A method, device and medium for identifying hidden road defects using three-dimensional ground penetrating radar

Abstract

The present application discloses a method, device and medium for identifying hidden road defects using three-dimensional ground penetrating radar, and relates to the field of road defect identification. The method includes: inputting the B-SCAN maps corresponding to each road position in the target road area into a road signal recognition model to obtain the corresponding road signal; determining the first judgment situation based on the road signals corresponding to the continuous road positions in the target road area, which is the presence of hidden defects, the presence of structures or no abnormality; matching the C-SCAN maps corresponding to each road position based on the preset abnormal area color threshold range to determine the road area to be judged; determining the second judgment situation based on the C-SCAN maps corresponding to the road area to be judged; determining the final hidden disease situation corresponding to the target road area based on the first judgment situation and the second judgment situation based on the preset multi-dimensional information cross-judgment rule. The present application can realize efficient and accurate recognition of hidden road defects.

Description

Technical Field

[0001] The present application relates to the field of road defect identification, and in particular to a method, device and medium for identifying hidden road defects using three-dimensional ground penetrating radar. Background Art

[0002] Due to the large differences in bearing capacity, stiffness and other characteristics between pavement and subgrade materials, as the service time of roads increases, hidden road defects such as voids and cavities are prone to occur under the combined effects of rain, groundwater, temperature, vibration and other factors. These defects are highly concealed and can extend to the ground if not repaired in time, leading to cracks on the road surface, road subsidence, and even collapse, causing major safety accidents.

[0003] Ground Penetrating Radar (GPR) is a non-destructive detection instrument that uses high-frequency electromagnetic wave signals to detect the distribution of underground soil media. It has the advantages of high efficiency and high precision. When using three-dimensional ground penetrating radar, multiple transmitting-receiving devices are generally arranged in parallel on the radar vehicle, and interpolation and other methods are used to combine various two-dimensional data to present horizontal and vertical dimensional information (the vertical dimension is the B-SCAN map, and the horizontal dimension is the C-SCAN map). The three-dimensional ground penetrating radar can easily and accurately determine abnormal points. However, the single detection distance of the three-dimensional ground penetrating radar is basically more than 10km. The subsequent interpretation of the radar map requires maintenance personnel to spend a lot of time and energy, and the data processing efficiency and accuracy are difficult to guarantee. Summary of the invention

[0004] The purpose of this application is to provide a three-dimensional ground penetrating radar road hidden disease identification method, equipment and medium, which can realize efficient and accurate road hidden disease identification.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides a method for identifying hidden road defects using a three-dimensional ground penetrating radar, comprising:

[0007] Acquire three-dimensional ground penetrating radar detection data of the target road area; the three-dimensional ground penetrating radar detection data includes a B-SCAN map and a C-SCAN map corresponding to any road position;

[0008] Inputting the B-SCAN graph corresponding to each of the road positions into a road signal recognition model to obtain a corresponding road signal; the road signal recognition model is a pre-trained YOLO model; the road signal includes a normal signal, a suspected abnormal signal and a structure signal;

[0009] Determining a first judgment condition of the target road area according to road signals corresponding to continuous road positions in the target road area; the first judgment condition is the presence of hidden diseases, the presence of structures, or no abnormality;

[0010] Based on the preset abnormal area color threshold range, the C-SCAN maps corresponding to each of the road positions are matched to determine the road area to be judged;

[0011] Determine a second interpretation of the target road area according to the C-SCAN map corresponding to the road area to be interpreted; the second interpretation is the presence of hidden diseases, the presence of structures or no abnormality;

[0012] Based on the preset multi-dimensional information cross-interpretation rules, the final hidden disease situation corresponding to the target road area is determined according to the first interpretation situation and the second interpretation situation; the final hidden disease situation includes the road position corresponding to the hidden disease.

[0013] In a second aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a method for identifying hidden road defects using a three-dimensional ground penetrating radar.

[0014] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for identifying hidden road defects using three-dimensional ground penetrating radar.

