Urban road ground penetrating radar scale general survey method based on change detection

By introducing change detection methods into ground penetrating radar technology, combining high-precision positioning and data processing algorithms, the problem of traditional technologies being unable to fully detect road hidden dangers and relying on manual experience is solved, and a rapid census and automatic detection of large-scale road grassroots conditions has been achieved. , improves detection efficiency and accuracy, and prevents road collapses and accidents.

CN118962667AInactive Publication Date: 2024-11-15SUZHOU TONGZHI RUIDA TRANSPORTATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410728441.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-06-06
Publication Date
2024-11-15
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Traditional road detection technology cannot fully detect hidden dangers under the road. It relies on manual experience and has low signal interpretation efficiency. The result is short effective period, making it difficult to achieve continuous and systematic monitoring of large-scale urban roads.

Method used

Using the ground-penetrating radar scale census method based on change detection, the ground-penetrating radar equipment is combined with high-precision positioning system, and the road base structure is regularly scanned to perform data slicing processing, filtering and signal feature extraction, and the change detection technology is used to compare multiple detection data. , realize automatic detection and evaluation.

Benefits of technology

It significantly improves the speed and accuracy of data analysis, reduces dependence on professional personnel, and realizes a rapid census of the grassroots conditions of large-scale road network roads, promptly detects road hidden dangers and lesions, repairs potential safety problems, and prevents road collapses and related accidents.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118962667A_ABST
    Figure CN118962667A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of road detection, in particular to an urban road ground penetrating radar scale general survey method based on change detection, which mainly comprises the following steps of: based on ground penetrating radar equipment, combining a high-precision positioning system and an encoder, systematically and regularly scanning an urban road to obtain detailed data of a road base structure condition, storing according to a standard format; further performing slicing processing on the ground penetrating radar data at the same position in multiple times of detection to obtain slicing data; interference noise of the slice data is eliminated through filtering processing, and then signal features of the slice data are extracted; using a change detection technology to compare the ground penetrating radar slice data acquired by multiple times of detection and discover a change area; and finally, evaluating the road base evolution condition of the slice areas detected for multiple times. Compared with the prior art, the method has the advantages of improving the roadbase general survey efficiency, reducing the dependence on professionals, reducing the detection cost and the like.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of road detection technology, and in particular to a ground-penetrating radar scale survey method for urban roads based on change detection, which compares and analyzes ground-penetrating radar data obtained at different times to achieve a rapid survey of road base conditions in a large-scale road network and timely discover hidden dangers and lesions of roads. Background Art

[0003] Traditional road inspection technology mainly relies on visual inspection and local excavation inspection, which often cannot fully detect hidden dangers under the road, such as cavities, broken water pipes or other underground facilities. As a high-frequency electromagnetic wave detection method, ground penetrating radar has the advantages of continuous non-destructive, fast speed and high accuracy. It has become one of the most widely used technical means for urban cavity detection. However, it still faces many challenges in large-scale urban road flaw detection applications, such as reliance on manual experience, difficult signal interpretation, low processing efficiency and short validity period of results: First, high-cost equipment and operation limit its application in large-scale urban roads; second, due to the complexity and hidden multi-solution of radar maps, traditional data processing and analysis are highly dependent on the experience and judgment of professionals. An expert can only process 20 kilometers of data per day, which not only increases labor costs, but also limits the efficiency and frequency of detection; finally, since existing technologies usually focus on point-to-point detection, it is difficult to achieve continuous and systematic monitoring of roads throughout the city.

[0004] Therefore, in order to prevent collapse disasters and accurately detect and proactively deal with the potential collapse hazards of urban roads in a timely manner, it is necessary to carry out tracking detection of underground hidden dangers on urban roads. By comparing and analyzing the ground penetrating radar data obtained at different times, a rapid survey of the road base conditions in a large area of ​​road networks can be achieved, hidden road lesions can be discovered in a timely manner, and potential road safety problems can be repaired, thereby effectively preventing road collapse and related accidents. Summary of the invention

[0005] The present invention aims to solve the problems faced by the existing road detection technology: 1) it is often unable to fully detect hidden dangers under the road, such as cavities, broken water pipes or damage to other underground facilities; 2) it relies on manual experience, signals are difficult to interpret, processing efficiency is low, and the results are short-lived. A large-scale survey method of urban roads by ground-penetrating radar based on change detection is provided. The method compares and analyzes ground-penetrating radar data obtained by detection at different times to achieve a rapid survey of the road base conditions in a large area of ​​the road network. It can timely discover hidden dangers and lesions of roads and repair potential road safety problems, thereby effectively preventing road collapse and related accidents.

