Concrete internal crack detection method based on image-impact echo

Abstract

The invention relates to the technical field of concrete crack detection, in particular to a concrete internal crack detection method based on image-impact echoes, which comprises the following steps: acquiring and marking a concrete crack data set, and preprocessing through a data enhancement technology; capturing multi-scale features of the crack by using an RFB network module, and training a target detection model; the trained model is applied to detect target concrete, the crack degree and coordinates are obtained, and a two-dimensional crack graph is drawn to divide crack areas; then, based on a crack area layout monitoring point, recording an echo signal of the concrete by using an impact echo technology, extracting crack characteristics and inputting the crack characteristics into the machine learning model, and obtaining a trained model to analyze an internal crack of the target concrete; and expanding the two-dimensional crack graph into a three-dimensional crack graph to obtain an actual crack degree. According to the method, the detection efficiency is improved, a visual effect is provided, and an effective technical means is provided for concrete structure crack monitoring.

Description

Technical Field

[0001] The invention relates to the technical field of concrete crack detection, in particular to a method for detecting internal cracks in concrete based on image-impact echo. Background Art

[0002] With the development of modern architecture and infrastructure, concrete is widely used in various projects as the main building material. However, concrete often develops internal cracks during long-term use due to factors such as load changes, environmental impacts and material aging. These cracks may cause serious structural safety hazards and even cause accidents. Therefore, accurately detecting internal cracks in concrete is crucial to ensure the safety and durability of buildings.

[0003] Traditional crack detection methods use methods such as drilling and cutting, which often cause physical damage to concrete structures, which not only increases the cost of repair, but may also affect the safety and stability of the structure. Although some methods have improved detection efficiency by using image detection technology in recent years, many methods still rely on a single technical means and lack comprehensive analysis of multi-source information. At the same time, fixed detection devices limit the flexibility and adaptability of detection and cannot fully cover concrete structures, making the evaluation efficiency of concrete structures still insufficient.

[0004] Therefore, a concrete internal crack detection method based on image-impact echo was proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a method for detecting internal cracks in concrete based on image-impact echo, by acquiring and marking a concrete crack data set, and preprocessing it through data enhancement technology; using the RFB network module to capture the multi-scale characteristics of the cracks, and training the target detection model; using the trained model to detect the target concrete, obtain the crack degree and coordinates, and draw a two-dimensional crack map to divide the crack area; based on the layout of monitoring points in the crack area, using the impact echo technology to record the echo signal of the concrete, and extracting the crack features and inputting them into the machine learning model, a trained model is obtained to analyze the internal cracks of the target concrete; the two-dimensional crack map is expanded into a three-dimensional crack map, so as to obtain the actual crack degree. This method improves the detection efficiency and reliability, provides a visualization effect, and thus provides an effective technical means for concrete structure health monitoring.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for detecting internal cracks in concrete based on image-impact echo, comprising:

[0008] Acquire a concrete crack data set, mark the crack degree of the concrete crack data set to obtain a second data set, and preprocess the second data set using data enhancement to obtain a first data set;

[0009] Introducing an RFB network module into the target detection model, wherein the RFB network module is used to capture multi-scale features of cracks; and training the target detection model according to the first data set;

[0010] Acquire target concrete data, input the target concrete data into the trained target detection model, obtain the crack degree and crack coordinates, draw a two-dimensional crack map according to the crack degree and the crack coordinates, and divide the crack area;

[0011] Layout monitoring points of the target concrete according to the crack area, detect the monitoring points using impact echoes, and record echo signals of the target concrete;

[0012] Extracting crack features of echo signals from the concrete crack data set using digital signal processing technology, and training a machine learning model based on the crack features;

[0013] Inputting the target concrete echo signal into the trained machine learning model to obtain the internal crack results of the target concrete;

[0014] The two-dimensional crack map is expanded into a three-dimensional crack map according to the internal crack results of the target concrete, and the overall crack defect area of ​​the target concrete is determined according to the three-dimensional crack map to obtain the actual crack degree.

[0015] Furthermore, the crack degree acquisition process includes:

[0016] Using image processing software to measure the concrete cracks according to the concrete crack data set, and obtaining the crack size according to the starting point and the end point of the concrete crack;

[0017] Obtain the RGB value of the concrete crack by using a color analysis tool, and obtain the brightness and saturation according to the concrete background color;

[0018] Using Gabor filter to extract texture features of the concrete cracks to obtain crack complexity;

[0019] The crack degrees of the concrete crack dataset are divided and marked according to the crack size, the brightness, the saturation, the crack complexity and expert experience, and the crack degrees include obvious cracks, slight cracks and no cracks.

[0020] Furthermore, the pre-processing process includes:

[0021] The data enhancement includes geometric change, color transformation and perspective transformation. After enhancing the second data set N times, a newly generated image is obtained. The newly generated image is merged with the image in the concrete crack data set to obtain a third data set.

[0022] resizing the third data set, wherein the resizing includes: if the third data set is larger than a preset requirement, clipping it according to the preset requirement using a sliding window method, and filling the area of ​​the clipped third data set smaller than the preset requirement with white; otherwise, clipping is not performed and the area of ​​the third data set smaller than the preset requirement is directly filled with white;

[0023] The first data set is obtained according to the adjusted third data set.

[0024] Furthermore, the target detection model training step includes:

[0025] Step S20: inputting the first data set into the target detection model;

[0026] Step S21: introducing the feature extraction layer of the target detection model into the RFB network module, the RFB network module processes the first data set in parallel to obtain a crack feature map, and performs feature fusion on the crack feature map to obtain a crack multi-scale feature map;

[0027] Step S22: using a detection layer to output the crack coordinates, crack confidence and crack degree according to the crack multi-scale feature map;

[0028] Step S23: using a loss function to calculate the gap between the predicted value and the true label according to the crack coordinates, the crack confidence and the crack degree, to obtain a crack loss value;

[0029] Step S24: Calculate parameter gradients using the chain method according to the crack loss value, and update the target detection model according to the parameter gradients;

[0030] Step S25: If the number of iterations reaches the first threshold, the training is terminated; otherwise, steps S20 to S25 are repeated.

