Method, system, device, medium and program for positioning depth of subgrade disease

Abstract

The application relates to a subgrade disease depth positioning method, system, device, storage medium and computer program. The method comprises the following steps: acquiring a radar image corresponding to a subgrade disease, and performing image preprocessing on the radar image to obtain a target radar image; performing horizon identification on the target radar image to obtain each subgrade layer of the railway subgrade in the target radar image; calculating the target thickness of each subgrade layer according to the preset dielectric constant corresponding to each subgrade layer; and calculating the depth of the subgrade disease depth corresponding to each subgrade layer according to the target thickness to obtain the subgrade disease depth. The accuracy of railway subgrade disease detection is improved, and accurate positioning of the subgrade disease depth is realized.

Description

Technical Field

[0001] This application relates to the field of radar image processing technology, and in particular to a method, system, device, medium and program for locating the depth of roadbed defects. Background Technology

[0002] Using vehicle-mounted ground-penetrating radar (GPR) to inspect railway subgrade is a very important aspect. Railway subgrade inspection typically utilizes GPR technology to detect the distribution characteristics of subgrade defects by transmitting and receiving high-frequency electromagnetic waves through antennas, thereby obtaining radar images. Then, the radar images are used for layer identification to locate subgrade defects based on each subgrade layer. However, in the process of implementation, the different media of each subgrade layer lead to insufficient accuracy in layer identification, which may result in the inability to accurately locate subgrade defects.

[0003] Current technologies for railway subgrade inspection mostly utilize ground-penetrating radar (GPR) to detect subgrade defects, obtaining radar images. These images are then used for layer identification to determine the individual subgrade layers. However, interference from factors such as the heterogeneity of the underground soil medium, GPR noise itself, and environmental noise severely disrupts the received two-dimensional echo images, resulting in inaccurate layer identification. Subgrade thickness is calculated based on the dielectric constant of each subgrade layer to determine the defect depth. However, because the subgrade surface has a multi-layered structure with different media and dielectric constants, the accuracy of the calculated defect depth is insufficient, leading to significant errors. Therefore, improving the accuracy of railway subgrade defect detection and achieving precise location of subgrade defect depth are urgent problems to be solved. Summary of the Invention

[0004] This application provides a method, system, equipment, medium, and program for locating the depth of subgrade defects, in order to solve the problems of insufficient accuracy in subgrade defect detection and inaccurate calculation of subgrade defect depth.

[0005] Firstly, this application provides a method for locating the depth of roadbed defects, including: Acquire radar images corresponding to roadbed defects, and perform image preprocessing on the radar images to obtain target radar images; Layer identification is performed on the target radar image to obtain the base layers of the railway subgrade in the target radar image; The target thickness of each substrate is calculated based on the preset dielectric constant of each substrate. The depth of subgrade defects is calculated based on the target thickness to obtain the subgrade defect depth.

[0006] In some embodiments, the step of performing image preprocessing on the radar image to obtain a target radar image includes: The radar image is subjected to noise removal using Gaussian filtering to obtain a denoised radar image; The denoised radar image is then enhanced with texture to obtain the target radar image.

[0007] In some embodiments, the step of removing noise from the radar image using Gaussian filtering to obtain a denoised radar image includes: Obtain the preset Gaussian kernel size and standard deviation; Generate a Gaussian kernel matrix based on the size and standard deviation of the Gaussian kernel; The edge pixels of the radar image are filled according to the Gaussian kernel matrix to obtain a filled radar image; The Gaussian kernel and the filled radar image are convolved to obtain several pixel values; The aforementioned pixel values ​​are aggregated into a denoised radar image.

[0008] In some embodiments, the step of performing layer identification on the radar image to obtain the base layers of the target railway subgrade includes: The radar image is horizontally segmented and sampled to obtain a set of horizontally sliced ​​radar images; Each horizontal slice image in the set of horizontal slice images is subjected to layer detection and annotation using a preset target detection model to obtain the target horizontal slice image and the annotation text file; Based on the layer annotation text in the annotation text file, determine the layer coordinates of the corresponding position in the target horizontal slice image; The layer coordinates of the aforementioned layer points are spliced ​​together to obtain the layer line of the railway subgrade.