[0006] The technical solution adopted by the present invention is:

[0007] A method for urban road ground penetrating radar scale survey based on change detection, comprising the following steps:

[0008] S1. Based on ground penetrating radar equipment, combined with high-precision positioning systems and encoders, systematically scan urban roads regularly to obtain detailed data on the road base structure and store it in a standard format;

[0009] S2, determining a GPR data slicing site corresponding to a change detection position, and slicing the GPR data at the same position in multiple detections to obtain slicing data;

[0010] S3, filtering the slice data and extracting signal features of the slice data;

[0011] S4. Based on the extracted signal features, using change detection technology, the GPR slice data collected by multiple tests are compared and the changed areas are found;

[0012] S5. Evaluate the road base evolution status based on the change detection results.

[0013] Furthermore, the standard format in step S1 is a two-dimensional numerical matrix, in which the magnitude of the numerical value represents the amplitude of the radar wave, and the positive and negative values ​​represent the phase information.

[0014] Furthermore, the step S2 includes the following sub-steps:

[0015] S21: After the first detection, determine the change detection position, and use a high-precision positioning system to obtain the GPS coordinates of the two slice sites before and after the change detection position, as well as the encoder distance of each point from the fixed marker point;

[0016] S22: In subsequent multiple detections, the locations whose spatial distance is less than the distance threshold are matched according to the GPS coordinates, and the same location of the ground penetrating radar data in the multiple detections is matched to obtain adjacent areas whose distance is less than the threshold;

[0017] S23, according to the encoder distance between the site obtained in S21 and the fixed marker point, the adjacent area obtained in S22 is divided more accurately, and the columns of the ground penetrating radar slice data in multiple detections are determined, that is, the columns corresponding to the two-dimensional numerical matrix;

[0018] S24, slicing the ground penetrating radar data at the same position, dividing the two-dimensional numerical matrix along the columns according to a preset number of columns, and obtaining slice data of the detection and comparison position.

[0019] Furthermore, the step S3 includes the following sub-steps:

[0020] S31, using filtering and noise reduction methods such as background elimination and direct wave removal to process the noise in the slice data to filter out signal interference in urban scenes;

[0021] S32, extracting features of radar slice data, which may include instantaneous frequency IF, instantaneous phase IP, instantaneous amplitude IA of the radar signal, or multi-attribute features of IF, IP, IA fused in pairs.

[0022] Furthermore, the change detection technology in step S4 is image difference technology, twin neural network or generative adversarial network to achieve automatic detection of changes in road base conditions.

[0023] Furthermore, the evaluation of the road base evolution condition in step S5 is that when the comparison finds that the slice area has no obvious change compared with the previous detection, it is determined that the current road base area has no obvious deterioration; when the comparison finds that the slice area has obvious change compared with the previous detection, the slice area is further identified and the corresponding road base condition is evaluated; the identification and evaluation is carried out by automatic algorithm recognition, professional review or a combination of the two.

[0024] Furthermore, the ground penetrating radar equipment is air-coupled or ground-coupled, the radar frequency is 200MHz to 1GHz, the acquisition frequency of the high-precision positioning system is ≥1Hz, and the positioning accuracy is ≤10m.

[0025] The beneficial effects of the present invention are:

[0026] (1) The present invention uses advanced ground-penetrating radar technology and intelligent data processing algorithms to achieve large-scale, high-efficiency scanning and automatic change detection of urban roads, significantly improving the speed and accuracy of data analysis.

[0027] (2) The present invention significantly reduces the reliance on professionals through algorithm optimization and automated processes, thereby reducing the marginal cost of repeated inspections and providing an effective approach for large-scale and high-frequency surveys of urban road infrastructure.

[0028] In summary, the urban road ground-penetrating radar large-scale survey method based on change detection provided by the present invention has the advantages of realizing a rapid survey of the road base conditions of a large-scale road network by comparing and analyzing the ground-penetrating radar data obtained at different times, and can timely discover hidden dangers and lesions of roads, repair potential road safety problems, and thus effectively prevent road collapse and related accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a flow chart of a method for urban road ground penetrating radar scale survey based on change detection provided by the present invention;

[0030] Figure 2 It is a schematic diagram for comparing changes in multiple detection results of step 4 in Example 1 of a method for large-scale survey of urban roads by ground penetrating radar based on change detection provided by the present invention;

[0031] Figure 3 This is a chart of evaluation results of the evolution of the road base layer in Example 1 of the urban road ground penetrating radar scale survey method based on change detection provided by the present invention. DETAILED DESCRIPTION

[0032] The core of the present invention is to provide a large-scale survey method of urban roads by ground-penetrating radar based on change detection. The method compares and analyzes the ground-penetrating radar data obtained at different times to achieve a rapid survey of the road base conditions in a large area of ​​the road network, and can timely discover hidden road lesions and repair potential road safety problems, thereby effectively preventing road collapse and related accidents.