[0015] According to the specific embodiments provided by the present application, the present application discloses the following technical effects: the present application provides a method, device and medium for identifying hidden defects of roads by three-dimensional ground penetrating radar, and obtains corresponding road signals according to the B-SCAN map and road signal recognition model corresponding to each road position in the target road area, and then determines the first judgment situation according to the road signals corresponding to the continuous road positions, that is, the hidden defect situation of the road determined according to the B-SCAN map is obtained at this time, which may be the situation of the presence of hidden defects, the presence of structures and the absence of abnormalities. At the same time, according to the C-SCAN map corresponding to each road position and the preset abnormal area color threshold range, the screening of the road area to be judged is realized, and then the area is judged to obtain the second judgment situation, that is, the hidden defect situation of the road determined according to the C-SCAN map is obtained at this time. Finally, according to the preset multi-dimensional information cross-interpretation rules, the final hidden defect situation is determined from the above two judgment situations, thereby overcoming the problem that a single map causes some abnormal signals to be missed in the actual road ground penetrating radar detection work, so as to hinder the road repair project. That is to say, this application combines the information contained in the B-SCAN map and the C-SCAN map corresponding to each road location, overcoming the bottleneck of the low efficiency of traditional ground penetrating radar maps, which mainly rely on manual interpretation, and also improves the problem of low accuracy and credibility in judging the existence of diseases based on a single map, thus achieving efficient and accurate identification of hidden road diseases. Furthermore, this application can also provide a basis for the evaluation of the underground soil status of road traffic and ensure the safety of urban road traffic networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 This is a diagram of the application environment of the method for identifying hidden road defects using three-dimensional ground penetrating radar in an embodiment of the present application.

[0018] Figure 2 It is a schematic flow chart of the method for identifying hidden road defects using three-dimensional ground penetrating radar in an embodiment of the present application.

[0019] Figure 3 A logical schematic diagram of the preset multi-dimensional information cross-interpretation rules provided in an embodiment of the present application.

[0020] Figure 4 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0022] This application analyzes the C-SCAN map to identify possible hidden diseases, and determines the areas where hidden diseases are suspected to exist in multiple dimensions; based on the B-SCAN map of the area suspected of having hidden diseases, accurate classification is performed, and multi-dimensional information is fully utilized to determine the hidden diseases and corresponding areas. Finally, the abnormal signal judgment logic of the B-SCAN map and the C-SCAN map is combined to construct a preset multi-dimensional information cross-interpretation rule. This interpretation rule can reduce the omission of abnormal signals, ensure the accuracy of abnormal signal recognition, and effectively alleviate the omission and misreading of the three-dimensional ground penetrating radar map interpretation process.

[0023] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0024] The three-dimensional ground penetrating radar road hidden disease identification method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. The terminal 102 can send the three-dimensional ground penetrating radar detection data of the target road area to the server 104. After receiving the data, the server 104 first determines two interpretations based on the B-SCAN map and the C-SCAN map, and then selects the two interpretations based on the preset multi-dimensional information cross-interpretation rules to obtain the final hidden disease situation. The server 104 can feedback the final hidden disease situation to the terminal 102. In addition, in some embodiments, the three-dimensional ground penetrating radar road hidden disease identification method can also be implemented by the server 104 or the terminal 102 alone.

[0025] The terminal 102 may be, but is not limited to, various desktop computers, laptop computers, smart phones, tablet computers, etc. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers, or may be a cloud server.

[0026] In an exemplary embodiment, Figure 2As shown, a method for identifying hidden road defects using a three-dimensional ground penetrating radar is provided. The method is executed by a computer device, and can be executed by a computer device such as a terminal or a server alone, or by a terminal and a server together. In the embodiment of the present application, the method is applied to Figure 1 The server 104 in the example is used for explanation, and the steps include the following steps 201 to 206.

[0027] Step 201, obtaining three-dimensional ground penetrating radar detection data of a target road area; the three-dimensional ground penetrating radar detection data includes a B-SCAN map and a C-SCAN map corresponding to any road position.

[0028] In a practical application, 3D ground penetrating radar detection data is derived from the 3D ground penetrating radar, which includes the B-SCAN spectrum, C-SCAN spectrum, and the location information of the spectrum area. During the movement of the 3D ground penetrating radar, multiple signal transmitters simultaneously transmit signals underground, and the information spectra in multiple depth directions of the reflection are linearly interpolated to obtain a 3D spectrum. The information spectrum formed by the reflected signal in the depth direction is the B-SCAN spectrum, and the signal spectrum in the horizontal ground direction is the C-SACN spectrum.

