B-Scan atlas optimization data sparse method in time mode of ground penetrating radar and related device

Abstract

The invention discloses a B-Scan atlas optimization data sparse method in a ground penetrating radar time mode and a related device, and relates to the technical field of ground penetrating radar signal processing, and the method comprises the steps: determining the instantaneous dragging speed of a ground penetrating radar during collection based on a front A-Scan data association model and a rear A-Scan data association model; establishing a nominal speed judgment model according to the collection frequency and the number of sampling points; smooth filtering is carried out on the instantaneous dragging speed set, the average speed and the standard deviation are calculated, and the speed confidence coefficient is calculated; calculating a comprehensive difference value of the adjacent data frames, and removing one frame in the adjacent data frames based on a self-adaptive redundancy judgment threshold function; calculating a sparse coefficient based on the nominal speed judgment model and the instantaneous dragging speed set; and if the sparse coefficient threshold value is greater than or equal to the sparse coefficient threshold value, performing sparseness and data optimization on the data frame to obtain an optimized A-Scan data sequence, and generating a B-Scan atlas with a high signal-to-noise ratio.

Description

Technical Field

[0001] This application relates to the field of ground penetrating radar signal processing technology, and in particular to a method and related apparatus for optimizing B-Scan map sparsity in ground penetrating radar time mode. Background Technology

[0002] Ground-penetrating radar (GPR) typically uses a range-triggered mode, acquiring A-scan data with equal intervals. After median or mean filtering, high-quality B-scan images can be obtained. However, due to terrain or other constraints, GPR needs to operate in time-triggered mode. In time-triggered mode, GPR is usually pushed or pulled by a person for detection. Since the sampling frequency of GPR is fixed, and the speed of the person dragging the radar is difficult to maintain consistently, the A-scan data acquired by GPR no longer has equal intervals. This leads to two problems: First, when the dragging speed is too fast, the GPR acquires too little data per unit distance, rendering the data unanalyzable and invalid. Second, when the dragging speed is too slow, the number of sampling frames per unit distance is too high, resulting in overly dense data. This leads to minimal waveform differences between adjacent A-scan data, causing the reflection characteristic signal of underground targets to be excessively stretched in the B-scan. This causes the filtering algorithm to incorrectly treat the target signal as background noise, resulting in feature degradation, where the reflection characteristic waveform of underground targets cannot be displayed in the radar image. Summary of the Invention

[0003] The purpose of this application is to provide a method and related apparatus for optimizing the sparsity of B-Scan maps in time-mode ground penetrating radar, which solves the problem of feature degradation caused by excessive data density in B-Scan maps in time-mode, and enhances the clarity and identifiability of the reflection features of underground targets.

[0004] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a method for optimizing B-Scan map sparsity in ground-penetrating radar time mode, including: Acquire A-Scan data frames from the ground-penetrating radar operating in a fixed time interval mode; For the current A-Scan data frame, the instantaneous drag speed of the ground penetrating radar during data acquisition is determined based on the correlation model of previous and subsequent A-Scan data. A nominal velocity determination model is established based on the acquisition frequency of the ground penetrating radar and the number of sampling points in each A-Scan data frame; The instantaneous drag speed set is smoothed and filtered, and the average speed and standard deviation of the filtered instantaneous drag speed set are calculated; the instantaneous drag speed set is the set of instantaneous drag speeds determined for each A-Scan data frame; Based on the average speed and standard deviation, calculate the speed confidence level; if the confidence level is higher than a preset threshold, determine that the A-Scan data frame is reliable and proceed to the next step; Calculate the overall difference value of adjacent A-Scan data frames, and based on the adaptive redundancy determination threshold function, remove one of the adjacent A-Scan data frames whose overall difference value is less than the adaptive redundancy determination threshold. Based on the nominal speed determination model and the filtered instantaneous drag speed set, the sparsity coefficient is calculated; If the sparsity coefficient is greater than or equal to the sparsity coefficient threshold, then the A-Scan data frame is sparsified to obtain the A-Scan data sequence; if the sparsity coefficient is less than the threshold, then the data in the A-Scan data frame is not sparsified. The sparsely processed A-Scan data frames are optimized to obtain optimized A-Scan data sequences, and high signal-to-noise ratio B-Scan maps are generated.