[0031] Furthermore, the two-dimensional crack map establishment process includes:

[0032] Taking the upper left corner of the target concrete as the origin of the coordinate system, a two-dimensional coordinate system is established according to the actual size of the target concrete, wherein the two-dimensional coordinate system includes an X-axis and a Y-axis, wherein the X-axis extends horizontally to the right and the Y-axis extends vertically downward;

[0033] Check whether the crack coordinates are the clipped coordinates. If the crack coordinates are the clipped coordinates, add the sliding window offset to the crack coordinates to obtain the first crack coordinates; otherwise, directly assign the crack coordinates to the first crack coordinates;

[0034] The actual coordinates are calculated according to the first crack coordinates using a scaling ratio, and the two-dimensional crack map is obtained according to the actual coordinates.

[0035] Furthermore, the crack region division process includes:

[0036] Calculate the center coordinates of the crack according to the crack coordinates;

[0037] The distance between the cracks is calculated using the Euclidean distance formula according to the central coordinates of the cracks, and the cracks whose distance is less than a second threshold value are divided into the same area to obtain a clustering result;

[0038] According to the clustering result, the number of cracks is calculated. If the number of cracks in the area exceeds the third threshold or at least one crack is an obvious crack, a surface crack area is obtained and the surface crack area is marked as a defect area; otherwise, it is marked as a non-defective area.

[0039] Furthermore, the monitoring point layout includes:

[0040] According to the size of the defect area and the number of cracks, the number of monitoring points in the defect area is obtained;

[0041] According to the size of the defect-free area and the number of cracks, the number of monitoring points in the defect-free area is obtained;

[0042] Arranging monitoring points at the junction of the defective area and the non-defective area to obtain the number of monitoring points at the junction;

[0043] The monitoring point spacing is calculated according to the number of monitoring points in the defective area, the number of monitoring points in the non-defective area and the number of monitoring points at the junction, and the monitoring point positions are marked according to the monitoring point spacing.

[0044] Furthermore, the crack feature acquisition process includes:

[0045] Acquire echo signal data according to the concrete crack data set;

[0046] Using a high-pass filter to remove background noise according to the echo signal data to obtain a first echo signal;

[0047] Performing signal segmentation using an overlapping segmentation method according to the first echo signal to obtain a second echo signal;

[0048] Calculate the signal peak value and the time when the impact reaches the monitoring point according to the second echo signal to obtain the time domain feature;

[0049] Extracting frequency amplitude using Fourier transform according to the second echo signal to obtain frequency domain features;

[0050] Extracting a signal using wavelet transform according to the second echo signal to obtain a time-frequency feature;

[0051] The time domain feature, the frequency domain feature and the time-frequency feature are combined to obtain the crack feature.

[0052] Furthermore, the machine learning model training process includes:

[0053] Acquire a historical echo signal and a historical echo signal label according to the concrete crack data set, wherein the historical echo signal label includes an internal crack depth and width;

[0054] The crack characteristics are input into the machine learning model, and the trained machine learning model is obtained using a first loss function according to the crack characteristics and the historical echo signal labels.

[0055] Furthermore, the actual crack degree acquisition process includes:

[0056] Extracting crack information according to the three-dimensional crack map;

[0057] The surface crack cross-sectional area and the internal crack cross-sectional area are calculated according to the crack information, and the actual crack degree is obtained by performing weighted summation based on the surface crack cross-sectional area and the internal crack cross-sectional area.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] 1. In order to improve the efficiency and accuracy of concrete surface crack detection, especially in applications under complex environments, the image is preprocessed using data enhancement technology and the multi-scale features of cracks are captured by the RFB network module. This can improve the sensitivity and accuracy of the detection model to cracks. The target detection model can accurately assess the degree of cracks and locate crack coordinates, and a two-dimensional crack map can be drawn based on this. The crack distribution and characteristics are clearly displayed, and the precise division of the surface crack area is achieved, thereby improving the efficiency and reliability of crack detection.

[0060] 2. In the internal detection of concrete, traditional impact echo detection technology cannot quickly and accurately locate deep cracks. By arranging monitoring points in different regions of the target concrete, accurate monitoring of the crack area can be achieved to ensure comprehensive and flexible detection coverage. In addition, the use of digital signal processing technology to extract crack features in echo signals not only improves the accuracy of feature extraction, but also effectively removes background noise. The extracted crack features are input into the machine learning model, which can train an efficient and accurate model, realize accurate identification of cracks inside concrete, and improve the efficiency and reliability of crack detection.

[0061] 3. Accurately dividing the defect area of ​​concrete cracks is crucial for structural inspection and maintenance. By expanding the two-dimensional crack map into a three-dimensional crack map, the three-dimensional characteristics of the internal cracks in the concrete can be fully displayed, making the assessment of the crack defect area more accurate. In addition, the three-dimensional crack map helps to improve the efficiency of dividing the overall crack defect area, and calculates the internal and external crack area areas based on the internal crack depth, width and crack coordinates, thereby obtaining an accurate actual crack degree, improving the efficiency and reliability of crack detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 A schematic diagram of a flow chart of a method for detecting internal cracks in concrete based on image-impact echo provided by the present invention;

[0063] Figure 2 A schematic diagram of a crack degree classification structure provided by an embodiment of the present invention;

[0064] Figure 3 A schematic diagram of the target detection model training process provided by an embodiment of the present invention;

[0065] Figure 4 A schematic diagram of the monitoring point layout provided by an embodiment of the present invention;

[0066] Figure 5 A schematic diagram of an overall crack defect area provided by an embodiment of the present invention.

[0067] In the figure: 1. Non-defective area; 2. Monitoring point; 3. Crack; 4. Defective area; 5. First crack area; 6. Second crack area. DETAILED DESCRIPTION

[0068] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0069] In modern industrial environments, concrete is the main building material, and its structural safety is crucial to the operation of the plant. Over time, concrete structures are susceptible to various factors, such as temperature changes, humidity, chemical corrosion, and mechanical loads, which can lead to the formation of internal cracks. Traditional concrete inspection methods, such as manual inspection and acoustic inspection, although effective, have defects such as low efficiency, limited coverage, and possible damage to the structure. Although some methods have been proposed to improve the inspection efficiency, they still rely on fixed inspection devices, which limits flexibility and adaptability, and are also demanding in terms of capital and time. Therefore, there is an urgent need to develop a novel, efficient and non-destructive inspection method to further improve the inspection efficiency so as to more comprehensively evaluate the safety of the internal structure of concrete.

