Ballast bed thickness identification method and device based on geological radar data

Abstract

The invention discloses a ballast bed thickness identification method and device based on geological radar data. The method comprises the following steps: acquiring geological radar data of a railroad bed; the geological radar data comprises waveform amplitude data of different horizontal positions and different geological horizon positions of a railroad bed; based on a preset algorithm, performing feature extraction on the geological radar data to obtain an incident surface layer bit line and a reflecting surface layer bit line; the preset algorithm is used for extracting an incident plane horizon bit line based on the absolute value of waveform amplitude data of geological radar data and the light volatility characteristic of an incident plane; extracting a reflecting surface horizon bit line based on the signal similarity of geological radar data and standard Ricker wavelets; the standard Ricker wavelet frequency is the same as the radar frequency; and determining the thickness of the ballast bed layer by utilizing the depth difference between the bit line of the incident surface layer and the bit line of the reflecting surface layer. The accuracy and efficiency of horizon information extraction can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of railway roadbed detection, and in particular to a roadbed thickness identification method and device based on geological radar data. Background Art

[0002] This section is intended to provide a background or context to the embodiments of the invention recited in the claims. No admission is made that the description herein is prior art by inclusion in this section.

[0003] Problems such as deformation, settlement, and water accumulation in the railway subgrade will affect the safe operation of the railway. Regular inspection of the railway subgrade status is one of the infrastructure maintenance tasks. At present, the detection of railway subgrade mainly adopts geological radar technology, which has the advantages of fast, non-destructive and continuous. During the detection data acquisition process, the geological radar emits electromagnetic waves to the underground medium and records the waveform reflected from the medium. The data interpretation work analyzes the characteristics of the reflected waves to obtain the stratigraphic information of the underground medium, thereby realizing the judgment of the deformation, settlement, water accumulation and other problems of the subgrade. Therefore, the data analysis of railway subgrade detection based on geological radar focuses on extracting the stratigraphic information of the subgrade and identifying the thickness of the roadbed layer.

[0004] At present, the analysis of detection data usually needs to be done manually, and the reflection layers in the data are manually identified and the layer lines are manually drawn. The information such as layer depth and layer signal strength is used to determine defects such as water content and subsidence. Therefore, there are still some challenges and limitations in practical applications.

[0005] The current technical limitations are: First, geological radar often requires professionals to perform complex data interpretation and analysis. The amount of data obtained is huge and complex. How to efficiently and accurately extract and analyze the stratigraphic information in the data has become a difficult problem that needs to be solved. Secondly, the existing data processing methods usually rely on the experience and technology of professionals and lack standardized and automated processes. This not only increases the cost of detection, but may also lead to inconsistencies in data interpretation, which in turn affects the assessment of the overall condition of the railway subgrade. In addition, with the continuous increase in railway transportation demand, the increase in detection frequency and the sharp increase in data volume, traditional manual analysis methods have been unable to meet the needs of modern railway safe operation. Summary of the invention

[0006] The embodiment of the present invention provides a method for identifying the thickness of a roadbed based on geological radar data, which is used to improve the accuracy and efficiency of extracting layer information and reduce labor costs. The method includes:

[0007] Acquire geological radar data of the railway embankment; the geological radar data includes waveform amplitude data of different horizontal positions and different geological layers of the railway embankment;

[0008] Based on the preset algorithm, feature extraction is performed on the geological radar data to obtain the incident surface layer line and the reflection surface layer line; the preset algorithm is used to: extract the incident surface layer line based on the absolute value of the waveform amplitude data of the geological radar data and the light wave characteristics of the incident surface; extract the reflection surface layer line based on the signal similarity between the geological radar data and the standard Ricker wavelet; the frequency of the standard Ricker wavelet is the same as the radar frequency;

[0009] The thickness of the roadbed layer is determined by using the depth difference between the incident surface layer line and the reflected surface layer line.

