A nondestructive testing method and system for detecting the depth of concrete open cracks

Abstract

The present application relates to the technical field of crack detection in construction, and in particular to a nondestructive detection method and system for detecting the depth of concrete open cracks, wherein the method comprises: a database establishment step, an image acquisition step, an image processing step, a three-dimensional visualization model establishment step, a depth initial acquisition step, a health determination step, a comprehensive health index determination step, a reminder step, and an alarm step; wherein the system comprises a database, an image acquisition module, an image processing module, a model building module, a depth initial acquisition module, a health determination module, a comprehensive health index determination module, a reminder module, and an alarm module. The present application has the effect of facilitating the control of the timing of crack repair, reducing the accident rate, and reducing losses.

Description

Technical Field

[0001] The present application relates to the technical field of crack detection in building construction, and in particular to a nondestructive detection method and system for detecting the depth of open cracks in concrete. Background Art

[0002] At present, concrete, as an important civil engineering material, plays a vital role in controlling the quality of engineering projects. At present, a large number of important infrastructure and large-scale civil engineering projects are under the interaction of structural loads, environmental erosion and other factors for a long time. Such concrete civil engineering structures will inevitably have various concrete defects, resulting in a significant decline in the ability of concrete civil engineering to resist sudden natural disasters and normal working life.

[0003] Among the various defects that damage concrete structures, cracks are the most common. Initially, many cracks are micro cracks that are invisible to the naked eye. However, the initial micro cracks often lead to further cracking, and then macro cracks visible to the naked eye appear. These cracks will have a significant impact on the durability and bearing capacity of concrete structures.

[0004] In the prior art, a master's thesis of Chang'an University entitled "Three-dimensional Modeling and Repair Parameter Calculation of Pavement Cracks Based on Machine Vision" records that a depth image is obtained through a depth camera, and then the internal and external parameters of the camera are obtained by calibrating the camera, and then the error source and noise of the original pavement crack depth image are analyzed, and the noise is processed by combining multiple algorithms to enhance the depth image contrast. After performing regional segmentation and uniform illumination to distinguish the cracks from the pavement background, the three-dimensional coordinates of the cracks are obtained to perform three-dimensional reconstruction of the cracks and obtain the three-dimensional information of the cracks, and then the crack width, depth and length are calculated; after the depth of the crack reaches the set threshold, the crack needs to be processed. However, the expansion of the crack is affected by many factors, and a single judgment element can easily lead to excessive concrete cracks, which cannot be repaired or are too difficult to repair, causing accidents or other losses. Summary of the invention

[0005] In order to facilitate the control of the timing of crack repair, reduce the accident rate and minimize losses, the present application provides a non-destructive detection method and system for detecting the depth of open cracks in concrete.

[0006] In the first aspect, the present application provides a non-destructive testing method for detecting the depth of concrete open cracks, which adopts the following technical solution:

[0007] A nondestructive testing method for detecting the depth of concrete open cracks comprises the following steps:

[0008] Database establishment: Establish a three-dimensional coordinate system based on the world coordinate system, measure based on the real position of the concrete building, obtain the real coordinates of the concrete building through measurement, mark the measured coordinates as historical coordinates, set multiple coordinate verification points and measured points, and store the above information in the database;

[0009] Image acquisition: Use tools to calibrate the camera and use the camera to collect depth images of the cracks;

[0010] Image processing: Eliminate black holes and remove noise from the crack depth image, segment the image to obtain a binary image, and then perform mask processing;

[0011] Establishment of 3D visualization model: transform the processed depth image into coordinates to obtain the coordinates of each point of the crack, then establish the 3D visualization model and construct the crack model in the 3D visualization model;

[0012] Initial depth acquisition: Get the coordinates of the crack location and form a crack coordinate set , obtain the coordinates of the concrete surface and form a concrete surface coordinate set , crack coordinate set With concrete surface coordinate set One-to-one mapping, and the ordinate difference between the one-to-one corresponding crack coordinates and the concrete coordinates is recorded as the initial depth set ;

[0013] Health determination: Determine the initial depth set Is there a depth greater than the preset safety threshold? , if yes, then execute the alarm step; otherwise, execute the comprehensive health index judgment step;

[0014] Comprehensive health index judgment: including data acquisition steps, index calculation steps and index comparison steps;

[0015] Data acquisition: Get the maximum width of the crack ,length and the maximum depth ;

[0016] Index calculation: Calculate the comprehensive health index , including the comprehensive health index The calculation model is as follows: ; In the formula, , To set the threshold, To preset a safe depth threshold;

[0017] Index comparison: judging the comprehensive health index Is it greater than 1? If so, execute the reminder step; otherwise, execute the alarm step;

[0018] Reminder: Generate data reports and remind construction personnel to check;

[0019] Alarm: Alarm construction personnel and send relevant information.

[0020] By adopting the above technical solution, through the acquisition and processing of depth images, and then through the construction of a three-dimensional visualization model and a crack model, the initial depth set is calculated, and then the depth is judged. When the depth is greater than the preset safety depth threshold, an alarm is issued, and when it is less than the preset safety depth threshold, the comprehensive health index is calculated, and then it is judged whether the comprehensive health index is greater than 1. If so, the reminder step is executed, otherwise, the alarm step is executed. Through the setting of the health judgment step and the comprehensive health index judgment step, the harmfulness of the cracks is judged and analyzed from multiple angles, and the safety degree of the cracks is comprehensively judged, so that the control of the timing of crack repair is more accurate, and the probability of accidents is reduced. At the same time, it can facilitate subsequent repairs while ensuring safe use, reducing repair costs and other costs.

