AR-based bridge inspection method and system

Abstract

The invention relates to the technical field of real-time image superposition, in particular to a bridge inspection method and system based on AR, and the method comprises the following steps: shooting a bridge on site, capturing a real-time image of a structure, and recording vibration frequency, displacement change and stress data by using acceleration, displacement and strain sensors; and synchronously integrating the structure real-time image, the vibration frequency, the displacement change and the stress data, and naming the data as multi-dimensional dynamic monitoring data. According to the method, the vibration frequency, the displacement change and the stress data are integrated in real time, so that the real-time performance and the analysis depth of the multi-dimensional dynamic monitoring data are remarkably improved, the high fusion of a real-time image and a three-dimensional model is realized, the accuracy of data visualization is enhanced, an abnormal region is directly visible through the change of colors and textures, and the accuracy of data visualization is improved. According to the method, the structure safety monitoring intuitiveness and interactivity are improved, in addition, through time sequence analysis, structure changes under different environment conditions are accurately simulated, and prospective risk prediction is provided.

Description

Technical Field

[0001] The present invention relates to the technical field of real-time image overlay, and in particular to an AR-based bridge inspection method and system. Background Art

[0002] Real-time image overlay technology is a key component in the field of augmented reality (AR), which enhances the user's perception and interactive experience by overlaying computer-generated images, videos or data layers onto the real-world vision. This technology usually relies on cameras to capture real-world scenes and merge virtual content with them to display on the user's device, such as smart glasses, mobile phones or tablet screens. The key to real-time image overlay is that it can quickly and accurately align virtual and actual visual elements in unpredictable dynamic environments, ensuring that virtual content is visually integrated with the actual environment, thereby providing an intuitive interactive interface and information display. This technology is widely used in many fields, including medical, entertainment, maintenance and industrial design, to provide real-time visual enhancement information to help users better understand and operate the real world.

[0003] Among them, the AR-based bridge inspection method refers to a modern inspection method that uses augmented reality technology to assist in visual and structural inspections of bridges. This method can enhance inspection efficiency and accuracy by superimposing detailed data and structural analysis information in the inspector's field of view. Inspectors can see real-time image overlays through specific equipment, such as AR glasses, which may include the precise location of cracks, stress distribution, historical maintenance data, and predictive maintenance information. Such a system not only improves the accuracy of inspections, but also greatly reduces the time and labor intensity required for inspections, making bridge maintenance more efficient and systematic. In addition, the AR-based inspection method can also provide remote assistance functions, allowing experts to remotely view real-time data and analysis of the same scene in different locations, so as to make more effective decisions and guidance.

[0004] Existing technologies usually rely on static data display in bridge inspection applications, lacking the integration of real-time dynamic data, which limits the comprehensive assessment of the immediate status of the bridge. In actual operations, the matching of data with three-dimensional models often lacks accuracy and flexibility, which is particularly prominent in dynamic environments, and may lead to inaccurate judgments of structural defects and affect the reliability of risk assessment. These technical limitations not only reduce the efficiency of inspection work, but may also threaten the safety of the structure due to judgment errors, affecting the overall effectiveness of maintenance work. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose an AR-based bridge inspection method and system.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a bridge inspection method based on AR, comprising the following steps:

[0007] S1: Take photos of the bridge on site to capture real-time images of the structure, and use acceleration, displacement, and strain sensors to record vibration frequency, displacement change, and stress data. The real-time images of the structure and the vibration frequency, displacement change, and stress data are synchronously integrated and named as multi-dimensional dynamic monitoring data;

[0008] S2: Based on the multi-dimensional dynamic monitoring data, load the three-dimensional model of the bridge in the AR device, map the vibration frequency, displacement, and stress data to the corresponding position of the model, match the real-time image with the model coordinates, adjust the viewing angle and scale to make the image coincide with the model, overlay the real-time data in the form of color or texture, and generate a real-time overlay visualization image;

[0009] S3: Monitor color and texture changes through the real-time superposition of visualized images, identify abnormal areas, use vibration frequency, displacement, and stress data that exceed the threshold to locate corresponding structural parts, compare structural part parameters, screen out areas with abnormal values, and generate key risk area data;

[0010] S4: Using the key risk area data, the parameters exceeding the threshold are arranged in time series, parameter trend analysis is performed on the key areas, parameter changes at future moments are calculated, changes under differentiated load and environmental conditions are simulated, and numerical differences in multiple scenarios are compared to obtain future risk prediction results.

