Real-time monitoring system for arch axis of arch bridge based on big data

Abstract

The invention provides an arch bridge axis real-time monitoring system based on big data, and the system comprises a plurality of laser scanners which are respectively arranged at a plurality of sites and are used for scanning an arch bridge under the current construction progress, and obtaining a plurality of pieces of point cloud data; the arch axis extraction equipment is connected with the plurality of laser scanners and is used for extracting a part of arch axes of the arch bridge in the current construction state from the point cloud data; the arch axis supplementing equipment is used for determining a supplementing part of the partial arch axis according to a big data technology to obtain a complete arch axis; and the arch axis deviation early warning equipment is used for comparing the complete arch axis with the standard arch axis and giving an alarm when the deviation degree exceeds a preset value. By means of the system, the arch axis of the arch bridge can be monitored in the construction process, whether the arch axis deforms or deviates in the construction process or not is determined, and the safety of the arch bridge construction process and the bridge construction quality are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of bridge monitoring, and in particular relates to a real-time monitoring system for an arch axis of an arch bridge based on big data. Background Art

[0002] Arch bridges are one of the most common types of bridges. As a traditional bridge structure, they have great advantages in spanning capacity and structural efficiency. Therefore, arch bridges are widely seen over rivers or canyons. Arch axis monitoring of arch bridges is of great significance to ensure the safety of bridge structures and construction quality.

[0003] In the related art, the arch axis of the completed arch bridge is generally monitored to determine the deformation of the arch axis during use. In fact, the deformation of the arch axis caused by the construction of the arch bridge will not only increase the difficulty of construction, but also have a more far-reaching impact on the subsequent safety of the arch bridge. Summary of the invention

[0004] In view of this, an object of the present invention is to provide a real-time monitoring system and device for the arch axis of an arch bridge based on big data, so as to meet the demand for monitoring the deformation of the arch axis during the construction process.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] The present invention provides a real-time monitoring system for the arch axis of an arch bridge based on big data, comprising: a plurality of laser scanners, respectively arranged at a plurality of sites, for scanning the arch bridge under the current construction progress to obtain a plurality of point cloud data; an arch axis extraction device, connected to the plurality of laser scanners, for extracting a part of the arch axis of the arch bridge under the current construction state from the point cloud data; an arch axis filling device, for determining the filling part of the part of the arch axis according to the big data technology to obtain the complete arch axis; an arch axis deviation early warning device, for comparing the complete arch axis with the standard arch axis, and issuing an alarm when the deviation exceeds a preset value.

[0007] Optionally, the arch axis line completion device includes: a first big data processing module, used to filter the completion data of arch bridges of the same type in the database, and extract the partial arch axis line of the arch bridge under the current construction progress; a first complete arch axis line determination module, used to match the partial arch axis lines of multiple arch bridges of the same type under the current construction progress with the partial arch axis lines in the current construction state, so as to use the arch axis line of the completed arch bridge of the same type with the highest matching accuracy as the complete arch axis line.

[0008] Optionally, a real-time monitoring system for the arch axis of an arch bridge based on big data also includes: an image acquisition device for collecting the current construction picture; a construction material receiving device for receiving the preset construction materials and their usage for each step input by the user; an arch axis completion device, including: a construction progress analysis module for analyzing the current construction picture and determining the completed construction steps; a construction step impact determination module for determining the historical construction data of multiple arch bridges of the same type in the database using big data technology, and analyzing the various construction steps that affect the deformation of the arch axis from the current construction step to the bridge completion process based on the historical construction data of multiple arch bridges of the same type; a second big data processing module for adjusting the various construction steps that affect the deformation of the arch axis in a non-essential construction sequence under the restriction of a necessary construction sequence to obtain multiple construction path data sets, and adjusting the digital twin model of the arch bridge under the current construction step based on the multiple construction path data sets and the construction materials and their usage corresponding to each construction step, and predicting the completed arch axis corresponding to the multiple construction path data sets; a second complete arch axis determination module for merging the multiple completed parts and the partial arch axis into multiple complete arch axis lines.

[0009] Optionally, a real-time monitoring system for arch axis lines of arch bridges based on big data also includes: an optimal matching degree confirmation module, which is used to match multiple complete arch axis lines with a standard arch axis line, and determine the matching degree of multiple complete arch axis lines with the standard arch axis line; a construction sequence screening module, which is used to select the complete arch axis line with the highest matching degree, and retrieve the construction path data set corresponding to the complete arch axis line as a target construction path data set, and search for the next construction step of the current construction step in the target construction path data set; a construction suggestion display module, which is used to display the next construction step of the current construction step.