[0005] Secondly, this application provides an apparatus for optimizing B-Scan map sparsity data in ground-penetrating radar time mode, comprising: The data acquisition module is used to acquire A-Scan data frames from the ground-penetrating radar in a fixed time interval operating mode; The instantaneous drag speed calculation module is used to determine the instantaneous drag speed of the ground penetrating radar during data acquisition based on the correlation model of previous and subsequent A-Scan data for the current A-Scan data frame. The nominal velocity determination model construction module is used to establish a nominal velocity determination model based on the acquisition frequency of the ground penetrating radar and the number of sampling points in each A-Scan data frame. The average speed and standard deviation calculation module is used to smooth and filter the instantaneous drag speed set, and calculate the average speed and standard deviation of the filtered instantaneous drag speed set; the instantaneous drag speed set is a set of instantaneous drag speeds determined for each A-Scan data frame; The speed confidence calculation module is used to calculate the speed confidence based on the average speed and standard deviation; if the confidence is higher than a preset threshold, the A-Scan data frame is determined to be reliable, and the next step is executed. The elimination module is used to calculate the comprehensive difference value of adjacent A-Scan data frames and, based on the adaptive redundancy determination threshold function, eliminate one of the adjacent A-Scan data frames whose comprehensive difference value is less than the adaptive redundancy determination threshold. The sparsity coefficient calculation module is used to calculate the sparsity coefficient based on the nominal speed determination model and the filtered instantaneous drag speed set. The sparsity module is used to sparsify the A-Scan data frame to obtain the A-Scan data sequence if the sparsity coefficient is greater than or equal to the sparsity coefficient threshold; otherwise, the data in the A-Scan data frame is not sparsified. The output module is used to optimize the sparsely processed A-Scan data frames to obtain optimized A-Scan data sequences and generate high signal-to-noise ratio B-Scan maps.

[0006] Thirdly, this application 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 executes the computer program to implement the B-Scan map optimization data sparsity method for ground penetrating radar time mode as described in any one of the above.

[0007] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the B-Scan map optimization data sparsity method for ground-penetrating radar time mode as described above.