[0010] The embodiment of the present invention further provides a device for identifying the thickness of a roadbed based on geological radar data, the device comprising:

[0011] A data acquisition module is used to acquire geological radar data of the railway embankment; the geological radar data includes waveform amplitude data of different horizontal positions and different geological layers of the railway embankment;

[0012] The feature extraction module is used to extract features from the geological radar data based on a preset algorithm to obtain the incident surface layer line and the reflection surface layer line; the preset algorithm is used to: extract the incident surface layer line based on the absolute value of the waveform amplitude data of the geological radar data and the light wave characteristics of the incident surface; extract the reflection surface layer line based on the signal similarity between the geological radar data and the standard Ricker wavelet; the frequency of the standard Ricker wavelet is the same as the radar frequency;

[0013] The thickness determination module is used to determine the thickness of the roadbed layer by using the depth difference between the incident surface layer position line and the reflected surface layer position line.

[0014] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for identifying the thickness of a roadbed based on geological radar data when executing the computer program.

[0015] An embodiment of the present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned roadbed thickness identification method based on geological radar data is implemented.

[0016] An embodiment of the present invention also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the above-mentioned roadbed thickness identification method based on geological radar data.

[0017] The geological radar data acquired in the embodiment of the present invention includes waveform amplitude data of different horizontal positions and different geological layers of the railway roadbed. Based on the preset algorithm, the geological radar data is feature extracted to obtain the incident surface layer line and the reflection surface layer line. The depth difference between the incident surface layer line and the reflection surface layer line is used to determine the thickness of the roadbed layer. Compared with the prior art, the embodiment of the present invention can greatly reduce the time of manual analysis, reduce labor costs, and quickly obtain detection results. In addition, the embodiment of the present invention constructs a preset algorithm, which extracts the incident surface layer line based on the absolute value of the waveform amplitude data of the geological radar data and the light wave wave characteristics of the incident surface; extracts the reflection surface layer line based on the signal similarity between the geological radar data and the standard Ricker wavelet; the standard Ricker wavelet frequency is the same as the radar frequency, which can improve the accuracy and efficiency of high-level information extraction, reduce data errors caused by human factors, and improve the reliability of detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0019] Figure 1 It is a schematic diagram of the process of the roadbed thickness identification method based on geological radar data in an embodiment of the present invention;

[0020] Figure 2 The original geological radar data of the railway embankment obtained in the embodiment of the present invention;

[0021] Figure 3 This is a schematic diagram of pre-processed data in an embodiment of the present invention;

[0022] Figure 4 Schematic diagram of incident surface layer bit lines extracted in an embodiment of the present invention;

[0023] Figure 5 Schematic diagram of the reflection surface layer line extracted in an embodiment of the present invention;

[0024] Figure 6 Schematic diagram of thickness results extracted in an embodiment of the present invention;

[0025] Figure 7 It is a schematic diagram of a roadbed thickness identification device based on geological radar data in an embodiment of the present invention. DETAILED DESCRIPTION

[0026] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0027] The embodiment of the present invention provides a railway roadbed detection method based on geological radar technology, specifically a technical solution for automatically extracting layer information from railway roadbed geological radar detection data and giving layer thickness. The technical solution aims to improve the efficiency and accuracy of railway roadbed detection and ensure the safe operation of the railway.

[0028] Figure 1 FIG. 1 is a flow chart of a method for identifying the thickness of a roadbed based on geological radar data in an embodiment of the present invention. Figure 1 As shown, the method includes:

[0029] Step 101, obtaining geological radar data of the railway embankment; the geological radar data includes waveform amplitude data of different horizontal positions and different geological layers of the railway embankment;

[0030] Step 102: Based on a preset algorithm, feature extraction is performed on the geological radar data to obtain incident surface layer lines and reflection surface layer lines; the preset algorithm is used to: extract the incident surface layer lines based on the absolute value of the waveform amplitude data of the geological radar data and the light wave characteristics of the incident surface; extract the reflection surface layer lines based on the signal similarity between the geological radar data and the standard Ricker wavelet; the frequency of the standard Ricker wavelet is the same as the radar frequency;

[0031] Step 103: Determine the thickness of the ballast layer by using the depth difference between the incident surface layer line and the reflected surface layer line.