[0010] Optionally, the image acquisition device includes: a panoramic camera for acquiring the overall construction picture of the bridge; a construction change determination module for determining the construction change position in two overall construction pictures taken at adjacent time intervals; a detail picture acquisition device mobilization module for determining the target detail picture acquisition device and the rotation angle according to the construction change position and the shooting range of each detail picture acquisition device, and rotating the target detail picture acquisition device according to the rotation angle, and shooting the construction detail picture; a construction progress analysis module including: a progress verification table extraction module for extracting the progress verification table, which contains multiple levels of classification, and each level of classification contains a classification description; a picture description module for describing multiple construction detail pictures according to a pre-trained neural network model, and the description vocabulary is a vocabulary preset in the vocabulary library; a key vocabulary matching module for capturing keywords in the picture description, and performing multi-level matching of the keywords with the classification description in the progress verification table, and when the keywords match the corresponding classification description in the progress verification table, the classification description in the progress verification table is marked as completed.

[0011] Optionally, the arch axis extraction device includes: a target color calibration device, which is used to perform target color calibration at the arch foot of the arch bridge and the highest point of the arch in the current construction state; a panoramic camera, which is used to shoot the overall construction picture of the bridge; extracting part of the arch axis of the arch bridge in the current construction state from the point cloud data, including: splicing multiple point cloud data to obtain a three-dimensional point cloud model of the arch bridge in the current construction state; combining the three-dimensional point cloud model of the arch bridge in the current construction state with the photographed overall construction picture of the bridge to obtain a three-dimensional point cloud model with color; extracting the target color calibration position from the three-dimensional point cloud model with color, taking any target color calibration position as the starting point, and extracting the corresponding line segment in the three-dimensional point cloud model; judging whether the end point of the line segment taking any target color calibration position as the starting point is the target color calibration position, and if so, the line segment is part of the arch axis in the current construction state.

[0012] Optionally, the complete arch axis is compared with the standard arch axis, including: dividing the line segment formed by the two endpoints of the complete arch axis and the standard arch axis respectively according to the target equal division to obtain multiple equidistant nodes; extending a vertical line upward from each equidistant node on the line segment formed by the two endpoints of the complete arch axis to obtain multiple intersection points of the vertical lines with the complete arch axis as the first intersection point; extending a vertical line upward from each equidistant node on the line segment formed by the two endpoints of the standard arch axis to obtain multiple intersection points of the vertical lines with the standard arch axis as the second intersection point; calculating the average value of the distance between each first intersection point and the corresponding second intersection point, and using the average value as a factor of the degree of deviation.

[0013] The present embodiment provides a real-time monitoring system for the arch axis of an arch bridge based on big data. The system scans the arch bridge under the current construction progress through multiple laser scanners to obtain multiple point cloud data, and extracts the partial arch axis of the arch bridge under the current construction state from the point cloud data. The arch axis completing device determines the completed part of the partial arch axis to obtain the complete arch axis. Then, the complete arch axis is compared with the standard arch axis. When the deviation exceeds the preset value, an alarm is issued. The system can monitor the arch axis of the arch bridge during the construction process, determine whether the arch axis is deformed or offset during the construction process, and improve the safety of the arch bridge construction process and the quality of the bridge construction.

[0014] Other advantages, objectives and features of the present invention will be described in the following description and will be apparent to those skilled in the art to some extent, or those skilled in the art may be taught from the practice of the present invention. The objectives and other advantages of the present invention may be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to make the purpose, technical solution and beneficial effects of the present invention clearer, the present invention provides the following drawings for illustration:

[0016] Figure 1 It is a structural schematic diagram of a real-time monitoring system of an arch axis of an arch bridge based on big data in the present invention;

[0017] Figure 2 This is a schematic diagram of the first intersection in the present invention. DETAILED DESCRIPTION

[0018] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, it can also be the internal connection of two components, it can be a wireless connection, or it can be a wired connection. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0020] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0021] The embodiment of the present invention provides a real-time monitoring system of the arch axis of an arch bridge based on big data, such as Figure 1 As shown, including:

[0022] A plurality of laser scanners 101 are respectively arranged at a plurality of sites, and are used to scan the arch bridge under the current construction progress to obtain a plurality of point cloud data;

[0023] The arch axis extraction device 102 is connected to the plurality of laser scanners and is used to extract part of the arch axis of the arch bridge in the current construction state from the point cloud data;

[0024] The arch axis filling device 103 is used to determine the filling part according to the big data technology, and merge the filling part and the partial arch axis into a complete arch axis;

[0025] The arch axis deviation warning device 104 is used to compare the complete arch axis with the standard arch axis, and issue an alarm when the deviation exceeds a preset value.