[0021] Optionally, a warning classification step is provided between the index comparison step and the reminder step;

[0022] Warning classification: The warning reminder levels are divided into Ⅰ, Ⅱ, and Ⅲ, corresponding to the comprehensive health index, and the reminder steps are executed according to the classification:

[0023] like < , then the warning level is Ⅰ;

[0024] like < < , then the warning level is II;

[0025] like > , the warning level is III.

[0026] By adopting the above technical solution and grading the warnings, maintenance personnel can choose the appropriate time for repair according to the warning level, give priority to more urgent maintenance tasks, arrange tasks reasonably, and reduce labor and time costs.

[0027] Optionally, after the health index determination step, a regular monitoring step and an abnormal growth determination step are further provided;

[0028] Periodic monitoring: set the crack detection within the t period Second-rate, The calculation model is as follows: ;

[0029] Abnormal growth judgment: including the initial judgment step, the secondary judgment step and the correction step;

[0030] Initial judgment: Calculate the depth growth difference between two consecutive detections of the same position and form a difference set , determine whether there are multiple points of continuous growth, if so, execute the secondary judgment step, otherwise, execute the correction step;

[0031] Second judgment: count the number of points in continuous positions and form a number set , determine the number of sets Is there a value greater than the upper threshold? If yes, then execute the alarm step;

[0032] Correction: If the point is an abnormal data point, make the depth value of the point equal to the mean of the two adjacent points, and update the initial depth set .

[0033] By adopting the above technical solution, for concrete buildings that can still be used, the expansion status of concrete cracks can be monitored through regular monitoring, thereby extending the service life of concrete buildings and reducing the frequency of maintenance. The expansion of cracks in concrete buildings can be discovered in time, and by judging whether multiple points are growing continuously, abnormal data points can be screened to reduce the interference of the points on the timing of maintenance, thereby improving the accuracy of the judgment of the timing of maintenance.

[0034] Optionally, a periodic updating step is further provided after the index calculation step;

[0035] Periodic update: Get crack data, update The calculation model is as follows:

[0036] ,in," " means taking the larger value, and Take the integer.

[0037] By adopting the above technical solution and changing the number of detections within the t cycle, more intensive detection is performed on cracks that have expanded or extended, thereby reducing the occurrence of accidents caused by cracks that are too large during the period of no detection, making it easier to promptly discover hidden dangers of cracks and eliminate them in a timely manner.

[0038] Optionally, after the depth initial acquisition step, a trend drawing step is also provided;

[0039] Trend drawing: Based on the initial depth set , draw a crack depth trend graph;

[0040] Actual coordinate acquisition: According to the crack depth trend chart, the coordinates of both sides of the crack at the z troughs are obtained. and , and then measure the actual coordinates of the crack positions on both sides of the z valleys and record them as and , and the historical coordinates of both sides of the crack position at the z trough when no cracks appear are recorded as and , measure and obtain the coordinates of the measured points ;

[0041] Coordinate correction: according to the measured point coordinates And the actual coordinates on both sides of the crack position at the z valley and , correct the three-dimensional coordinates formed by the depth image;

[0042] Recalculation: Based on the updated 3D coordinates, a final depth value set is formed ,make , update the initial depth set .

[0043] By adopting the above technical solution, the three-dimensional coordinates obtained through the depth image are calibrated through manual detection and verification of the world coordinates, so that the data after detection and calculation are more accurate, which facilitates the timely discovery of the limit value of crack expansion and more accurate control of the timing of crack repair.

[0044] Optionally, a crack environment detection step is further provided after the recalculation step;

[0045] Crack environment detection: determine whether there is water or ice in the crack. If so, execute the depth estimation calculation step, otherwise execute the health determination step;

[0046] Depth estimation calculation: The depth estimation is obtained by estimating the actual coordinates on both sides of the crack position at the z valleys. ;

[0047] Depth value update: ,in, is the initial depth value of the crack position at the zth valley, and the updated Update to initial depth set .

[0048] By adopting the above technical solution, in rainy or humid areas or in winter, there may be water or ice in the cracks, which will affect the acquisition of depth images. It will also affect traditional ultrasonic testing, resulting in inaccurate data actually obtained, affecting the judgment of the timing of crack repair. During the repair, it is inconvenient to prepare repair materials, resulting in a waste of time, which may cause accidents or increase maintenance costs.

[0049] Optionally, a crack property determination step is provided between the crack environment detection step and the depth estimation calculation step;

[0050] Crack property judgment: extract multiple vertical sections of crack positions from the 3D visualization model, and judge whether the change is uniform based on the geometric parameters of the vertical sections. If so, perform the depth estimation calculation step; otherwise, perform the re-judgment step;

[0051] Determine again: determine whether the change is gradually reduced through the geometric parameters of the vertical section, if so, perform the adjustment step;

[0052] Adjustment: includes the initial acquisition step and the depth recalculation step;

[0053] Initial acquisition: Get the edge depth coordinates on both sides of the crack at the z trough and , get the depth of f positions in the depth direction And the corresponding width ;

[0054] Depth recalculation: Calculate the depth estimate , and execute the depth value update step;

[0055] The calculation model is as follows:

[0056] ; In the formula, is the depth of f positions The mean of The width corresponding to the f positions The mean of .

[0057] By adopting the above technical solution, the depth estimation is different for different types of cracks. When the crack is a uniform crack, the development of the crack is predictable. Therefore, the estimation is made through the coordinates and historical coordinates on both sides of the crack, which has high reliability, so that the calculated depth is more accurate than the result obtained by simply relying on depth image processing. When the crack width decreases with increasing depth, the estimation of underwater or under-ice cracks is more reliable and accurate through the establishment of a three-dimensional model, compared with the depth value obtained by relying on depth image processing, the timing of maintenance is more accurate, and the preparation of maintenance materials is more sufficient and accurate. At the same time, it can improve safety and extend the service life of current concrete buildings.