[0032] Below Figure 1 The roadbed thickness identification method based on geological radar data is explained in detail.

[0033] First, geological radar data of the railway embankment is obtained; the geological radar data includes waveform amplitude data of different horizontal positions and different geological layers of the railway embankment.

[0034] During the implementation, high-resolution geological radar equipment is used to conduct a comprehensive scan of the railway subgrade, and the reflected signal data inside the subgrade is obtained in real time. These data contain geological information, physical characteristics and potential defect characteristics of different layers.

[0035] For example, the radar scans horizontally along the railway embankment to obtain data one by one. All the data are combined, including horizontal and vertical waveform amplitude data, for example, expressed in the form of a data matrix.

[0036] Figure 2The geological radar raw data of the railway roadbed obtained in the embodiment of the present invention can exist in the form of a grayscale image, including multiple rows and columns of data, and the data of each point is a waveform amplitude value.

[0037] In one embodiment, in order to improve data quality and facilitate subsequent analysis, before extracting features from the geological radar data, the method may also include: preprocessing the geological radar data of the railway roadbed to obtain preprocessed data; the preprocessing includes zero drift removal, background denoising, bandpass filtering, normalization, one or any combination thereof.

[0038] For example, preprocessing the geological radar data of the railway embankment may include: performing the following zero drift removal on the geological radar data of the railway embankment:

[0039] For each column of data x in the M-row and N-column geological radar data matrix n , the zero drift value Z of the column can be obtained by calculating the mean of the 100 points at the end of the column data, taking the i-th column data as an example:

[0040] Using the data of the last 100 points of the column, the zero drift value Z of the column is calculated according to the following formula:

[0041]

[0042] Among them, x i is the geological radar data of the i-th point;

[0043] Subtract the zero drift value Z from all elements of the column data to obtain the data after removing the zero drift.

[0044] For example, preprocessing the geological radar data of the railway embankment may include: performing the following background denoising on the geological radar data of the railway embankment:

[0045] For a geological radar data matrix with M rows and N columns, a first sliding window with M rows and w columns is set in the horizontal direction. For each column of geological radar data in the first sliding window, the background denoised data is calculated according to the following formula:

[0046]

[0047] Among them, x k is the kth column data in the sliding window before denoising, y i is the denoised data of the i-th column, x i is the i-th column data before denoising;

[0048] That is, for each column of geological radar data in the first sliding window, the mean of the w columns of data centered on the column of data is subtracted to obtain the data after background denoising.

[0049] In one embodiment, the length w of the first sliding window is proportional to the total number of measuring lines between the two sleepers.

[0050] Preferably, the length column number w of the first sliding window is a preset multiple of the total number of measuring lines contained between the two sleepers.

[0051] For example, the length of w is 10 times the total number of survey lines between the two sleepers. When the survey line density is such that there are 20 data lines between the two sleepers, w is 200 and the sliding step length is 50 accordingly.

[0052] For example, when bandpass filtering is performed on geological radar data, the lower frequency limit is 1 / 2 of the antenna main frequency, and the upper frequency limit is 2 times of the antenna main frequency. For example, the antenna main frequency is 400MHz, the lower frequency limit is 200MHz, and the upper frequency limit is 800MHz.

[0053] During normalization: First, calculate the absolute value of the data matrix after bandpass filtering and obtain the maximum value M X , and then all the data with absolute values ​​greater than 0.8×M X The data is set to 0.8×M X Finally, calculate the correlation between each point in the data matrix and 0.8M X The normalized result is obtained by taking the ratio of .