[0009] In some embodiments, determining the layer coordinates corresponding to the location of the target horizontal slice image based on the layer annotation text in the annotation text file includes: Traverse the annotation text file. If the corresponding layer annotation text is found in the annotation text file, read the layer annotation text to obtain the layer coordinates of the corresponding position in the horizontal slice image of the radar image. If the corresponding layer label text is not found in the label text file, the layer coordinates of the corresponding position in the horizontal slice of the radar image are set to zero.

[0010] In some embodiments, calculating the target thickness of each substrate based on the preset dielectric constant of each substrate includes: Determine the dielectric constant of each subgrade and the radar data of the target railway subgrade; Establish the relationship between the dielectric constant, the radar data, and the target thickness corresponding to each layer; The target thickness of each base layer is calculated based on the given formula.

[0011] Secondly, this application provides a system for locating the depth of roadbed defects, characterized in that the system comprises: The image preprocessing module is used to acquire radar images corresponding to roadbed defects and perform image preprocessing on the radar images to obtain target radar images; The layer identification module is used to perform layer identification on the target radar image to obtain the various base layers of the railway subgrade in the target radar image; The target thickness calculation module is used to calculate the target thickness of each substrate according to the preset dielectric constant of each substrate. The disease depth calculation module is used to calculate the depth of subgrade disease corresponding to each road base layer based on the target thickness, so as to obtain the subgrade disease depth.

[0012] Thirdly, this application provides a computer device including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method described above.

[0013] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0014] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the methods described above.

[0015] This invention utilizes ground-penetrating radar (GPR) technology to detect defects in railway subgrades, acquiring corresponding radar images. Image preprocessing effectively removes noise and enhances contrast, improving image quality and subsequent target identification accuracy. Lateral segmentation of the target radar image yields multiple lateral slices, which are then labeled using a target detection model to obtain the coordinates of each subgrade layer location. This improves the accuracy and efficiency of layer identification. The coordinates of each layer location are then stitched together and visualized to obtain the subgrade layer lines of the target railway subgrade, achieving non-destructive identification and ensuring the integrity and reliability of railway subgrade identification. Precise calculation of the subgrade defect depth accurately determines the internal defect situation, further improving defect detection accuracy and enabling precise location of subgrade defects, providing a reliable basis for subsequent treatment. Attached Figure Description

[0016] The present application will be described in more detail below based on embodiments and with reference to the accompanying drawings: Figure 1 A flowchart illustrating a method for locating the depth of roadbed defects, provided in an embodiment of this application; Figure 2 A schematic diagram of the functional modules of a roadbed distress depth positioning system provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device for locating the depth of roadbed defects, provided in an embodiment of this application.

[0017] In the accompanying drawings, the same parts are referred to by the same reference numerals, and the drawings are not drawn to scale. Detailed Implementation

[0018] To enable those skilled in the art to better understand the technical solutions of this application, and to fully understand and implement the process of how this application uses technical means to solve technical problems and achieve corresponding technical effects, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, not all of them. The embodiments of this application and the various features within them can be combined with each other without conflict, and the resulting technical solutions are all within the protection scope of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of this application.

[0019] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0020] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0021] This application provides a method for locating the depth of roadbed defects. The execution subject of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the system provided in this application: a server, a terminal, etc. In other words, the method for locating the depth of roadbed defects can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks, and big data and artificial intelligence platforms.

[0022] Example 1 Figure 1 A flowchart illustrating a method for locating the depth of roadbed defects provided in this application embodiment is shown below. Figure 1 As shown, a method for locating the depth of roadbed defects includes: S1. Obtain radar images corresponding to roadbed defects, and perform image preprocessing on the radar images to obtain target radar images.

[0023] In this embodiment of the invention, the image preprocessing includes median filtering, geometric correction, and image enhancement of the radar image, thereby improving image quality and the accuracy of subsequent target recognition.

[0024] In detail, this invention uses ground-penetrating radar (GPR) technology to detect defects in railway subgrades, thereby obtaining radar images corresponding to the defects. The GPR technology refers to the fact that when a GPR is working, it emits electromagnetic waves into the ground. Electromagnetic waves propagate in different ways in different media and are reflected at the interface between different media. The radar's receiving antenna is used to receive the reflected signals, and then the radar signals are processed to obtain radar images.

[0025] In this embodiment of the invention, the step of performing image preprocessing on the radar image to obtain the target radar image includes: The radar image is subjected to noise removal using Gaussian filtering to obtain a denoised radar image; The denoised radar image is then enhanced with texture to obtain the target radar image.