[0026] For example, the laser scanner 101 may be a three-dimensional laser scanner, and scanning stations are set at multiple locations to obtain multi-directional point cloud data of the arch bridge under the current construction progress. The arch axis extraction device 102 may be any device with an arch axis extraction function, such as a mobile phone, a computer, etc., which is not limited in this embodiment.

[0027] The arch axis extraction device 102 first splices the multi-directional point cloud data to obtain a three-dimensional point cloud model of the arch bridge in the current construction state. The specific method of extracting part of the arch axis of the arch bridge in the current construction state from the point cloud data in this embodiment can be first, the arch axis extraction device receives a small amount of manually calibrated point cloud data and the original designed arch axis formula input by the user. The small amount of point cloud data can be a partial arch axis point cloud based on manual judgment, and then extracts part of the arch axis in the three-dimensional point cloud model according to the boundary point constraint condition based on the small amount of manually calibrated point cloud data. The boundary point constraint condition can be that the angle between the connecting line between the calibrated point cloud and the adjacent point cloud and the slope of the curve represented by the original designed arch axis formula of the calibrated point cloud is less than a preset angle. The preset angle can be 20 degrees, which is not limited in this embodiment.

[0028] The arch axis completion device 103 uses big data technology to predict the missing arch axis part and obtain the complete arch axis. Specifically, a machine learning model can be used to learn a large number of bridge images of the same type of arch bridges at various construction stages, as well as arch bridge images under the completed bridge, so that when a bridge image at any construction stage is input, the corresponding completed arch bridge can be output. The machine learning model can use a multi-layer perceptron (MLP) and a long short-term memory (LSTM) network to learn from the data, form a nonlinear mapping between the bridge image at any construction stage and the completed arch bridge, and make full use of their capabilities to handle complex nonlinear problems.

[0029] The arch axis deviation warning device 104 can be an alarm device placed at the construction site, or it can be a mobile device of the construction manager. This embodiment does not limit the arch axis deviation warning device. It should be noted that the arch axis extraction device 102, the arch axis completion device 103 and the arch axis deviation warning device 104 can be the same device with computing capabilities or different devices with computing capabilities. The arch axis deviation warning device 104 compares the complete arch axis with the standard arch axis during design, and the comparison method can be the Frechet distance. If the deviation between the actual arch axis and the standard arch axis exceeds the preset value, the device will sound an alarm to prompt the construction team to make adjustments to ensure that the construction of the arch bridge meets the design requirements.

[0030] As an optional implementation, the complete arch axis is compared with the standard arch axis, including: dividing the line segment formed by the two endpoints of the complete arch axis and the standard arch axis into equal parts according to the target to obtain multiple equidistant nodes; extending a vertical line upward from each equidistant node on the line segment formed by the two endpoints of the complete arch axis to obtain multiple intersections of the vertical lines with the complete arch axis as the first intersection; extending a vertical line upward from each equidistant node on the line segment formed by the two endpoints of the standard arch axis to obtain multiple intersections of the vertical lines with the standard arch axis as the second intersection; calculating the average value of the distance between each first intersection point and the corresponding second intersection point, and using the average value as a factor of the degree of deviation.

[0031] like Figure 2As shown in the figure, it is a schematic diagram of the first intersection. The arc in the figure is the complete arch axis. The line segment formed by the two endpoints of the complete arch axis is divided into eight equal parts to obtain eight first intersections. Similarly, the line segment formed by the two endpoints of the standard arch axis is divided into eight equal parts to obtain eight second intersections. Thus, the distance between the extreme first intersection and the second intersection can be calculated by taking the standard arch axis and the complete arch axis as the origin, establishing a coordinate system, and obtaining the coordinates of each first intersection and each second intersection. According to the coordinates of the first intersection and the second intersection, the Euclidean distance is calculated, and then the average value of the Euclidean distance is determined based on multiple Euclidean distance calculation results. The degree of deviation is evaluated by the average value. The higher the average value, the higher the degree of deviation, and the lower the average value, the lower the degree of deviation.