[0070] Embodiment 1

[0071] Take a factory as an example, and inspect the concrete blocks in the factory. Through comprehensive inspection of concrete, potential structural problems can be discovered in time, so that appropriate maintenance measures can be taken to ensure the safe operation and production efficiency of the factory. This method can not only improve the inspection efficiency, but also provide important data support for the long-term monitoring and management of concrete.

[0072] See also Figures 1 to 5 The present invention provides a method for detecting internal cracks in concrete based on image-impact echo, and the technical solution is as follows:

[0073] like Figure 1 As shown, a method for detecting internal cracks in concrete based on image-impact echo comprises:

[0074] Step S1: Acquire a concrete crack data set, mark the crack degree of the concrete crack data set to obtain a second data set, and use data enhancement to preprocess the second data set to obtain a first data set.

[0075] Among them, the concrete crack dataset includes concrete photo set and concrete echo signal data.

[0076] Specifically, a variety of concrete structures were selected in the factory, and high-definition cameras were used to take photos of cracks under different conditions to ensure that different types of cracks (such as horizontal, vertical, staggered, etc.) and interference items (such as plant branches, artificial graffiti, and water flow marks, etc.) were included to increase the diversity of the data set. In addition, a signal exciter and a high-frequency sensor were used. The signal exciter was brought close to the concrete surface and an impact force was applied. The sensor collected the reflected wave signal and recorded the time and intensity of the reflected wave. The computer software and experts analyzed it and calculated the path of the sound wave from the surface to the crack and reflected back, thereby determining the crack depth and crack width.

[0077] Furthermore, the crack degree acquisition process includes:

[0078] The concrete cracks are measured using image processing software according to the concrete crack data set. According to the starting point and the ending point of the concrete crack, the software can calculate the distance between the two points to obtain the crack size; for example, in ImageJ, the "line segment tool" can be used to mark the crack, and then the length data can be obtained in the "measurement" function;

[0079] The RGB value of the concrete crack is obtained by a color analysis tool, and the brightness and saturation are obtained according to the concrete background color; for example, in Photoshop, the RGB value of the selected area can be viewed using the "Information" panel, and the brightness value is calculated according to the RGB value using the brightness calculation formula, and the saturation of the background color is calculated by HSV color space conversion;

[0080] The Gabor filter can capture texture information of different frequencies and directions, and the Gabor filter is used to extract the texture features of the concrete cracks to obtain the crack complexity;

[0081] According to the crack size, the brightness, the saturation, the crack complexity and expert experience, the crack degree of the concrete crack dataset is divided and marked, such as Figure 2 As shown, the crack degrees include obvious cracks, slight cracks and no cracks.

[0082] The conditions for division are:

[0083] Obvious cracks: length greater than 15 cm; width greater than 3 mm; brightness less than or equal to 150; saturation less than or equal to 30; height irregular, with obvious peeling or falling off;

[0084] Slight cracks: length less than or equal to 15 cm and greater than or equal to 5 cm; width less than or equal to 3 mm and less than or equal to 1 mm; brightness greater than 150 and less than or equal to 200; saturation greater than 30 and less than or equal to 60; blurred boundaries and branches;

[0085] No cracks: length less than 5 cm, width less than 1 mm; brightness greater than 200; saturation greater than 60; regular texture and clear boundaries.

[0086] Mark the category of each crack in the table and draw a rectangular box of the crack on the image to generate a labeling file.

[0087] By systematically measuring and classifying concrete cracks, accurate assessment of crack status is achieved. In addition, the generated annotation files record the location, category, size and characteristics of the cracks, providing a comprehensive data basis for subsequent analysis, thereby improving the efficiency and reliability of crack detection.

[0088] Furthermore, the pre-processing process includes:

[0089] The data enhancement includes geometric change, color transformation and perspective transformation. After enhancing the second data set N times, a newly generated image is obtained. The newly generated image is merged with the image in the concrete crack data set to obtain a third data set.

[0090] Among them, N can be set to 5 to 10 times, and the generated enhanced images are randomly checked to ensure that the crack characteristics of the enhanced images are still obvious and consistent with the actual situation.

[0091] Specifically, the geometric transformation is to randomly rotate, translate and scale the image to simulate cracks at different perspectives; the color transformation is to randomly change the brightness and hue of the image to simulate different lighting conditions; the perspective transformation is to simulate the impact of different shooting angles on crack characteristics to ensure that the model can handle cracks at different perspectives;

[0092] resizing the third data set, wherein the resizing includes: if the third data set is larger than a preset requirement, clipping it according to the preset requirement using a sliding window method, and filling the area of ​​the clipped third data set smaller than the preset requirement with white; otherwise, clipping is not performed and the area of ​​the third data set smaller than the preset requirement is directly filled with white;

[0093] The first data set is obtained according to the adjusted third data set.

[0094] Specifically, assuming that the captured image size is 1024×768 pixels and the preset requirement is 608×608 pixels, the step size is set to 64 pixels, and multiple 608×608 pixel areas are cropped from left to right and from top to bottom. After cropping, the area smaller than the preset requirement is filled with white. For each cropped image, the bounding box coordinates of the crack need to be updated to ensure its correspondence with the new image. For example, if the bounding box coordinates of a crack in the original image are (100, 150, 200, 300), and the position after cropping becomes (0, 0, 608, 608), the position of the bounding box in the new image needs to be updated and recorded.

[0095] By implementing the above preprocessing process, the quality and diversity of the concrete crack dataset are improved. The data enhancement technology is used to effectively generate diverse samples and enrich the training data. At the same time, the use of size adjustment ensures that the image maintains its original features without distortion and can accurately capture details, thereby maximizing the retention of crack feature information and improving the efficiency and reliability of crack detection.

[0096] Step S2: introducing an RFB network module into the target detection model, wherein the RFB network module is used to capture multi-scale features of cracks; and training the target detection model according to the first data set.