[0054] During implementation, the geological radar data of the railway subgrade can be subjected to zero drift removal, background denoising, bandpass filtering, and normalization in sequence to obtain preprocessed data.

[0055] Furthermore, extracting features from geological radar data may include: extracting features from preprocessed data.

[0056] Figure 3 This is a schematic diagram of preprocessed data in an embodiment of the present invention. Figure 3 Shows the Figure 2 The original data is the preprocessed data.

[0057] Afterwards, based on the preset algorithm, feature extraction is performed on the geological radar data to obtain the incident surface layer line and the reflection surface layer line. The preset algorithm is used to: extract the incident surface layer line based on the absolute value of the waveform amplitude data of the geological radar data and the light wave characteristics of the incident surface; extract the reflection surface layer line based on the signal similarity between the geological radar data and the standard Ricker wavelet; the frequency of the standard Ricker wavelet is the same as the radar frequency.

[0058] Specifically, the image processing technology is used to extract features from the preprocessed radar data, and the layer information is automatically identified and extracted. The process includes two steps: incident surface extraction and reflection surface extraction.

[0059] In one embodiment, based on a preset algorithm, feature extraction is performed on geological radar data to obtain incident surface layer lines, which may include the following three steps.

[0060] The first step is horizontal smoothing:

[0061] The geological radar data matrix with M rows and N columns is smoothed in the horizontal direction to obtain a smoothed data matrix.

[0062] For example, perform horizontal Gaussian smoothing of a window of specified length on the data matrix to remove horizontal jitter.

[0063] The second step is to preliminarily pick the incident surface:

[0064] The element-wise absolute value calculation is performed on the smoothed data matrix, and the maximum value of each column is marked as the incident surface point vector of 1 row and M columns.

[0065] The third step is to organize the incident surface:

[0066] First, set the first sliding window with M rows and w columns, and calculate the volatility of the incident surface point vector in each column in the first sliding window according to the following formula:

[0067]

[0068] Among them, E i represents the volatility of the incident surface point vector in the i-th column, M i represents the value of the incident surface point vector of the i-th column, represents the mean value of the position vectors of all incident surface points in the first sliding window;

[0069] When Ei of the i-th column>0.1, the value of the incident surface point vector in the i-th column is modified to the value of the incident surface point vector in the i-1-th column.

[0070] After the above three steps, the incident surface layer line is obtained.

[0071] After obtaining the incident surface layer line, in one embodiment, based on a preset algorithm, feature extraction is performed on the geological radar data to obtain the reflection surface layer line, which may include:

[0072] For the geological radar data matrix with M rows and N columns, the part above the incident surface layer line is set to 0, and the part below the incident surface layer line is normalized to between -1 and +1;

[0073] The second sliding window is set in the horizontal direction, and the Pearson correlation coefficient between the standard Ricker wavelet and the geological radar data waveform in the second sliding window is calculated. The points with a Pearson correlation coefficient greater than the set threshold are marked as reflection surface points. After the second sliding window slides from the first column to the last column, the reflection surface layer line is obtained.

[0074] For example, first, the part of the M-row and N-column matrix whose row number is less than the incident surface point is set to 0. Next, a window with a fixed length of L is used to normalize the waveform in the window to between -1 and +1 below the incident surface point. At the same time, a standard Ricker wavelet with the same radar frequency is used to calculate the Pearson correlation coefficient with the waveform in the window. Finally, the L window is slid M minus L times with a step length of 1, and the first value greater than 0.7 in each column is recorded as the reflection surface vector.

[0075] Figure 4 is a schematic diagram of incident surface layer bit lines extracted in an embodiment of the present invention, Figure 4 The red mark shows the Figure 3 Based on the extracted incident surface layer line, Figure 5 is a schematic diagram of the reflection surface layer line extracted in an embodiment of the present invention, Figure 5 The red mark shows the Figure 4 Based on the extracted reflection surface layer lines.

[0076] Afterwards, the thickness of the ballast layer can be determined by the depth difference between the incident surface layer line and the reflected surface layer line.