[0058] Optionally, an environment monitoring step is further provided between the regular monitoring step and the depth value updating step;

[0059] Environmental monitoring: Obtain rainwater conditions and determine whether there is a rise in crack liquid level. If so, execute the actual coordinate acquisition step; otherwise, execute the value-added step;

[0060] Increment: Calculate the value increment ,in, The calculation model is as follows:

[0061] ,make .

[0062] By adopting the above technical solution, when it is not raining, there is no obvious water level rise phenomenon. It is possible to obtain the difference through historical calculation, obtain the numerical increment, and obtain the approximate height of the water level or ice, which is convenient for calculating the depth of the cracks. Of course, the expansion of the cracks may be irregular, and the water level will change or fluctuate. The calculated numerical increment can only be used as a rough estimate, which is more accurate than the original algorithm, thereby making the timing of maintenance more accurate, and the preparation of maintenance materials more sufficient and accurate, while improving safety and extending the service life of current concrete buildings.

[0063] In a second aspect, the present application provides a nondestructive testing system for detecting the depth of concrete open cracks, which adopts the following technical solution:

[0064] A nondestructive testing system for detecting the depth of concrete open cracks includes the following modules:.

[0065] Database: It stores the historical real coordinates of concrete buildings, coordinate verification points and measured point information, as well as historical inspection data;

[0066] Image acquisition module: The output end is connected to the input end of the database, used to collect the depth image of the crack and upload it to the database;

[0067] Image processing module: the input end is connected to the output end of the database and the output end of the image acquisition module, and is used to process the depth image;

[0068] Model building module: The input end is connected to the output end of the image processing module, and is used to build a three-dimensional visualization model, convert the processed depth image into world coordinates, and build a crack model in the three-dimensional visualization model;

[0069] Depth initial acquisition module: The input end is connected to the output end of the model building module to calculate and obtain the initial depth set ;

[0070] Health judgment module: The input end is connected to the output end of the depth initial acquisition module to judge the initial depth set Is there a depth greater than the preset safety threshold? situation;

[0071] Comprehensive health index judgment module: the input end is connected to the output end of the health judgment module, and is used to judge whether the comprehensive health index needs to trigger the reminder module or the alarm module;

[0072] Reminder module: The input end is connected to the output end of the comprehensive health index judgment module to generate data reports and remind construction personnel to check;

[0073] Alarm: The input end is connected to the output end of the comprehensive health index judgment module to alarm the construction personnel and send relevant information.

[0074] By adopting the above technical solution, the database is constructed by acquiring world coordinates, establishing a three-dimensional coordinate system, setting coordinate verification points and storing historical data, and the depth image is collected through the image acquisition module. Then, the model construction module is executed to complete image processing and the construction of the crack model. The initial depth is calculated, and then the comprehensive health index is calculated based on the initial depth, length and width. Then, it is determined whether the alarm module or reminder module needs to be executed. The above settings facilitate the execution of the crack detection method, and at the same time, the cracks can be detected and evaluated to see whether they can still be used and repaired at an appropriate time. The data can be used to estimate the amount of repair material, which facilitates the rapid repair of cracks and reduces the probability of safety accidents.

[0075] In summary, the present application includes at least one of the following beneficial technical effects:

[0076] 1. Through the acquisition and processing of depth images, the construction of a three-dimensional visualization model and a crack model, the initial depth set is calculated, and then the depth is judged. When the depth is greater than the preset safety depth threshold, an alarm is issued, and when it is less than the preset safety depth threshold, the comprehensive health index is calculated, and then it is judged whether the comprehensive health index is greater than 1. If so, the reminder step is executed, otherwise, the alarm step is executed. Through the setting of the health judgment step and the comprehensive health index judgment step, the harmfulness of the crack is judged and analyzed from multiple angles, and the safety degree of the crack is comprehensively judged, so that the control of the crack repair timing is more accurate, the probability of accidents is reduced, and at the same time, it can facilitate subsequent repairs while ensuring safe use, reducing repair costs and other costs;

[0077] 2. Through the alarm classification, maintenance personnel can choose the appropriate repair time according to the alarm level, give priority to more urgent maintenance tasks, arrange tasks reasonably, and reduce labor and time costs;

[0078] 3. For concrete buildings that can still be used, the expansion status of concrete cracks is monitored through regular monitoring, which can extend the service life of concrete buildings and reduce the frequency of maintenance. The expansion of cracks in concrete buildings can be discovered in time, and by judging whether multiple points are growing continuously, abnormal data points can be screened to reduce the interference of this point on the timing of maintenance, thereby improving the accuracy of judgment on the timing of maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 is a flow chart of the nondestructive testing method in Example 1 of the present application;

[0080] Figure 2 It is a flow chart of the regular monitoring step and the abnormal growth judgment step in Example 2 of the present application;

[0081] Figure 3 is a flowchart from the trend drawing step to the recalculation step in Example 2 of the present application;

[0082] Figure 4 It is a flow chart from the crack environment detection step to the depth value update step in Example 2 of the present application. DETAILED DESCRIPTION

[0083] The following combination Figures 1 to 4 This application is described in further detail.

[0084] This embodiment discloses a nondestructive testing method for detecting the depth of concrete open cracks.