[0011] As a further solution of the present invention, the step of acquiring the multi-dimensional dynamic monitoring data is specifically as follows:

[0012] S111: Use a high-definition camera to photograph the bridge, capture real-time images of the structure, conduct preliminary analysis through automatic image recognition technology, verify key structural elements in the image, and generate real-time structural image data;

[0013] S112: Acceleration sensors, displacement sensors and strain sensors are installed at multiple key locations of the bridge structure to record vibration frequency, displacement change and stress data in real time. The data are sent to the data processing center through real-time data transmission technology to generate real-time monitoring data of the sensors;

[0014] S113: Integrate the real-time structural image data and the real-time monitoring data of the sensor, and use a multi-dimensional data fusion algorithm to perform data alignment, synchronization and fusion, using the formula:

[0015]

[0016] Generate multi-dimensional dynamic monitoring data;

[0017] Among them, Dmulti represents multi-dimensional dynamic monitoring data, w i represents the weight factor of the i-th data source, which is used to adjust the influence of the data source in the overall synthesis result. i Represents the real-time monitoring value of the i-th data source, including vibration frequency, displacement change, and stress data obtained from multi-sensor and image analysis.

[0018] As a further solution of the present invention, the step of acquiring the real-time superimposed visualization image is specifically as follows:

[0019] S211: Based on the multi-dimensional dynamic monitoring data, a three-dimensional model of the bridge is loaded into the AR device, and a structural image part of the multi-dimensional dynamic monitoring data is called to generate three-dimensional model data of the bridge;

[0020] S212: Utilize the vibration frequency, displacement change and stress data in the multi-dimensional dynamic monitoring data and map them according to the coordinate system of the three-dimensional bridge model using the formula:

[0021]

[0022] Calculate and generate mapping data;

[0023] Among them, w1 represents the weight coefficient of vibration frequency data, w2 represents the weight coefficient of displacement change data, w3 represents the weight coefficient of stress data, w4 represents the weight coefficient of temperature data, f represents vibration frequency data, d represents displacement change data, s represents stress data, t represents temperature data, p represents proportional factor, and q represents adjustment parameter;

[0024] S213: Based on the bridge three-dimensional model data and the mapping data, coordinate matching is performed, the viewing angle and scale are adjusted, and the mapping data is superimposed in the form of color or texture to generate a real-time superimposed visualization image.

[0025] As a further solution of the present invention, the step of acquiring the key risk area data is specifically as follows:

[0026] S311: monitoring the changes in color and texture from the real-time superimposed visualization image, identifying abnormal areas, and generating preliminary identified abnormal area data;

[0027] S312: Using the initially identified abnormal area data, combined with the vibration frequency, displacement, and stress data exceeding the preset threshold, the bridge structure part of the abnormal area is located using the formula:

[0028]

[0029] Calculate the parameter deviation of the structural parts and generate the structural parts positioning data;

[0030] Among them, x f,i 、x d,i 、x s,i Respectively represent the actual measured values ​​of vibration frequency, displacement and stress of the i-th measuring point, μ f , μ d , μ s They represent the normal thresholds of vibration frequency, displacement and stress respectively, and n represents the total number of measurement points;

[0031] S313: Call the structural location data, perform numerical comparison and analysis, combine threshold judgment and abnormal monitoring data, determine the risk level, and generate key risk area data.

[0032] As a further solution of the present invention, the steps of obtaining the future risk prediction results are specifically as follows:

[0033] S411: using the key risk area data, arranging the parameters exceeding the threshold value in time series, laying a foundation for parameter trend analysis, and generating time series arrangement data;

[0034] S412: Perform parameter trend analysis on the time series arrangement data, and use a statistical model to estimate the changes of parameters at multiple moments in the future, using the formula:

[0035] P t =α·X t-1 +(1-α)·P t-1

[0036] Calculate the parameter prediction values ​​in the future time period and generate the future parameter change prediction data;

[0037] Among them, P t represents the predicted value at time t, X t-1 represents the actual observation value at time t-1, α is the smoothing constant, P t-1 is the predicted value at the previous moment;

[0038] S413: simulating parameter changes under differentiated load and environmental conditions, and calculating and generating multi-scenario numerical difference data using the future parameter change prediction data;

[0039] S414: Compare the multi-scenario numerical difference data, determine the risk level under the multi-scenario, integrate the data, and obtain future risk prediction results.

