upper or lower edge region. i Let y′ be the ordinate of the i-th outer contour point in the upper or lower edge region of the current frame's binarized mask image. i It represents the ordinate of the i-th point in the upper or lower edge region of the previous frame's binarized mask image.

10. An automatic monitoring system for the full-area deformation of reinforced concrete beams, characterized in that, include: processor; A memory storing computer-readable instructions, which, when executed by the processor, implement the automatic monitoring method for full-area deformation of reinforced concrete beams as described in any one of claims 1 to 9.

Images