[0085] Example 1: Reference Figure 1 , a nondestructive testing method for detecting the depth of concrete open cracks, comprising the following steps:

[0086] S1: Database establishment: A three-dimensional coordinate system is established based on the world coordinate system, and the real position of the concrete building is measured. The real coordinates of the concrete building are obtained through measurement, and the measured coordinates are marked as historical coordinates. Multiple coordinate verification points and actual measurement points are set, and the above information is stored in the database;

[0087] Specifically, the database is established to store the world coordinates of historical concrete buildings during construction, and a three-dimensional coordinate system is established using the world coordinate system to simplify the coordinates for ease of calculation. Multiple coordinate verification points and measured points are set on the concrete buildings, as well as historical measurement data.

[0088] S2: Image acquisition: Use the tool to calibrate the camera and use the camera to collect the depth image of the crack;

[0089] S3: Image processing: Eliminate black holes and remove noise from the crack depth image, segment the image to obtain a binary image, and then perform mask processing;

[0090] S4: Establishment of three-dimensional visualization model: transform the processed depth image into coordinates to obtain the coordinates of each point of the crack, then establish a three-dimensional visualization model, and construct a crack model in the three-dimensional visualization model;

[0091] S5: Initial depth acquisition: Get the coordinates of the crack location and form a crack coordinate set , obtain the coordinates of the concrete surface and form a concrete surface coordinate set , crack coordinate set With concrete surface coordinate set One-to-one mapping, and the ordinate difference between the one-to-one corresponding crack coordinates and the concrete coordinates is recorded as the initial depth set ;

[0092] Specifically, the methods of image acquisition, image processing, the establishment of a three-dimensional visualization model, the construction of a crack model, and the acquisition and calculation of the initial depth, width, and length are all existing technologies. Please refer to the master's thesis of Chang'an University "Three-dimensional Modeling and Repair Parameter Calculation of Pavement Cracks Based on Machine Vision", in which image acquisition relies on a Kinect camera. Image processing first eliminates black holes. Black holes are eliminated by using a black hole filling method to judge each column of data in the depth image, and then the zero point is eliminated by combining the pixel points around the zero point; then filtering is performed to remove noise, and the filtering process uses an average Any one of value filtering, median filtering, Gaussian filtering and bilateral filtering, preferably bilateral filtering; after the acquired depth image is opened in Matlab, its grayscale value range is between 0-255, and 256 grayscale values ​​correspond to 25.6 cm in Matlab. Here, the grayscale value of a certain point in the image is recorded as g, and the actual distance h between the point and the Kinect camera is equal to the distance H between the ground and the camera divided by the integer part of 25.6 multiplied by 25.6 plus the grayscale value g read in Matlab, and the depth value of a certain point of the crack in the real event coordinate system is obtained. The calculation formula is as follows:

[0093] ;

[0094] Where a is the height of the camera from the ground. The mutual conversion relationship between the image coordinate system and the real-world coordinate system derived from the camera calibration can be used to obtain the coordinate value in the real-world coordinate system, and a 3D visualization model can be established. Then, the 3D reconstruction is preprocessed. The crack image is uniformly illuminated and then the watershed segmentation algorithm is used to perform regional segmentation on the crack depth image to obtain a mask image. Based on the combination of the mask image and the crack depth image, the influence of the unevenness of the undamaged road surface is eliminated, and a 3D visualization model of the crack is established.

[0095] S6: Health determination: Determine the initial depth set Is there a depth greater than the preset safety threshold? , if yes, then execute the alarm step; otherwise, execute the comprehensive health index judgment step;

[0096] Specifically, the initial depth is first compared with the preset safety depth threshold. When the crack depth is too deep, the alarm step is triggered in time. When the depth is within the safe range, the comprehensive health index judgment step is performed to make a multi-dimensional judgment on the expansion of the crack. The preset safety depth threshold The value of is determined by parameters such as the concrete material, mortar mixing ratio, and working environment. For example, in a dam, the working environment is a water environment. When the cracks are too deep, water leakage may occur.

[0097] S7: Comprehensive health index determination: including data acquisition step S7-1, index calculation step S7-2 and index comparison step S7-3;

[0098] S7-1: Data acquisition: Get the maximum width of the crack ,length and the maximum depth ;

[0099] S7-2: Index calculation: Calculate the comprehensive health index , including the comprehensive health index The calculation model is as follows: ; In the formula, , To set the threshold, To preset a safe depth threshold;

[0100] Specifically, The value of can be obtained through the historical destruction data of similar concrete buildings, or determined by parameters such as concrete material, mortar mix ratio, and working environment.

[0101] S7-3: Index comparison: judging the comprehensive health index Is it greater than 1? If so, execute the warning classification step; otherwise, execute the alarm step;

[0102] S8: Warning classification: The warning reminder levels are divided into Ⅰ, Ⅱ, and Ⅲ, corresponding to the comprehensive health index, and the reminder steps are executed according to the classification:

[0103] like < , then the warning level is Ⅰ;

[0104] like < < , then the warning level is II;

[0105] like > , the warning level is III.

[0106] S9: Reminder: Generate data report and remind construction personnel to check;

[0107] S10: Alarm: Alarm the construction personnel and send relevant information.