[0097] Furthermore, if Figure 3 As shown, the target detection model training steps include:

[0098] Step S20: inputting the first data set into the target detection model;

[0099] Step S21: introducing the feature extraction layer of the target detection model into the RFB network module, the RFB (Receptive Field Block) network module processes the first data set in parallel to obtain a crack feature map, and performs feature fusion on the crack feature map to obtain a crack multi-scale feature map; for example, using different convolution kernel sizes (such as 3×3, 5×5, 7×7) and fusion of the output feature maps to form a crack multi-scale feature map, which is expressed as:

[0100]

[0101] Among them, F 3×3 、F 5×5 and F 7×7 They are the feature maps obtained by different convolution kernels. Represents the feature fusion operation.

[0102] Step S22: Output the crack coordinates, crack confidence and crack degree using the detection layer according to the crack multi-scale feature map, expressed as:

[0103] output={(x min ,y min ,x max ,y max ,p,v)};

[0104] Among them, (x min ,y min ) is the coordinate of the upper left corner of the crack, (x max ,y max ) is the coordinate of the lower right corner of the crack, p is the confidence level, and v is the type of the crack.

[0105] Step S23: using a loss function to calculate the gap between the predicted value and the true label according to the crack coordinates, the crack confidence and the crack degree, to obtain a crack loss value;

[0106] Among them, the loss function is expressed as:

[0107]

[0108] Among them, y i is the true label, is the predicted value, N is the number of samples, and i is the sample number.

[0109] Step S24: Calculate parameter gradients using the chain method according to the crack loss value, and update the target detection model according to the parameter gradients, which is expressed as:

[0110]

[0111] Among them, c is the model parameter, a is the learning rate, which can be set to 0.001. is the gradient of the loss function with respect to the parameters.

[0112] Step S25: If the number of iterations reaches the first threshold, the training is terminated; otherwise, steps S20 to S25 are repeated.

[0113] Specifically, for example, YOLOv5 is selected as the target detection model, and a first data set containing 1,000 concrete crack images is used. In the feature extraction layer, the model can process some images of obvious cracks, such as cracks with a length of 20 cm and a width of 3 mm. The detection layer outputs the crack coordinates as (150, 300, 155, 320) based on the multi-scale feature map of the cracks, with a confidence of 0.95, and is marked as "obvious cracks". After the first iteration, the loss value calculated by the loss function is 0.05, indicating that the model's prediction is relatively accurate. After 800 iterations, the loss value gradually decreases to 0.02. Through the target detection model, the multi-scale features of the cracks can be accurately captured, which improves the accuracy and efficiency of crack detection. When identifying obvious cracks, the confidence can reach up to 0.98, indicating that the model can stably identify cracks of different degrees in practical applications. In addition, the reduction in the loss value also shows that the learning effect of the model is good, which further improves the efficiency and reliability of crack detection.

[0114] Step S3: Acquire target concrete data, input the target concrete data into the trained target detection model, obtain the crack degree and crack coordinates, draw a two-dimensional crack map according to the crack degree and the crack coordinates, and divide the crack area.

[0115] Furthermore, the two-dimensional crack map establishment process includes:

[0116] Taking the upper left corner of the target concrete as the origin of the coordinate system, a two-dimensional coordinate system is established according to the actual size of the target concrete, wherein the two-dimensional coordinate system includes an X-axis and a Y-axis, wherein the X-axis extends horizontally to the right and the Y-axis extends vertically downward;

[0117] Check whether the crack coordinates are the clipped coordinates. If the crack coordinates are the clipped coordinates, add the sliding window offset to the crack coordinates to obtain the first crack coordinates; otherwise, directly assign the crack coordinates to the first crack coordinates;

[0118] The actual coordinates are calculated according to the first crack coordinates using a scaling ratio, and the two-dimensional crack map is obtained according to the actual coordinates.

[0119] Specifically, assuming that the actual size of the target concrete is 100 cm × 50 cm, the established X-axis extends from 0 to 100 cm, and the Y-axis extends from 0 to 50 cm; assuming that the detected crack coordinates are (15, 25, 35, 45), since this coordinate is in the cropped image, if it is offset by the sliding window (the offset is (5, 5)), the new upper left corner coordinate is (15+5, 25+5)=(20, 30), and the lower right corner coordinate is (35+5, 45+5)=(40, 50); assuming that the scaling ratio is 2, the actual coordinates are (40, 60, 80, 100); according to the actual coordinates, the position of the crack in the two-dimensional image is drawn, which is represented as a rectangular area from (40, 60) to (80, 100); the crack shape in the rectangular area is converted into a binary image, including black pixels and white pixels, and the widest and longest values ​​of the black pixels are recorded as the surface crack width and length, and are drawn in the two-dimensional image in equal proportion. By continuously drawing crack rectangular areas, the specific location and size of cracks on the target concrete can be clearly indicated, providing a detailed visual reference for subsequent structural evaluation and maintenance, and improving the efficiency and reliability of crack detection.

[0120] Furthermore, the crack region division process includes:

[0121] The center coordinates of the crack are calculated according to the crack coordinates, which are expressed as:

[0122]

[0123] Among them, (C x ,C y ) is the center coordinate of the crack.

[0124] The distance between the cracks is calculated using the Euclidean distance formula according to the center coordinates of the cracks, and the cracks whose distance is less than the second threshold are divided into the same area to obtain a clustering result, which is expressed as:

[0125]

[0126] Where D is the distance between the center coordinates of the two slits, (C x1 ,C y1 ) and (C x2 ,C y2 ) are the center coordinates of the two cracks; in addition, the second threshold can be set to 30 mm.

[0127] According to the clustering result, the number of cracks is calculated. If the number of cracks in the area exceeds the third threshold or at least one crack is an obvious crack, a surface crack area is obtained and the surface crack area is marked as a defect area; otherwise, it is marked as a non-defective area.

[0128] Among them, the third threshold can be set to 3.