[0040] An AR-based bridge inspection system, the AR-based bridge inspection system is used to perform the above-mentioned AR-based bridge inspection method, the system comprising:

[0041] The data capture module takes pictures of the structure at the bridge site, collects real-time images of the structure, uses acceleration sensors to record vibration frequency, displacement sensors to record displacement changes, and strain sensors to record stress data. The vibration frequency, displacement changes, and stress data are recorded synchronously in time and integrated into multi-dimensional dynamic monitoring data.

[0042] The model synchronization module loads the three-dimensional model of the bridge in the AR device, extracts the vibration frequency, displacement, and stress data in the multi-dimensional dynamic monitoring data, marks the corresponding positions of the data according to the coordinate system of the three-dimensional model, adjusts the AR perspective and scale to align the real-time image with the three-dimensional model, and uses color and texture to overlay the real-time data on the model to generate a real-time overlay visualization image;

[0043] The anomaly detection module monitors the changes in color and texture in the real-time superimposed visualization image, identifies the area that is different from the standard model, extracts the vibration frequency, displacement, and stress data that exceed the preset threshold, locates the corresponding structural parts, compares the current data with the normal data, and screens the key risk area data;

[0044] The trend analysis module arranges the vibration frequency, displacement, and stress parameters exceeding the threshold in the key risk area data in chronological order, analyzes the change trend of multiple parameters, simulates the change of parameters under differentiated loads and environmental conditions, compares the differences of parameters under multiple scenarios, and obtains the parameter change trend analysis results;

[0045] The risk prediction module predicts the vibration frequency, displacement, and stress changes of multiple structural parts at future moments based on the parameter change trend analysis results, evaluates the overall risk level of the structure, and generates future risk prediction results.

[0046] Compared with the prior art, the advantages and positive effects of the present invention are:

[0047] In the present invention, the real-time performance and analysis depth of multi-dimensional dynamic monitoring data are significantly improved by integrating vibration frequency, displacement change and stress data in real time. The real-time image and three-dimensional model are highly integrated, the accuracy of data visualization is enhanced, and the abnormal area is directly visible through the change of color and texture, which improves the intuitiveness and interactivity of structural safety monitoring. In addition, through time series analysis, the structural changes under different environmental conditions are accurately simulated, and forward-looking risk prediction is provided, thereby effectively improving the preventive and systematic nature of bridge maintenance work. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0049] Figure 2 This is a flow chart of the steps for obtaining multi-dimensional dynamic monitoring data of the present invention;

[0050] Figure 3 A flowchart of the steps for obtaining real-time superimposed visualization images of the present invention;

[0051] Figure 4 This is a flow chart of the steps for obtaining key risk area data of the present invention;

[0052] Figure 5 This is a flow chart of the steps for obtaining future risk prediction results of the present invention. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0054] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0055] Embodiment 1

[0056] See also Figure 1 The present invention provides a technical solution: an AR-based bridge inspection method, comprising the following steps:

[0057] S1: Take photos of the bridge on site to capture real-time images of the structure, and use acceleration, displacement, and strain sensors to record vibration frequency, displacement change, and stress data. The real-time images of the structure and the vibration frequency, displacement change, and stress data are synchronously integrated and named as multi-dimensional dynamic monitoring data;

[0058] S2: Based on the multi-dimensional dynamic monitoring data, load the three-dimensional model of the bridge in the AR device, map the vibration frequency, displacement, and stress data to the corresponding position of the model, match the real-time image with the model coordinates, adjust the viewing angle and scale to make the image coincide with the model, overlay the real-time data in the form of color or texture, and generate a real-time overlay visualization image;

[0059] S3: Monitor color and texture changes through real-time superposition of visual images, identify abnormal areas, use vibration frequency, displacement, and stress data that exceed the threshold to locate the corresponding structural parts, compare the parameters of the structural parts, screen out areas with abnormal values, and generate key risk area data;

[0060] S4: Using the data of key risk areas, arrange the parameters exceeding the threshold in time series, perform parameter trend analysis on key areas, calculate parameter changes at future moments, simulate changes under differentiated load and environmental conditions, compare the differences in multiple scenario values, and obtain future risk prediction results.

[0061] The multi-dimensional dynamic monitoring data specifically include real-time images of the structure, vibration frequency, displacement changes, and stress data. The real-time superimposed visualization images include color changes, texture changes, and the matching degree between the model and the image. The key risk area data specifically refer to abnormal vibration frequency, abnormal displacement, and abnormal stress data. The future risk prediction results include parameter change trends and multi-scenario numerical differences.