[0032] The present embodiment provides a real-time monitoring system for the arch axis of an arch bridge based on big data. The system scans the arch bridge under the current construction progress through multiple laser scanners to obtain multiple point cloud data, and extracts the partial arch axis of the arch bridge under the current construction state from the point cloud data. The arch axis completing device determines the completed part of the partial arch axis to obtain the complete arch axis. Then, the complete arch axis is compared with the standard arch axis. When the deviation exceeds the preset value, an alarm is issued. The system can monitor the arch axis of the arch bridge during the construction process, determine whether the arch axis is deformed or offset during the construction process, and improve the safety of the arch bridge construction process and the quality of the bridge construction.

[0033] As an optional implementation, the arch axis filling device includes:

[0034] The first big data processing module is used to filter the data of completed arch bridges of the same type in the database and extract the arch axis of the arch bridge under the current construction progress;

[0035] The first complete arch axis determination module is used to match the partial arch axis lines of multiple arch bridges of the same type under current construction progress with the partial arch axis lines under current construction status, so as to match the arch axis line of the same type of arch bridge with the highest matching accuracy as the complete arch axis line.

[0036] Exemplarily, the first data processing module screens the data of completed arch bridges of the same type in the database, and extracts the arch axis of the completed arch bridges of the same type. Specifically, first, the abnormal data is eliminated, the sample database is established, and the data is subjected to feature standardization processing to eliminate the influence of the sample data dimension; then, the database management system is used to screen out the same type of arch bridges according to the design parameters of the arch bridge (such as arch shape, material, construction method, etc.), and obtain its construction process image, find the arch bridge image corresponding to the current construction progress, and use image processing technology to extract part of the arch axis from the arch bridge image. Among them, the same type of arch bridge can be the result of classification according to a single method, or it can be classified according to multiple methods, and the cross-classification results obtained, for example, according to a single classification method, if classified by the form of arch building, including solid arch bridge, hollow arch bridge, combined system arch bridge, classified by the type of arch axis, including circular arch bridge, parabolic arch bridge, catenary arch bridge, classified by different building materials, including stone arch bridge, concrete arch bridge and steel arch bridge. The cross-classification results can be obtained by classifying in a variety of ways, and can be cross-classified by any two or three of the arch building, the arch axis type, and the building material. This embodiment does not limit the specific classification, and those skilled in the art can determine it according to needs.

[0037] The first complete arch axis determination module will match the partial arch axis lines of multiple arch bridges of the same type under the current construction progress with the partial arch axis lines under the current construction state, so as to match the arch axis line of the same type of arch bridge with the highest matching accuracy as the complete arch axis line.

[0038] The present embodiment provides a real-time monitoring system for the arch axis of an arch bridge based on big data. The first big data processing module can screen the completion data of arch bridges of the same type and extract part of the arch axis of the arch bridge under the current construction progress. The system uses big data technology for data-driven decision support, which can improve the scientificity and accuracy of decision-making, thereby improving the accuracy and reliability of arch axis extraction.

[0039] As an optional implementation, a real-time monitoring system for an arch axis of an arch bridge based on big data further includes:

[0040] Image acquisition equipment, used to collect the current construction picture;

[0041] A construction material receiving device, used to receive the preset construction materials and usage amounts of each step input by the user;

[0042] Arch axis filling equipment, including:

[0043] The construction progress analysis module is used to analyze the current construction screen and determine the completed construction steps;

[0044] The construction step impact determination module is used to determine the historical construction data of multiple arch bridges of the same type in the database using big data technology, and analyze the various construction steps that affect the deformation of the arch axis from the current construction step to the completion of the bridge based on the historical construction data of multiple arch bridges of the same type;

[0045] The second data processing module is used to adjust the various construction steps that affect the deformation of the arch axis in the non-essential construction sequence under the restriction of the necessary construction sequence, obtain multiple construction path data sets, and adjust the digital twin model of the arch bridge under the current construction step according to the multiple construction path data sets and the construction materials and usage amounts corresponding to each construction step, and predict the completed part of the arch axis corresponding to the multiple construction path data sets;

[0046] The second complete arch axis determination module merges the multiple supplementary parts and partial arch axes into multiple complete arch axes respectively.

[0047] Exemplarily, the image acquisition device may be a high-definition camera, a camera mounted on a drone, or other image acquisition device. The image acquisition device can capture images of the construction site in real time and provide basic data for subsequent analysis. The construction material receiving device may be an intelligent device including a human-computer interaction interface, which may be a software display interface that allows a user to input a list of materials and estimated usage.

