Road surface cavity water-containing disease detection method based on radar detection

Abstract

The invention relates to the technical field of quantum radar, in particular to a pavement cavity water-containing disease detection method based on radar detection, and the method comprises the steps: analyzing the extreme value of a ballistic reflection wave at each position of a to-be-detected pavement, and obtaining a significant extreme value, a reflection wave sub-column and an extreme value sequence; analyzing the difference characteristics between the significant extreme values and the length distribution condition of the reflected wave sub-columns, and obtaining a radar distribution coefficient in combination with the element distribution of the extreme value sequence and the shape characteristics of the fitting curve; analyzing the similarity between the difference overall distribution characteristics of the significant extremum and the overall distribution of the environmental humidity data to obtain a radar correlation coefficient; simulating a radar signal of each position of a to-be-detected road surface, performing statistical analysis on the radar signal and a signal acquired in a current round, and obtaining a disease probability value and a detection result of a corresponding position in combination with a threshold segmentation result of radar correlation coefficients of a preset number of samples. The invention aims to improve the detection accuracy of the pavement cavity water-containing disease.

Description

Technical Field

[0001] This application relates to the field of quantum radar technology, specifically to a method for detecting water-bearing road surface cavities based on radar detection. Background Technology

[0002] Road defects are a significant factor affecting public transportation safety, especially internal road defects, which are more difficult to detect, often more concealed, and can pose greater risks. With the introduction of radar detection technology into road inspection, more sensitive and accurate detection of internal road defects has become possible.

[0003] When ground-penetrating radar (GPR) performs non-destructive testing of internal road surface defects, it needs to utilize the attenuation and reflection of electromagnetic waves within the road surface to determine the condition inside. Because electromagnetic waves of different frequencies are difficult to perfectly balance in terms of detection resolution and depth, often one aspect's quality is sacrificed to improve another; and current technologies often only analyze the radar signals acquired during detection, making it difficult to accurately detect minute defects deep within the road. Summary of the Invention

[0004] In view of the above, it is necessary to provide a radar-based method for detecting water-bearing cavities in road surfaces to solve the aforementioned problems.

[0005] One embodiment of this application provides a method for detecting water-bearing cavities in road surfaces based on radar detection, the method comprising:

[0006] Acquire the ambient humidity data of the road surface to be tested and the single-channel reflected wave received at each location in the current cycle;

[0007] The extreme values ​​in the ballistic reflected waves of each position in the current round are filtered to obtain significant extreme values; all extreme values ​​of the reflected waves between two adjacent significant extreme values ​​are formed into a reflected wave subsequence; the number of significant extreme values ​​of each position in all current rounds is used to form the extreme value sequence of each position in the current round, and the extreme value sequence is fitted to obtain a fitting curve; the differences between significant extreme values ​​in all current rounds of each position, the length distribution of the reflected wave subsequence, and the element distribution of the extreme value sequence obtained in the current round of each position, combined with the shape characteristics of the fitting curve, are used to obtain the radar distribution coefficient of each position in the current round.

[0008] By analyzing the overall distribution characteristics of the differences in significant extreme values ​​in all rounds at each location and the similarity to the overall distribution of environmental humidity data, the humidity lag at each location is obtained. Combined with the radar distribution coefficient, the radar correlation coefficient of each location in the current round is obtained.

[0009] The radar signal at each location on the road surface to be inspected is simulated and statistically analyzed with the signal collected in the current round. Combined with the threshold segmentation results of the radar correlation coefficient of a preset number of samples, the probability value of the defect at each location on the road surface to be inspected is obtained, and it is determined whether there is a cavity and water-bearing defect at each location.

[0010] Specifically, obtaining significant extreme values ​​means:

[0011] For each position and each round of single-channel reflected wave, the absolute values ​​of all extreme values ​​are used to form a reflected wave sequence. The upper quartile of the reflected wave sequence is taken, and the elements greater than the upper quartile are recorded as significant extreme values.

[0012] Specifically, obtaining the radar distribution coefficient for each location in the current round is as follows:

[0013] Analyze the difference distribution between adjacent significant extreme values ​​in all rounds at each location to obtain the average change value of reflected wave attenuation in the current round at each location;

[0014] Based on the length distribution of the reflected wavelet subsequence obtained in all rounds at each position, the minimum length value of the current round is obtained;

[0015] Analyze the element distribution of the extreme value sequence obtained in the current round at each position and the shape characteristics of the fitted curve to obtain the trend of the number of extreme values ​​in the current round at each position;

[0016] The negative correlation mapping result of the minimum length value of the current round is positively fused with the average change value of the reflected wave attenuation, and then added to the trend value of the extreme value quantity of the current round at each position to obtain the radar distribution coefficient of the current round at each position.

