A cable-stayed bridge construction monitoring system and method based on machine vision
Through machine vision technology combined with finite element simulation data, the stress and deformation of the bridge tower during cable-stayed bridge construction is monitored in real time, which solves the construction risk problem that traditional methods are difficult to monitor and respond quickly, and achieves the safety and controllability of the construction process.
Patent Information
- Application Number
- CN202410939800.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-15
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-07-15
AI Technical Summary
The prior art is difficult to effectively monitor the stress and deformation of the bridge tower during the construction of cable-stayed bridges, and the traditional methods are complex in operation and have a blank window period, which increases construction risks and makes it difficult to quickly detect and resolve abnormal situations.
The construction monitoring system based on machine vision is adopted, and through the combination of image acquisition, processing and finite element simulation data, the deformation and lifting operations of steel tower components are monitored in real time, monitoring reports are generated and alarms are issued, ensuring effective control of the force and deformation of the bridge tower.
Real-time monitoring of the stress and deformation of the bridge tower is achieved, construction efficiency and safety are improved, and the overall structure is safe and controllable during the lifting process.
Smart Images

Figure CN118482770B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of machine vision, and in particular to a machine vision-based cable-stayed bridge construction monitoring system and method. Background Art
[0002] As a typical composite bridge system, the cable-stayed bridge tower is an important load-bearing and force-transmitting component supporting the overall structure. During the construction phase, the control of the component quality and the control of the tower force and deformation are crucial to the construction quality and personnel safety. However, the traditional monitoring methods currently used are complicated to operate, and there is a blank window period during the measurement, which increases the potential risks in the construction process. In addition, if there are abnormal forces and deformations during the construction phase, it is difficult for traditional methods to quickly find the cause of the abnormality and take timely remedial measures in a short period of time.
[0003] Therefore, the present invention provides a machine vision-based cable-stayed bridge construction monitoring system and method to ensure effective monitoring of the stress and deformation of the bridge tower during the hoisting process, thereby ensuring construction efficiency and safety. Summary of the invention
[0004] The present invention provides a machine vision-based cable-stayed bridge construction monitoring system and method, which are used to obtain target image data by processing the image data of the cable-stayed bridge construction site, and then monitor the deformation of steel tower components, hoisting operations, and the stress and deformation changes of the bridge tower in combination with preset finite element simulation data, and issue an alarm based on the monitoring results to ensure effective monitoring of the stress and deformation of the bridge tower during the hoisting process, thereby ensuring the safety and controllability of the overall structural system during the hoisting construction of the bridge tower segment.
[0005] The present invention provides a machine vision-based cable-stayed bridge construction monitoring system, comprising:
[0006] Image acquisition module: used to take real-time photos of the cable-stayed bridge construction site using a target camera installed in a preset shooting area to obtain first image data;
[0007] Image processing module: used for processing the first image data to obtain target image data;
[0008] Monitoring module: used to combine and analyze preset finite element simulation data with target image data, monitor the deformation of steel tower components, hoisting operations, and the stress and deformation changes of bridge towers in real time, and generate corresponding monitoring reports;
[0009] Alarm module: used to receive the monitoring report and perform analysis and alarm.
[0010] Preferably, the image acquisition module includes:
[0011] Calibration unit: used for calibrating all target camera devices installed in a preset shooting area to obtain a first parameter result;
[0012] The acquisition unit is used to use the target camera to shoot the special-shaped steel tower segments that reach the preset shooting area according to several preset rotation angles to obtain the first image data and collect them to obtain the first image group.
[0013] Preferably, the preset shooting area refers to the pre-assembly area and the hoisting area of the special-shaped steel tower segment.
[0014] Preferably, the image processing module includes:
[0015] A primary processing unit: performing distortion correction on all first image data in each first image group based on a first parameter result of a corresponding target camera device;
[0016] Selecting an orthophoto image from the distortion-corrected first image data as a reference image, and then projecting the remaining images onto the reference image according to a corresponding preset rotation angle to obtain second image data;
[0017] Rotate the acquired second image data according to a preset rotation angle to obtain a second image group;
[0018] A stitching unit: used for relatively orienting the second image data in the second image group, and then stitching the images adjacent to the reference image in a direction from the middle to both sides to obtain a new reference image;
[0019] The new reference image is spliced with all the remaining images in the same group and then image enhancement processing is performed to obtain target image data.