[0033] A method for urban road ground penetrating radar scale survey based on change detection, the specific process is as follows Figure 1 As shown, the implementation steps mainly include the following:

[0034] S1. Based on ground penetrating radar equipment and combined with a high-precision positioning system, urban roads are systematically scanned regularly to obtain detailed data on the road base structure status, and the collection distance is recorded by an encoder installed on the wheels of the ground penetrating radar vehicle. Ground penetrating radar equipment can be air-coupled or ground-coupled, with a radar frequency of 200MHz to 1GHz to provide high-quality detection data under different road types and conditions. The acquisition frequency of the high-precision positioning system is ≥1Hz, and the positioning accuracy is ≤10m. After the acquisition is completed, the ground penetrating radar data is stored in a standard format, that is, a two-dimensional numerical matrix, where the numerical value represents the amplitude of the radar wave, and the positive and negative numerical values ​​represent the phase information.

[0035] S2. In order to compare the detection data of the same position in multiple detections, it is necessary to slice the two-dimensional numerical matrix obtained by radar detection. The slice site is the column corresponding to the two-dimensional numerical matrix. Determine the two slice sites corresponding to the detection position, slice the ground penetrating radar data of the same position in multiple detections to obtain slice data. The specific sub-steps are as follows:

[0036] S21. After the first detection, determine the change detection position, and use a high-precision positioning system to obtain the GPS coordinates of the two slice sites before and after the change detection position, as well as the encoder distance of each point from the fixed marker point;

[0037] S22, in subsequent multiple detections, according to the GPS coordinate matching of the sites whose spatial distance is less than the distance threshold, the same location of the ground penetrating radar data in the multiple detections is matched to obtain adjacent areas whose distance is less than the threshold, wherein the distance threshold is usually selected as 10 meters;

[0038] S23. According to the encoder distance between the site obtained in S21 and the fixed marker point, the adjacent area obtained in S22 is divided more accurately to determine the columns of the ground penetrating radar slice data in multiple detections, that is, the columns corresponding to the two-dimensional numerical matrix.

[0039] S24, slicing the ground penetrating radar data at the same position, dividing the two-dimensional numerical matrix along the columns according to a preset number of columns, and obtaining slice data at the change detection position.

[0040] S3, filtering the slice data and extracting the signal features of the slice data. The specific sub-steps are as follows:

[0041] S31. Use background elimination, direct wave removal and other filtering noise reduction methods to process the noise in the slice data to filter out signal interference in urban scenes.

[0042] S32, extracting the features of the radar slice data, the instantaneous frequency (IF), instantaneous phase (IP), instantaneous amplitude (IA) of the radar signal, or multi-attribute features of IF, IP, IA fused in pairs, can be selected, and the calculation is as follows:

[0043]

[0044]

[0045]

[0046]

[0047] Where x(t) is the original signal, t is the time variable, ν(t) is the Hilbert transform result of x(t), α(t) is the instantaneous amplitude, θ(t) is the instantaneous phase, and ω(t) is the instantaneous frequency.

[0048] S4. Based on the extracted signal features, the change detection technology is used to compare the GPR slice data collected by multiple tests and find the changed areas. The change detection technology can be image difference technology, twin neural network, generative adversarial network, etc., to realize the automatic detection of changes in road base conditions.

[0049] S5. Evaluate the evolution of the road base in the slice area. When the comparison shows that the slice area has no significant changes compared with the last detection, it is determined that the current road base area has no significant degradation; when the comparison shows that the slice area has significant changes compared with the last detection, further identify the slice area and evaluate the corresponding road base condition.

[0050] The above identification and evaluation can be carried out by automatic algorithm identification, professional review or a combination of both.

[0051] 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. The described embodiments are part of the embodiments of the present invention, but 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 should fall within the scope of protection of the present invention.

[0052] The specific process of this embodiment is as follows Figure 1 This embodiment can be divided into the following six main steps:

[0053] Step 1: Select a road section in Xuhui District, Shanghai, and use LTD-2100 ground-coupled ground penetrating radar equipment to regularly scan the road base structure at intervals of two months, and store the collected data in a standard two-dimensional numerical matrix format. The radar antenna frequency is selected to be 400MHz, with a longitudinal accuracy of 0.02-0.05m, a lateral accuracy of 0.1-0.2m, and a maximum detection depth of 4 meters.

[0054] Step 2: Slice the suspected lesion area in the road section, use the high-precision positioning system and the vehicle encoder to determine the range of the slice area, slice the two-dimensional numerical matrix collected by multiple tests, and obtain the slice data.