[0062] See also Figure 2 ,The specific steps for obtaining multi-dimensional dynamic monitoring data are:

[0063] S111: Use a high-definition camera to photograph the bridge, capture real-time images of the structure, conduct preliminary analysis through automatic image recognition technology, verify key structural elements in the image, and generate real-time structural image data;

[0064] On-site staff use high-definition cameras to capture real-time images of the bridge structure, conduct preliminary analysis through automatic image recognition technology, and verify key structural elements in the image. This process involves identifying the color, edge and texture features of the image to ensure that the identified structural elements represent the actual physical state of the bridge. Through the data, staff can understand the structural state of the bridge at a specific point in time, which provides a basis for subsequent data analysis and structural evaluation, and obtains real-time structural image data.

[0065] S112: Acceleration sensors, displacement sensors and strain sensors are installed at multiple key locations of the bridge structure to record vibration frequency, displacement change and stress data in real time. The data are sent to the data processing center through real-time data transmission technology to generate real-time monitoring data of the sensors;

[0066] Acceleration sensors, displacement sensors and strain sensors are installed at key locations of the bridge structure and begin to record data in real time. The data collected by these sensors include information on multiple dimensions such as vibration frequency, displacement changes and stress. The sensors send the data to the data processing center in real time through wireless transmission technology. The data processing center uses high-speed data processing chips to process the collected information immediately. These processes include data cleaning, normalization and preliminary analysis to remove abnormal data and extract useful monitoring indicators, thereby generating real-time sensor monitoring data.

[0067] S113: Integrate real-time structural image data and real-time sensor monitoring data, use multi-dimensional data fusion algorithm to align, synchronize and fuse data, using the formula:

[0068]

[0069] Generate multi-dimensional dynamic monitoring data;

[0070] Among them, D multi represents multi-dimensional dynamic monitoring data, w i represents the weight factor of the i-th data source, which is used to adjust the influence of the data source in the overall synthesis result. i Represents the real-time monitoring value of the i-th data source, including vibration frequency, displacement change, and stress data obtained from multi-sensor and image analysis.

[0071] formula:

[0072]

[0073] The benefit of the formula is that by assigning differentiated weight factors w to differentiated monitoring data i , can be adjusted according to the importance and influence of multiple data points, thereby effectively integrating and optimizing the analysis of bridge monitoring data.

[0074] Detailed explanation of the formula and the process of formula calculation and derivation:

[0075] The data points x1, x2, and x3 collected at the three key monitoring points of the bridge are set to vibration frequency 5Hz, displacement change 0.02m, and stress 2000N / m^2. The weight factors w1, w2, and w3 are set to 0.5, 0.3, and 0.2 based on data and expert experience. The calculation formula is:

[0076] D multi =0.5·5+0.3·0.02+0.2·2000

[0077] D multi =2.5+0.006+400

[0078] Dmulti =402.506

[0079] The results show that the dynamic monitoring data index after comprehensive weight adjustment is 402.506. This value will be used for structural health monitoring analysis, indicating that the current state of the bridge is safe within the monitoring threshold. It is also convenient for comparing data changes at different time points, thereby conducting long-term health assessments.

[0080] See also Figure 3 , the specific steps for obtaining real-time superimposed visualization images are:

[0081] S211: Based on the multi-dimensional dynamic monitoring data, the three-dimensional model of the bridge is loaded into the AR device, and the structural image part of the multi-dimensional dynamic monitoring data is called to generate the three-dimensional model data of the bridge;

[0082] Based on the multi-dimensional dynamic monitoring data, the process of loading the three-dimensional bridge model in the AR device first needs to determine the compatibility of the model data with the multi-dimensional dynamic monitoring data. This process involves the conversion of data formats and the adjustment of parameters to match the processing capabilities of the AR device. The model data must include sufficient details to support data overlay operations. By adjusting the spatial positioning parameters and scale factors of the model, the accurate representation of the model in the AR device is ensured. At the same time, the geometric accuracy and structural integrity of the model need to be verified to prevent data mismatch problems in subsequent processes, to ensure that the model can correspond to the real-time monitoring data, so as to meet the requirements of real-time display and generate bridge three-dimensional model data.