[0077] In order to further improve the reliability of the extracted thickness information, in a preferred embodiment, before determining the thickness of the ballast layer by using the depth difference between the incident surface layer line and the reflected surface layer line, the method may further include:

[0078] By using the correlation between the incident surface layer bit line and the reflection surface layer bit line, the mirror reflection interference in the reflection surface layer bit line is removed;

[0079] And / or, removing invalid points in the incident surface layer bit line and the reflection surface layer bit line; the invalid points have data abnormalities; the data abnormalities are, for example, smaller or larger data.

[0080] During implementation, based on the incident surface layer line and the reflection surface layer line, the layer arrangement and thickness extraction are carried out.

[0081] (1) Mirror reflection removal: Use a window w with the same length as before to calculate the Pearson correlation coefficient between the incident surface vector and the reflection surface vector in the window. For windows with a coefficient greater than 0.8, they are considered to be mirror reflection interference. The part of the M-row and N-column matrix with a row number less than the reflection surface point is set to 0, and the reflection surface extraction step is performed again.

[0082] (2) Removal of invalid information: Take the absolute value of the numerical value at the reflection surface layer in the M-row and N-column matrix, and regard the value less than 0.2 as invalid data and set it to blank.

[0083] Finally, the depth difference between the incident surface and the reflecting surface is used to obtain the roadbed thickness information.

[0084] After the above three steps, the roadbed thickness information is obtained.

[0085] Figure 6 Schematic diagram of the thickness results extracted in the embodiment of the present invention, such as Figure 6 As shown, the thickness of the roadbed layer at different measuring points is displayed.

[0086] In the embodiment of the present invention, the geological radar data is arranged according to different horizontal positions and directions of different geological layers, and processed based on a preset algorithm to extract the thickness of the roadbed layer. A large amount of manual detection work is avoided, labor costs are reduced, detection efficiency is improved, and manual detection errors are avoided. At the same time, in the embodiment of the present invention, zero drift removal, background denoising, bandpass filtering, and normalization processing are performed on the geological radar data to improve the accuracy of data calculation. An algorithm flow for extracting the incident surface and reflection surface layer information and thickness from the geological radar data is proposed, especially in complex geological environments, which has better adaptability and higher accuracy. The embodiment of the present invention also removes mirror reflection interference through a correlation algorithm, thereby ensuring the validity of the data and ensuring the accurate extraction of the reflection surface information.

[0087] The present invention also provides a device for identifying the thickness of a roadbed based on geological radar data, as described in the following embodiments. Since the principle of the device to solve the problem is similar to that of the method for identifying the thickness of a roadbed based on geological radar data, the implementation of the device can refer to the implementation of the method for identifying the thickness of a roadbed based on geological radar data, and the repeated parts will not be repeated.

[0088] Figure 7 FIG. 1 is a schematic diagram of a device for identifying the thickness of a roadbed based on geological radar data in an embodiment of the present invention. Figure 7 As shown, the device 700 includes:

[0089] The data acquisition module 701 is used to acquire geological radar data of the railway embankment; the geological radar data includes waveform amplitude data of different horizontal positions and different geological layers of the railway embankment;

[0090] The feature extraction module 702 is used to extract features from the geological radar data based on a preset algorithm to obtain the incident surface layer line and the reflection surface layer line; the preset algorithm is used to: extract the incident surface layer line based on the absolute value of the waveform amplitude data of the geological radar data and the light wave wave characteristics of the incident surface; extract the reflection surface layer line based on the signal similarity between the geological radar data and the standard Ricker wavelet; the frequency of the standard Ricker wavelet is the same as the radar frequency;

[0091] The thickness determination module 703 is used to determine the thickness of the roadbed layer by using the depth difference between the incident surface layer line and the reflected surface layer line.

[0092] In one embodiment, the apparatus 700 further includes a pre-processing module.