[0017] Specifically, obtaining the average change in attenuation of the reflected wave at each position in the current round is as follows:

[0018] Obtain the degree of dispersion of the difference between all adjacent significant extreme values ​​at each position in each round, and record the average value of the difference between the degree of dispersion of all adjacent rounds before the current round at each position as the average change value of the reflected wave attenuation in the current round.

[0019] Specifically, obtaining the minimum length value for the current round is as follows:

[0020] Obtain the minimum length of the reflected wavelet subsequence at each position and in each round. Average the minimum lengths obtained in all rounds prior to the current round at each position to obtain the minimum length value for the current round.

[0021] Specifically, the trend of the number of extreme values ​​at each position in the current round is obtained as follows:

[0022] Obtain the maximum value of the extreme value sequence at each position, and obtain the percentage of data points on the curve to the left of the maximum value whose derivative values ​​are positive, denoted as the first percentage.

[0023] The percentage of data points on the curve to the right of the maximum value whose derivative values ​​are negative is recorded as the second percentage.

[0024] The average of the first proportion and the second proportion is used as the trend of the number of extreme values ​​in the current round for each position.

[0025] Specifically, obtaining the humidity hysteresis at each location is as follows:

[0026] The environmental humidity data of each location in all current rounds are combined into a humidity sequence;

[0027] Calculate the mean of the differences between all adjacent significant extreme values ​​in each round, and use the obtained mean to form a decay sequence;

[0028] Cross-correlation analysis was performed on the humidity series and attenuation series to obtain the cross-correlation coefficient corresponding to each hysteresis; the hysteresis corresponding to the maximum cross-correlation coefficient was taken as the humidity hysteresis at each location.

[0029] The specific process for obtaining the radar correlation coefficient for each location in the current round is as follows:

[0030] For the current round at each location, the dispersion of all elements in the decay sequence is obtained, and the negative correlation mapping results with the humidity hysteresis at each location are positively fused to obtain the decay fluctuation index.

[0031] The product of the radar correlation coefficient of the current round and the attenuation fluctuation index is used as the radar correlation coefficient for each round.

[0032] The specific formula for obtaining the probability value of defects at each location of the road surface to be inspected is as follows: P is the disease probability value of each location in the current round, B is the radar correlation coefficient of each location in the current round, λ is the radar correlation coefficient segmentation threshold of a preset number of samples, P0 is the probability value of statistical analysis for each location, and min{} represents the minimum value function.

[0033] Specifically, determining whether a cavity or water-bearing defect exists at each location involves:

[0034] If the probability value of the disease in the current round at each location is greater than the preset threshold, it is determined that there is a cavity water-bearing disease at the corresponding location; otherwise, there is no cavity water-bearing disease.

[0035] This application has at least the following beneficial effects:

[0036] This application first analyzes the changing characteristics of radar signals and, combined with the development characteristics of pavement cavities and water-bearing defects, calculates a radar distribution coefficient. This coefficient, combining the temporal characteristics of radar signals and the development trend of defects, helps to enhance the accuracy of detecting pavement cavities and water-bearing defects. Then, it calculates a radar correlation coefficient by combining the correlation characteristics between radar signal changes and environmental changes. This coefficient, taking into account the influence of environmental humidity on pavement cavities and water-bearing defects, further enhances the sensitivity and accuracy of detection. In this way, by considering the temporal characteristics of radar signals and the development characteristics of pavement cavities and water-bearing defects, and incorporating environmental influences, sensitive and accurate detection can be achieved even for minor defects based on their development. Attached Figure Description

[0037] Figure 1 A flowchart of the radar-based method for detecting water-bearing cavities in road surfaces provided in this application;

[0038] Figure 2 A flowchart for obtaining the radar correlation coefficient provided in this application. Detailed Implementation

[0039] In the description of the embodiments in this application, the words "exemplary," "or," and "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary," "or," and "for example" is intended to present the relevant concepts in a specific manner.

[0040] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in this application's specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.

[0041] It should also be noted that the terms "first" and "second" in this application and its accompanying drawings are used to distinguish similar objects, rather than to describe a specific order or sequence. The methods disclosed in the embodiments of this application or the methods shown in the flowcharts include one or more steps for implementing the method. Without departing from the scope of protection of this application, the execution order of multiple steps can be interchanged, and some steps can also be deleted.