[0026] In detail, this invention improves image quality by removing noise through image preprocessing, specifically Gaussian filtering; it enhances the texture of the denoised radar image set by adjusting the radar image acquisition parameters and imaging geometry, including sampling rate and sampling interval. Setting a higher sampling rate and a smaller sampling interval allows for the capture of more image features from the radar image, ensuring correct image acquisition. Correcting image distortion based on imaging geometry helps improve image accuracy and further enhances the texture enhancement effect; by adjusting parameters such as contrast and brightness of the radar image, target features in the image are enhanced to improve the accuracy of subsequent target recognition; and by performing image preprocessing on the radar image, a target radar image is obtained.

[0027] In this embodiment of the invention, the step of removing noise from the radar image using Gaussian filtering to obtain a denoised radar image includes: Obtain the preset Gaussian kernel size and standard deviation; Generate a Gaussian kernel matrix based on the size and standard deviation of the Gaussian kernel; The edge pixels of the radar image are filled according to the Gaussian kernel matrix to obtain a filled radar image; The Gaussian kernel and the filled radar image are convolved to obtain several pixel values; The aforementioned pixel values ​​are aggregated into a denoised radar image.

[0028] In detail, Gaussian filtering is used to denoise the radar image. The size of the Gaussian kernel depends on the desired smoothness of the denoised radar image and the size of the radar image. The standard deviation of the Gaussian kernel affects the smoothness of the denoised radar image; the larger the standard deviation of the Gaussian kernel, the better the smoothing effect. Before applying Gaussian filtering to the radar image, the edge pixels of the radar image need to be filled to ensure that the radar image has complete edge pixels, so that the Gaussian kernel can accurately cover all parts of the radar image.

[0029] In detail, the central element of the Gaussian kernel matrix is ​​matched with the central pixel of the filled dataset, the Gaussian kernel matrix and the corresponding pixels in the filled radar image are weighted and averaged, and finally the weighted average results are added together to obtain the denoised radar image set.

[0030] Furthermore, the target radar image is a two-dimensional matrix grayscale image, with each column representing a radar data line. The amplitude of each radar data line is reflected in the image as the grayscale value of pixels at different depths. The x-axis of the radar image represents the mileage direction of the route, and the y-axis represents the direction from the surface to the depths. Each column of radar data essentially consists of 512 or 1024 sampling points. Each sampling point refers to the precise location of the radar image within the acquisition area. During image acquisition, the points selected when converting continuous radar images into discrete pixels are used. The sampling points correspond to the time window set during image acquisition. The radar data includes the speed of light in a vacuum, the dielectric constant of each substrate, the round-trip time of electromagnetic waves from transmission to reception, the sampling point position vector of the acquired radar image, the preset zero-line sampling point position vector, and the total number of sampling points contained in each radar data line.

[0031] In this embodiment of the invention, ground-penetrating radar technology is used to detect defects in railway subgrade to obtain radar images corresponding to the defects. The radar images are then preprocessed to effectively remove noise and enhance contrast, thereby improving the quality of the radar images and increasing the accuracy of subsequent target identification.

[0032] S2. Perform layer identification on the target radar image to obtain the base layers of the railway subgrade in the target radar image.

[0033] In this embodiment of the invention, the various base layers refer to the possible layer structures along the depth direction for railway subgrade testing: air-track interface (this layer is generally called the zero line), track-subgrade surface interface, and subgrade surface-subgrade bottom layer interface.

[0034] In this embodiment of the invention, the step of performing layer-by-layer identification on the radar image to obtain the base layers of the target railway subgrade includes: The radar image is horizontally segmented and sampled to obtain a set of horizontally sliced ​​radar images; Each horizontal slice image in the set of horizontal slice images is subjected to layer detection and annotation using a preset target detection model to obtain the target horizontal slice image and the annotation text file; Based on the layer annotation text in the annotation text file, determine the layer coordinates of the corresponding position in the target horizontal slice image; The layer coordinates of the aforementioned layer points are spliced ​​together to obtain the layer line of the railway subgrade.

[0035] In this embodiment of the invention, the vertical direction of the radar image represents the depth of the railway subgrade, and the horizontal direction of the radar image represents the mileage of the railway subgrade; the annotation text file includes layer annotation text corresponding to multiple horizontal slice images of the annotated radar image.