[0020] Preferably, the monitoring module includes:
[0021] Component deformation monitoring unit: used for extracting target image data captured in the pre-assembly area from the target image data and treating it as the first target image data;
[0022] The grayscale feature difference algorithm is used to screen the feature points of the prefabricated assembly image and the first target image data extracted from the prefabrication plant image database, obtain the first feature point and extract the FAST feature;
[0023] Get the direction of the feature points of FAST, and then use the greedy algorithm to determine the first feature descriptor;
[0024] Perform feature point matching on the prefabricated assembly image and the first target image data based on the first feature descriptor to obtain a first matching result;
[0025] According to the first matching result, using the twin network to perform difference analysis on the component splicing area image and the component non-splicing area image in the prefabricated assembly image and the first target image respectively;
[0026] When the difference analysis result is greater than the preset difference threshold, it is determined that the current component is deformed, and a component deformation monitoring report is generated and transmitted to the alarm module;
[0027] Lifting operation identification unit:
[0028] Data acquisition block: used to filter out the hoisting operation image data and the hoisting point position image data from the target image data captured in the hoisting area extracted from the target image data, and regard them as the second target image data;
[0029] Extract the target lifting scheme from the preset finite element simulation database to obtain the preset lifting operation behavior and preset lifting point position data;
[0030] Recognition block: used for inputting the second target image data into an abnormality recognition model pre-trained based on the preset hoisting operation behavior and preset hoisting point position data, and performing abnormality recognition on the actual hoisting operation and hoisting point position during the construction process;
[0031] When abnormal lifting operation steps or abnormal lifting point positions are identified, a corresponding abnormal operation monitoring report is generated and transmitted to the alarm module;
[0032] Bridge tower force monitoring unit:
[0033] Data acquisition block: used for receiving in real time the actual force data sent by the monitor installed at the first monitoring point;
[0034] The actual force data is compared with the set force threshold. According to the comparison result, if the actual force data is greater than the set force threshold, a force abnormality monitoring report is generated and transmitted to the alarm module;
[0035] Otherwise, the simulated force data are extracted from the preset finite element simulation database;
[0036] Construct a force difference change curve between the actual force data and the simulated force data of the current first monitoring point within a preset time period, and obtain corresponding curve index data;
[0037] The comparison result between the actual stress data and the set stress threshold is combined with the curve index data to calculate the difference risk value;
[0038] The calculated difference risk value is compared with the preset difference risk threshold, and when the difference risk value is greater than the preset difference risk threshold, a stress abnormality monitoring report is generated and transmitted to the alarm module;
[0039] Bridge tower deformation monitoring unit: used to filter out the bridge tower deformation area image from the target image data captured in the hoisting area extracted from the target image data;
[0040] Obtaining the pixel coordinates of each preset deformation calculation point in the deformation area image of the bridge tower at the current moment, and comparing them with the pixel coordinates at the previous moment in sequence to obtain the pixel coordinate difference;
[0041] Restoring the actual distance of the pixel coordinate difference of each preset deformation calculation point in the deformation area image of the same bridge tower in the object space coordinate system, and then collectively constructing and obtaining the first deformation curve;
[0042] Extracting a simulated deformation curve of the deformation area image of the bridge tower from a preset finite element simulation database, and comparing and analyzing it with the first deformation curve to obtain a deformation difference curve;
[0043] Respectively obtaining key characteristic data of the deformation difference curve and the first deformation curve, and then performing weighted calculation to obtain a deformation risk value;
[0044] The calculated deformation risk value is compared with the preset difference risk threshold, and when the deformation risk value is greater than the preset difference risk threshold, a deformation anomaly monitoring report is generated and transmitted to the alarm module.
[0045] Preferably, the calculation formula of the differential risk value is as follows:
[0046] ; In the formula, It is expressed as the difference risk value of the first monitoring point i; It is represented as the j-th curve index data of the force difference change curve of the i-th first monitoring point; It is expressed as the influence weight of the j-th curve index of the force difference change curve of the i-th first monitoring point; It is expressed as the difference between the actual monitoring data and the theoretical simulation data of the hth influencing factor; It is expressed as the deviation adjustment factor of the hth influencing force factor on the difference between actual and simulated force data; It is expressed as the influence factor of the difference between the actual stress data and the simulated stress data on the calculated difference risk value; It is expressed as the difference between the actual force value of the first monitoring point i at the current time t and the set force threshold; It is expressed as the difference between the actual force value of the first monitoring point i at time t-1 and the set force threshold; 2 represents the differential risk factor, where .