[0093] A preprocessing module is used to preprocess the geological radar data of the railway roadbed before the data acquisition module 701 extracts features from the geological radar data to obtain preprocessed data; the preprocessing includes one or any combination of zero drift removal, background denoising, bandpass filtering, and normalization;

[0094] The feature extraction module 702 is specifically used to extract features from the preprocessed data.

[0095] In one embodiment, the preprocessing module is specifically used to perform the following zero drift removal on the geological radar data of the railway embankment:

[0096] For each column of data x in the M-row and N-column geological radar data matrix n :

[0097] Using the data of the last 100 points of the column, the zero drift value Z of the column is calculated according to the following formula:

[0098]

[0099] Among them, x i is the geological radar data of the i-th point in this column of data;

[0100] Subtract the zero drift value Z from all elements of the column data to obtain the data after removing the zero drift.

[0101] In one embodiment, the preprocessing module is specifically used to perform the following background denoising on the geological radar data of the railway embankment:

[0102] For a geological radar data matrix with M rows and N columns, set the first sliding window with M rows and w columns. For each column of geological radar data in the first sliding window, calculate the background denoised data according to the following formula:

[0103]

[0104] Among them, x k is the mean of the kth column data in the first sliding window, y i is the denoised data of the i-th column, x i is the i-th column data before denoising.

[0105] In one embodiment, the length column number w of the first sliding window is a preset multiple of the total number of survey lines included between the two sleepers.

[0106] In one embodiment, the feature extraction module 702 is specifically used for:

[0107] Perform horizontal smoothing on the geological radar data matrix of M rows and N columns to obtain a smoothed data matrix;

[0108] Perform element-wise absolute value calculation on the smoothed data matrix, and mark the maximum value of each column as the incident surface point vector of 1 row and M columns;

[0109] Set the first sliding window with M rows and w columns, and calculate the volatility of the incident surface point vector in each column in the first sliding window according to the following formula:

[0110]

[0111] Among them, E i represents the volatility of the incident surface point vector in the i-th column, M i represents the value of the incident surface point vector of the i-th column, represents the mean value of the position vectors of all incident surface points in the first sliding window;

[0112] When Ei of the i-th column>0.1, the value of the incident surface point vector in the i-th column is modified to the value of the incident surface point vector in the i-1-th column.

[0113] In one embodiment, the feature extraction module 702 is specifically used for:

[0114] For the geological radar data matrix with M rows and N columns, the part above the incident surface layer line is set to 0, and the part below the incident surface layer line is normalized to between -1 and +1;

[0115] The second sliding window is set in the horizontal direction, and the Pearson correlation coefficient between the standard Ricker wavelet and the geological radar data waveform in the second sliding window is calculated. The points with a Pearson correlation coefficient greater than the set threshold are marked as reflection surface points. After the second sliding window slides from the first column to the last column, the reflection surface layer line is obtained.

[0116] In one embodiment, the apparatus 700 may further include:

[0117] The layer arrangement module is used to use the correlation between the incident surface layer line and the reflection surface layer line to remove the mirror reflection interference in the reflection surface layer line before the thickness determination module 703 uses the depth difference between the incident surface layer line and the reflection surface layer line to determine the thickness of the roadbed layer; and / or, to remove invalid points in the incident surface layer line and the reflection surface layer line; the invalid point data is abnormal.

[0118] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for identifying the thickness of a roadbed based on geological radar data when executing the computer program.

[0119] An embodiment of the present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned roadbed thickness identification method based on geological radar data is implemented.

[0120] An embodiment of the present invention also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the above-mentioned roadbed thickness identification method based on geological radar data.