[0129] Specifically, assuming that there are three cracks with coordinates of (10, 20, 30, 40), (32, 22, 50, 42) and (70, 80, 90, 100), the center coordinates are calculated to be (20, 30), (41, 32) and (80, 90). According to the center coordinates, the distance between crack 1 and crack 2 is 21.1, the distance between crack 1 and crack 3 is 84.9, and the distance between crack 2 and crack 3 is 69.9. Since the distance between crack 1 and crack 2 is less than 30, they are divided into area A and crack 3 is area B. Since crack 1 is an obvious crack and crack 3 is a slight crack in area A, area A is marked as a defective area and area B is a non-defective area. Through this process, defective areas in concrete structures can be effectively identified and marked, and the accuracy of detection can be enhanced. In addition, the set threshold helps to simplify the association judgment between cracks, providing a reliable basis for the subsequent setting of impact echo monitoring points, thereby improving the efficiency and reliability of crack detection.

[0130] Step S4: Layout monitoring points of the target concrete according to the crack area, detect the monitoring points using impact echoes, and record target concrete echo signals.

[0131] Furthermore, the monitoring point layout includes:

[0132] According to the size of the defect area and the number of cracks, the number of monitoring points in the defect area is obtained;

[0133] According to the size of the defect-free area and the number of cracks, the number of monitoring points in the defect-free area is obtained;

[0134] Arranging monitoring points at the junction of the defective area and the non-defective area to obtain the number of monitoring points at the junction;

[0135] The monitoring point spacing is calculated according to the number of monitoring points in the defective area, the number of monitoring points in the non-defective area and the number of monitoring points at the junction, and the monitoring point positions are marked according to the monitoring point spacing.

[0136] Specifically, Figure 4 As shown in the figure, assuming that the actual size of the target concrete is 100 cm × 50 cm, the size of the defective area 4 is 40 cm × 20 cm, the number of cracks 3 is 3, the defect-free area 1 is the remaining part, the number of cracks 3 is 0, and the coverage area of ​​each monitoring point 2 is 20 cm × 10 cm, then the number of monitoring points 2 in the defect-free area 1 is calculated to be 16, the number of monitoring points 2 in the defective area 4 is 4, and according to the number of cracks 3, the number of monitoring points 2 is increased to 12 in the defective area 4 at 1 cm to 2 cm above, below, left and right of the cracks 3, and 4 monitoring points 2 are arranged at the junction of the defective area 4 and the defect-free area 1. If the distance between the monitoring points 2 does not exceed 5 cm, no monitoring point 2 is added. The uniform grid layout of the defect-free area 1 ensures the basic monitoring of the entire concrete structure and provides global data support. In addition, the density of monitoring points 2 is increased in the defective area 4, so that the monitoring of the defective area 4 is more refined, and potential problems can be discovered in time, thereby improving the efficiency and reliability of crack 3 detection.

[0137] Step S5: Using digital signal processing technology to extract crack features of echo signals from the concrete crack data set, and training a machine learning model based on the crack features.

[0138] Furthermore, the crack feature acquisition process includes:

[0139] Acquire echo signal data according to the concrete crack data set;

[0140] Using a high-pass filter to remove background noise according to the echo signal data to obtain a first echo signal;

[0141] Performing signal segmentation using an overlapping segmentation method according to the first echo signal to obtain a second echo signal;

[0142] Calculate the signal peak value and the time when the impact reaches the monitoring point according to the second echo signal to obtain the time domain feature;

[0143] Extracting frequency amplitude using Fourier transform according to the second echo signal to obtain frequency domain features;

[0144] Extracting a signal using wavelet transform according to the second echo signal to obtain a time-frequency feature;

[0145] The time domain feature, the frequency domain feature and the time-frequency feature are combined to obtain the crack feature.

[0146] Specifically, assuming that the original echo signal is [0.1, 0.3, 0.5, 0.4, 0.2, 0.1], after high-pass filtering, the first echo signal is [0.2, 0.4, 0.3]. Assuming that the signal is divided into 10ms segments, the second echo signal may be Segment1[0.2, 0.4] and Segment2[0.4, 0.3]. For Segment1, the peak is 0.4 and the time is 0ms. The frequency domain feature obtained by Fourier transform is [0.5, 0.3], and the time-frequency feature is [0.6, 0.4]. Combining the time domain feature, frequency domain feature and time-frequency feature, the final crack feature may be [0.4, 0, 0.5, 0.6]. In this way, the signal-to-noise ratio of the signal can be effectively improved, and the useful high-frequency components can be retained. The final crack feature contains multi-dimensional information in the time domain, frequency domain and time-frequency, which is helpful for the effective training of the subsequent machine learning model and improves the model's recognition ability and accuracy for concrete cracks.

[0147] Furthermore, the machine learning model training process includes:

[0148] Acquire a historical echo signal and a historical echo signal label according to the concrete crack data set, wherein the historical echo signal label includes an internal crack depth and width;

[0149] Specifically, assuming there are 100 historical echo signal data, for signal 1, the depth is 5 cm and the width is 2 mm, for signal 2, the depth is 10 cm and the width is 3 mm, the format of the label is expressed as:

[0150]

[0151] Among them, Label is a label, and each row represents the depth and width of a historical echo signal;

[0152] The previously extracted fracture features are organized into a format suitable for input into the machine learning model, which can be expressed as:

[0153]

[0154] Among them, Feature is a feature matrix, and each row represents a set of crack features;

[0155] The crack characteristics are input into the machine learning model, and the trained machine learning model is obtained using a first loss function according to the crack characteristics and the historical echo signal labels.

[0156] Among them, machine learning models can choose linear regression, support vector machine, random forest or neural network, etc.

[0157] Specifically, assuming that random forest is selected as the machine learning model, the mean square error is selected as the first loss function, which is expressed as:

[0158]

[0159] Among them, MSE is the mean square error, n is the total number of labels, i is the label number, is the depth and width predicted by the model, e i is the label of the historical echo signal;

[0160] Specifically, the crack features and corresponding labels of 80 historical echo signals were used for training, the number of iterations was set to 1000, and the model parameters were gradually adjusted to reduce the loss value. The crack features and corresponding labels of 20 historical echo signals were used for verification, and the mean square error was 0.15, indicating that the machine learning model has good predictive performance on unseen data, can effectively capture the characteristics of cracks and make accurate assessments, thereby improving the efficiency and reliability of crack detection.

[0161] Step S6: Obtain the internal crack results of the target concrete using the trained machine learning model according to the target concrete echo signal.