[0047] Preferably, the alarm module includes:
[0048] General alarm unit: used to generate corresponding alarm instructions according to the component deformation monitoring report and abnormal operation monitoring report transmitted by the monitoring module, and transmit them to the designated staff in real time;
[0049] Risk warning unit: used to evaluate the risk level according to the corresponding risk values in the stress abnormality monitoring report and deformation abnormality monitoring report transmitted by the monitoring module;
[0050] The corresponding alarm instructions are matched according to the risk level results and transmitted to the designated staff.
[0051] The present invention provides a cable-stayed bridge construction monitoring method based on machine vision, comprising:
[0052] Step 1: Use a target camera installed in a preset shooting area to shoot the cable-stayed bridge construction site in real time to obtain first image data;
[0053] Step 2: Processing the first image data to obtain target image data;
[0054] Step 3: Combine and analyze the preset finite element simulation data with the target image data to monitor the deformation of steel tower components, hoisting operations, and the stress and deformation changes of the bridge tower in real time, and generate corresponding monitoring reports;
[0055] Step 4: Receive the monitoring report and perform analysis and alarm.
[0056] Compared with the prior art, the present invention has the following beneficial effects:
[0057] The target image data is obtained by processing the image data taken at the cable-stayed bridge construction site, and then combined with the preset finite element simulation data, the deformation of steel tower components, hoisting operations, and the stress and deformation changes of the bridge tower are monitored. Alarms are issued based on the monitoring results to ensure effective monitoring of the stress and deformation of the bridge tower during the hoisting process, thereby ensuring the safety and controllability of the overall structural system during the hoisting construction of the bridge tower segment.
[0058] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.
[0059] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0060] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:
[0061] Figure 1 It is a structural diagram of a cable-stayed bridge construction monitoring system based on machine vision in an embodiment of the present invention;
[0062] Figure 2 The present invention is a flowchart of a cable-stayed bridge construction monitoring method based on machine vision in an embodiment of the present invention. DETAILED DESCRIPTION
[0063] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.
[0064] The embodiment of the present invention provides a cable-stayed bridge construction monitoring system based on machine vision, such as Figure 1 As shown, including:
[0065] Image acquisition module: used to take real-time photos of the cable-stayed bridge construction site using a target camera installed in a preset shooting area to obtain first image data;
[0066] Image processing module: used for processing the first image data to obtain target image data;
[0067] Monitoring module: used to combine and analyze preset finite element simulation data with target image data, monitor the deformation of steel tower components, hoisting operations, and the stress and deformation changes of bridge towers in real time, and generate corresponding monitoring reports;
[0068] Alarm module: used to receive the monitoring report and perform analysis and alarm.
[0069] In this embodiment, the preset shooting area refers to the pre-assembly area and the hoisting area of the special-shaped steel tower segment, wherein the pre-assembly area is used to pre-assemble the steel tower segments transported to the site; the hoisting area is used to complete the hoisting of each segment; the target camera device refers to the rotating camera; the first image data is an image obtained by using the rotating camera to shoot the cable-stayed bridge construction site in the preset shooting area.
[0070] In this embodiment, the target image data is an image obtained after the first image data is subjected to distortion correction, projection transformation, relative orientation and splicing, and image enhancement processing.
[0071] In this embodiment, the preset finite element simulation data refers to data generated by simulating the construction process of a cable-stayed bridge using a preset finite element model established using finite element software such as Midas, such as bridge tower simulation force data, preset hanging point position data, and the like.
[0072] In this embodiment, the monitoring report includes a component deformation monitoring report, a stress abnormality monitoring report, and a deformation abnormality monitoring report.
[0073] The beneficial effect of the above technical solution is: by processing the image data of the cable-stayed bridge construction site, the target image data is obtained, and then combined with the preset finite element simulation data, the deformation of the steel tower components, the lifting operation and the stress and deformation changes of the bridge tower are monitored, and an alarm is issued according to the monitoring results to ensure the effective monitoring of the stress and deformation of the bridge tower during the lifting process, thereby ensuring the safety and controllability of the overall structural system during the lifting construction of the bridge tower segment.