[0055] Step 3: Design a two-dimensional filter, perform a two-dimensional inverse Fourier transform on the slice data to remove the direct wave part of the signal, and then use the mean subtraction method to eliminate the background clutter of the ground penetrating radar data and perform data filtering. Then, perform a Hibbert transform on the ground penetrating radar data that has passed the signal gain to extract the instantaneous frequency, instantaneous phase, and instantaneous amplitude characteristics of the radar signal.

[0056] Step 4: Based on historical ground penetrating radar detection, extract multiple pairs of radar image data collected at the same location at different times. Each pair of images needs to be marked with known change areas and preprocessed to construct training samples, and then train to obtain a change detection model based on the twin network. Based on the multi-attribute features of the radar signal extracted in step 3, use the trained change detection model to perform change detection on the slice data extracted in step 2, and automatically compare and identify the changes in the slice area in multiple detections. The comparison results are as follows: Figure 2 shown.

[0057] Step 5: Use the target detection algorithm to identify the damage in the slice area, and combine the review and judgment of professionals to evaluate the evolution of the road base in the slice area that has been tested multiple times. The evaluation results are as follows: Figure 3 shown.

[0058] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for urban road ground penetrating radar scale survey based on change detection, characterized in that: The following steps are involved: S1. Based on ground penetrating radar equipment, combined with high-precision positioning systems and encoders, urban roads are systematically scanned regularly to obtain detailed data on the road base structure and store it in a standard format; S2, determining a GPR data slicing site corresponding to a change detection position, and slicing the GPR data at the same position in multiple detections to obtain slicing data; S3, filtering the slice data and extracting signal features of the slice data; S4. Based on the extracted signal features, using change detection technology, the GPR slice data collected by multiple tests are compared and the changed areas are found; S5. Evaluate the road base evolution status based on the change detection results.

2. The urban road ground penetrating radar scale survey method based on change detection according to claim 1 is characterized by: The standard format in step S1 is a two-dimensional numerical matrix, in which the magnitude of the numerical value represents the amplitude of the radar wave, and the positive and negative values ​​represent the phase information.

3. The urban road ground penetrating radar scale survey method based on change detection according to claim 1 is characterized by: The step S2 includes the following sub-steps: S21: After the first detection, determine the change detection position, and use a high-precision positioning system to obtain the GPS coordinates of the two slice sites before and after the change detection position, as well as the encoder distance of each point from the fixed marker point; S22: In subsequent multiple detections, the locations whose spatial distance is less than the distance threshold are matched according to the GPS coordinates, and the same location of the ground penetrating radar data in the multiple detections is matched to obtain adjacent areas whose distance is less than the threshold; S23, according to the encoder distance between the site obtained in S21 and the fixed marker point, the adjacent area obtained in S22 is divided more accurately, and the columns of the ground penetrating radar slice data in multiple detections are determined, that is, the columns corresponding to the two-dimensional numerical matrix; S24, slicing the ground penetrating radar data at the same position, dividing the two-dimensional numerical matrix along the columns according to a preset number of columns, and obtaining slice data of the detection and comparison position.

4. The urban road ground penetrating radar scale survey method based on change detection according to claim 1 is characterized in that; The step S3 includes the following sub-steps: S31, using filtering and noise reduction methods such as background elimination and direct wave removal to process the noise in the slice data to filter out signal interference in urban scenes; S32, extracting features of radar slice data, which may include instantaneous frequency IF, instantaneous phase IP, instantaneous amplitude IA of the radar signal, or multi-attribute features of IF, IP, IA fused in pairs.

5. The urban road ground penetrating radar scale survey method based on change detection according to claim 1 is characterized by: The change detection technology in step S4 is image difference technology, twin neural network or generative adversarial network to achieve automatic detection of changes in road base conditions.

6. The urban road ground penetrating radar scale survey method based on change detection according to claim 1 is characterized in that: The evaluation of the road base evolution condition in step S5 is that when the comparison finds that the slice area has no obvious change compared with the previous detection, it is determined that the current road base area has no obvious deterioration; when the comparison finds that the slice area has obvious change compared with the previous detection, further identification is performed on the slice area and the corresponding road base condition is evaluated; the identification and evaluation are performed by automatic algorithm recognition, professional review or a combination of the two.

7. The urban road ground penetrating radar scale survey method based on change detection according to claim 1 is characterized in that: The ground penetrating radar equipment is of air coupling type or ground coupling type, the radar frequency is 200MHz-1GHz, the acquisition frequency of the high-precision positioning system is ≥1Hz, and the positioning accuracy is ≤10m.