[0108] Specifically, when judging the comprehensive health index, the crack length is first The length can be obtained by calculating the total number of pixels and the actual length of each pixel in the real world. Width calculation: The cracks are corroded and eliminated by the image corrosion algorithm to obtain the crack width. , The calculation model is as follows:

[0109] ; In the formula, is the width of each elimination, is the number of times the cracks are eliminated; the maximum width of the cracks is obtained by counting the maximum number of cracks eliminated here. The maximum depth is obtained by calculating the Z value of the depth coordinate at the deepest position of the crack and the Z value of the ground coordinate; then the index calculation is performed, and the index calculation is related to the product of the maximum width, maximum depth, and maximum length of the crack. The health index is judged by the degree of crack expansion. When the health index is greater than 1, the larger the health index, the higher the safety of the crack. Therefore, the health index greater than 1 is graded. , , and The value can be determined according to the external environment and usage of the concrete or calculated according to historical data. The urgency of the treatment can be determined by the warning level. When the warning level is I, it is necessary to prepare for crack treatment or give priority to it when there is no other emergency work; when the warning level is II, preparations can be made in advance for crack treatment or more attention can be paid to it, and other cracks can be given priority; when the warning level is III, there is no need to pay too much attention to the crack, which will not affect the normal use of the concrete building and will not expand rapidly; when the health index is less than 1, an alarm will be issued.

[0110] The implementation principle of Example 1 of the present application is as follows: through camera calibration, crack image acquisition is performed, and then through image processing and the establishment of a three-dimensional visualization model, coordinate conversion is completed, and a crack model is constructed in the three-dimensional visualization model. Then, the depth, width and length are obtained, and an initial judgment is made on the crack depth to determine whether there are cracks that are too deep, and repair intervention is performed in a timely manner. When the crack depth is less than the safety threshold, the health index is calculated to determine the treatment level of the crack, reasonably arrange the treatment time, and control the repair timing.

[0111] Example 2: Reference Figure 2 The difference between this embodiment and embodiment 1 is that:

[0112] After the database establishment step, a regular monitoring step S11 and an abnormal growth judgment step S12 are also provided;

[0113] S11: Periodic monitoring: Set the crack detection within the t period. Second-rate, The calculation model is as follows: ;

[0114] Specifically, T is the detection period when there is no crack initially, or the detection period when cracks just begin to appear. The value of is related to the design and construction organization requirements, or is determined according to parameters such as the concrete material, mortar mix ratio, working environment, etc., or is determined based on the degree of harm and expansion rate of cracks of different depths, lengths and widths in historical data.

[0115] S12: abnormal growth judgment: including the initial judgment step S12-1, the secondary judgment step S12-2 and the correction step S12-3;

[0116] S12-1: Initial judgment: Calculate the depth growth difference between two consecutive detections of the same position and form a difference set , determine whether there are multiple points of continuous growth, if so, execute the secondary judgment step, otherwise, execute the correction step;

[0117] S12-2: Second judgment: count the number of points at consecutive positions and form a number set , determine the number of sets Is there a value greater than the upper threshold? If yes, then execute the alarm step;

[0118] Specifically, The setting is derived from historical data, or , is the total number of crack length detection points.

[0119] S12-3: Correction: If the point is an abnormal data point, make the depth value of the point equal to the mean of the two adjacent points, and update the initial depth set .

[0120] Specifically, by setting the number of detections within a period of time t, that is, T detections are required within a period of time t; periodic detection of cracks is carried out to facilitate timely detection of changes in cracks and the timing of possible problems in concrete buildings, while ensuring that safety is ensured while minimizing the impact on use; regular monitoring is carried out to determine whether continuous growth occurs and whether cracks are continuously expanding and the expansion range is large, and timely intervention is carried out to reduce the occurrence of safety accidents. At the same time, by determining data anomalies, the data is corrected to reduce the impact of abnormal data caused by environmental factors or camera factors on the actual depth results, or the impact on the health index; Among them, is calculated as the depth value of the c-th detection of the m position minus the depth value of the c-1-th detection of the m position.

[0121] In other embodiments, a periodic updating step S13 is further provided after the index calculation step;

[0122] S13: Periodic update: Get crack data and update The calculation model is as follows:

[0123] ,in," " means taking the larger value, and Take the integer.

[0124] Specifically, as the health index decays, the possibility of crack expansion or safety problems increases. By increasing the number of detections and the frequency of crack detection, crack problems can be discovered and repaired in a timely manner. In other embodiments, a periodic comparison is also provided to determine whether the crack is Is it greater than the maximum value of the set cycle? If yes, then execute the alarm step, where ,and The value of depends on the application environment and degree of hazard of the concrete building.

[0125] refer to Figure 3 In other embodiments, after the initial depth acquisition step, a trend drawing step S14 is further provided;

[0126] S14: Trend drawing: according to the initial depth set , draw a crack depth trend graph;

[0127] S15: Acquisition of actual coordinates: Based on the crack depth trend graph, obtain the coordinates of both sides of the crack at the z troughs and , and then measure the actual coordinates of the crack positions on both sides of the z valleys and record them as and , and the historical coordinates of both sides of the crack position at the z trough when no cracks appear are recorded as and , measure and obtain the coordinates of the measured points ;

[0128] S16: Coordinate correction: according to the measured point coordinates And the actual coordinates on both sides of the crack position at the z valley and , correct the three-dimensional coordinates formed by the depth image;

[0129] S17: Recalculate: Based on the updated three-dimensional coordinates, form a depth final value set ,make , update the initial depth set .

[0130] Specifically, although the depth camera has been calibrated, there may be problems with the coordinates of the acquired depth image due to parameter anomalies or other reasons. The three-dimensional coordinates of the depth image are corrected by verifying the point coordinates and historical coordinates actually obtained on both sides of the crack, and then the depth is recalculated, and the initial depth set is updated to continue executing the health judgment step and the comprehensive health index judgment step.