[0036] In detail, the sampling distance for each slide is determined based on the preset number of sampling points and the preset length ratio of the transverse slice image. Based on this determined sliding distance, the radar image is transversely slid along the image to perform transverse segmentation sampling, thus obtaining a set of transverse slice images. Each transverse slice image in the set is then subjected to layer detection and annotation using a preset target detection model, resulting in a target transverse slice image and an annotation text file. The annotation text file includes layer annotation text corresponding to multiple transverse slice images of the annotated radar image. The annotation text file contains layer annotation text for the transverse slice images where the target detection model identifies peaks and troughs. The target detection model identifies and marks the positions of peaks and troughs using bounding boxes, thus obtaining the layer annotation text. For example, if no bounding boxes are marked in the output transverse slice image, it indicates that the target detection model has not detected it. Based on the position coordinates of peaks and troughs in the layer annotation text, the layer point coordinates corresponding to the positions in multiple transverse slice images of the radar image are determined. These layer point coordinates are then stitched together to obtain layer lines, thereby obtaining the base layers of the target railway subgrade.

[0037] In this embodiment of the invention, determining the layer coordinates corresponding to the location of the target horizontal slice image based on the layer annotation text in the annotation text file includes: Traverse the annotation text file. If the corresponding layer annotation text is found in the annotation text file, read the layer annotation text to obtain the layer coordinates of the corresponding position in the horizontal slice image of the radar image. If the corresponding layer label text is not found in the label text file, the layer coordinates of the corresponding position in the horizontal slice of the radar image are set to zero.

[0038] In detail, the present invention stores the layer coordinates of the corresponding positions of multiple horizontal slices of the radar image into a text file to obtain the layer line of the railway subgrade. The visualization result of the stitched layer line is shown in the figure.

[0039] In this embodiment of the invention, the target radar image is horizontally segmented and sampled to obtain horizontal slice images of multiple radar images. Then, the target detection model is used to perform layer labeling to obtain the layer point coordinates of each base course, which improves the accuracy and efficiency of layer identification. The layer point coordinates of each layer are stitched together and visualized to obtain the layer lines of each base course of the target railway subgrade, realizing non-destructive identification and ensuring the integrity and reliability of railway subgrade identification.

[0040] S3. Calculate the target thickness of each substrate according to the preset dielectric constant of each substrate.

[0041] In this embodiment of the invention, the dielectric constant refers to the composite electrical property parameter of different layers of media in an electric field. The various base layers of railway subgrade include different media such as track bed, subgrade bed, soil, and rock. Therefore, different media have different dielectric constants in different layers. For example, the dielectric constant of air is about 1.0006, which can be approximated as the dielectric constant of vacuum; while the dielectric constants of media such as water, rock, and soil are much greater than those of air.

[0042] In this embodiment of the invention, calculating the target thickness of each substrate based on a preset dielectric constant includes: Determine the dielectric constant of each subgrade and the radar data of the target railway subgrade; Establish the relationship between the dielectric constant, the radar data, and the target thickness corresponding to each layer; The target thickness of each base layer is calculated based on the given formula.

[0043] Specifically, the relation is as follows:

[0044] in, The speed of light in a vacuum, The dielectric constant of each of the aforementioned base layers is... The round-trip time of the electromagnetic wave from transmission to reception in the radar data. This refers to the location vector of the corresponding sampling points of each layer in the acquired radar image. This refers to the preset zero-line sampling point position vector in the acquired radar image. This refers to the total number of sampling points contained in each channel of the radar data. The target thickness of each of the aforementioned base layers.

[0045] In this embodiment of the invention, by presetting the dielectric constant and calculating the target thickness of the corresponding roadbed, error interference can be reduced, significantly improving the accuracy and reliability of the detection results, and laying the groundwork for subsequent calculation of the depth of roadbed defects.

[0046] S4. Calculate the depth of subgrade defects corresponding to each subgrade layer based on the target thickness to obtain the subgrade defect depth.

[0047] In this embodiment of the invention, the target thickness of the subgrade is calculated by determining the dielectric constant of the subgrade layer where the subgrade disease is located and by obtaining relevant subgrade radar data through ground penetrating radar technology. Then, the depth of the corresponding subgrade disease is calculated to obtain the subgrade disease depth.