[0083] S212: Utilize the vibration frequency, displacement change and stress data in the multi-dimensional dynamic monitoring data and map them according to the coordinate system of the three-dimensional bridge model using the formula:

[0084]

[0085] Calculate and generate mapping data;

[0086] Among them, w1 represents the weight coefficient of vibration frequency data, w2 represents the weight coefficient of displacement change data, w3 represents the weight coefficient of stress data, w4 represents the weight coefficient of temperature data, f represents vibration frequency data, d represents displacement change data, s represents stress data, t represents temperature data, p represents proportional factor, and q represents adjustment parameter;

[0087] formula:

[0088]

[0089] The benefit of the formula is that it combines four independent monitoring dimensions, namely vibration frequency, displacement change, stress and temperature, by means of weighted average, making the output mapping data more multidimensional and accurate, thus improving the efficiency of data utilization and the accuracy of monitoring.

[0090] Detailed explanation of the formula and the process of formula calculation and derivation:

[0091] The given parameters are: w1=0.25, w2=0.25, w3=0.25, w4=0.25, f=50 Hz, d=0.02 m, s=175 MPa, t=35°C, p=1.5, q=2.0.

[0092] According to the parameters, first calculate the weight denominator:

[0093]

[0094] The weight numerator is calculated as follows:

[0095]

[0096] The calculated value of V is:

[0097]

[0098] The results show that by comprehensively referring to four differentiated monitoring data, a comprehensive index of 273.6 was obtained. This value reflects the overall stability and safety of the bridge in its current state, providing a scientific basis for the next step of monitoring and maintenance.

[0099] S213: Based on the bridge three-dimensional model data and the mapping data, coordinate matching is performed, the viewing angle and scale are adjusted, and the mapping data is superimposed in the form of color or texture to generate a real-time superimposed visualization image.

[0100] The process of coordinate matching based on the bridge 3D model data and mapping data involves spatial calculations to ensure perfect alignment between the real-time image and the 3D model, and to adjust the viewing angle and scale so that the superimposed image and the model can be visually seamlessly matched. During the operation, it is necessary to control the data transmission and processing speed to achieve the effect of real-time update, and use color or texture changes to intuitively display data changes, increase user perception, and enable changes in monitoring data to be monitored instantly, thereby enhancing the interactivity and practicality of the monitoring system and generating real-time superimposed visualization images.

[0101] See also Figure 4 , the specific steps for obtaining key risk area data are:

[0102] S311: monitor the changes in color and texture from the real-time superimposed visualization image, identify abnormal areas, and generate preliminary identified abnormal area data;

[0103] Monitor changes in color and texture in real-time overlay visualization images. Through image analysis technology, machine learning algorithms are used to automatically identify abnormal areas. Areas may show color or texture changes that are significantly different from normal due to stress concentration or material fatigue. Technicians can identify structural defects or damage by analyzing data. This process relies on image data captured by high-definition cameras and analysis results processed in real time. Through this method, continuous monitoring and evaluation can be carried out without affecting the operation of the bridge, generating preliminary identified abnormal area data;

[0104] S312: Using the initially identified abnormal area data, combined with the vibration frequency, displacement, and stress data exceeding the preset threshold, locate the bridge structure part in the abnormal area, using the formula:

[0105]

[0106] Calculate the parameter deviation of the structural parts and generate the structural parts positioning data;

[0107] Among them, x f,i 、x d,i 、x s,i Respectively represent the actual measured values ​​of vibration frequency, displacement and stress of the i-th measuring point, μ f , μ d , μ s They represent the normal thresholds of vibration frequency, displacement and stress respectively, and n represents the total number of measurement points;

[0108] formula:

[0109]

[0110] The benefit of the formula is that it quantitatively evaluates the risk level of structural parts by calculating the deviation between the measured data and the preset threshold, providing a numerical evaluation standard for bridge maintenance, which helps to timely discover potential structural problems and repair them;

[0111] Detailed explanation of the formula and the process of formula calculation and derivation:

[0112] There are 5 monitoring points on a bridge. The normal thresholds of vibration frequency, displacement and stress are: μ f =50Hz,μ d =0.3cm, μ s =120MPa. The data obtained in one monitoring are as follows: f =

[0113] [48,52,49,51,53]Hz,x d =[0.28,0.35,0.29,0.32,0.34]cm,x s =

[0114] [125,117,122,118,121]MPa.