[0048] The arch axis completion equipment includes a construction progress analysis module, a construction step impact determination module, a second big data processing module, and a second complete arch axis determination module. The construction progress analysis module can use image recognition and machine learning technology to analyze the current construction picture and determine the completed construction steps. The construction step impact determination module uses big data technology to query the historical construction data of multiple arch bridges of the same type in the database, and by analyzing the historical construction data, it determines which construction steps will affect the deformation of the arch axis from the current construction step to the bridge completion process. It should be noted that all arch bridges in the historical construction data stored in the database meet the construction standards.

[0049] The historical construction data contains the arch axis data of the arch bridge at each construction stage. The specific recording method can be a picture method, that is, taking images of the arch bridge at each construction stage. The method of extracting the arch axis from the arch bridge images taken at each construction stage can be to use machine vision and image recognition algorithms to automatically identify and detect the arch axis of the bridge. Specifically, first obtain a digital image of the arch bridge, and then pre-process the image, including grayscale, edge detection, etc., to extract the arch axis contour information. Then, use morphological processing and feature extraction methods to analyze the arch axis and identify the linear features of the arch axis.

[0050] After the arch axis of each construction stage is obtained from the historical construction data, the changes in the arch axis of each construction stage are recorded, and the stages in which the arch axis changes in adjacent construction stages exceed a preset threshold are marked. Furthermore, the preset thresholds can be classified according to the severity of the changes. For example, multiple preset thresholds are set, the first preset threshold is less than the second preset threshold, and the second preset threshold is less than the third preset threshold, so that the construction stage of the arch axis change is refined. When computing resources are relatively abundant, smaller and more refined preset thresholds are selected to filter the construction stages, so that more construction steps that affect the deformation of the arch bridge can be determined. When computing resources are relatively scarce, larger and coarser preset thresholds can be selected to filter the construction stages to determine fewer construction steps that affect the deformation of the arch bridge.

[0051] The screened construction steps are classified according to whether they can be flexibly adjusted, and are classified into construction steps under a necessary construction sequence and construction steps under a non-necessary construction sequence. The classification method may be to query a pre-stored mapping table, which specifies which steps are fixed steps and which steps are flexibly adjusted steps. The construction steps under the necessary construction sequence may be steps that serve as the main framework, for example, the preliminary assessment and preparation step is earlier than the foundation treatment step, the foundation treatment step is earlier than the bridge slope excavation and infrastructure construction step, the bridge slope excavation and infrastructure construction step is earlier than the abutment and bridge formwork construction step, the abutment and bridge formwork construction step is earlier than the steel bar processing and installation step, and so on. The construction steps under a non-necessary construction sequence may be various construction steps in which various tasks can be carried out simultaneously under the construction steps under any necessary construction sequence. This embodiment does not limit the specific classification results, and those skilled in the art can determine as needed.

[0052] After classifying the construction steps under the necessary construction sequence and the construction steps under the non-essential construction sequence, the construction steps of the non-essential construction sequence can be adjusted under the framework of the construction steps under the necessary construction sequence, thereby obtaining multiple construction path data sets. This embodiment integrates BIM and three-dimensional reconstruction to construct a digital twin model of an arch bridge. Multiple construction path data sets and the construction materials and usage corresponding to each construction step are used as inputs to the digital twin model. The digital twin model processes the input according to the internally constructed neural network model and the pre-embedded physical principles, thereby deducing the arch bridge model under different construction path data sets, thereby extracting the corresponding arch axis completion part from the bridge model.

[0053] The second complete arch axis determination module merges the multiple filled-in parts and partial arch axes into multiple complete arch axis lines respectively. Any arch axis line can be selected as the input of the arch axis line deviation warning device 104, or multiple complete arch axis lines can be used as the input of the arch axis line deviation warning device 104 respectively, and then the average deviation of the multiple complete arch axis lines from the standard arch axis line is obtained as the quantitative result of the deviation degree.

[0054] The present embodiment provides a real-time monitoring system for the arch axis of an arch bridge based on big data. The system analyzes the influence of construction steps on the deformation of the arch bridge, recombines the construction steps, obtains multiple construction path data sets, and simulates the corresponding completed parts of the arch axis of the completed bridge according to the multiple construction path data sets. It is possible to predict multiple arch axis lines, thereby improving the accuracy of arch axis prediction.