[0008] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application provides a method and related apparatus for optimizing B-Scan map sparsity in ground-penetrating radar time mode. First, by determining the instantaneous drag velocity, a nominal velocity determination model is established, accurately grasping the radar's velocity during data acquisition and providing a reliable velocity basis for subsequent processing. The instantaneous drag velocity set is smoothed and filtered, and the average velocity, standard deviation, and velocity confidence are calculated to ensure data reliability and avoid data errors caused by velocity instability affecting subsequent analysis. In terms of data filtering, the comprehensive difference value of adjacent A-Scan data frames is calculated, and redundant data frames are removed using an adaptive redundancy threshold function, effectively solving the problem of excessive data density and reducing unnecessary data volume. Simultaneously, selective sparsity processing is performed on the data based on the sparsity coefficient, ensuring data analyzability while avoiding information loss due to excessive data sparsity. Finally, the sparsified data is optimized to generate a high signal-to-noise ratio B-Scan map. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1This is an application environment diagram of a B-Scan map optimization data sparsity method in a ground-penetrating radar time mode according to an embodiment of this application; Figure 2 A flowchart illustrating a method for optimizing B-Scan map sparsity in ground-penetrating radar time mode, provided as an embodiment of this application; Figure 3 This is a schematic diagram illustrating the steps of A-Scan acquisition and B-Scan generation provided in an embodiment of this application; Figure 4 A schematic diagram of velocity estimation based on an A-Scan inter-frame correlation model provided in an embodiment of this application; Figure 5 A schematic diagram of the relationship between the nominal velocity determination model and the radar center frequency provided in an embodiment of this application; Figure 6 This is a comparison of B-Scan image effects before and after sparsification provided in an embodiment of this application; Figure 7 A schematic diagram of the functional modules of a ground-penetrating radar time-mode B-Scan map sparse optimization device provided in an embodiment of this application; Figure 8 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0011] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0012] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0013] The B-Scan map optimization data sparsity method for ground-penetrating radar in time mode provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up independently, integrated into server 104, or placed in the cloud or on other servers. Terminal 102 can send A-Scan data frames acquired from ground-penetrating radar operating at fixed time intervals to server 104. After receiving the A-Scan data frames, server 104, for each A-Scan data frame, determines the instantaneous drag speed of the ground-penetrating radar during acquisition based on a correlation model of previous and subsequent A-Scan data; establishes a nominal speed determination model based on the acquisition frequency of the ground-penetrating radar and the number of sampling points in each A-Scan data frame; performs smoothing filtering on the instantaneous drag speed set, and calculates the average speed and standard deviation of the filtered instantaneous drag speed set; the instantaneous drag speed set is composed of the instantaneous drag speeds determined for each A-Scan data frame. The system first sets up a B-Scan data frame; then calculates the speed confidence level based on the average speed and standard deviation; if the confidence level is higher than a preset threshold, the A-Scan data frame is deemed reliable, and the next step is executed; the system calculates the comprehensive difference value of adjacent A-Scan data frames, and based on the adaptive redundancy judgment threshold function, removes one of the adjacent A-Scan data frames whose comprehensive difference value is less than the adaptive redundancy judgment threshold; based on the nominal speed judgment model and the filtered instantaneous drag speed set, the system calculates the sparsity coefficient; if the sparsity coefficient is greater than or equal to the sparsity coefficient threshold, the A-Scan data frame is sparsified to obtain an A-Scan data sequence; if the sparsity coefficient is less than the threshold, the data in the A-Scan data frame is not sparsified; the sparsified A-Scan data frame is optimized to obtain an optimized A-Scan data sequence, and a high signal-to-noise ratio B-Scan map is generated. The server 104 can feed back the obtained high signal-to-noise ratio B-Scan map to the terminal 102. Furthermore, in some embodiments, the B-Scan map optimization data sparsity method in the ground penetrating radar time mode can also be implemented separately by the server 104 or the terminal 102. For example, the terminal 102 can directly perform B-Scan map optimization data sparsity in the ground penetrating radar time mode on the A-Scan data frames obtained in the ground penetrating radar working mode at a fixed time interval. Alternatively, the server 104 can obtain the A-Scan data frames obtained in the ground penetrating radar working mode at a fixed time interval from the data storage system and perform B-Scan map optimization data sparsity in the ground penetrating radar time mode on the A-Scan data frames obtained in the ground penetrating radar working mode at a fixed time interval.

[0014] The terminal 102 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The server 104 can be implemented using a standalone server or a server cluster composed of multiple servers, or it can be a cloud server.

[0015] In one exemplary embodiment, such as Figure 2 As shown, a method for optimizing B-Scan map sparsity in ground-penetrating radar time mode is provided. This method is executed by a computer device, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps 201 to 209. Wherein: Step 201: Acquire A-Scan data frames from the ground-penetrating radar in a fixed time interval operating mode; Step 202: For the current A-Scan data frame, determine the instantaneous drag speed of the ground penetrating radar during data acquisition based on the correlation model of previous and subsequent A-Scan data; Step 203: Establish a nominal velocity determination model based on the acquisition frequency of the ground penetrating radar and the number of sampling points in each A-Scan data frame; Step 204: Perform smoothing filtering on the instantaneous drag speed set, and calculate the average speed and standard deviation in the filtered instantaneous drag speed set; the instantaneous drag speed set is a set of instantaneous drag speeds determined for each A-Scan data frame; Step 205: Calculate the speed confidence level based on the average speed and standard deviation; if the confidence level is higher than a preset threshold, the A-Scan data frame is determined to be reliable, and proceed to the next step. Step 206: Calculate the comprehensive difference value of adjacent A-Scan data frames, and based on the adaptive redundancy determination threshold function, remove one of the adjacent A-Scan data frames whose comprehensive difference value is less than the adaptive redundancy determination threshold. Step 207: Calculate the sparsity coefficients based on the nominal speed determination model and the filtered instantaneous drag speed set; Step 208: If the sparsity coefficient is greater than or equal to the sparsity coefficient threshold, then the A-Scan data frame is sparsified to obtain the A-Scan data sequence; if the sparsity coefficient is less than the threshold, then the data in the A-Scan data frame is not sparsified. Step 209: Optimize the sparsely processed A-Scan data frames to obtain optimized A-Scan data sequences and generate high signal-to-noise ratio B-Scan maps.