[0029] In order to facilitate the calculation of subsequent steps and improve the calculation efficiency, after the step of obtaining the three-dimensional ground penetrating radar detection data of the target road area, the B-SCAN map and C-SCAN map corresponding to each road position are intercepted from the three-dimensional ground penetrating radar detection data, and are numbered and named and stored in the B-SCAN map database and the C-SCAN map database respectively, and the actual coordinate information of all road positions is stored in a preset TXT format file. For example, for the B-SCAN map, it is intercepted and named: channel 1.jpg, channel 2.jpg...channel N.jpg; for the C-SCAN map, it is intercepted and named: depth 0.1.jpg, depth 0.2.jpg...depth N.jpg.

[0030] Step 202, inputting the B-SCAN graph corresponding to each of the road positions into a road signal recognition model to obtain a corresponding road signal; the road signal recognition model is a pre-trained YOLO model; the road signal includes a normal signal, a suspected abnormal signal and a structure signal.

[0031] In a specific application, the training process of the road signal recognition model includes steps (21) to (23), as shown below:

[0032] (21) Obtaining a set of atlas samples; any atlas sample in the atlas sample set includes a B-SCAN atlas and a road signal label corresponding to each road position in the historical three-dimensional ground penetrating radar detection data; the road signal label is any one of a normal signal, a suspected abnormal signal, and a structure signal, and is obtained by using LabelMe software to annotate the B-SCAN atlas with anchor frames according to actual survey results. Among them, the structure category corresponding to the structure signal can be a well, a pipeline, and other structures. In order to facilitate subsequent model training, the atlas sample set can be stored in the form of a VOC2007 format data set for backup.

[0033] (22) Build the YOLO V8 model, which is used to identify abnormal echo features in the B-SCAN map, and detect and classify them to obtain the corresponding road signals.

[0034] (23) Inputting the atlas sample set into the YOLO V8 model for training until the preset training conditions are met, and obtaining an optimized YOLO V8 model; marking the optimized YOLO V8 model as a road signal recognition model. Specifically, during the training process, the YOLO V8s pre-trained model is used for 1000 rounds of training, and the learning rate is set to 0.01, so that the trained model can accurately recognize the abnormal echo features in the B-SCAN atlas. When verifying the road signal recognition model obtained through the above training, it can be found that: especially in the recognition of invisible road diseases such as voids and holes, the mAP50 of the road signal recognition model of the present application is 0.962, and the mAP50-95 is 0.626, and the recognition accuracy index mAP (Mean Average Precision) of pipelines, wells and other non-abnormal structure signals is relatively high, which shows that the road signal recognition model of the present application can effectively recognize the abnormal echo features in the B-SCAN atlas.

[0035] Step 203, determining a first judgment condition of the target road area according to road signals corresponding to continuous road positions in the target road area; the first judgment condition is the presence of hidden defects, the presence of structures, or no abnormality.

[0036] According to actual engineering experience, the width of abnormal signals usually exceeds 0.3m. Based on this, if the signal of a structure such as a well or a suspected abnormal signal appears in three or more consecutive B-SCAN maps (the width of three consecutive B-SCAN scans is 0.25m), it is determined that there are invisible diseases in the target road area. Based on this, in a specific application, step 203 of this application includes:

[0037] (31) When the road signals corresponding to at least three consecutive road positions in the target road area are all suspected abnormal signals, the first judgment of the target road area is marked as the presence of hidden defects. In order to facilitate subsequent processing, the road positions corresponding to the suspected abnormalities can be determined first for backup.

[0038] (32) When the road signals corresponding to at least three consecutive road positions in the target road area are all structure signals, the first judgment situation of the target road area is marked as the existence of a structure. In order to facilitate subsequent processing, the road position corresponding to the structure can be determined first for backup.

[0039] (33) When the road signals corresponding to at most two consecutive road positions in the target road area are suspected abnormal signals or structure signals, the first judgment situation of the target road area is marked as no abnormality.

[0040] Step 204, based on the preset abnormal area color threshold range, the C-SCAN maps corresponding to the various road positions are matched to determine the road area to be judged.

[0041] The process of determining the preset color threshold range of the abnormal area includes:

[0042] (41) Based on the manual judgment experience of 3D ground penetrating radar detection data, the color of the abnormal area on the C-SCAN map is determined; in actual application, the colors of the abnormal area on the C-SCAN map are red and orange.