[0042] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0043] This application proposes a radar-based method for detecting water-filled cavities in road surfaces, applied in the field of quantum radar technology. (See attached document.) Figure 1 The method includes the following steps:

[0044] S1: Acquire the ambient humidity data of the road surface to be detected and the single-channel reflected wave received at each location in the current cycle.

[0045] This application takes into account that the development and changes of road surface defects have temporal characteristics and are strongly related to various physical parameters. Therefore, the influence of these factors is introduced into the analysis of radar signals to further improve the sensitivity and accuracy of detecting water-bearing defects in road surface cavities.

[0046] First, a testing vehicle equipped with ground-penetrating radar (GPR) conducts non-destructive testing on the road section to be tested. A Ricker wavelet with a center frequency of 1500MHz is used as the excitation source, with a transmit / receive interval of 2cm and a time window of 12ns. Electromagnetic waves are transmitted towards the road every 0.5m, and the vehicle moves sequentially along the road section, receiving single-channel reflected waves. Simultaneously, a humidity sensor within the testing system collects environmental humidity data for the road section under test.

[0047] S2: Filter the extreme values ​​in the ballistic reflected waves of each position in the current round to obtain significant extreme values; form a reflected wave subsequence by combining all extreme values ​​of the reflected waves between two adjacent significant extreme values; form an extreme value sequence for each position in the current round by combining the number of significant extreme values ​​of each position in all current rounds, and fit the extreme value sequence to obtain a fitting curve; analyze the difference characteristics between significant extreme values ​​in all current rounds of each position, the length distribution of the reflected wave subsequence, and combine the element distribution of the extreme value sequence obtained for each position in the current round with the shape characteristics of the fitting curve to obtain the radar distribution coefficient for each position in the current round.

[0048] Ground-penetrating radar (GPR) detects underground structures by emitting electromagnetic waves and measuring the changes in reflection intensity and waveform when these waves encounter road structural layers with different dielectric constants. For normal roads, the internal material distribution is highly uniform; even if it is not uniform, the difference in dielectric constant is relatively small. This results in fewer reflected waves in the radar signal from normal roads, and the electromagnetic wave attenuation is relatively consistent.

[0049] When pavement cavities and water-containing defects occur, changes in humidity lead to significant variations in the dielectric constant, resulting in noticeable changes in reflection and signal attenuation at the defect site. In the early stages of pavement cavities, multiple small cavities may exist within the pavement. These cavities exhibit high dispersion, and due to their small size, the differences in electromagnetic wave attenuation caused by them are also small. As the cavities gradually develop, the small cavities merge, forming larger cavities. The appearance of larger cavities results in less electromagnetic wave reflection, but greater attenuation of electromagnetic waves within the cavities. Furthermore, larger cavities may contain water deposits and air. Since the dielectric constants of these two media differ significantly, and the impurity content in air and water is lower than in solid road materials, there will be significant differences in reflection and attenuation within the wavelength band corresponding to the suspected cavity. However, except at the interface between the two media, the overall wave fluctuations within the band are more stable.

[0050] In a single-channel reflected wave, the absolute values ​​of all peaks and troughs are arranged in chronological order along the horizontal axis to form a reflected wave sequence. The upper quartile of the reflected wave sequence is taken, and peaks or troughs greater than the upper quartile are designated as significant extrema. The upper quartile is a well-known technique and will not be elaborated further. The peaks and troughs of the reflected waves between any two adjacent significant extrema are arranged in chronological order to form a subsequence of reflected waves between these two significant extrema.

[0051] When analyzing the temporal characteristics of radar signals, it is necessary to use the historical detection results of each location on the road segment to be detected. The significant extreme values ​​of each historical detection result of a detection location are obtained in the manner described above. The number of significant extreme values ​​in the radar signal at each detection time is arranged in the order of detection time to form the extreme value sequence of that location. Then, the elements in the extreme value sequence are used as the vertical axis and the index of the element is used as the horizontal axis to perform quadratic curve fitting and output the fitted curve function. Quadratic curve fitting is a well-known technique and will not be elaborated further.