[0074] An embodiment of the present invention provides a cable-stayed bridge construction monitoring system based on machine vision, wherein the image acquisition module comprises:
[0075] Calibration unit: used for calibrating all target camera devices installed in a preset shooting area to obtain a first parameter result;
[0076] The acquisition unit is used to use the target camera to shoot the special-shaped steel tower segments that reach the preset shooting area according to several preset rotation angles to obtain the first image data and collect them to obtain the first image group.
[0077] In this embodiment, the preset shooting area refers to the pre-assembly area of the special-shaped steel tower segment and the set hoisting area; the target camera device refers to the rotating camera; the purpose of camera calibration is to obtain accurate spatial image information from the image and convert the pixel coordinates of the image into actual spatial coordinates; the first parameter result includes distortion parameters, focal length, etc.
[0078] In this embodiment, the preset rotation angle is determined in advance, such as 5° to the left and 5° to the right; the first image data refers to the image obtained by photographing the cable-stayed bridge construction site in the preset shooting area using a rotating camera; the first image group is a collection of the first images taken at all preset rotation angles for the same operation scene or special-shaped steel tower segment.
[0079] The beneficial effect of the above technical solution is: after the camera is calibrated for the target camera device installed in the preset shooting area, image data is obtained by shooting at a preset rotation angle, thereby providing effective data support for subsequent monitoring.
[0080] The embodiment of the present invention provides a cable-stayed bridge construction monitoring system based on machine vision, wherein the image processing module comprises:
[0081] A primary processing unit: performing distortion correction on all first image data in each first image group based on a first parameter result of a corresponding target camera device;
[0082] Selecting an orthophoto image from the distortion-corrected first image data as a reference image, and then projecting the remaining images onto the reference image according to a corresponding preset rotation angle to obtain second image data;
[0083] Rotate the acquired second image data according to a preset rotation angle to obtain a second image group;
[0084] A stitching unit: used for relatively orienting the second image data in the second image group, and then stitching the images adjacent to the reference image in a direction from the middle to both sides to obtain a new reference image;
[0085] The new reference image is spliced with all the remaining images in the same group and then image enhancement processing is performed to obtain target image data.
[0086] In this embodiment, distortion correction refers to performing distortion correction on an image according to a distortion parameter, the purpose of which is to eliminate deformation caused by lens distortion, such as cropping or stretching; the reference image is an orthographic image selected from the first image group.
[0087] In this embodiment, the second image data refers to an image obtained after distortion correction and projection transformation of the first image, wherein the projection transformation is an extension of the affine transformation and can keep the shape of the transformed image unchanged.
[0088] In this embodiment, the purpose of relative orientation is to eliminate the influence of slight deviations produced during the rotation of the camera; the new reference image is obtained by splicing the reference image with adjacent images; the target image data is an image obtained after distortion correction, projection transformation, relative orientation and splicing of the first image data, and image enhancement processing.
[0089] The beneficial effect of the above technical solution is: by performing distortion correction, projection transformation, relative orientation and stitching, and image enhancement processing on the first image data, the target image data is obtained, the interference of the image is removed, the original information of the image is restored as much as possible, and the quality of the image is improved, thereby providing an effective data basis for subsequent monitoring.