[0131] refer to Figure 4 In other embodiments, a crack environment detection step S18 is further provided after the recalculation step;

[0132] S18: Crack environment detection: determine whether there is water or ice in the crack, if so, execute the crack property determination step, otherwise execute the health determination step;

[0133] S19: Determine the nature of cracks: extract vertical sections of multiple crack positions from the three-dimensional visualization model, and determine whether the vertical sections are uniformly changed based on the geometric parameters of the vertical sections. If so, perform the depth estimation calculation step; otherwise, perform the re-determination step;

[0134] S20: Depth estimation calculation: The depth estimation is obtained by estimating the actual coordinates on both sides of the crack position at the z valleys. ;

[0135] S21: Depth value update: ,in, is the initial depth value of the crack position at the zth valley, and the updated Update to initial depth set .

[0136] S22: judging again: judging whether the change is gradually reduced by the geometric parameters of the vertical section, and if so, executing the adjustment step;

[0137] S23: Adjustment: including an initial acquisition step S23-1 and a depth recalculation step S23-2;

[0138] S23-1: Initial acquisition: Get the edge depth coordinates on both sides of the crack at the z valleys and , get the depth of f positions in the depth direction And the corresponding width ;

[0139] S23-2: Depth recalculation: Calculate and obtain depth estimation , and execute the depth value update step;

[0140] The calculation model is as follows:

[0141] ; In the formula, is the depth of f positions The mean of The width corresponding to the f positions The mean of .

[0142] Specifically, when there is water or ice in the cracks, it will interfere with the depth image taken by the depth camera and the ultrasonic detection, which will lead to inaccurate depth acquisition. It is possible that the repair time of the cracks will be delayed or the preparation of materials will be insufficient due to inaccurate depth calculation. In order to increase the accuracy, the cross-section of the constructed crack model is analyzed to determine the development trend of the cracks, whether it is a uniform crack or a gradually smaller crack, and then calculate them separately. For uniform cracks, a depth valuation calculation model can be established when estimating the depth. The calculation model is as follows:

[0143] ; In the formula, for The x-coordinate value of the point. for The x-coordinate value of the point. for The y-coordinate value of the point. for The y-coordinate value of the point. for The coordinate value of the point in the x direction; the calculation is performed using the above formula, or the crack depth is estimated by manually inserting a tool into the crack, or other methods in which the water level has less impact on the detection are used to estimate; when the crack gradually shrinks, the approximate change pattern of the crack is obtained by calculation, and a depth estimation calculation model is established through function fitting to calculate the depth estimation; through the above intervention, the detection results are optimized, the impact of water on depth image analysis is reduced, and the accuracy of depth and health index calculation is improved, thereby increasing the probability of timely detection of crack changes and making the actual control of repair more accurate.

[0144] In other embodiments, an environment monitoring step S26 is further provided between the regular monitoring step and the depth value updating step;

[0145] S24: Environmental monitoring: Obtain rainwater conditions and determine whether there is a rise in the crack liquid level. If so, execute the actual coordinate acquisition step; otherwise, execute the value-added step;

[0146] S25: Increment: Calculate the value increment ,in, The calculation model is as follows:

[0147] ,make .

[0148] Specifically, by judging whether the liquid level changes again, if there is a liquid level change, manual intervention is required again to correct the detection accuracy. Otherwise, the numerical increment is obtained according to the historical calculation, and the current liquid level is roughly estimated. Then, the depth value obtained by the new calculation is intervened, so that the calculation of the health index and the depth value are more accurate, which is convenient for judging the safety of the crack.

[0149] The implementation principle of Example 2 of the present application is: by setting periodic detection, the real-time safety of cracks is monitored, and at the same time, according to the degree of harm of the cracks, the monitoring cycle is updated in time to facilitate timely detection of crack expansion and timely repair treatment.

[0150] When it rains a lot or there is water in the cracks due to other reasons, the depth data obtained through image analysis is corrected through manual correction and depth estimation, and then the health index is corrected, making the control of repair timing and material conditions more accurate.

[0151] In the absence of temporary rainfall or other circumstances, the liquid level may drop, but in most cases it will not rise. Therefore, through liquid level estimation, the crack depth obtained by image analysis is corrected to make the crack depth closer to the actual one, which makes it easier to discover the harmfulness of the cracks and make timely repairs.

[0152] The embodiment of the present application also discloses a nondestructive testing system for detecting the depth of open cracks in concrete.

[0153] Reference Figure 2 , a non-destructive testing system for detecting the depth of open cracks in concrete, includes the following modules:.

[0154] Database: It stores the historical real coordinates of concrete buildings, coordinate verification points and measured point information, as well as historical inspection data;

[0155] Image acquisition module: The output end is connected to the input end of the database, used to collect the depth image of the crack and upload it to the database;

[0156] Image processing module: the input end is connected to the output end of the database and the output end of the image acquisition module, and is used to process the depth image;

[0157] Model building module: The input end is connected to the output end of the image processing module, and is used to build a three-dimensional visualization model, convert the processed depth image into world coordinates, and build a crack model in the three-dimensional visualization model;

[0158] Depth initial acquisition module: The input end is connected to the output end of the model building module to calculate and obtain the initial depth set ;

[0159] Health judgment module: The input end is connected to the output end of the depth initial acquisition module to judge the initial depth set Is there a depth greater than the preset safety threshold? situation;

[0160] Comprehensive health index judgment module: the input end is connected to the output end of the health judgment module, and is used to judge whether the comprehensive health index needs to trigger the reminder module or the alarm module;

[0161] Warning classification module: The input end is connected to the output end of the comprehensive health index judgment module, and is used to classify the cracks according to the health index, so that the repair personnel can intuitively determine the priority of the repair;

[0162] Regular monitoring module: the input end is connected to the output end of the comprehensive health index judgment module, and is used to control the depth camera to perform regular monitoring of cracks;

[0163] Cycle update module: The input end is connected to the output end of the comprehensive health index judgment module, and is used to adjust the cycle according to the degree of damage of the crack;

[0164] Abnormal growth judgment module: The input end is connected to the output end of the comprehensive health index judgment module, and is used to judge whether there are multiple points of continuous growth, large-scale growth or data abnormal points, and to correct the depth value or execute the alarm module;

[0165] Reminder module: The input end is connected to the output end of the comprehensive health index judgment module to generate data reports and remind construction personnel to check;

[0166] Alarm module: The input end is connected to the output end of the comprehensive health index judgment module to alarm the construction personnel and send relevant information.