[0043] (42) Based on the HSV color dictionary, determine the color area minimum limit and the color area maximum limit according to the color of the abnormal area on the C-SCAN map; corresponding to the red and orange in the above text, the color area minimum limit is (0, 43, 46), and the color area maximum limit is (12, 255, 255).

[0044] (43) Determine a preset abnormal area color threshold range based on the color area minimum limit value and the color area maximum limit value.

[0045] Step 205, determining a second interpretation of the target road area according to the C-SCAN map corresponding to the road area to be interpreted; the second interpretation is the presence of hidden diseases, the presence of structures or no abnormality.

[0046] In a specific application, step 205 includes:

[0047] (51) Determine the depth value of the signal in the road area to be determined based on the C-SCAN map corresponding to the road area to be determined. Specifically, traverse all C-SCAN maps of different depths in the road area to be determined to obtain the corresponding specific depth value.

[0048] (52) When the depth value is lower than the first preset depth value, the second judgment situation is marked as no abnormality. According to the experience of actual engineering cases, for general abnormal areas, the depth will exceed 0.3m; for water wells and other structural areas, the depth will exceed 1.2m, so the first preset depth value is set to 0.3m and the second preset depth value is set to 1.2m. That is, when the road area to be judged is continuously distributed along the depth direction, that is, the depth value does not exceed 0.3m (that is, the road area to be judged appears in no more than three consecutive C-SCAN maps), it can be determined that the road area to be judged only contains changes in the density of the soil layer, and there is no special signal. Therefore, the second judgment situation is marked as no abnormality.

[0049] (53) When the depth value is higher than the first preset depth value and lower than the second preset depth value, the second judgment situation is marked as the presence of hidden defects. Specifically, when the road area to be judged is continuously distributed along the depth direction, that is, the depth value exceeds 0.3m and does not exceed 1.2m, it can be determined that there is a suspected abnormal signal in the road area to be judged, and therefore, the second judgment situation is marked as the presence of hidden defects.

[0050] (54) When the depth value is higher than the second preset depth value, the second judgment situation is marked as the existence of a structure. Specifically, when the road area to be judged is continuously distributed along the depth direction, that is, the depth value exceeds 1.2m, it can be determined that there is a structure such as a well in the road area to be judged, and therefore, the second judgment situation is marked as the existence of a structure.

[0051] Through the above steps (51) to (54), it can be determined whether there are hidden diseases in the road area to be judged. In order to facilitate subsequent processing, it is necessary to determine the road location corresponding to the hidden disease.

[0052] Step 206, based on the preset multi-dimensional information cross-interpretation rules, according to the first interpretation situation and the second interpretation situation, determine the final hidden disease situation corresponding to the target road area; the final hidden disease situation includes the road position corresponding to the hidden disease.

[0053] In a specific application, the preset multi-dimensional information cross-reading rule is:

[0054] (61) When the first judgment situation and the second judgment situation are the same, the first judgment situation or the second judgment situation is marked as the final hidden disease situation corresponding to the target road area. For example, when the first judgment situation and the second judgment situation are both normal, the corresponding final hidden disease situation is also normal, that is, when the two judgment situations are consistent, the consistent result is output.

[0055] (62) When the first judgment situation and the second judgment situation are different, and the first judgment situation or the second judgment situation is normal, the corresponding other judgment situation is marked as the final hidden disease situation corresponding to the target road area.

[0056] (63) When the first judgment situation and the second judgment situation are different, and both the first judgment situation and the second judgment situation are not no abnormality, the first judgment situation is marked as the final hidden disease situation corresponding to the target road area. This is because the misjudgment rate of hidden diseases in the C-SCAN map is relatively high, so the judgment situation corresponding to the B-SCAN map is selected as the final hidden disease situation. This judgment logic can reduce the missed judgment of hidden diseases and ensure the recognition accuracy.

[0057] In specific applications, there are two data processing orders based on the above-mentioned preset multi-dimensional information cross-interpretation rules:

[0058] The first method is to first process the B-SCAN map (execute steps 202-203) to obtain a first interpretation, then or simultaneously process the C-SCAN map (execute steps 204 and 205) to obtain a second interpretation, and finally combine the two interpretations to obtain the final hidden disease situation.