[0090] An embodiment of the present invention provides a cable-stayed bridge construction monitoring system based on machine vision, wherein the monitoring module comprises:
[0091] Component deformation monitoring unit: used for extracting target image data captured in the pre-assembly area from the target image data and treating it as the first target image data;
[0092] The grayscale feature difference algorithm is used to screen the feature points of the prefabricated assembly image and the first target image data extracted from the prefabrication plant image database, obtain the first feature point and extract the FAST feature;
[0093] Get the direction of the feature points of FAST, and then use the greedy algorithm to determine the first feature descriptor;
[0094] Perform feature point matching on the prefabricated assembly image and the first target image data based on the first feature descriptor to obtain a first matching result;
[0095] According to the first matching result, using the twin network to perform difference analysis on the component splicing area image and the component non-splicing area image in the prefabricated assembly image and the first target image respectively;
[0096] When the difference analysis result is greater than the preset difference threshold, it is determined that the current component is deformed, and a component deformation monitoring report is generated and transmitted to the alarm module;
[0097] Lifting operation identification unit:
[0098] Data acquisition block: used to filter out the hoisting operation image data and the hoisting point position image data from the target image data captured in the hoisting area extracted from the target image data, and regard them as the second target image data;
[0099] Extract the target lifting scheme from the preset finite element simulation database to obtain the preset lifting operation behavior and preset lifting point position data;
[0100] Recognition block: used for inputting the second target image data into an abnormality recognition model pre-trained based on the preset hoisting operation behavior and preset hoisting point position data, and performing abnormality recognition on the actual hoisting operation and hoisting point position during the construction process;
[0101] When abnormal lifting operation steps or abnormal lifting point positions are identified, a corresponding abnormal operation monitoring report is generated and transmitted to the alarm module;
[0102] Bridge tower force monitoring unit:
[0103] Data acquisition block: used for receiving in real time the actual force data sent by the monitor installed at the first monitoring point;
[0104] The actual force data is compared with the set force threshold. According to the comparison result, if the actual force data is greater than the set force threshold, a force abnormality monitoring report is generated and transmitted to the alarm module;
[0105] Otherwise, the simulated force data are extracted from the preset finite element simulation database;
[0106] Construct a force difference change curve between the actual force data and the simulated force data of the current first monitoring point within a preset time period, and obtain corresponding curve index data;
[0107] The comparison result between the actual stress data and the set stress threshold is combined with the curve index data to calculate the difference risk value;
[0108] The calculated difference risk value is compared with the preset difference risk threshold, and when the difference risk value is greater than the preset difference risk threshold, a stress abnormality monitoring report is generated and transmitted to the alarm module;
[0109] Bridge tower deformation monitoring unit: used to filter out the bridge tower deformation area image from the target image data captured in the hoisting area extracted from the target image data;
[0110] Obtaining the pixel coordinates of each preset deformation calculation point in the deformation area image of the bridge tower at the current moment, and comparing them with the pixel coordinates at the previous moment in sequence to obtain the pixel coordinate difference;
[0111] Restoring the actual distance of the pixel coordinate difference of each preset deformation calculation point in the deformation area image of the same bridge tower in the object space coordinate system, and then collectively constructing and obtaining the first deformation curve;
[0112] Extracting a simulated deformation curve of the deformation area image of the bridge tower from a preset finite element simulation database, and comparing and analyzing it with the first deformation curve to obtain a deformation difference curve;
[0113] Respectively obtaining key characteristic data of the deformation difference curve and the first deformation curve, and then performing weighted calculation to obtain a deformation risk value;
[0114] The calculated deformation risk value is compared with the preset difference risk threshold, and when the deformation risk value is greater than the preset difference risk threshold, a deformation anomaly monitoring report is generated and transmitted to the alarm module.
[0115] In this embodiment, the grayscale feature difference algorithm is used to screen the pending feature points; the prefabrication plant image database is composed of the splicing images of the components pre-assembled in the prefabrication plant and the component names and numbers; the prefabricated assembly image refers to the splicing images of the components pre-assembled in the prefabrication plant.
[0116] In this embodiment, the direction of the feature point is obtained by calculating the grayscale of the corner point and the offset of the center of mass; the first feature descriptor is used to describe the detected feature; the first matching result refers to the result after the feature point matching is successful; the twin network is a network structure; the component splicing area image refers to the splicing part of the component to be assembled; the component non-splicing area image refers to the remaining part of the component to be assembled except the component splicing area; the difference analysis result refers to the image difference value between the prefabricated assembly image and the component splicing area image in the first target image and the image difference value of the component non-splicing area image; the preset difference threshold is set in advance; the component deformation monitoring report includes the component name, number, deformation area and description, difference value, etc.
[0117] In this embodiment, the second target image data refers to the lifting operation image data and the lifting point position image data screened out from the target image data taken in the lifting area; the preset finite element simulation database is a database composed of data generated by simulating the construction process of a preset finite element model, wherein the preset finite element model is established using finite element software such as Midas.
[0118] In this embodiment, the target lifting plan includes lifting components, equipment data, lifting operation behaviors and layout content data, such as lifting point positions and lifting lug positions; the preset lifting operation behaviors refer to the optimal lifting operation steps obtained after finite element model simulation; the preset lifting point position data are the best lifting position data obtained after finite element model simulation; the first monitoring point refers to the position point for monitoring the stress of the bridge tower, which is generally set at the middle section of the tower body, the top section of the tower and the root section, etc.