[0048] In this embodiment of the invention, the dielectric constant of the subgrade layer where the subgrade defect is located and the radar data of the target railway subgrade are determined, and the relationship between the dielectric constant and the radar data and the target thickness of each subgrade layer is established, thereby obtaining the target thickness of each subgrade layer. Then, the depth of the subgrade defect corresponding to each subgrade layer is obtained by addition calculation of the target thickness of each subgrade layer.

[0049] Specifically, the relation is as follows:

[0050]

[0051] in, The speed of light in a vacuum, Let be the dielectric constant of the subgrade layer where the subgrade defects are located. The round-trip time of the electromagnetic wave from transmission to reception in the radar data. This refers to the location vector of the corresponding sampling points of each layer in the acquired radar image. This is the location vector of the sampling points on the surface of the subgrade layer where the subgrade defects are located. This refers to the total number of sampling points contained in each channel of the radar data. The target thickness of each of the aforementioned base layers, The target thickness of the subgrade layer where the subgrade defects are located. The depth of the roadbed defects is described.

[0052] In this embodiment of the invention, by accurately calculating the depth of roadbed defects, the condition of defects inside the roadbed can be accurately determined, thereby improving the accuracy of defect detection, achieving precise location of roadbed defects, and providing a reliable basis for subsequent treatment.

[0053] This invention utilizes ground-penetrating radar (GPR) technology to detect defects in railway subgrades, acquiring corresponding radar images. Image preprocessing effectively removes noise and enhances contrast, improving image quality and subsequent target identification accuracy. Lateral segmentation of the target radar image yields multiple lateral slices, which are then labeled using a target detection model to obtain the coordinates of each subgrade layer location. This improves the accuracy and efficiency of layer identification. The coordinates of each layer location are then stitched together and visualized to obtain the subgrade layer lines of the target railway subgrade, achieving non-destructive identification and ensuring the integrity and reliability of railway subgrade identification. Precise calculation of the subgrade defect depth accurately determines the internal defect situation, further improving defect detection accuracy and enabling precise location of subgrade defects, providing a reliable basis for subsequent treatment.

[0054] Example 2 like Figure 2 The diagram shown is a functional block diagram of a roadbed defect depth positioning system 100 provided for this embodiment.

[0055] The roadbed defect depth positioning system 100 of this invention can be installed in an electronic device. Depending on the functions implemented, the roadbed defect depth positioning system 100 may include an image preprocessing module 101, a layer identification module 102, a target thickness calculation module 103, and a defect depth calculation module 104. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and which are stored in the memory of the electronic device.

[0056] In this embodiment, the functions of each module / unit are as follows: The image preprocessing module is used to acquire radar images corresponding to roadbed defects and perform image preprocessing on the radar images to obtain target radar images; The layer identification module is used to perform layer identification on the target radar image to obtain the various base layers of the railway subgrade in the target radar image; The target thickness calculation module is used to calculate the target thickness of each substrate according to the preset dielectric constant of each substrate. The disease depth calculation module is used to calculate the depth of subgrade disease corresponding to each road base layer based on the target thickness, so as to obtain the subgrade disease depth.

[0057] Example 3 Figure 3 This is a schematic diagram of the structure of an electronic device for locating the depth of roadbed defects, provided in an embodiment of this application.

[0058] Based on the above embodiments, this embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method described in the above embodiments.

[0059] In some embodiments of this example, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements the steps of the method described in the above embodiments.

[0060] In some embodiments of this example, a computer program product is provided, including a computer program, characterized in that the computer program, when executed by a processor, implements the steps of the method described in the above embodiments.

[0061] The processor may include, but is not limited to, one or more processors or microprocessors. Each processor may be implemented as an Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), controller, microcontroller, microprocessor, or other electronic component, for executing the methods in the above embodiments.

[0062] Computer-readable storage media can be implemented by any type of volatile or non-volatile storage device or a combination thereof, including but not limited to, random access memory (RAM), read-only memory (ROM), flash memory, EPROM memory, EEPROM memory, registers, and computer storage media (e.g., hard disks, floppy disks, solid-state drives, removable disks, CD-ROMs, DVD-ROMs, Blu-ray discs, etc.).

[0063] Computer-readable storage media may also store at least one computer-executable program, such as computer-readable instructions. Computer-readable storage media include, but are not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Computer-readable storage media may include, for example, read-only memory (ROM), hard disk, flash memory, etc. For example, a non-transitory computer-readable storage medium may be connected to a computing device such as a computer, and then, when the computing device executes the computer-readable instructions stored on the computer-readable storage medium, the various methods described above can be performed.