[0115] Calculate the deviation of the point and apply the formula:

[0116]

[0117] Calculate the sum of each deviation and get:

[0118]

[0119] The results show that the overall health of the structure has fluctuated slightly, but is still within acceptable limits, and can be used to determine whether emergency repairs or routine inspections are needed.

[0120] S313: Call the structural location data, perform numerical comparison and analysis, combine threshold judgment and abnormal monitoring data, determine the risk level, and generate key risk area data.

[0121] Through the above positioning data, combined with threshold analysis and historical trend comparison, numerical comparative analysis is performed. This process involves processing the newly acquired data and comparing it with the existing data. The analysis results are presented using statistical analysis methods to facilitate engineers' interpretation and decision-making. Through this method, the health of the bridge can be monitored, the identified risk areas can be graded, and key risk area data can be generated. These data provide a basis for subsequent maintenance work or reinforcement measures to ensure the long-term safe operation of the bridge.

[0122] See also Figure 5 , the specific steps for obtaining future risk prediction results are:

[0123] S411: Using the key risk area data, the parameters exceeding the threshold are arranged in time series, laying the foundation for parameter trend analysis and generating time series arrangement data;

[0124] Using the key risk area data, the parameters that exceed the threshold are arranged in time series to lay the foundation for parameter trend analysis. This is a key step because it directly affects the accuracy of data analysis and the reliability of risk prediction. By arranging the data, the trend of parameter changes over time can be monitored, so as to better predict and simulate changes in future time periods. This process involves collecting data from each detection point to ensure the integrity and accuracy of the data. At the same time, the data is pre-processed, including cleaning and sorting, so that the data can be effectively used in subsequent steps. In this way, it can be ensured that all data are correctly arranged in chronological order to maximize the effect of the analysis, thereby providing a solid foundation for the generated time series arrangement data.

[0125] S412: Perform parameter trend analysis on the time series data, and use statistical models to estimate the changes in parameters at multiple moments in the future, using the formula:

[0126] P t =α·X t-1 +(1-α)·P t-1

[0127] Calculate the parameter prediction values ​​in the future time period and generate the future parameter change prediction data;

[0128] Among them, P t represents the predicted value at time t, X t-1 represents the actual observation value at time t-1, α is the smoothing constant, P t-1 is the predicted value at the previous moment;

[0129] formula:

[0130] P t =α·X t-1 +(1-α)·P t-1

[0131] The benefit of the formula is that it predicts future parameter values ​​through exponential smoothing, referring to observed values ​​and previous forecast results, so that the forecast can be adjusted according to the current data, improving the flexibility and accuracy of the forecast.

[0132] Detailed explanation of the formula and the process of formula calculation and derivation:

[0133] Set α = 0.2 and set the actual observed value X at the previous moment t-1 =50, the predicted value P at the previous moment t-1 =48,

[0134] The calculation process is as follows:

[0135] P t =0.2·50+0.8·48

[0136] P t =10+38.4

[0137] P t =48.4

[0138] The results show that under a given smoothing constant, the predicted value at the next moment is 48.4, which helps to analyze how parameter values ​​develop in actual situations of cyclic changes and thus predict risks in practical applications.

[0139] S413: simulating parameter changes under differentiated load and environmental conditions, using future parameter change prediction data, and calculating and generating multi-scenario numerical difference data;

[0140] Simulate parameter changes under differentiated load and environmental conditions, use future parameter change prediction data, and calculate numerical differences under various scenarios. This step is an extremely critical part of risk management because it directly affects the analysis and preparation of differentiated risk scenarios. By setting differentiated environmental conditions and load parameters, simulate the performance of parameters under these conditions, such as temperature changes, humidity adjustments, and load increases and decreases. These changes in conditions may have a significant impact on structural safety. Therefore, it is very necessary to carefully analyze the data and calculate its potential impact on safety. In this way, identify the key parameters that are estimated to cause structural failure under target conditions, and then take preventive measures in actual operations to reduce potential risks and losses.

[0141] S414: Compare the numerical difference data of multiple scenarios, determine the risk level under multiple scenarios, integrate the data, and obtain future risk prediction results.

[0142] Compare the numerical differences of multiple scenarios, determine the risk levels under multiple scenarios, integrate data, and obtain future risk prediction results. This process is a summary and refinement of the analysis in the previous steps, which helps decision makers understand the estimated risk level under differentiated situations. By comparing the numerical differences under differentiated scenarios, you can see which parameters are most likely to cause problems and the estimated severity of the problems. This information is critical for formulating response measures and prevention strategies to ensure that you can respond quickly and effectively when encountering adverse situations, thereby protecting the safety of the structure and extending its service life.