[0052] Based on the above analysis, a radar distribution coefficient is calculated to measure the impact of road surface defects on radar signals. Specifically: the dispersion of the differences between all adjacent significant extreme values ​​at each location and in each cycle is obtained; the average of the differences in dispersion between all adjacent cycles before the current cycle at each location is recorded as the average change value of reflected wave attenuation in the current cycle; the minimum value of the reflected wave subsequence length obtained at each location and in each cycle is obtained; the average of the minimum values ​​obtained at each location and in all cycles before the current cycle is calculated to obtain the minimum length value in the current cycle; the maximum value of the extreme value sequence at each location is obtained; the proportion of data points on the curve to the left of the maximum value whose derivative values ​​are positive is recorded as the first proportion; the proportion of data points on the curve to the right of the maximum value whose derivative values ​​are negative is recorded as the second proportion; the average of the first proportion and the second proportion is used as the trend of the number of extreme values ​​in the current cycle at each location; the negative correlation mapping result of the minimum length value in the current cycle is positively fused with the average change value of reflected wave attenuation, and then added to the trend value of the number of extreme values ​​in the current cycle at each location to obtain the radar distribution coefficient for the current cycle at each location. In this embodiment, the dispersion of a variable is calculated using the standard deviation; multiple variables are positively fused using multiplication; and the negative correlation mapping result of a variable is calculated using the reciprocal of the variable.

[0053] Understandably, for normal road surfaces, the attenuation of reflected waves in radar detection is relatively consistent, and the number of reflected waves corresponding to significant extrema does not exhibit a clear monotonic trend. Furthermore, due to the complex composition of impurities within solid materials, there are more reflected waves, meaning the lengths of reflected wavelets are relatively long, resulting in a smaller radar distribution coefficient. For road sections with pavement cavities and water-bearing defects, as the defects worsen over time, the number of significant extrema initially increases and then decreases. As the cavities grow larger, the water-bearing cavities cause increasingly severe attenuation of electromagnetic waves. However, since the constituent media within the cavities are mostly water and air, the impurity content of these two elements is much lower than that of the solid material, causing the number of reflected waves in the corresponding waveband to gradually decrease, resulting in a larger radar distribution coefficient.

[0054] S3: Analyze the overall distribution characteristics of the differences in significant extreme values ​​in all rounds at each location and the similarity to the overall distribution of environmental humidity data to obtain the humidity lag at each location. Combine this with the radar distribution coefficient to obtain the radar correlation coefficient for the current round at each location.

[0055] Radar distribution coefficients attempt to reveal the distribution and temporal changes of cavities within roads by analyzing the direction of radar signal variations. However, due to the complexity of road structures and the difficulty in accurately predicting changes, relying solely on radar signal fluctuations is insufficient for accurate detection of road surface cavities and water-containing defects. The primary cause of radar signal variations is the change in dielectric constant of different media, with humidity changes leading to significant alterations in the dielectric constant and severely impacting cavities. Since humidity exhibits a high degree of similarity to environmental factors, measuring the relationship between radar signals and environmental changes is crucial for improving the accuracy and sensitivity of road surface cavity and water-containing defect detection.

[0056] For normal road surfaces, due to their relatively intact structure and low porosity, moisture migration is relatively slow, resulting in smaller humidity fluctuations. Furthermore, humidity changes are typically more drastic closer to the surface, leading to a greater rate of change in electromagnetic wave attenuation closer to the surface. Secondly, the slow moisture migration results in a slower response to changes in environmental humidity, with humidity changes being relatively delayed. However, for pavement cavities and water-bearing defects, the presence of cavities and water-bearing areas, coupled with a loose structure and high porosity, allows moisture to easily accumulate and migrate, leading to larger humidity fluctuations. The cavities provide a closed or semi-closed space for moisture accumulation, and as moisture gradually accumulates, humidity at deeper locations may also be significantly affected. Secondly, the loose structure of pavement cavities makes them more susceptible to environmental influences, resulting in a faster response time.

[0057] The ambient humidity data collected during all radar detection cycles at each location are arranged in chronological order to form a humidity sequence. The mean of the differences between all adjacent significant extreme values ​​during each detection cycle is calculated, and these mean values ​​are arranged in chronological order to form an attenuation sequence. Then, using the humidity sequence and attenuation sequence as input, cross-correlation analysis is employed to output the cross-correlation coefficient corresponding to each hysteresis. The hysteresis corresponding to the maximum cross-correlation coefficient is taken as the humidity hysteresis for each location. Cross-correlation analysis is a well-known technique and will not be elaborated further.

[0058] Based on the above analysis, the radar correlation coefficient is calculated to measure the impact of environmental changes on radar signals. Specifically, the dispersion of all elements in the attenuation sequence is obtained, and the negative correlation mapping results with the humidity lag at each location are positively fused to obtain the attenuation fluctuation index. The product of the radar correlation coefficient of the current round and the attenuation fluctuation index is used as the radar correlation coefficient of each round.