[0119] In this embodiment, the stress threshold is set in advance; the stress difference change curve refers to a curve composed of the difference between the actual stress data of the current first monitoring point within a preset time period and the simulated stress data; the curve indicators include peak value, lifting speed, etc.; the difference risk value is used to express the influence of the actual difference in the stress of the bridge tower on the construction effect and safety; the preset difference risk threshold is set in advance.
[0120] In this embodiment, the deformation area image of the bridge tower refers to the cross-sectional image of the bridge tower whose deformation needs to be monitored, such as the middle section of the tower body, the top section and the root section; the preset deformation calculation points are set in advance; the first deformation curve refers to the actual distance of the pixel coordinate difference of each preset deformation calculation point in the deformation area image of the same bridge tower in the object space coordinate system, and then the collection is constructed; the simulated deformation curve refers to a curve composed of the deformation values of the bridge tower calculated by simulating the construction process using a preset finite element model; the deformation difference curve is obtained by comparing the first deformation curve with the simulated deformation curve; the key feature data include maximum value, minimum value, lifting speed, etc.; the deformation risk value is used to express the influence of the deformation change of the bridge tower on the construction effect and safety; the preset difference risk threshold is set in advance; the deformation abnormality monitoring report is composed of deformation abnormality data, deformed bridge tower number and name, deformation risk value, etc.
[0121] The beneficial effect of the above technical solution is: by combining and analyzing the preset finite element simulation data with the target image data, the deformation of the steel tower components, the lifting operation, and the stress and deformation changes of the bridge tower can be monitored in real time, which helps to ensure the safety and controllability of the overall structural system during the lifting construction of the bridge tower segment.
[0122] The embodiment of the present invention provides a cable-stayed bridge construction monitoring system based on machine vision, and the calculation formula of the difference risk value is as follows:
[0123] ; In the formula, It is expressed as the difference risk value of the first monitoring point i; It is represented as the j-th curve index data of the force difference change curve of the i-th first monitoring point; It is expressed as the influence weight of the j-th curve index of the force difference change curve of the i-th first monitoring point; It is expressed as the difference between the actual monitoring data and the theoretical simulation data of the hth influencing factor; It is expressed as the deviation adjustment factor of the hth influencing force factor on the difference between actual and simulated force data; It is expressed as the influence factor of the difference between the actual stress data and the simulated stress data on the calculated difference risk value; It is expressed as the difference between the actual force value of the first monitoring point i at the current time t and the set force threshold; It is expressed as the difference between the actual force value of the first monitoring point i at time t-1 and the set force threshold; 2 represents the differential risk factor, where .
[0124] In this embodiment, the first monitoring point refers to the position point for monitoring the stress of the bridge tower, which is generally set at the middle section of the tower body, the top section of the tower and the root section; factors affecting the stress include temperature, wind speed, the position of the hanging point and the angle of the hanging rope, etc.
[0125] The beneficial effect of the above technical solution is: by comparing the actual force data with the set force threshold, and combining it with the curve index data of the force difference change curve to calculate the difference risk value, it effectively reflects the degree of dangerous impact of the current actual force of the bridge tower on the construction effect and safety.
[0126] The embodiment of the present invention provides a cable-stayed bridge construction monitoring system based on machine vision, wherein the alarm module comprises:
[0127] General alarm unit: used to generate corresponding alarm instructions according to the component deformation monitoring report and abnormal operation monitoring report transmitted by the monitoring module, and transmit them to the designated staff in real time;
[0128] Risk warning unit: used to evaluate the risk level according to the corresponding risk values in the stress abnormality monitoring report and deformation abnormality monitoring report transmitted by the monitoring module;
[0129] The corresponding alarm instructions are matched according to the risk level results and transmitted to the designated staff.
[0130] In this embodiment, the component deformation monitoring report consists of the component name, number, deformation area and description, and monitoring time; the abnormal operation monitoring report consists of monitoring time, abnormal cause, abnormal data, etc.