[0016] Specifically, the B-Scan map data sparsity optimization method provided in this embodiment is used to establish the relationship between nominal velocity and actual velocity when the ground penetrating radar acquires data in time mode by combining the A-Scan frame similarity model and the velocity determination model. The sampling density is adaptively adjusted according to the differences and velocity to achieve dynamic sparsification of redundant data in time mode. Furthermore, the density of the ground penetrating radar A-Scan data frames is optimized by modeling nominal velocity to improve the quality of B-Scan images.

[0017] In one exemplary embodiment, when performing steps 201-209, as follows: Figure 3 As shown, the specific details are as follows: First, data acquisition is performed: the ground-penetrating radar operates in time mode with a fixed sampling period. Each collection A-Scan data frames are obtained by using data points, and the timestamp of each frame is recorded. A series of A-Scan data were obtained. ， … … ; This is the sampling time interval for A-Scan data, measured in seconds, and its value is equal to the A-Scan data acquisition frequency. The reciprocal: Sampling frequency Provided by the radar hardware sampler or controller, in units of .

[0018] Then for the first A-Scan data frames are used to determine the instantaneous drag speed using a correlation model of preceding and following A-Scan data. Estimate. This is achieved by calculating the cross-correlation coefficients between adjacent A-Scan data frames. The calculation method is as follows: .

[0019] Among them, the cross-correlation coefficient sequence There will be a peak value in discrete implementations. τ Perform a limited search to find the peak position. (Unit: number of sampling points) This represents the time offset required to make two A-Scan data frames most aligned and similar. This means the second A-Scan signal lags behind the first, corresponding to a forward movement of the radar antenna, where n is the number of A-Scan data points. Then, a time delay is applied. Converted to spatial displacement : ; In the formula, To equivalently map the sampling point index to a calibration factor (m / point) of displacement along the radar's forward direction, it was obtained through experimental calibration (dragging a fixed known distance and observing the displacement of the cross-correlation peak), as follows: Figure 4 As shown, the two traces demonstrate the offset of adjacent A-Scan waveforms, which can be calculated and converted into displacement through cross-correlation.

[0020] Finally, consider the time interval. Calculate the instantaneous drag speed: ; In the formula, The empirical correction factor needs to be determined by combining factors such as instrument geometry and actual sliding friction.

[0021] Then, based on the ground-penetrating radar acquisition frequency Number of sampling points per frame of A-Scan data Establish nominal speed Determine the model, nominal speed With radar center frequency (like Figure 5 As shown, common center frequencies are 100MHz, 200MHz, 400MHz, 600MHz, 800MHz, and 1GHz. This information is used to reflect the spatial resolution characteristics and sampling matching relationship of the system in different frequency bands.

[0022] The nominal speed can be determined empirically through experimental calibration: ; In the formula, This is the frequency scaling factor (determined through system debugging or experience). The empirical method for determining this is as follows: In a ground-penetrating radar system, the target spatial resolution... It is the smallest spatial feature size that the system can resolve, determined by the radar center frequency and medium characteristics. The sampling interval between two A-Scan data frames should not exceed [a certain value]. Otherwise, spatial undersampling will occur. Relatedly, the wavelength of radar waves in the geological medium... ,in At the speed of light, Let be the relative permittivity. The empirical formula for spatial resolution is usually: ; Because, under the condition of meeting spatial sampling requirements, the maximum permissible moving speed (generally also the nominal speed) )for: ; We can obtain: ; ; When the user When set to a fixed value (e.g., 5mm, 10mm), , ; Then, velocity preprocessing and confidence assessment are performed: This embodiment processes a series of A-Scan data frames to obtain the velocity corresponding to each frame. and instantaneous drag speed set , The instantaneous drag velocity set is obtained by performing moving average filtering or Kalman filtering. Then calculate the average speed of this instantaneous drag speed set. : ; Simultaneously calculate the standard deviation of the set of instantaneous drag velocities after filtering. : ; Finally, calculate the velocity confidence level. : ; In the formula, To avoid division by zero, small positive numbers are usually chosen as 10. -6 ~10 -3 .