[0059] The second method is to process the C-SCAN map first (execute steps 204 and 205), and after obtaining the second interpretation, only process the B-SCAN maps corresponding to the two road locations (execute steps 202-203) based on the road locations corresponding to the hidden diseases or the road locations corresponding to the structures included in the second interpretation, so as to obtain the final hidden disease situation. Compared with the first method, this processing method reduces the total data processing amount.

[0060] In a specific application, corresponding to the second data processing order above, you can Figure 3The logic of represents the preset multi-dimensional information cross-interpretation rule, which is expressed in words as follows: first determine the C-SCAN recognition result, that is, the second interpretation situation; if the second interpretation situation determines that there is no special signal (that is, the second interpretation situation is no abnormality), then determine the B-SCAN recognition result, that is, determine the first interpretation situation; regardless of whether the first interpretation situation is no abnormality or other, the first interpretation situation is used as the final hidden disease situation. If the second interpretation situation determines that there is a special signal (that is, the second interpretation situation is the existence of a structure or the existence of a hidden disease), then determine the B-SCAN interpretation result, that is, determine the first interpretation situation; if the first interpretation situation is no abnormality, then use the C-SCAN interpretation result as the final hidden disease situation; if the first interpretation situation is the existence of a structure or the existence of a hidden disease, then use the B-SCAN interpretation result as the final hidden disease situation.

[0061] In another specific application, the present application also provides 50 cases, giving a comparison of the B-SCAN interpretation results, C-SCAN interpretation results, multi-dimensional information cross-judgment results determined by the preset multi-dimensional information cross-judgment rules of the present application, and actual survey results, which are presented in the form of the following Table 1. In Table 1, the corresponding interpretation of "suspected abnormality" is the presence of hidden diseases, the corresponding interpretation of "wells, pipelines, and other structures" is the presence of structures, and the corresponding interpretation of "no special signals" is no abnormality. It can be seen from Table 1 that if only a single B-SCAN or C-SCAN is used for judgment, the corresponding B-SCAN interpretation results and C-SCAN interpretation results have error rates of 16% and 24%. When the preset multi-dimensional information cross-judgment rules of the present application are used, the error rate is only 10%, which improves the accuracy by 6%, and there are no cases in which the situation that is actually a hidden disease is judged as no abnormality, which effectively alleviates the phenomenon of missed reading.

[0062] Table 1

[0063]

[0064] In summary, compared with the prior art, the present application has the following advantages:

[0065] (1) This application combines the actual specific situation and establishes a preset abnormal area color threshold range based on the HSV color dictionary, which can read the road area to be judged with suspected abnormal points in the C-SCAN map; and based on the engineering experience summary, combined with the depth value, the interpretation of hidden diseases can be realized.

[0066] (2) This application combines the multi-dimensional information of B-SCAN and C-SCAN in three-dimensional ground penetrating radar for cross-interpretation, thereby realizing intelligent identification of hidden road defects. Compared with traditional methods, it can effectively improve the accuracy and efficiency of identifying hidden road defects, and has great application potential in urban road safety assessment and repair.

[0067] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a three-dimensional ground penetrating radar road hidden disease identification method is implemented.

[0068] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0069] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0070] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0071] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0072] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0073] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0074] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0075] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for identifying hidden road defects using three-dimensional ground penetrating radar, characterized in that: The three-dimensional ground penetrating radar road hidden disease identification method comprises: Acquire three-dimensional ground penetrating radar detection data of the target road area; the three-dimensional ground penetrating radar detection data includes a B-SCAN map and a C-SCAN map corresponding to any road position; Inputting the B-SCAN graph corresponding to each of the road positions into a road signal recognition model to obtain a corresponding road signal; the road signal recognition model is a pre-trained YOLO model; the road signal includes a normal signal, a suspected abnormal signal and a structure signal; Determining a first judgment condition of the target road area according to road signals corresponding to continuous road positions in the target road area; the first judgment condition is the presence of hidden diseases, the presence of structures, or no abnormality; Based on the preset abnormal area color threshold range, the C-SCAN maps corresponding to each of the road positions are matched to determine the road area to be judged; Determine a second interpretation of the target road area according to the C-SCAN map corresponding to the road area to be interpreted; the second interpretation is the presence of hidden diseases, the presence of structures or no abnormality; Based on the preset multi-dimensional information cross-interpretation rule, the final hidden disease situation corresponding to the target road area is determined according to the first interpretation situation and the second interpretation situation; the final hidden disease situation includes the road position corresponding to the hidden disease; The preset multi-dimensional information cross-reading rule is: When the first interpretation situation and the second interpretation situation are the same, marking the first interpretation situation or the second interpretation situation as the final hidden disease situation corresponding to the target road area; When the first judgment situation and the second judgment situation are different, and the first judgment situation or the second judgment situation is normal, marking the corresponding other judgment situation as the final hidden disease situation corresponding to the target road area; When the first judgment situation and the second judgment situation are different, and both the first judgment situation and the second judgment situation are not without abnormality, the first judgment situation is marked as the final hidden disease situation corresponding to the target road area.