[0059] In this embodiment, the radar correlation coefficient of each location in the current round is denoted as B, and its formula is as follows: In the formula, B is the radar correlation coefficient, A is the radar distribution coefficient, σ is the variance of the attenuation sequence, and Z is the humidity hysteresis. It is denoted as the decaying volatility index.

[0060] The flowchart for obtaining the radar correlation coefficient is as follows: Figure 2 As shown.

[0061] It is understandable that the radar distribution coefficient of normal road surfaces is relatively small. At the same time, due to the small fluctuation range of humidity, the fluctuation of the dielectric constant of the road surface caused by humidity is relatively small. Therefore, the attenuation fluctuation of electromagnetic waves under different humidity environments is relatively small. In addition, the response speed of normal road surfaces to environmental changes is relatively slow, resulting in a relatively large hysteresis. Conversely, when there are cavities and water-containing defects in the road surface, the more severe the defects, the larger the radar correlation coefficient.

[0062] S4: Simulate the radar signal at each location on the road surface to be inspected, and perform statistical analysis with the signal collected in the current round. Combine the threshold segmentation results of the radar correlation coefficient of a preset number of samples to obtain the probability value of defects at each location on the road surface to be inspected, and determine whether there are cavities and water-bearing defects at each location.

[0063] Ground-penetrating radar (GPR) signals from γ samples are collected. These γ samples include normal road surfaces and road surfaces with varying degrees of severity and moisture content, including those with cavities and water-containing defects. The radar correlation coefficient for each location is calculated. It should be noted that the γ samples refer to all detection locations on the road segment to be inspected. In this embodiment, γ is set to 200, but the implementer can set it according to actual conditions. The radar correlation coefficients of the γ samples are used as input, and the Otsu threshold segmentation method is applied to output a segmentation threshold. The Otsu threshold segmentation method is a well-known technique and will not be described in detail here.

[0064] When detecting pavement cavities and water-bearing defects at a specific location, a simulation is first performed using the finite-difference time-domain method to output simulated radar signal data. Then, statistical analysis is performed on the simulated radar signal data and the actual ground-penetrating radar signals to obtain the probability of pavement cavities and water-bearing defects existing at that location. In this embodiment, the statistical analysis is performed using Bayesian statistics, a well-known technique that will not be elaborated upon here. The above probability only considers the characteristics of the radar signal; its accuracy needs improvement when significantly affected by environmental factors or for minor defects. Therefore, further optimization is performed by incorporating radar correlation coefficients. In the formula, P is the probability value of the disease at each location in the current round, B is the radar correlation coefficient at each location in the current round, θ is the radar correlation coefficient segmentation threshold for a preset number of samples, P0 is the probability value of the statistical analysis at each location, and min{} represents the minimum value function. When the adjusted probability value is greater than the preset threshold, it is determined that there is a cavity water-bearing disease; otherwise, there is no cavity water-bearing disease. In this embodiment, the preset threshold is 0.7, and the implementer can set it according to the actual situation.

[0065] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description; sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0066] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for detecting pavement cavities and water-bearing defects based on radar detection, characterized in that, The method includes the following steps: Acquire the ambient humidity data of the road surface to be tested and the single-channel reflected wave received at each location in the current cycle; The extreme values ​​in the ballistic reflected waves of each position in the current round are filtered to obtain significant extreme values; all extreme values ​​of the reflected waves between two adjacent significant extreme values ​​are formed into a reflected wave subsequence; the number of significant extreme values ​​of each position in all current rounds is used to form the extreme value sequence of each position in the current round, and the extreme value sequence is fitted to obtain a fitting curve; the differences between significant extreme values ​​in all current rounds of each position, the length distribution of the reflected wave subsequence, and the element distribution of the extreme value sequence obtained in the current round of each position, combined with the shape characteristics of the fitting curve, are used to obtain the radar distribution coefficient of each position in the current round. By analyzing the overall distribution characteristics of the differences in significant extreme values ​​in all rounds at each location and the similarity to the overall distribution of environmental humidity data, the humidity lag at each location is obtained. Combined with the radar distribution coefficient, the radar correlation coefficient of each location in the current round is obtained. The radar signal at each location on the road surface to be inspected is simulated and statistically analyzed with the signal collected in the current round. Combined with the threshold segmentation results of the radar correlation coefficient of a preset number of samples, the probability value of the defect at each location on the road surface to be inspected is obtained, and it is determined whether there is a cavity and water-bearing defect at each location.