[0064] In addition, the computer device may include (but is not limited to) a data bus, an input / output (I / O) bus, a display, and input / output devices (e.g., keyboard, mouse, speakers, etc.).

[0065] The processor can communicate with external devices via the communication interface of the I / O bus through wired or wireless networks.

[0066] In one embodiment, the at least one computer-executable instruction may also be compiled into or comprise a software product / computer program product, wherein one or more computer-executable instructions are executed by a processor to perform the steps of the various functions and / or methods in the embodiments described herein.

[0067] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can also be implemented in other ways. The system embodiments described above are merely illustrative. For example, 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 various 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. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked 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. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, 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.

[0068] It should be noted that, in this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element limited by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0069] Although the embodiments disclosed in this application are as described above, the above content is merely for the purpose of facilitating understanding of this application and is not intended to limit this application. Any person skilled in the art to which this application pertains may make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in this application; however, the scope of patent protection of this application shall still be determined by the scope defined in the appended claims.

Claims

1. A method for locating the depth of roadbed defects, characterized in that, include: Acquire radar images corresponding to roadbed defects, and perform image preprocessing on the radar images to obtain target radar images; Layer identification is performed on the target radar image to obtain the base layers of the railway subgrade in the target radar image; The target thickness of each substrate is calculated based on the preset dielectric constant of each substrate. The depth of subgrade defects is calculated based on the target thickness to obtain the subgrade defect depth.

2. The method according to claim 1, characterized in that, The step of preprocessing the radar image to obtain the target radar image includes: The radar image is subjected to noise removal using Gaussian filtering to obtain a denoised radar image; The denoised radar image is then enhanced with texture to obtain the target radar image.

3. The method according to claim 2, characterized in that, The step of removing noise from the radar image using Gaussian filtering to obtain a denoised radar image includes: Obtain the preset Gaussian kernel size and standard deviation; Generate a Gaussian kernel matrix based on the size and standard deviation of the Gaussian kernel; The edge pixels of the radar image are filled according to the Gaussian kernel matrix to obtain a filled radar image; The Gaussian kernel and the filled radar image are convolved to obtain several pixel values; The aforementioned pixel values ​​are aggregated into a denoised radar image.

4. The method according to claim 1, characterized in that, The process of performing layer-by-layer identification on the radar image to obtain the various base layers of the target railway subgrade includes: The radar image is horizontally segmented and sampled to obtain a set of horizontally sliced ​​radar images; Each horizontal slice image in the set of horizontal slice images is subjected to layer detection and annotation using a preset target detection model to obtain the target horizontal slice image and the annotation text file; Based on the layer annotation text in the annotation text file, determine the layer coordinates of the corresponding position in the target horizontal slice image; The layer coordinates of the aforementioned layer points are spliced ​​together to obtain the layer line of the railway subgrade.

5. The method according to claim 4, characterized in that, The step of determining the layer coordinates corresponding to the location of the target horizontal slice image based on the layer annotation text in the annotation text file includes: Traverse the annotation text file. If the layer annotation text corresponding to the target horizontal slice image is found in the annotation text file, then read the layer annotation text to obtain the layer coordinates of the corresponding position of the horizontal slice image. If the layer label text corresponding to the target horizontal slice image is not found in the label text file, then the layer coordinates of the corresponding position in the horizontal slice image of the radar image are set to zero.

6. The method according to claim 1, characterized in that, The step of calculating the target thickness of each substrate based on the preset dielectric constant of each substrate includes: Determine the dielectric constant of each subgrade and the radar data of the target railway subgrade; Establish the relationship between the dielectric constant, the radar data, and the target thickness corresponding to each layer; The target thickness of each base layer is calculated based on the given formula.

7. A system for locating the depth of roadbed defects, characterized in that, The system includes: The image preprocessing module is used to acquire radar images corresponding to roadbed defects and perform image preprocessing on the radar images to obtain target radar images; The layer identification module is used to perform layer identification on the target radar image to obtain the various base layers of the railway subgrade in the target radar image; The target thickness calculation module is used to calculate the target thickness of each substrate according to the preset dielectric constant of each substrate. The disease depth calculation module is used to calculate the depth of subgrade disease corresponding to each road base layer based on the target thickness, so as to obtain the subgrade disease depth.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the method according to any one of claims 1 to 6.

Images