[0055] As an optional implementation, a real-time monitoring system for an arch axis of an arch bridge based on big data further includes:

[0056] An optimal matching degree confirmation module is used to match the multiple complete arch axis lines with the standard arch axis lines and determine the matching degree of the multiple complete arch axis lines with the standard arch axis lines;

[0057] The construction sequence screening module is used to select the complete arch axis with the highest matching degree, and retrieve the construction path data set corresponding to the complete arch axis as the target construction path data set, and search for the next construction step of the current construction step in the target construction path data set;

[0058] The construction suggestion display module is used to display the next construction step of the current construction step.

[0059] Exemplarily, the optimal matching degree confirmation module uses a geometric analysis method to calculate the maximum deviation between the two curves to determine the matching degree between the complete arch axis and the standard arch axis. The specific method of calculating the maximum deviation between the two curves can be to use a statistical method, such as calculating the standard deviation or variance of the deviations of each corresponding point on the two curves, so as to quantify the matching degree.

[0060] According to the results of the optimal matching degree confirmation module, all complete arch axes are sorted, and the arch axis with the highest matching degree is selected as the target arch axis. Then, the corresponding construction path data set is retrieved in the database according to the target arch axis. In the target construction path data set, the next construction step is determined according to the current construction step. The construction suggestion display module can be a display interface on a mobile phone or a display interface on a computer terminal, which is used to display the next construction suggestion step.

[0061] The present embodiment provides a real-time monitoring system for the arch axis of an arch bridge based on big data. By simulating different construction path data sets and comparing the corresponding completed arch axis of the bridge with the standard arch axis, the construction path data set corresponding to the optimal completed part is selected. This can effectively guide the user's construction steps and ensure the quality of arch bridge construction while improving construction efficiency.

[0062] As an optional implementation, a real-time monitoring system for an arch axis of an arch bridge based on big data, an image acquisition device, includes:

[0063] Panoramic camera, used to obtain the overall construction picture of the bridge;

[0064] A construction change determination module is used to determine the location of construction changes in two overall construction pictures taken at two adjacent time intervals;

[0065] The detail picture acquisition device mobilization module is used to determine the target detail picture acquisition device and the rotation angle according to the construction change position and the shooting range of each detail picture acquisition device, rotate the target detail picture acquisition device according to the rotation angle, and shoot the construction detail picture;

[0066] Construction progress analysis module, including:

[0067] The progress verification table extraction module is used to extract the progress verification table, which contains multiple levels of classification, and each level of classification contains a classification description;

[0068] A picture description module is used to describe multiple construction detail pictures according to a pre-trained neural network model, and the description vocabulary is a vocabulary preset in the vocabulary library;

[0069] The keyword matching module is used to capture keywords in the screen description and perform multi-level matching between the keywords and the category descriptions in the progress verification table. When the keywords match the corresponding category descriptions in the progress verification table, the category descriptions in the progress verification table are marked as completed.

[0070] Exemplarily, the panoramic camera can be a 360-degree panoramic camera that can provide a comprehensive view of the construction site. It can be fixedly installed at a high point or on a mobile platform at the construction site to take panoramic images of the construction site at regular intervals. The construction change determination module compares the differences between two images and identifies the location of construction changes through image processing technology, such as feature point matching, optical flow method or deep learning model. The detail image acquisition device mobilization module can use a robot or a remotely controlled camera to calculate the position and rotation angle of the detail image acquisition device that needs to be mobilized according to the coordinates of the construction change position, and then remotely control the device to shoot to obtain the construction detail map.

[0071] The construction progress analysis module includes a progress verification table extraction module, a screen description module and a key word matching module, wherein the progress verification table in the progress verification table extraction module can be a predefined construction plan stored in a database, and the progress verification table can be divided into multiple levels according to the construction level, and each level contains a classification description corresponding to the level. Furthermore, the classification description can be processed, the classification description vocabulary can be collected, and a vocabulary library can be established based on the classification description vocabulary.

[0072] The screen description module reads the progress verification form from the database and represents it in a structured form. The screen description module can use a deep learning model for image recognition and description generation (such as CNN+RNN) to extract features and generate descriptions of the construction detail screen. The descriptive words in the training process are words in a pre-established vocabulary library, which is a vocabulary library built based on the descriptive words in the progress verification form. The key vocabulary matching module uses natural language processing technology, such as TF-IDF, Word2Vec or BERT model, to extract keywords from the screen description and match them with the classification description in the progress verification form. After a successful match, the corresponding classification description in the progress verification form is marked as completed.