[0143] The AR-based bridge inspection system is used to perform the above-mentioned AR-based bridge inspection method, and the system includes:

[0144] The data capture module takes pictures of the structure at the bridge site, collects real-time images of the structure, uses acceleration sensors to record vibration frequency, displacement sensors to record displacement changes, and strain sensors to record stress data. The vibration frequency, displacement changes, and stress data are recorded synchronously in time and integrated into multi-dimensional dynamic monitoring data.

[0145] The model synchronization module loads the 3D model of the bridge into the AR device, extracts the vibration frequency, displacement, and stress data from the multi-dimensional dynamic monitoring data, marks the corresponding positions of the data according to the coordinate system of the 3D model, adjusts the AR perspective and scale to align the real-time image with the 3D model, and uses color and texture to overlay the real-time data onto the model to generate a real-time overlay visualization image.

[0146] The anomaly detection module monitors the changes in color and texture in the real-time superimposed visualization image, identifies areas that differ from the standard model, extracts vibration frequency, displacement, and stress data that exceed the preset threshold, locates the corresponding structural parts, compares the current data with the normal data, and screens the data of key risk areas;

[0147] The trend analysis module arranges the vibration frequency, displacement, and stress parameters that exceed the threshold in the key risk area data in chronological order, analyzes the change trend of multiple parameters, simulates the change of parameters under differentiated loads and environmental conditions, compares the differences in parameters under multiple scenarios, and obtains the parameter change trend analysis results;

[0148] The risk prediction module predicts the vibration frequency, displacement, and stress changes of multiple structural parts at future times based on the parameter change trend analysis results, evaluates the overall risk level of the structure, and generates future risk prediction results.

[0149] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. The bridge inspection method based on AR is characterized by: The following steps are involved: Photograph the bridge on site to capture real-time images of the structure, and use acceleration, displacement, and strain sensors to record vibration frequency, displacement change, and stress data. The real-time images of the structure and the vibration frequency, displacement change, and stress data are synchronously integrated and named as multi-dimensional dynamic monitoring data; Based on the multi-dimensional dynamic monitoring data, a three-dimensional model of the bridge is loaded into the AR device, the vibration frequency, displacement, and stress data are mapped to the corresponding positions of the model, the real-time image is matched with the model coordinates, the viewing angle and scale are adjusted to make the image coincide with the model, and the real-time data is superimposed in the form of color or texture to generate a real-time superimposed visualization image; Through the real-time superposition of visualized images, color and texture changes are monitored, abnormal areas are identified, and the corresponding structural parts are located using the vibration frequency, displacement, and stress data that exceed the threshold, and the parameters of the structural parts are compared to screen out the areas with abnormal values ​​and generate key risk area data; Using the key risk area data, the parameters that exceed the threshold are arranged in time series, parameter trend analysis is performed on the key areas, parameter changes at future times are calculated, changes under differentiated load and environmental conditions are simulated, and numerical differences in multiple scenarios are compared to obtain future risk prediction results.

2. The AR-based bridge inspection method according to claim 1, characterized in that: The steps for obtaining the multi-dimensional dynamic monitoring data are specifically as follows: Use high-definition cameras to photograph the bridge, capture real-time images of the structure, conduct preliminary analysis through automatic image recognition technology, verify key structural elements in the image, and generate real-time structural image data; Acceleration sensors, displacement sensors and strain sensors are installed at multiple key locations of the bridge structure to record vibration frequency, displacement change and stress data in real time. The data is sent to the data processing center through real-time data transmission technology to generate real-time monitoring data of the sensors. Integrate the real-time structural image data and the real-time monitoring data of the sensor, use a multi-dimensional data fusion algorithm to perform data alignment, synchronization and fusion, and use the formula: Generate multi-dimensional dynamic monitoring data; Among them, D multi represents multi-dimensional dynamic monitoring data, w i represents the weight factor of the i-th data source, which is used to adjust the influence of the data source in the overall synthesis result. i Represents the real-time monitoring value of the i-th data source, including vibration frequency, displacement change, and stress data obtained from multi-sensor and image analysis.