[0023] However, the speed confidence right The sensitivity is too high, when When it is close to 0, even a very small amount It will also lead to A sharp drop can be observed, at which point a variation of the confidence level can be used: ; like If the value is greater than 0.7, then the calculated speed of the A-Scan data frame is considered to be... It is credible.

[0024] When performing similarity measurement and redundancy determination, adjacent A-Scan data frames are calculated. A i , A j Overall differences between It can be as follows: ; In the formula, For time-domain cross-correlation, For frequency domain spectral difference, For energy difference, For spectral entropy difference, , , , For the weight values, the recommended initial value is... (Can be optimized based on on-site calibration). This represents the difference in amplitude spectrum shape between two frames of A-Scan data in the frequency domain. It is typically calculated using the Euclidean distance of the logarithmic amplitude spectrum to enhance sensitivity to the relative shape of the spectrum and mitigate the impact of overall energy differences. The calculation steps are as follows: First, regarding the first Frame A-Scan data Perform a Discrete Fourier Transform (DFT) to obtain the spectrum. And take the amplitude spectrum: ; Then calculate the logarithmic magnitude spectrum. : ; in Small positive numbers are used to avoid zero values.

[0025] Final calculation : ; In the formula, This represents the difference in energy distribution between two frames of A-Scan data in the time domain. By segmenting the signal and comparing the differences in energy distribution across segments, the local energy variation characteristics of the signal can be captured. The specific calculation steps are as follows: First, the signal is segmented, and the first segment is... The A-Scan data of the frame is uniformly divided into Segments (non-overlapping, if) Cannot be Divisible by integers, zeros can be added to the last segment. The segment signal is denoted as: ; in The length of each segment.

[0026] Then calculate the energy of each signal segment: ; Normalize its vectors: ; Final calculation : ; In the formula, The difference in spectral complexity between two A-Scan frames is represented by the absolute value of the difference in their respective spectral entropies. It reflects the uniformity of the spectral energy distribution. A larger entropy value indicates a more uniform energy distribution (such as noise), while a smaller entropy value indicates a more concentrated energy distribution (such as a single-frequency signal). The specific calculation steps are as follows: Calculate the frequency spectrum power spectrum : ; Normalize it to obtain the probability distribution : ; Then calculate the spectral entropy. : ; Finally, the spectral entropy difference is calculated: ; Calculate the overall difference value Then, the adaptive redundancy determination threshold function is used: ; Then, it dynamically determines whether two frames of A-Scan data are similar enough to remove one of them. If a frame is found to be redundant, it can be removed.

[0027] In the adaptive redundancy determination threshold function It is the speed influence coefficient, which determines the sensitivity of the threshold to changes in speed. It is the basic threshold, representing the redundancy judgment threshold when the speed is zero. and It needs to be determined through experiments or system debugging.

[0028] Finally, sparsity is applied to the A-Scan data: Calculate the sparsity coefficient The expression is: ; like If, then no sparsity is performed; if Then calculate the retention interval. It indicates that it is for k Rounding is performed.

[0029] Combined with comprehensive difference value Further decision: If And if the frame is not a key frame, then it can be edited according to... Discard at intervals; if If the conditions of the keyframe are met, then retain it.

[0030] The conditions for satisfying the keyframe are as follows: 1) In calculation At that time, among them Mutations can occur, including: ; in, It is the average energy segment of the data in that frame. For the preset proportional coefficient, such as Adjustments will be made based on the actual situation.

[0031] 2) Calculate the spectrum At that time, it was found that the amplitude of the main frequency peak was several times higher than the background noise. ; in It's a multiple, and will be adjusted according to the actual situation. It is the frequency amplitude of the background noise.

[0032] 3) Calculation At that time, it was found that the spectral entropy dropped sharply, and: ; in, These are experience values, which can be adjusted based on actual circumstances.

[0033] These values, which need to be adjusted according to the actual situation, should be calibrated using labeled test data during the initial commissioning of the project.

[0034] The reconstruction results are processed by edge enhancement, noise suppression and smoothing, and the output is a sparsified A-Scan sequence that can be directly used for median or mean filtering background removal algorithms.