[0167] Trend drawing module: the input end is connected to the output end of the depth initial acquisition module, and is used to determine the trough position of the depth according to the depth value;

[0168] Actual coordinate acquisition module: The input end is connected to the output end of the trend drawing module for manual detection of coordinates;

[0169] Coordinate correction module: The input end is connected to the output end of the actual coordinate acquisition module, and is used to update the depth image coordinates according to the manually detected coordinates;

[0170] Recalculation module: The input end is connected to the output end of the coordinate correction module, and is used to recalculate the crack depth, width, and length according to the updated depth image coordinates;

[0171] Crack environment detection module: the input end is connected to the output end of the database to determine whether there is water or ice in the crack;

[0172] Crack property judgment module: the input end is connected to the output end of the crack environment detection module and the output end of the model construction module, and is used to judge the development of the crack according to the cross section of the crack model;

[0173] Depth estimation calculation module: The input end is connected to the output end of the crack property judgment module, and is used to manually estimate the depth value of uniform cracks or estimate the depth value based on the historical coordinates and actual coordinates of the cracks, and update the initial depth set;

[0174] Depth recalculation module: the input end is connected to the output end of the crack property judgment module, and is used to judge whether the crack is a crack that is gradually shrinking, and if it is judged to be so, the depth is estimated according to the change trend of the crack;

[0175] Environmental judgment module: The input end is connected to the output end of the periodic monitoring module, and is used to determine whether there is a liquid level rise. If there is no rise, it will increase the value and estimate the depth based on the depth obtained by image analysis.

[0176] The implementation principle of a nondestructive testing system for detecting the depth of open cracks in concrete according to an embodiment of the present application is as follows: crack image information is obtained through an image acquisition module, and then an image processing module, a model building module, and a depth initial acquisition module are executed in sequence to obtain a depth value, and then a health judgment module is executed to determine whether to execute an alarm module or a comprehensive health index judgment module. When executing the comprehensive health index judgment module, a warning classification module is executed according to the health index for classification, and a reminder module is executed according to the classification.

[0177] When cracks appear, it is necessary to execute the periodic monitoring module to perform periodic monitoring on the cracks, and execute the periodic updating module through the acquired data to update the monitoring period; and at the same time execute the abnormal growth judgment module to judge whether there are abnormal growth points.

[0178] The trend drawing module is executed by obtaining the initial depth set, and the trough position is determined according to the trend. Then, the actual coordinate acquisition module is executed, and the coordinate correction module and the recalculation module are executed according to the actually obtained coordinates to update the initial depth set.

[0179] Through the execution of the crack environment detection module and the crack property judgment module, the cracks are analyzed, the depth estimation calculation module or the depth recalculation module is executed, the depth value at the trough is recalculated, and the initial depth set is updated; when the periodic monitoring module is executed, the environment judgment module is executed at the same time to determine whether there is a liquid level rise in the crack, the acquired image analysis depth value is incremented, the new depth value at the trough is obtained, and the initial depth set is updated.

[0180] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A nondestructive testing method for detecting the depth of concrete open cracks, characterized in that: The following steps are involved: Database establishment: Establish a three-dimensional coordinate system based on the world coordinate system, measure based on the real position of the concrete building, obtain the real coordinates of the concrete building through measurement, mark the measured coordinates as historical coordinates, set multiple coordinate verification points and measured points, and store the above information in the database; Image acquisition: Use tools to calibrate the camera and use the camera to collect depth images of the cracks; Image processing: Eliminate black holes and remove noise from the crack depth image, segment the image to obtain a binary image, and then perform mask processing; Establishment of 3D visualization model: transform the processed depth image into coordinates to obtain the coordinates of each point of the crack, then establish the 3D visualization model and construct the crack model in the 3D visualization model; Initial depth acquisition: Get the coordinates of the crack location and form a crack coordinate set , obtain the coordinates of the concrete surface and form a concrete surface coordinate set , crack coordinate set With concrete surface coordinate set One-to-one mapping, and the ordinate difference between the one-to-one corresponding crack coordinates and the concrete coordinates is recorded as the initial depth set ; Health determination: Determine the initial depth set Is there a depth greater than the preset safety threshold? , if yes, then execute the alarm step; otherwise, execute the comprehensive health index judgment step; Comprehensive health index judgment: including data acquisition steps, index calculation steps and index comparison steps; Data acquisition: Get the maximum width of the crack ,length and the maximum depth ; Index calculation: Calculate the comprehensive health index , including the comprehensive health index The calculation model is as follows: ; In the formula, , To set the threshold, To preset a safe depth threshold; Index comparison: judging the comprehensive health index Is it greater than 1? If so, execute the reminder step; otherwise, execute the alarm step; Reminder: Generate data reports and remind construction personnel to check; Alarm: Alarm construction personnel and send relevant information.