[0162] Step S7: expanding the two-dimensional crack map into a three-dimensional crack map according to the internal crack results of the target concrete, determining the overall crack defect area of ​​the target concrete according to the three-dimensional crack map, and obtaining the actual crack degree.

[0163] Furthermore, the actual crack degree acquisition process includes:

[0164] Input the surface crack coordinates and the depth and width information of the internal cracks into the modeling software to generate a three-dimensional crack map. In this process, the cracks can be assumed to be uniform in depth and width, or adjusted according to the specific situation;

[0165] Extracting crack information according to the three-dimensional crack map, including the coordinates, depth and width of the crack; marking the area containing the crack as an overall crack defect area to facilitate subsequent analysis;

[0166] According to the overall crack area, the surface crack and the internal crack extending downward are marked as the first crack area, and all the overall crack areas are marked in this way;

[0167] The surface crack cross-sectional area of ​​the first crack region and the internal crack cross-sectional area corresponding to the first crack region are calculated according to the crack information, and the actual crack degree is obtained by weighted summing the surface crack cross-sectional area and the internal crack cross-sectional area.

[0168] Specifically, Figure 5As shown, for the first crack region 5, assuming that the surface crack width is 10 mm and the length is 100 mm, the cross-sectional area of ​​the surface crack is 1000 mm. 2 , assuming the depth is 20mm and the width is 5mm, the corresponding cross-sectional area of ​​the internal crack is 100mm 2 , the surface crack cross-sectional area and the internal crack cross-sectional area are weighted and summed. Assuming the weights are 0.3 and 0.7, the actual crack degree is 370. For the actual crack degree not exceeding 500, it can be marked as an overall slight crack area, and the second crack area 6 is also marked in the same way. The visualization of the three-dimensional crack map provides a scientific basis for maintenance and reinforcement decisions and timely discovers crack problems. In addition, giving priority to the overall severe crack area helps optimize resource allocation and reduce costs, thereby improving the efficiency and reliability of crack detection.

[0169] In summary, by acquiring and marking concrete crack data sets, combined with data enhancement technology, data can be preprocessed efficiently to generate a high-quality first data set; the RFB network module is used to extract the multi-scale features of cracks, and the target detection model is trained to accurately identify the degree and coordinates of cracks in the target concrete, thereby drawing a detailed two-dimensional crack map and dividing the crack area. This provides a scientific basis for the subsequent layout of monitoring points. Through impact echo detection and recording echo signals, crack features are further extracted and applied to machine learning models, making the detection of internal crack results more accurate; the two-dimensional crack map is expanded into a three-dimensional crack map, and the overall crack defect area is determined, so as to obtain the actual crack degree, improve the accuracy and efficiency of crack detection, and provide reliable support for the health management of concrete structures.

[0170] Embodiment 2

[0171] Taking a factory as an example, the four walls of the factory are used for concrete internal crack detection. The walls of the factory are concrete structures with a height of 4 meters, a width of 10 meters, and a thickness of 0.3 meters. They are located in a humid environment and have been affected by moisture erosion and temperature changes for a long time, resulting in multiple cracks in the walls, affecting the safety and service life of the structure. An image-impact echo concrete internal crack detection method is used to improve the efficiency and reliability of crack detection, thereby efficiently and accurately detecting the safety of the wall.

[0172] A method for detecting internal cracks in concrete based on image-impact echo, comprising:

[0173] Acquire a concrete crack data set, including 400 wall images at different angles and lighting conditions and 200 historical echo signal samples, mark the crack degrees of the concrete crack data set, including obvious cracks (60 locations), slight cracks (90 locations) and no cracks (40 locations), obtain a second data set, and use data enhancement to preprocess the second data set to obtain a first data set;

[0174] Introducing an RFB network module into the target detection model, wherein the RFB network module is used to capture multi-scale features of cracks; and training the target detection model according to the first data set;

[0175] Acquire target concrete data, input the target concrete data into the trained target detection model, obtain the crack degree and crack coordinates, draw a two-dimensional crack map according to the crack degree and the crack coordinates, and divide the crack area;

[0176] Layout monitoring points of the target concrete according to the crack area, detect the monitoring points using impact echoes, and record echo signals of the target concrete;

[0177] Extracting crack features of echo signals from the concrete crack data set using digital signal processing technology, and training a machine learning model based on the crack features;

[0178] Inputting the target concrete echo signal into the trained machine learning model to obtain the internal crack results of the target concrete;

[0179] The two-dimensional crack map is expanded into a three-dimensional crack map according to the internal crack results of the target concrete, and the overall crack defect area of ​​the target concrete is determined according to the three-dimensional crack map to obtain the actual crack degree.

[0180] Furthermore, the crack degree acquisition process includes:

[0181] Using image processing software to measure the concrete cracks according to the concrete crack data set, and obtaining the crack size according to the starting point and the end point of the concrete crack;

[0182] Obtain the RGB value of the concrete crack by using a color analysis tool, and obtain the brightness and saturation according to the concrete background color;

[0183] Using Gabor filter to extract texture features of the concrete cracks to obtain crack complexity;

[0184] The crack degrees of the concrete crack dataset are divided and marked according to the crack size, the brightness, the saturation, the crack complexity and expert experience, and the crack degrees include obvious cracks, slight cracks and no cracks.

[0185] Furthermore, the pre-processing process includes:

[0186] The data enhancement includes geometric change, color transformation and perspective transformation. After enhancing the second data set three times, 1200 newly generated images are obtained. The newly generated images are merged with the images in the concrete crack data set to obtain a third data set.

[0187] resizing the third data set, wherein the resizing includes: if the third data set is larger than a preset requirement, clipping it according to the preset requirement using a sliding window method, and filling the area of ​​the clipped third data set smaller than the preset requirement with white; otherwise, clipping is not performed and the area of ​​the third data set smaller than the preset requirement is directly filled with white;

[0188] For easier operation, the photo size is selected as 608×608, so there is no need to crop or fill.

[0189] The first data set is obtained according to the adjusted third data set.