2. The method for detecting water-bearing road surface cavities based on radar detection as described in claim 1, characterized in that, The significant extreme value obtained is specifically as follows: For each position and each round of single-channel reflected wave, the absolute values ​​of all extreme values ​​are used to form a reflected wave sequence. The upper quartile of the reflected wave sequence is taken, and the elements greater than the upper quartile are recorded as significant extreme values.

3. The method for detecting water-bearing road surface cavities based on radar detection as described in claim 1, characterized in that, The radar distribution coefficient for each location in the current round is obtained as follows: Analyze the difference distribution between adjacent significant extreme values ​​in all rounds at each location to obtain the average change value of reflected wave attenuation in the current round at each location; Based on the length distribution of the reflected wavelet subsequence obtained in all rounds at each position, the minimum length value of the current round is obtained; Analyze the element distribution of the extreme value sequence obtained in the current round at each position and the shape characteristics of the fitted curve to obtain the trend of the number of extreme values ​​in the current round at each position; The negative correlation mapping result of the minimum length value of the current round is positively fused with the average change value of the reflected wave attenuation, and then added to the trend value of the extreme value quantity of the current round at each position to obtain the radar distribution coefficient of the current round at each position.

4. The method for detecting water-bearing road surface cavities based on radar detection as described in claim 3, characterized in that, The method for obtaining the average change in attenuation of the reflected wave at each position in the current round is as follows: Obtain the degree of dispersion of the difference between all adjacent significant extreme values ​​at each position in each round, and record the average value of the difference between the degree of dispersion of all adjacent rounds before the current round at each position as the average change value of the reflected wave attenuation in the current round.

5. The method for detecting water-bearing road surface cavities based on radar detection as described in claim 3, characterized in that, The minimum length value for the current round is obtained as follows: Obtain the minimum length of the reflected wavelet subsequence at each position and in each round. Average the minimum lengths obtained in all rounds prior to the current round at each position to obtain the minimum length value for the current round.

6. The method for detecting water-bearing road surface cavities based on radar detection as described in claim 3, characterized in that, The trend of the number of extreme values ​​at each position in the current round is obtained as follows: Obtain the maximum value of the extreme value sequence at each position, and obtain the percentage of data points on the curve to the left of the maximum value whose derivative values ​​are positive, denoted as the first percentage. The percentage of data points on the curve to the right of the maximum value whose derivative values ​​are negative is recorded as the second percentage. The average of the first proportion and the second proportion is used as the trend of the number of extreme values ​​in the current round for each position.

7. The method for detecting water-bearing road surface cavities based on radar detection as described in claim 1, characterized in that, The humidity hysteresis at each location is obtained as follows: The environmental humidity data of each location in all current rounds are combined into a humidity sequence; Calculate the mean of the differences between all adjacent significant extreme values ​​in each round, and use the obtained mean to form a decay sequence; Cross-correlation analysis was performed on the humidity series and the attenuation series to obtain the cross-correlation coefficient for each hysteresis quantity. The hysteresis corresponding to the maximum cross-correlation coefficient is taken as the humidity hysteresis for each location.

8. The method for detecting water-bearing road surface cavities based on radar detection as described in claim 7, characterized in that, The specific process for obtaining the radar correlation coefficient for each location in the current round is as follows: For the current round at each location, the dispersion of all elements in the decay sequence is obtained, and the negative correlation mapping results with the humidity hysteresis at each location are positively fused to obtain the decay fluctuation index. The product of the radar correlation coefficient of the current round and the attenuation fluctuation index is used as the radar correlation coefficient for each round.

9. The method for detecting water-bearing road surface cavities based on radar detection as described in claim 1, characterized in that, The specific formula for obtaining the probability value of defects at each location of the road surface to be inspected is as follows: P is the disease probability value of each location in the current round, B is the radar correlation coefficient of each location in the current round, λ is the radar correlation coefficient segmentation threshold of a preset number of samples, P0 is the probability value of statistical analysis for each location, and min{} represents the minimum value function.

10. The method for detecting water-bearing road surface cavities based on radar detection as described in claim 1, characterized in that, The determination of whether there is a cavity containing water at each location is specifically as follows: If the probability value of the disease in the current round at each location is greater than the preset threshold, it is determined that there is a cavity water-bearing disease at the corresponding location; otherwise, there is no cavity water-bearing disease.

Images