3. The AR-based bridge inspection method according to claim 2, characterized in that: The steps of obtaining the real-time superimposed visualization image are specifically as follows: Based on the multi-dimensional dynamic monitoring data, a three-dimensional model of the bridge is loaded into the AR device, and the structural image part of the multi-dimensional dynamic monitoring data is called to generate the three-dimensional model data of the bridge; The vibration frequency, displacement change and stress data in the multi-dimensional dynamic monitoring data are mapped according to the coordinate system of the three-dimensional bridge model using the formula: Calculate and generate mapping data; Among them, w1 represents the weight coefficient of vibration frequency data, w2 represents the weight coefficient of displacement change data, w3 represents the weight coefficient of stress data, w4 represents the weight coefficient of temperature data, f represents vibration frequency data, d represents displacement change data, s represents stress data, t represents temperature data, p represents proportional factor, and q represents adjustment parameter; Based on the bridge three-dimensional model data and the mapping data, coordinate matching is performed, the viewing angle and the scale are adjusted, and the mapping data is superimposed in the form of color or texture to generate a real-time superimposed visualization image.

4. The AR-based bridge inspection method according to claim 3, characterized in that: The steps for obtaining the key risk area data are specifically as follows: Monitoring color and texture changes from the real-time overlaid visualization image, identifying abnormal areas, and generating preliminary identified abnormal area data; The abnormal area data initially identified are used to locate the bridge structure part in the abnormal area in combination with the vibration frequency, displacement and stress data exceeding the preset threshold value, and the formula is used: Calculate the parameter deviation of the structural parts and generate the structural parts positioning data; Among them, x f,i 、x d,i 、x s,i Respectively represent the actual measured values ​​of vibration frequency, displacement and stress of the i-th measuring point, μ f , μ d , μ s They represent the normal thresholds of vibration frequency, displacement and stress respectively, and n represents the total number of measurement points; The structural location data is called to perform numerical comparison and analysis, and the risk level is determined by combining threshold judgment and abnormal monitoring data to generate key risk area data.

5. The AR-based bridge inspection method according to claim 4, characterized in that: The steps for obtaining the future risk prediction results are specifically as follows: Using the key risk area data, the parameters exceeding the threshold are arranged in time series, laying a foundation for parameter trend analysis and generating time series arrangement data; The parameter trend analysis is performed on the time series arrangement data, and the statistical model is used to estimate the changes of the parameters at multiple moments in the future, using the formula: P t =α·X t-1 +(1-a)·P t-1 Calculate the parameter prediction values ​​in the future time period and generate the future parameter change prediction data; Among them, P t represents the predicted value at time t, X t-1 represents the actual observation value at time t-1, α is the smoothing constant, P t-1 is the predicted value at the previous moment; Simulating parameter changes under differentiated load and environmental conditions, and using the future parameter change prediction data to calculate and generate multi-scenario numerical difference data; Compare the multi-scenario numerical difference data, determine the risk level under multiple scenarios, integrate the data, and obtain future risk prediction results.

6. AR-based bridge inspection system, characterized by: According to the AR-based bridge inspection method according to any one of claims 1 to 5, the system comprises: The data capture module takes pictures of the structure at the bridge site, collects real-time images of the structure, uses acceleration sensors to record vibration frequency, displacement sensors to record displacement changes, and strain sensors to record stress data. The vibration frequency, displacement changes, and stress data are recorded synchronously in time and integrated into multi-dimensional dynamic monitoring data. The model synchronization module loads the three-dimensional model of the bridge in the AR device, extracts the vibration frequency, displacement, and stress data in the multi-dimensional dynamic monitoring data, marks the corresponding positions of the data according to the coordinate system of the three-dimensional model, adjusts the AR perspective and scale to align the real-time image with the three-dimensional model, and uses color and texture to overlay the real-time data on the model to generate a real-time overlay visualization image; The anomaly detection module monitors the changes in color and texture in the real-time superimposed visualization image, identifies the area that is different from the standard model, extracts the vibration frequency, displacement, and stress data that exceed the preset threshold, locates the corresponding structural parts, compares the current data with the normal data, and screens the key risk area data; The trend analysis module arranges the vibration frequency, displacement, and stress parameters exceeding the threshold in the key risk area data in chronological order, analyzes the change trend of multiple parameters, simulates the change of parameters under differentiated loads and environmental conditions, compares the differences of parameters under multiple scenarios, and obtains the parameter change trend analysis results; The risk prediction module predicts the vibration frequency, displacement, and stress changes of multiple structural parts at future moments based on the parameter change trend analysis results, evaluates the overall risk level of the structure, and generates future risk prediction results.

Images