[0190] Furthermore, the target detection model training step includes:

[0191] Step S20: inputting the first data set into the target detection model;

[0192] Step S21: introducing the feature extraction layer of the target detection model into the RFB network module, the RFB network module processes the first data set in parallel to obtain a crack feature map, and performs feature fusion on the crack feature map to obtain a crack multi-scale feature map;

[0193] Step S22: using a detection layer to output the crack coordinates, crack confidence and crack degree according to the crack multi-scale feature map;

[0194] Step S23: using a loss function to calculate the gap between the predicted value and the true label according to the crack coordinates, the crack confidence and the crack degree, to obtain a crack loss value;

[0195] Step S24: Calculate parameter gradients using the chain method according to the crack loss value, and update the target detection model according to the parameter gradients;

[0196] Step S25: If the number of iterations reaches the first threshold, the training is terminated; otherwise, steps S20 to S25 are repeated.

[0197] Specifically, YOLOV5 was used as the target detection model, the first threshold was 500, and after 500 iterations, the loss value was reduced to 0.12. After the training was completed, 60 obvious cracks, 90 minor cracks, and 32 no cracks were finally identified.

[0198] Furthermore, the two-dimensional crack map establishment process includes:

[0199] Taking the upper left corner of the target concrete as the origin of the coordinate system, a two-dimensional coordinate system is established according to the actual size of the target concrete, wherein the two-dimensional coordinate system includes an X-axis and a Y-axis, wherein the X-axis extends horizontally to the right and the Y-axis extends vertically downward;

[0200] Check whether the crack coordinates are the clipped coordinates. If the crack coordinates are the clipped coordinates, add the sliding window offset to the crack coordinates to obtain the first crack coordinates; otherwise, directly assign the crack coordinates to the first crack coordinates;

[0201] The actual coordinates are calculated according to the first crack coordinates using a scaling ratio, and the two-dimensional crack map is obtained according to the actual coordinates.

[0202] Furthermore, the crack region division process includes:

[0203] Calculate the center coordinates of the crack according to the crack coordinates;

[0204] The distance between the cracks is calculated using the Euclidean distance formula according to the central coordinates of the cracks, and the cracks whose distance is less than a second threshold value are divided into the same area to obtain a clustering result;

[0205] The second threshold is set to 50 mm.

[0206] According to the clustering result, the number of cracks is calculated. If the number of cracks in the area exceeds the third threshold or at least one crack is an obvious crack, a surface crack area is obtained and the surface crack area is marked as a defect area; otherwise, it is marked as a non-defective area.

[0207] Among them, the third threshold is set to 5.

[0208] Furthermore, the monitoring point layout includes:

[0209] According to the size of the defect area and the number of cracks, the number of monitoring points in the defect area is obtained;

[0210] According to the size of the defect-free area and the number of cracks, the number of monitoring points in the defect-free area is obtained;

[0211] Arranging monitoring points at the junction of the defective area and the non-defective area to obtain the number of monitoring points at the junction;

[0212] The monitoring point spacing is calculated according to the number of monitoring points in the defective area, the number of monitoring points in the non-defective area and the number of monitoring points at the junction, and the monitoring point positions are marked according to the monitoring point spacing.

[0213] Specifically, wall 4 has more cracks than other walls, and there are two obvious cracks. The number of monitoring points set on the four walls are 25, 20, 24 and 35 respectively. A total of 6 additional monitoring points are set at the junction of defective areas and non-defective areas, with a total of 110 monitoring points.

[0214] Furthermore, the crack feature acquisition process includes:

[0215] Acquire echo signal data according to the concrete crack data set;

[0216] Using a high-pass filter to remove background noise according to the echo signal data to obtain a first echo signal;

[0217] Performing signal segmentation using an overlapping segmentation method according to the first echo signal to obtain a second echo signal;

[0218] Calculate the signal peak value and the time when the impact reaches the monitoring point according to the second echo signal to obtain the time domain feature;

[0219] Extracting frequency amplitude using Fourier transform according to the second echo signal to obtain frequency domain features;

[0220] Extracting a signal using wavelet transform according to the second echo signal to obtain a time-frequency feature;

[0221] The time domain feature, the frequency domain feature and the time-frequency feature are combined to obtain the crack feature.

[0222] Furthermore, the machine learning model training process includes:

[0223] Acquire a historical echo signal and a historical echo signal label according to the concrete crack data set, wherein the historical echo signal label includes an internal crack depth and width;

[0224] The crack characteristics are input into the machine learning model, and the trained machine learning model is obtained using a first loss function according to the crack characteristics and the historical echo signal labels.

[0225] Specifically, a decision tree is used as the machine learning model with a maximum depth of 5, a minimum sample split number of 10, and a minimum sample leaf number of 5. The mean square error is used as the first loss function, and 200 historical echo signal samples are used for training and testing. The mean square error during testing is 0.13, which can accurately predict the depth and width of the crack, thereby improving the efficiency and reliability of crack detection.

[0226] Furthermore, the actual crack degree acquisition process includes:

[0227] Extracting crack information according to the three-dimensional crack map;

[0228] The surface crack cross-sectional area and the internal crack cross-sectional area are calculated according to the crack information, and the actual crack degree is obtained by performing weighted summation based on the surface crack cross-sectional area and the internal crack cross-sectional area.

[0229] Specifically, for wall 4, there are obvious surface cracks and internal cracks, and the weights for surface cracks and internal cracks are set to 0.4 and 0.6 respectively. The final actual crack degree is 528. Because the actual crack degree exceeds 500, it is marked as an overall serious crack area. By calculating the actual crack degree, the safety of the concrete structure can be better evaluated, helping to decide whether reinforcement or repair is needed, thereby improving the efficiency and reliability of crack detection.

[0230] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for detecting internal cracks in concrete based on image-impact echo, characterized in that: include: Acquire a concrete crack data set, mark the crack degree of the concrete crack data set to obtain a second data set, and preprocess the second data set using data enhancement to obtain a first data set; An RFB network module is introduced into the target detection model, and the RFB network module is used to capture multi-scale features of cracks; Training the target detection model according to the first data set; Acquire target concrete data, input the target concrete data into the trained target detection model, obtain the crack degree and crack coordinates, draw a two-dimensional crack map according to the crack degree and the crack coordinates, and divide the crack area; Layout monitoring points of the target concrete according to the crack area, detect the monitoring points using impact echoes, and record echo signals of the target concrete; Extracting crack features of echo signals from the concrete crack data set using digital signal processing technology, and training a machine learning model based on the crack features; Inputting the target concrete echo signal into the trained machine learning model to obtain the internal crack results of the target concrete; The two-dimensional crack map is expanded into a three-dimensional crack map according to the internal crack results of the target concrete, and the overall crack defect area of ​​the target concrete is determined according to the three-dimensional crack map to obtain the actual crack degree.