[0073] This embodiment provides a real-time monitoring system for the arch axis of an arch bridge based on big data. Through the collaborative work of these modules, the bridge construction monitoring system can monitor the construction progress in real time, automatically identify construction changes, and compare them with the construction plan, thereby realizing accurate identification of the construction progress.

[0074] As an optional embodiment, the arch axis extraction device includes:

[0075] Target color calibration equipment, used to calibrate the target color at the arch foot and the highest point of the arch under the current construction status;

[0076] Panoramic camera, used to capture the overall construction picture of the bridge;

[0077] Extracting some arch axes of the arch bridge in the current construction state from the point cloud data includes:

[0078] Multiple point cloud data are spliced ​​together to obtain a three-dimensional point cloud model of the arch bridge under the current construction status;

[0079] The 3D point cloud model of the arch bridge under current construction status is combined with the photographed overall construction picture of the bridge to obtain a 3D point cloud model with color;

[0080] Extracting the target color calibration position from the three-dimensional point cloud model with color, taking any target color calibration position as the starting point, extracting the corresponding line segment in the three-dimensional point cloud model;

[0081] It is determined whether the end point of a line segment starting from any target color calibration position is the target color calibration position. If yes, the line segment is a partial arch axis in the current construction state.

[0082] Exemplarily, the target color calibration device may be a movable device that carries a target object, and the target object may be color-marked on the arch bridge. The target object may be a magnetic block of the target color, or an inkjet device that can spray the target color. The target color calibration device may be a robot that can receive a user's route instruction and move according to the route instruction, and can receive a user's magnetic block placement instruction or inkjet instruction to place a magnetic block or spray the target color at the target location.

[0083] The panoramic camera shoots the construction scene with target color calibration. After obtaining the construction scene with color and the three-dimensional point cloud model, the image coordinates of the target color calibration position are obtained by Canny edge detection and sub-pixel positioning algorithm, and then the improved direct linear transformation method is used to solve the camera's relative world coordinate system from the world coordinate value and pixel coordinate value of the coded marker. The mapping relationship between the point cloud and the color image pixel is established through the camera imaging model. Finally, the color of the local point cloud is calculated by the interpolation method and the color fusion of the entire point cloud is completed to obtain a three-dimensional point cloud model with color. The target color calibration position is extracted from the three-dimensional point cloud model with color. Taking any target color calibration position as the starting point, the corresponding line segment is extracted in the three-dimensional point cloud model. It is judged whether the end point of the line segment starting from any target color calibration position is the target color calibration position. If yes, the line segment is part of the arch axis under the current construction state.

[0084] The present embodiment provides a real-time monitoring system for the arch axis of an arch bridge based on big data. The target color calibration device can perform accurate color calibration at key positions of the arch bridge (arch foot and arch top). The three-dimensional point cloud model is combined with the image data taken by the panoramic camera to obtain a three-dimensional model with color information. The target color calibration position is automatically extracted from the three-dimensional point cloud model with color, and the corresponding line segment is extracted from the model. This process can be automated, reducing manual intervention and improving efficiency. In addition, the arch axis can be accurately identified by judging whether the starting point and end point of the line segment are the target color calibration positions.

[0085] Finally, it should be noted that the above preferred embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail through the above preferred embodiments, those skilled in the art should understand that various changes can be made in form and details without departing from the scope defined by the claims of the present invention.

Claims

1. A real-time monitoring system for the arch axis of an arch bridge based on big data, characterized in that: include: Multiple laser scanners are set up at multiple sites to scan the arch bridge under the current construction progress and obtain multiple point cloud data; An arch axis extraction device, connected to multiple laser scanners, is used to extract part of the arch axis of the arch bridge in the current construction state from the point cloud data; The arch axis filling device is used to determine the filling part of the partial arch axis according to the big data technology to obtain the complete arch axis; The arch axis deviation warning device is used to compare the complete arch axis with the standard arch axis. When the deviation exceeds the preset value, an alarm is issued.

2. The real-time monitoring system of the arch axis of an arch bridge based on big data according to claim 1 is characterized in that: Arch axis filling equipment, including: The first big data processing module is used to filter the data of completed arch bridges of the same type in the database and extract the arch axis of the arch bridge under the current construction progress; The first complete arch axis determination module is used to match the partial arch axis lines of multiple arch bridges of the same type under current construction progress with the partial arch axis lines under current construction status, so as to use the arch axis line of the same type of arch bridge with the highest matching accuracy as the complete arch axis line.