[0035] like Figure 6 As shown in the figure, (a) is the simulated B-Scan before sparsification (containing noise and with superbolic reflection features), and (b) is the schematic diagram after processing, where the reflection is more prominent and the background noise is reduced.

[0036] This application also provides a detailed numerical example. The following uses a typical field test as an example to give the step-by-step calculation from parameter reading to sparse decision-making. All arithmetic is expanded and calculated step by step to ensure reproducibility.

[0037] 1) Set the operating parameters of the ground penetrating radar system: Target spatial resolution Δd = 0.005 m (5 mm); Scan data sampling rate f s =100Hz, that is Δt =0.01s; Number of Scan data sampling points N s = 1024 points; initial weight value w 1 =0.4, w 2 =0.25, w 3 =0.2, w 4 =0.15; Adaptive redundancy determination threshold function parameter α =0.5, β =0.02; confidence threshold C thresh =0.7; Small constant ε = 1 × 10 -6 .

[0038] 2) Assuming estimated speed v(t) =0.2m / s, which can be obtained through a distance measuring wheel or a cross-correlation algorithm.

[0039] 3) Calculate the sparsity coefficient k .

[0040] v ref =0.005×100=0.5m / s.

[0041] k=max ( v ref / v(t), 1 )= max (0.5 / 0.2, 1) = 2.500, retain interval M=round(k) =3, meaning that one frame is retained for every three frames of A-Scan data (one frame is retained for every four frames).

[0042] 4) Calculate the difference value. θ=α·v(t)+β =0.5×0.2+0.02=0.120.

[0043] 5) Calculate the metrics for two adjacent frames (assuming sample results, for example): Rij =0.95 (normalized cross-correlation peak value); D freq =0.04, D E =0.02, D H =0.01 Δ(i,j) =0.4×(1-0.95)+0.25×0.04+0.2×0.02+0.15×0.01=0.035500; 6) Comparison Δ(i,j) With θ: Δ=0.035500<θ=0.120, result: can be removed (depending on keyframe rules).

[0044] 7) Empirical Results: In the field tests of this embodiment (using a 400MHz intermediate frequency antenna), f s (Taking 100Hz as an example), after sparse processing and median filtering to remove background, the overall signal-to-noise ratio of B-Scan is improved by about 30%, and the target boundary is clearer. This value is derived from the statistical average of the signal peak to background noise ratio (SNR) at the target boundary before and after the comparison processing.

[0045] In this case, the implementation considerations and engineering recommendations are as follows: It is recommended to perform a calibration of the cross-correlation displacement to the actual drag distance during equipment manufacturing or on-site installation to determine... or On-site, weights can be automatically adjusted using a small window (e.g., 1 second). w and threshold θ To adapt to different geological and noise environments. Cross-correlation, FFT (used for...) The main computational overhead is 0.5%, but the use of sliding window and FFT pre-computation ensures real-time performance on the embedded platform. The original A-Scan and sparse decision logs are retained for easy replay and reproduction, and algorithm fine-tuning.

[0046] Based on the same inventive concept, this application also provides a ground-penetrating radar (GPR) time-mode B-Scan map sparse optimization device for implementing the aforementioned ground-penetrating radar (GPR) time-mode B-Scan map sparse optimization method. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more GPR time-mode B-Scan map sparse optimization device embodiments provided below can be found in the limitations of the GPR time-mode B-Scan map sparse optimization method described above, and will not be repeated here.