2. The method for detecting internal cracks in concrete based on image-impact echo according to claim 1, characterized in that: The crack degree acquisition process includes: Using image processing software to measure the concrete cracks according to the concrete crack data set, and obtaining the crack size according to the starting point and the end point of the concrete crack; Obtain the RGB value of the concrete crack by using a color analysis tool, and obtain the brightness and saturation according to the concrete background color; Using Gabor filter to extract texture features of the concrete cracks to obtain crack complexity; The crack degrees of the concrete crack dataset are divided and marked according to the crack size, the brightness, the saturation, the crack complexity and expert experience, and the crack degrees include obvious cracks, slight cracks and no cracks.

3. The method for detecting internal cracks in concrete based on image-impact echo according to claim 1, characterized in that: The pre-treatment process includes: The data enhancement includes geometric change, color transformation and perspective transformation. After enhancing the second data set N times, a newly generated image is obtained. The newly generated image is merged with the image in the concrete crack data set to obtain a third data set. resizing the third data set, wherein the resizing includes: if the third data set is larger than a preset requirement, clipping it according to the preset requirement using a sliding window method, and filling the area of ​​the clipped third data set smaller than the preset requirement with white; otherwise, clipping is not performed and the area of ​​the third data set smaller than the preset requirement is directly filled with white; The first data set is obtained according to the adjusted third data set.

4. The method for detecting internal cracks in concrete based on image-impact echo according to claim 1, characterized in that: The target detection model training step includes: Step S20: inputting the first data set into the target detection model; Step S21: introducing the feature extraction layer of the target detection model into the RFB network module, the RFB network module processes the first data set in parallel to obtain a crack feature map, and performs feature fusion on the crack feature map to obtain a crack multi-scale feature map; Step S22: using a detection layer to output the crack coordinates, crack confidence and crack degree according to the crack multi-scale feature map; Step S23: using a loss function to calculate a crack loss value according to the crack coordinates, the crack confidence and the crack degree; Step S24: Calculate parameter gradients using the chain method according to the crack loss value, and update the target detection model according to the parameter gradients; Step S25: If the number of iterations reaches the first threshold, the training is terminated; otherwise, steps S20 to S25 are repeated.

5. The method for detecting internal cracks in concrete based on image-impact echo according to claim 1, characterized in that: The two-dimensional crack map establishment process includes: Taking the upper left corner of the target concrete as the origin of the coordinate system, a two-dimensional coordinate system is established according to the actual size of the target concrete, wherein the two-dimensional coordinate system includes an X-axis and a Y-axis, wherein the X-axis extends horizontally to the right and the Y-axis extends vertically downward; Check whether the crack coordinates are the clipped coordinates. If the crack coordinates are the clipped coordinates, add the sliding window offset to the crack coordinates to obtain the first crack coordinates; otherwise, directly assign the crack coordinates to the first crack coordinates; The actual coordinates are calculated according to the first crack coordinates using a scaling ratio, and the two-dimensional crack map is obtained according to the actual coordinates.

6. The method for detecting internal cracks in concrete based on image-impact echo according to claim 1, characterized in that: The crack area division process includes: Calculate the center coordinates of the crack according to the crack coordinates; The distance between the cracks is calculated using the Euclidean distance formula according to the central coordinates of the cracks, and the cracks whose distance is less than a second threshold value are divided into the same area to obtain a clustering result; According to the clustering result, the number of cracks is calculated. If the number of cracks in the area exceeds the third threshold or at least one crack is an obvious crack, a surface crack area is obtained and the surface crack area is marked as a defect area; otherwise, it is marked as a non-defective area.

7. The method for detecting internal cracks in concrete based on image-impact echo according to claim 1, characterized in that: The monitoring point layout includes: According to the size of the defect area and the number of cracks, the number of monitoring points in the defect area is obtained; According to the size of the defect-free area and the number of cracks, the number of monitoring points in the defect-free area is obtained; Arranging monitoring points at the junction of the defective area and the non-defective area to obtain the number of monitoring points at the junction; The monitoring point spacing is calculated according to the number of monitoring points in the defective area, the number of monitoring points in the non-defective area and the number of monitoring points at the junction, and the monitoring point positions are marked according to the monitoring point spacing.

8. The method for detecting internal cracks in concrete based on image-impact echo according to claim 1, characterized in that: The crack feature acquisition process includes: Acquire echo signal data according to the concrete crack data set; Using a high-pass filter to remove background noise according to the echo signal data to obtain a first echo signal; Performing signal segmentation using an overlapping segmentation method according to the first echo signal to obtain a second echo signal; Calculate the signal peak value and the time when the impact reaches the monitoring point according to the second echo signal to obtain the time domain feature; Extracting frequency amplitude using Fourier transform according to the second echo signal to obtain frequency domain features; Extracting a signal using wavelet transform according to the second echo signal to obtain a time-frequency feature; The time domain feature, the frequency domain feature and the time-frequency feature are combined to obtain the crack feature.

9. The method for detecting internal cracks in concrete based on image-impact echo according to claim 1, characterized in that: The machine learning model training process includes: Acquire a historical echo signal and a historical echo signal label according to the concrete crack data set, wherein the historical echo signal label includes an internal crack depth and width; The crack characteristics are input into the machine learning model, and the trained machine learning model is obtained using a first loss function according to the crack characteristics and the historical echo signal labels.

10. The method for detecting internal cracks in concrete based on image-impact echo according to claim 1, characterized in that: The actual crack degree acquisition process includes: Extracting crack information according to the three-dimensional crack map; The surface crack cross-sectional area and the internal crack cross-sectional area are calculated according to the crack information, and the actual crack degree is obtained by performing weighted summation based on the surface crack cross-sectional area and the internal crack cross-sectional area.

Images