[0131] In this embodiment, the stress abnormality monitoring report is composed of stress abnormality data, the location of stress abnormality, and differential risk value; the deformation abnormality monitoring report is composed of deformation abnormality data, deformed bridge tower number and name, deformation risk value, etc.; the risk value includes differential risk value and deformation risk value; the risk level includes general, greater risk and major risk.
[0132] The beneficial effect of the above technical solution is: by analyzing the monitoring report sent by the monitoring module, matching the corresponding alarm instructions and transmitting them to the designated staff in a timely manner, the safety and controllability of the overall structural system during the bridge tower segment hoisting construction process can be guaranteed.
[0133] The embodiment of the present invention provides a cable-stayed bridge construction monitoring method based on machine vision, such as Figure 2 As shown, including:
[0134] Step 1: Use a target camera installed in a preset shooting area to shoot the cable-stayed bridge construction site in real time to obtain first image data;
[0135] Step 2: Processing the first image data to obtain target image data;
[0136] Step 3: Combine and analyze the preset finite element simulation data with the target image data to monitor the deformation of steel tower components, hoisting operations, and stress and deformation changes of bridge towers in real time, and generate corresponding monitoring reports;
[0137] Step 4: Receive the monitoring report and perform analysis and alarm.
[0138] The beneficial effect of the above technical solution is: by processing the image data of the cable-stayed bridge construction site, the target image data is obtained, and then combined with the preset finite element simulation data, the deformation of the steel tower components, the lifting operation and the stress and deformation changes of the bridge tower are monitored, and an alarm is issued according to the monitoring results to ensure the effective monitoring of the stress and deformation of the bridge tower during the lifting process, thereby ensuring the safety and controllability of the overall structural system during the lifting construction of the bridge tower segment.
[0139] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.
Claims
1. A cable-stayed bridge construction monitoring system based on machine vision, It is characterized in that include: Image acquisition module: used to take real-time photos of the cable-stayed bridge construction site using a target camera installed in a preset shooting area to obtain first image data; Image processing module: used for processing the first image data to obtain target image data; Monitoring module: used to combine and analyze preset finite element simulation data with target image data, monitor the deformation of steel tower components, hoisting operations, and the stress and deformation changes of bridge towers in real time, and generate corresponding monitoring reports; Alarm module: used to receive the monitoring report and perform analysis and alarm; The image acquisition module comprises: Calibration unit: used for calibrating all target camera devices installed in a preset shooting area to obtain a first parameter result; Acquisition unit: used for photographing the special-shaped steel tower segments reaching the preset shooting area at a plurality of preset rotation angles using a target camera device, obtaining first image data, and collecting to obtain a first image group; The preset shooting area refers to the pre-assembly area and the hoisting area of the special-shaped steel tower segment; The image processing module comprises: A primary processing unit: performing distortion correction on all first image data in each first image group based on a first parameter result of a corresponding target camera device; Selecting an orthophoto image from the distortion-corrected first image data as a reference image, and then projecting the remaining images onto the reference image according to a corresponding preset rotation angle to obtain second image data; Rotate the acquired second image data according to a preset rotation angle to obtain a second image group; A stitching unit: used for relatively orienting the second image data in the second image group, and then stitching the images adjacent to the reference image in a direction from the middle to both sides to obtain a new reference image; The new reference image is spliced with all the remaining images in the same group and then image enhancement is performed to obtain target image data; The monitoring module comprises: Component deformation monitoring unit: used for extracting target image data captured in the pre-assembly area from the target image data and treating it as the first target image data; The grayscale feature difference algorithm is used to screen the feature points of the prefabricated assembly image and the first target image data extracted from the prefabrication plant image database, obtain the first feature point and extract the FAST feature; Get the direction of the feature points of FAST, and then use the greedy algorithm to determine the first feature descriptor; Perform feature point matching on the prefabricated assembly image and the first target image data based on the first feature descriptor to obtain a first matching result; According to the first matching result, using the twin network to perform difference analysis on the component splicing area image and the component non-splicing area image in the prefabricated assembly image and the first target image respectively; When the difference analysis result is greater than the preset difference threshold, it is determined that the current component is deformed, and a component deformation monitoring report is generated and transmitted to the alarm module; Lifting operation identification unit: Data acquisition block: used to filter out the hoisting operation image data and the hoisting point position image data from the target image data captured in the hoisting area extracted