[0121] The geological radar data acquired in the embodiment of the present invention includes waveform amplitude data of different horizontal positions and different geological layers of the railway roadbed. Based on the preset algorithm, the geological radar data is feature extracted to obtain the incident surface layer line and the reflection surface layer line. The depth difference between the incident surface layer line and the reflection surface layer line is used to determine the thickness of the roadbed layer. Compared with the prior art, the embodiment of the present invention can greatly reduce the time of manual analysis, reduce labor costs, and quickly obtain detection results. In addition, the embodiment of the present invention constructs a preset algorithm, which extracts the incident surface layer line based on the absolute value of the waveform amplitude data of the geological radar data and the light wave wave characteristics of the incident surface; extracts the reflection surface layer line based on the signal similarity between the geological radar data and the standard Ricker wavelet; the standard Ricker wavelet frequency is the same as the radar frequency, which can improve the accuracy and efficiency of high-level information extraction, reduce data errors caused by human factors, and improve the reliability of detection.

[0122] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0123] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0124] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0125] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0126] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for identifying the thickness of a roadbed based on geological radar data, characterized in that: include: Obtaining geological radar data of railway embankments; The geological radar data include waveform amplitude data of different horizontal positions and different geological layers of the railway embankment; Based on the preset algorithm, feature extraction is performed on the geological radar data to obtain the incident surface layer line and the reflection surface layer line; the preset algorithm is used to: extract the incident surface layer line based on the absolute value of the waveform amplitude data of the geological radar data and the light wave characteristics of the incident surface; extract the reflection surface layer line based on the signal similarity between the geological radar data and the standard Ricker wavelet; the frequency of the standard Ricker wavelet is the same as the radar frequency; The thickness of the roadbed layer is determined by using the depth difference between the incident surface layer line and the reflected surface layer line.

2. The method according to claim 1, characterized in that Before extracting features from geological radar data, the following steps are also required: Preprocessing the geological radar data of the railway embankment to obtain preprocessed data; the preprocessing includes one or any combination of zero drift removal, background denoising, bandpass filtering, and normalization; Feature extraction of geological radar data, including: Perform feature extraction on the preprocessed data.

3. The method according to claim 2, characterized in that Preprocessing of the geological radar data of the railway embankment includes: performing the following zero drift removal on the geological radar data of the railway embankment: For each column of data in the M-row and N-column geological radar data matrix: Using the data of the last 100 points of the column, the zero drift value Z of the column is calculated according to the following formula: Among them, x i is the geological radar data of the i-th point in this column of data; Subtract the zero drift value Z from all elements of the column data to obtain the data after removing the zero drift.

4. The method according to claim 2, characterized in that Preprocessing of the geological radar data of the railway embankment includes: background denoising of the geological radar data of the railway embankment as follows: For a geological radar data matrix with M rows and N columns, set the first sliding window with M rows and w columns. For each column of geological radar data in the first sliding window, calculate the background denoised data according to the following formula: Among them, x k is the mean of the kth column data in the first sliding window, y i is the denoised data of the i-th column, x i is the i-th column data before denoising.

5. The method according to claim 4, characterized in that The length column number w of the first sliding window is a preset multiple of the total number of measuring lines contained between the two sleepers.

6. The method according to claim 1, characterized in that Based on the preset algorithm, feature extraction is performed on the geological radar data to obtain the incident surface layer line, including: Perform horizontal smoothing on the geological radar data matrix of M rows and N columns to obtain a smoothed data matrix; Perform element-wise absolute value calculation on the smoothed data matrix, and mark the maximum value of each column as the incident surface point vector of 1 row and M columns; Set the first sliding window with M rows and w columns, and calculate the volatility of the incident surface point vector in each column in the first sliding window according to the following formula: Among them, E i represents the volatility of the incident surface point vector in the i-th column, M i represents the value of the incident surface point vector of the i-th column, represents the mean value of the position vectors of all incident surface points in the first sliding window; When Ei of the i-th column>0.1, the value of the incident surface point vector in the i-th column is modified to the value of the incident surface point vector in the i-1-th column.