[0047] In one exemplary embodiment, such as Figure 7 As shown, a device for optimizing B-Scan map sparsity in ground-penetrating radar time mode is provided, comprising: Data acquisition module 701 is used to acquire A-Scan data frames of ground penetrating radar in a fixed time interval working mode; The instantaneous drag speed calculation module 702 is used to determine the instantaneous drag speed of the ground penetrating radar during acquisition based on the correlation model of previous and subsequent A-Scan data frames for the current A-Scan data frame. The nominal velocity determination model construction module 703 is used to establish a nominal velocity determination model based on the acquisition frequency of the ground penetrating radar and the number of sampling points in each A-Scan data frame. The average speed and standard deviation calculation module 704 is used to perform smoothing filtering on the instantaneous drag speed set and calculate the average speed and standard deviation in the filtered instantaneous drag speed set; the instantaneous drag speed set is a set of instantaneous drag speeds determined for each A-Scan data frame; The speed confidence calculation module 705 is used to calculate the speed confidence based on the average speed and standard deviation; if the confidence is higher than a preset threshold, the A-Scan data frame is determined to be reliable, and the next step is executed. The elimination module 706 is used to calculate the comprehensive difference value of adjacent A-Scan data frames and, based on the adaptive redundancy determination threshold function, eliminate one of the adjacent A-Scan data frames whose comprehensive difference value is less than the adaptive redundancy determination threshold. The sparse coefficient calculation module 707 is used to calculate the sparse coefficient based on the nominal speed determination model and the filtered instantaneous drag speed set. The sparse module 708 is used to sparse the A-Scan data frame to obtain the A-Scan data sequence if the sparse coefficient is greater than or equal to the sparse coefficient threshold; otherwise, it does not sparse the data in the A-Scan data frame. The output module 709 is used to optimize the sparsely processed A-Scan data frame to obtain the optimized A-Scan data sequence and generate a high signal-to-noise ratio B-Scan map.

[0048] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 8As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs in the non-volatile storage media to run. The database stores B-Scan maps. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements a method for optimizing the sparsity of B-Scan map data in a ground-penetrating radar time-mode configuration.

[0049] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0050] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0051] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0052] 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, data stored, data displayed, 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 the relevant data must comply with relevant regulations.

[0053] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media 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), magnetic 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 can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0054] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0055] In summary, this application has the following technical effects: This application combines a correlation model of A-Scan data before and after the measurement to determine velocity, and uses a nominal velocity determination model to determine velocity correction and sparsity ratio. By calculating the velocity adaptive sparsity coefficient, A-Scan data is selectively retained and interpolated for reconstruction, ensuring sufficient differences between adjacent waveforms. The signal-to-noise ratio of the generated B-Scan image after filtering and background removal is significantly improved, thereby enhancing the clarity and identifiability of underground target reflection features.

[0056] The technical features of the above embodiments can be combined in any way. For the sake of brevity, 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.

[0057] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for optimizing B-Scan map sparsity in ground-penetrating radar time mode, characterized in that, include: Acquire A-Scan data frames from the ground-penetrating radar operating in a fixed time interval mode; For the current A-Scan data frame, the instantaneous drag speed of the ground penetrating radar during data acquisition is determined based on the correlation model of previous and subsequent A-Scan data. A nominal velocity determination model is established based on the acquisition frequency of the ground penetrating radar and the number of sampling points in each A-Scan data frame; The instantaneous drag speed set is smoothed and filtered, and the average speed and standard deviation of the filtered instantaneous drag speed set are calculated; the instantaneous drag speed set is the set of instantaneous drag speeds determined for each A-Scan data frame; Based on the average speed and standard deviation, calculate the speed confidence level; if the confidence level is higher than a preset threshold, determine that the A-Scan data frame is reliable and proceed to the next step; Calculate the overall difference value of adjacent A-Scan data frames, and based on the adaptive redundancy determination threshold function, remove one of the adjacent A-Scan data frames whose overall difference value is less than the adaptive redundancy determination threshold. Based on the nominal speed determination model and the filtered instantaneous drag speed set, the sparsity coefficient is calculated; If the sparsity coefficient is greater than or equal to the sparsity coefficient threshold, then the A-Scan data frame is sparsified to obtain the A-Scan data sequence. If the sparsity coefficient is less than 1, then the data in the A-Scan data frame will not be sparsified. The sparsely processed A-Scan data frames are optimized to obtain optimized A-Scan data sequences, and high signal-to-noise ratio B-Scan maps are generated.

2. The method for optimizing B-Scan map sparsity in ground-penetrating radar time mode according to claim 1, characterized in that, The formula for calculating the instantaneous drag speed is: ； in, For instantaneous drag speed, To verify the correction coefficient, This represents the spatial displacement of the radar antenna. This represents the sampling time interval for A-Scan data.