2. The method for identifying invisible road defects using three-dimensional ground penetrating radar according to claim 1, characterized in that: The process of determining the preset abnormal area color threshold range includes: Based on the manual judgment experience of 3D ground penetrating radar detection data, the color of the abnormal area on the C-SCAN map is determined; Based on the HSV color dictionary, determine the color area minimum limit and the color area maximum limit according to the color of the abnormal area on the C-SCAN map; A preset abnormal area color threshold range is determined according to the color area minimum limit value and the color area maximum limit value.

3. The method for identifying invisible road defects using three-dimensional ground penetrating radar according to claim 2, characterized in that: The lower limit of the color area is (0, 43, 46), and the upper limit of the color area is (12, 255, 255).

4. The method for identifying invisible road defects using three-dimensional ground penetrating radar according to claim 1, characterized in that: Determining a second interpretation condition of the target road area according to the C-SCAN map corresponding to the road area to be interpreted includes: Determining a depth value of a signal in the road area to be determined based on a C-SCAN map corresponding to the road area to be determined; When the depth value is lower than the first preset depth value, marking the second judgment situation as normal; When the depth value is higher than the first preset depth value and lower than the second preset depth value, marking the second interpretation as the presence of a hidden disease; When the depth value is higher than a second preset depth value, the second judgment situation is marked as the existence of a structure.

5. The method for identifying invisible road defects using three-dimensional ground penetrating radar according to claim 1, characterized in that: The training process of the road signal recognition model includes: Acquire a set of atlas samples; any atlas sample in the atlas sample set includes a B-SCAN atlas and a road signal label corresponding to each road position in the historical three-dimensional ground penetrating radar detection data; the road signal label is any one of a normal signal, a suspected abnormal signal, and a structure signal; Build the YOLO V8 model; The atlas sample set is input into the YOLO V8 model for training until the preset training conditions are met, thereby obtaining an optimized YOLO V8 model; and the optimized YOLO V8 model is marked as a road signal recognition model.

6. The method for identifying invisible road defects using three-dimensional ground penetrating radar according to claim 5, characterized in that: Determining a first interpretation condition of the target road area according to road signals corresponding to continuous road positions in the target road area includes: When the road signals corresponding to at least three consecutive road positions in the target road area are all suspected abnormal signals, marking the first interpretation of the target road area as having hidden defects; When the road signals corresponding to at least three consecutive road positions in the target road area are all structure signals, marking the first interpretation situation of the target road area as the existence of a structure; When the road signals corresponding to at most two consecutive road positions in the target road area are both suspected abnormal signals or structure signals, the first interpretation situation of the target road area is marked as no abnormality.

7. The method for identifying invisible road defects using three-dimensional ground penetrating radar according to claim 1, characterized in that: After the step of acquiring the three-dimensional ground penetrating radar detection data of the target road area, the three-dimensional ground penetrating radar road invisible disease identification method further includes: The B-SCAN map and C-SCAN map corresponding to each road position are intercepted from the three-dimensional ground penetrating radar detection data, numbered and named, and stored in the B-SCAN map database and the C-SCAN map database respectively, and the actual coordinate information of all the road positions is stored in a preset TXT format file.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the three-dimensional ground penetrating radar road hidden defect identification method described in any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for identifying hidden road defects using three-dimensional ground penetrating radar described in any one of claims 1 to 7 is implemented.

Images