7. The method according to claim 6, characterized in that Based on the preset algorithm, feature extraction is performed on the geological radar data to obtain the reflection surface layer line, including: For the geological radar data matrix with M rows and N columns, the part above the incident surface layer line is set to 0, and the part below the incident surface layer line is normalized to between -1 and +1; The second sliding window is set in the horizontal direction, and the Pearson correlation coefficient between the standard Ricker wavelet and the geological radar data waveform in the second sliding window is calculated. The points with a Pearson correlation coefficient greater than the set threshold are marked as reflection surface points. After the second sliding window slides from the first column to the last column, the reflection surface layer line is obtained.

8. The method according to claim 1, characterized in that Before determining the thickness of the ballast layer by using the depth difference between the incident surface layer line and the reflected surface layer line, the following steps are also included: By using the correlation between the incident surface layer bit line and the reflection surface layer bit line, the mirror reflection interference in the reflection surface layer bit line is removed; And / or, removing invalid points in the incident surface layer bit line and the reflection surface layer bit line; the invalid point data is abnormal.

9. A device for identifying the thickness of a roadbed based on geological radar data, characterized in that: include: Data acquisition module, used to obtain geological radar data of railway subgrade; The geological radar data include waveform amplitude data of different horizontal positions and different geological layers of the railway embankment; The feature extraction module is used to extract features from the geological radar data based on a preset algorithm to obtain the incident surface layer line and the reflection surface layer line; the preset algorithm is used to: extract the incident surface layer line based on the absolute value of the waveform amplitude data of the geological radar data and the light wave characteristics of the incident surface; extract the reflection surface layer line based on the signal similarity between the geological radar data and the standard Ricker wavelet; the frequency of the standard Ricker wavelet is the same as the radar frequency; The thickness determination module is used to determine the thickness of the roadbed layer by using the depth difference between the incident surface layer position line and the reflected surface layer position line.

10. The device according to claim 9, characterized in that Also includes: The preprocessing module is used to preprocess the geological radar data of the railway roadbed before the data acquisition module extracts features from the geological radar data to obtain preprocessed data; The preprocessing includes one or any combination of zero drift removal, background denoising, bandpass filtering, and normalization; The feature extraction module is specifically used to extract features from preprocessed data.

11. The device according to claim 9, characterized in that The feature extraction module is specifically used for: Perform horizontal smoothing on the geological radar data matrix of M rows and N columns to obtain a smoothed data matrix; Perform element-wise absolute value calculation on the smoothed data matrix, and mark the maximum value of each column as the incident surface point vector of 1 row and M columns; Set the first sliding window with M rows and w columns, and calculate the volatility of the incident surface point vector in each column in the first sliding window according to the following formula: Among them, E i represents the volatility of the incident surface point vector in the i-th column, M i represents the value of the incident surface point vector of the i-th column, represents the mean value of the position vectors of all incident surface points in the first sliding window; When Ei of the i-th column>0.1, the value of the incident surface point vector in the i-th column is modified to the value of the incident surface point vector in the i-1-th column.

12. The device according to claim 11, characterized in that The feature extraction module is specifically used for: For the geological radar data matrix with M rows and N columns, the part above the incident surface layer line is set to 0, and the part below the incident surface layer line is normalized to between -1 and +1; The second sliding window is set in the horizontal direction, and the Pearson correlation coefficient between the standard Ricker wavelet and the geological radar data waveform in the second sliding window is calculated. The points with a Pearson correlation coefficient greater than the set threshold are marked as reflection surface points. After the second sliding window slides from the first column to the last column, the reflection surface layer line is obtained.

13. The device according to claim 9, characterized in that Also includes: The layer arrangement module is used to use the correlation between the incident surface layer line and the reflection surface layer line to remove the mirror reflection interference in the reflection surface layer line before the thickness determination module uses the depth difference between the incident surface layer line and the reflection surface layer line to determine the thickness of the roadbed layer; and / or remove invalid points in the incident surface layer line and the reflection surface layer line; the invalid point data is abnormal.

14. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 8 is implemented.

15. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

16. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images