from the target image data, and regard them as the second target image data; Extract the target lifting scheme from the preset finite element simulation database to obtain the preset lifting operation behavior and preset lifting point position data; Recognition block: used for inputting the second target image data into an abnormality recognition model pre-trained based on the preset hoisting operation behavior and preset hoisting point position data, and performing abnormality recognition on the actual hoisting operation and hoisting point position during the construction process; When abnormal lifting operation steps or abnormal lifting point positions are identified, a corresponding abnormal operation monitoring report is generated and transmitted to the alarm module; Bridge tower force monitoring unit: Data acquisition block: used for receiving in real time the actual force data sent by the monitor installed at the first monitoring point; The actual force data is compared with the set force threshold. According to the comparison result, if the actual force data is greater than the set force threshold, a force abnormality monitoring report is generated and transmitted to the alarm module; Otherwise, the simulated force data are extracted from the preset finite element simulation database; Construct a force difference change curve between the actual force data and the simulated force data of the current first monitoring point within a preset time period, and obtain corresponding curve index data; The comparison result between the actual stress data and the set stress threshold is combined with the curve index data to calculate the difference risk value; The calculated difference risk value is compared with the preset difference risk threshold, and when the difference risk value is greater than the preset difference risk threshold, a stress abnormality monitoring report is generated and transmitted to the alarm module; Bridge tower deformation monitoring unit: used to filter out the bridge tower deformation area image from the target image data captured in the hoisting area extracted from the target image data; Obtaining the pixel coordinates of each preset deformation calculation point in the deformation area image of the bridge tower at the current moment, and comparing them with the pixel coordinates at the previous moment in sequence to obtain the pixel coordinate difference; Restoring the actual distance of the pixel coordinate difference of each preset deformation calculation point in the deformation area image of the same bridge tower in the object space coordinate system, and then collectively constructing and obtaining the first deformation curve; Extracting a simulated deformation curve of the deformation area image of the bridge tower from a preset finite element simulation database, and comparing and analyzing it with the first deformation curve to obtain a deformation difference curve; Respectively obtaining key characteristic data of the deformation difference curve and the first deformation curve, and then performing weighted calculation to obtain a deformation risk value; The calculated deformation risk value is compared with the preset difference risk threshold, and when the deformation risk value is greater than the preset difference risk threshold, a deformation abnormality monitoring report is generated and transmitted to the alarm module; The calculation formula of the difference risk value is as follows: ; In the formula, It is expressed as the difference risk value of the first monitoring point i; It is represented as the j-th curve index data of the force difference change curve of the i-th first monitoring point; It is expressed as the influence weight of the j-th curve index of the force difference change curve of the i-th first monitoring point; It is expressed as the difference between the actual monitoring data and the theoretical simulation data of the hth influencing factor; It is expressed as the deviation adjustment factor of the hth influencing force factor on the difference between actual and simulated force data; It is expressed as the influence factor of the difference between the actual stress data and the simulated stress data on the calculated difference risk value; It is expressed as the difference between the actual force value of the first monitoring point i at the current time t and the set force threshold; It is expressed as the difference between the actual force value of the first monitoring point i at time t-1 and the set force threshold; 2 represents the differential risk factor, where .
2. According to claim 1, a cable-stayed bridge construction monitoring system based on machine vision, It is characterized in that The alarm module comprises: General alarm unit: used to generate corresponding alarm instructions according to the component deformation monitoring report and abnormal operation monitoring report transmitted by the monitoring module, and transmit them to the designated staff in real time; Risk warning unit: used to evaluate the risk level according to the corresponding risk values in the stress abnormality monitoring report and deformation abnormality monitoring report transmitted by the monitoring module; The corresponding alarm instructions are matched according to the risk level results and transmitted to the designated staff.
3. A cable-stayed bridge construction monitoring method based on machine vision, using the cable-stayed bridge construction monitoring system based on machine vision as claimed in claim 1, It is characterized in that include: Step 1: Use a target camera installed in a preset shooting area to shoot the cable-stayed bridge construction site in real time to obtain first image data; Step 2: Processing the first image data to obtain target image data; Step 3: Combine and analyze the preset finite element simulation data with the target image data to monitor the deformation of steel tower components, hoisting operations, and the stress and deformation changes of the bridge tower in real time, and generate corresponding monitoring reports; Step 4: Receive the monitoring report and perform analysis and alarm.