2. The nondestructive testing method for detecting the depth of concrete open cracks according to claim 1 is characterized in that: Between the index comparison step and the reminder step, there is also a warning classification step; Warning classification: The warning reminder levels are divided into Ⅰ, Ⅱ, and Ⅲ, corresponding to the comprehensive health index, and the reminder steps are executed according to the classification: like < , then the warning level is Ⅰ; like < < , then the warning level is II; like > , then the warning level is Ⅲ, where, and is the classification threshold, , .

3. The nondestructive testing method for detecting the depth of concrete open cracks according to claim 2 is characterized in that: After the health index determination step, a regular monitoring step and an abnormal growth determination step are also provided; Periodic monitoring: set the crack detection within the t period Second-rate, The calculation model is as follows: ; Abnormal growth judgment: including the initial judgment step, the secondary judgment step and the correction step; Initial judgment: Calculate the depth growth difference between two consecutive detections of the same position and form a difference set , determine whether there are multiple points of continuous growth, if so, execute the secondary judgment step, otherwise, execute the correction step; Second judgment: count the number of points in continuous positions and form a number set , determine the number of sets Is there a value greater than the upper threshold? If yes, then execute the alarm step; Correction: If the point is an abnormal data point, make the depth value of the point equal to the mean of the two adjacent points, and update the initial depth set .

4. The nondestructive testing method for detecting the depth of concrete open cracks according to claim 3 is characterized in that: A periodic updating step is also provided after the index calculation step; Periodic update: Get crack data, update The calculation model is as follows: ,in," " means taking the larger value, and Take the integer.

5. The nondestructive testing method for detecting the depth of concrete open cracks according to any one of claims 1 to 4, characterized in that: After the initial depth acquisition step, a trend drawing step is also provided; Trend drawing: Based on the initial depth set , draw a crack depth trend graph; Actual coordinate acquisition: According to the crack depth trend chart, the coordinates of both sides of the crack at the z troughs are obtained. and , and then measure the actual coordinates of the crack positions on both sides of the z valleys and record them as and , and the historical coordinates on both sides of the crack position at the z trough when no cracks appear are recorded as and , measure and obtain the coordinates of the measured points ; Coordinate correction: according to the measured point coordinates And the actual coordinates on both sides of the crack position at the z valley and , correct the three-dimensional coordinates formed by the depth image; Recalculation: Based on the updated 3D coordinates, a final depth value set is formed ,make , update the initial depth set .

6. The nondestructive testing method for detecting the depth of concrete open cracks according to claim 5 is characterized in that: A crack environment detection step is also provided after the recalculation step; Crack environment detection: determine whether there is water or ice in the crack. If so, execute the depth estimation calculation step, otherwise execute the health determination step; Depth estimation calculation: The depth estimation is obtained by estimating the actual coordinates on both sides of the crack position at the z valleys. ; Depth value update: ,in, is the initial depth value of the crack position at the zth valley, and the updated Update to initial depth set .

7. The nondestructive testing method for detecting the depth of concrete open cracks according to claim 6 is characterized in that: Between the crack environment detection step and the depth estimation calculation step, a crack property judgment step is also provided; Crack property judgment: extract multiple vertical sections of crack locations from the 3D visualization model, and judge whether the change is uniform based on the geometric parameters of the vertical sections. If so, perform the depth estimation calculation step; Otherwise, the re-judgment step is executed; Determine again: determine whether the change is gradually reduced through the geometric parameters of the vertical section, if so, perform the adjustment step; Adjustment: includes the initial acquisition step and the depth recalculation step; Initial acquisition: Get the edge depth coordinates on both sides of the crack at the z trough and , get the depth of f positions in the depth direction And the corresponding width ; Depth recalculation: Calculate the depth estimate , and execute the depth value update step; The calculation model is as follows: ; In the formula, is the depth of f positions The mean of The width corresponding to the f positions The mean of .

8. The nondestructive testing method for detecting the depth of concrete open cracks according to claim 7, characterized in that: An environment monitoring step is also provided between the regular monitoring step and the depth value updating step; Environmental monitoring: Obtain rainwater conditions and determine whether there is a rise in crack liquid level. If so, execute the actual coordinate acquisition step; otherwise, execute the value-added step; Increment: Calculate the value increment ,in, The calculation model is as follows: ,make .

9. A nondestructive testing system for detecting the depth of concrete open cracks, characterized in that: The nondestructive testing method for detecting the depth of concrete open cracks according to any one of claims 1 to 8 comprises the following modules: Database: It stores the historical real coordinates of concrete buildings, coordinate verification points and measured point information, as well as historical inspection data; Image acquisition module: The output end is connected to the input end of the database, used to collect the depth image of the crack and upload it to the database; Image processing module: the input end is connected to the output end of the database and the output end of the image acquisition module, and is used to process the depth image; Model building module: The input end is connected to the output end of the image processing module, and is used to build a three-dimensional visualization model, convert the processed depth image into world coordinates, and build a crack model in the three-dimensional visualization model; Depth initial acquisition module: The input end is connected to the output end of the model building module to calculate and obtain the initial depth set ; Health judgment module: The input end is connected to the output end of the depth initial acquisition module to judge the initial depth set Is there a depth greater than the preset safety threshold? situation; Comprehensive health index judgment module: the input end is connected to the output end of the health judgment module, and is used to judge whether the comprehensive health index needs to trigger the reminder module or the alarm module; Reminder module: The input end is connected to the output end of the comprehensive health index judgment module to generate data reports and remind construction personnel to check; Alarm module: The input end is connected to the output end of the comprehensive health index judgment module to alarm the construction personnel and send relevant information.

Images