3. The method for optimizing B-Scan map sparsity in ground-penetrating radar time mode according to claim 2, characterized in that, The formula for the nominal speed determination model is as follows: ； in, This is the frequency scaling factor. The acquisition frequency of the ground-penetrating radar. The center frequency of the radar. This is the nominal speed.

4. The method for optimizing B-Scan map sparsity in ground-penetrating radar time mode according to claim 3, characterized in that, The instantaneous drag speed set is smoothed and filtered, and the average speed and standard deviation of the filtered instantaneous drag speed set are calculated, specifically including: Process each A-Scan data frame to obtain the instantaneous drag speed and the set of instantaneous drag speeds corresponding to each A-Scan data frame. ; The instantaneous drag velocity set is obtained by performing a moving average filter or a Kalman filter on the instantaneous drag velocity set. ; According to the formula Calculate the average speed in the filtered instantaneous drag speed set; According to the formula Calculate the standard deviation of the filtered instantaneous drag speed set; In the formula, N is the number of sampling points; t is time.

5. The method for optimizing B-Scan map sparsity in ground-penetrating radar time mode according to claim 4, characterized in that, The formula for calculating the velocity confidence level is: ； in, It is a small positive number, with a value range of 10. -6 ~10 -3 .

6. The method for optimizing B-Scan map sparsity in ground-penetrating radar time mode according to claim 5, characterized in that, Calculate the overall difference value of adjacent A-Scan data frames, and based on the adaptive redundancy determination threshold function, remove one of the adjacent A-Scan data frames whose overall difference value is less than the adaptive redundancy determination threshold. Specifically, this includes: According to the formula Calculate the overall difference value between adjacent A-Scan data frames; where, For time-domain cross-correlation, For frequency domain spectral difference, For energy difference, For spectral entropy difference, , , , These are weight values; Based on the adaptive redundancy determination threshold function Determine the adaptive redundancy threshold for A-Scan data frames; where, It is the speed influence coefficient. It is the basic threshold; when At that time, one frame from the adjacent A-Scan data frames is removed.

7. The method for optimizing B-Scan map sparsity in ground-penetrating radar time mode according to claim 6, characterized in that, The formula for calculating the sparsity coefficient is as follows: 。 8. A device for optimizing B-Scan map sparsity data in ground-penetrating radar time mode, characterized in that, include: The data acquisition module is used to acquire A-Scan data frames from the ground-penetrating radar in a fixed time interval operating mode; The instantaneous drag speed calculation module is used to determine the instantaneous drag speed of the ground penetrating radar during data acquisition based on the correlation model of previous and subsequent A-Scan data for the current A-Scan data frame. The nominal velocity determination model construction module is used to establish a nominal velocity determination model based on the acquisition frequency of the ground penetrating radar and the number of sampling points in each A-Scan data frame. The average speed and standard deviation calculation module is used to smooth and filter the instantaneous drag speed set, and calculate the average speed and standard deviation of the filtered instantaneous drag speed set; the instantaneous drag speed set is a set of instantaneous drag speeds determined for each A-Scan data frame; The speed confidence calculation module is used to calculate the speed confidence based on the average speed and the standard deviation. If the confidence level is higher than the preset threshold, the A-Scan data frame is determined to be reliable, and the next step is executed. The elimination module is used to calculate the comprehensive difference value of adjacent A-Scan data frames and, based on the adaptive redundancy determination threshold function, eliminate one of the adjacent A-Scan data frames whose comprehensive difference value is less than the adaptive redundancy determination threshold. The sparsity coefficient calculation module is used to calculate the sparsity coefficient based on the nominal speed determination model and the filtered instantaneous drag speed set. The sparse module is used to sparse the A-Scan data frame if the sparse coefficient is greater than or equal to the sparse coefficient threshold, so as to obtain the A-Scan data sequence. If the sparsity coefficient is less than 1, then the data in the A-Scan data frame will not be sparsified. The output module is used to optimize the sparsely processed A-Scan data frames to obtain optimized A-Scan data sequences and generate high signal-to-noise ratio B-Scan maps.

9. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement a method for optimizing B-Scan map sparsity in ground-penetrating radar time mode according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements a method for optimizing B-Scan map sparsity in ground-penetrating radar time mode according to any one of claims 1-7.

Images