3. The real-time monitoring system of the arch axis of an arch bridge based on big data according to claim 1 is characterized in that: Also includes: Image acquisition equipment, used to collect the current construction picture; A construction material receiving device, used to receive the preset construction materials and usage amounts of each step input by the user; Arch axis filling equipment, including: The construction progress analysis module is used to analyze the current construction screen and determine the completed construction steps; The construction step impact determination module is used to determine the historical construction data of multiple arch bridges of the same type in the database using big data technology, and analyze the various construction steps that affect the deformation of the arch axis from the current construction step to the completion of the bridge based on the historical construction data of multiple arch bridges of the same type; The second data processing module is used to adjust the various construction steps that affect the deformation of the arch axis in the non-essential construction sequence under the restriction of the necessary construction sequence, obtain multiple construction path data sets, and adjust the digital twin model of the arch bridge under the current construction step according to the multiple construction path data sets and the construction materials and usage amounts corresponding to each construction step, and predict the completed part of the arch axis corresponding to the multiple construction path data sets; The second complete arch axis determination module merges the multiple supplementary parts and partial arch axes into multiple complete arch axes.

4. The real-time monitoring system of the arch axis of an arch bridge based on big data according to claim 3 is characterized in that: Also includes: An optimal matching degree confirmation module is used to match the multiple complete arch axis lines with the standard arch axis lines and determine the matching degree of the multiple complete arch axis lines with the standard arch axis lines; The construction sequence screening module is used to select the complete arch axis with the highest matching degree, and retrieve the construction path data set corresponding to the complete arch axis as the target construction path data set, and search for the next construction step of the current construction step in the target construction path data set; The construction suggestion display module is used to display the next construction step of the current construction step.

5. The real-time monitoring system of the arch axis of an arch bridge based on big data according to claim 3 is characterized in that: Image acquisition equipment, including: Panoramic camera, used to obtain the overall construction picture of the bridge; A construction change determination module is used to determine the location of construction changes in two overall construction pictures taken at two adjacent time intervals; The detail picture acquisition device mobilization module is used to determine the target detail picture acquisition device and the rotation angle according to the construction change position and the shooting range of each detail picture acquisition device, rotate the target detail picture acquisition device according to the rotation angle, and shoot the construction detail picture; Construction progress analysis module, including: The progress verification table extraction module is used to extract the progress verification table, which contains multiple levels of classification, and each level of classification contains a classification description; A picture description module is used to describe multiple construction detail pictures according to a pre-trained neural network model, and the description vocabulary is a vocabulary preset in the vocabulary library; The keyword matching module is used to capture keywords in the screen description and perform multi-level matching between the keywords and the category descriptions in the progress verification table. When the keywords match the corresponding category descriptions in the progress verification table, the category descriptions in the progress verification table are marked as completed.

6. The real-time monitoring system of the arch axis of an arch bridge based on big data according to claim 1 is characterized in that: Arch axis extraction equipment, including: Target color calibration equipment, used to calibrate the target color at the arch foot and the highest point of the arch under the current construction status; Panoramic camera, used to capture the overall construction picture of the bridge; Extracting some arch axes of the arch bridge in the current construction state from the point cloud data includes: Multiple point cloud data are spliced ​​together to obtain a three-dimensional point cloud model of the arch bridge under the current construction status; The 3D point cloud model of the arch bridge under current construction status is combined with the photographed overall construction picture of the bridge to obtain a 3D point cloud model with color; Extracting the target color calibration position from the three-dimensional point cloud model with color, taking any target color calibration position as the starting point, extracting the corresponding line segment in the three-dimensional point cloud model; It is determined whether the end point of a line segment starting from any target color calibration position is the target color calibration position. If yes, the line segment is a partial arch axis in the current construction state.

7. The real-time monitoring system of the arch axis of an arch bridge based on big data according to claim 1 is characterized in that: Compare the full arch axis to the standard arch axis, including: The line segments formed by the two endpoints of the complete arch axis and the standard arch axis are divided equally according to the target to obtain a plurality of equidistant nodes; A vertical line is extended upward from each equidistant node on the line segment formed by the two end points of the complete arch axis to obtain a plurality of intersection points of the vertical line and the complete arch axis as the first intersection point; A vertical line is extended upward from each equidistant node on the line segment formed by the two end points of the standard arch axis to obtain a plurality of intersection points of the vertical line and the standard arch axis as the second intersection point; The average value of the distances between each first intersection point and the corresponding second intersection point is calculated, and the average value is used as a factor of the degree of deviation.

Images