Phased array ground penetrating radar disease positioning method, device, equipment and medium

Abstract

This invention relates to the field of ground-penetrating radar (GPR) data inversion technology, and discloses a method, apparatus, equipment, and medium for locating defects using phased-array GPR. The method includes: acquiring multi-channel raw echo data through broadband scanning and transmission focusing, and initially identifying suspected defect locations; performing multi-layer medium model inversion on each profile based on the multi-channel raw echo data, wherein the multi-layer medium model includes the dielectric constant and interface depth field of the multi-layer medium; solving the propagation path and arrival time of electromagnetic waves from each channel to the suspected defect location based on the multi-layer medium model, obtaining the delay time of each channel; performing time alignment of the echo data of each channel based on the delay time of each channel, and superimposing the multi-channel echo data to obtain a superimposed image of each profile; and locating the defect within the neighborhood of the suspected defect location based on the superimposed image of each profile. This invention can improve the detection accuracy of deep, weak defects.

Description

Technical Field

[0001] This invention belongs to the field of ground penetrating radar data inversion technology, and particularly relates to a method, device, equipment and medium for locating defects in phased array ground penetrating radar. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Ground penetrating radar (GPR) is a non-destructive testing technology that uses electromagnetic waves to detect underground non-visual targets. To improve detection efficiency, positioning accuracy, and the ability to identify deep-seated defects, phased array technology has been integrated into the GPR system. By precisely controlling the transmission delay, beam deflection and focusing effects are achieved. At the same time, multi-channel receiving technology is used to achieve signal gain after coherent superposition, thereby enhancing detection performance.

[0004] However, in typical underground scenarios such as road structures, there are often layers of media, such as overburden and basement layers. Due to the difference in dielectric constant between the upper and lower layers, electromagnetic waves undergo refraction during propagation across layers, and the equivalent wave velocity also changes accordingly. Currently, most existing methods assume a uniform half-space or a straight path for electromagnetic waves for time-delay focusing and pointing control operations. This approach can easily lead to a mismatch between the propagation path and the time-delay model, resulting in deviations in the estimation of the depth and horizontal position of deep-seated defects. It can also cause problems such as reduced coherent superposition efficiency and a wider focusing response range, ultimately affecting the detection accuracy of deep, weak defects. Summary of the Invention

[0005] In view of this, the present invention provides a method, apparatus, equipment and medium for locating defects in phased array ground penetrating radar, in order to improve the detection accuracy of deep and weak defects.

[0006] One aspect of the present invention provides a method for locating defects in phased array ground-penetrating radar, comprising the following steps:

[0007] Multi-channel raw echo data was acquired through broadband scanning and transmission focusing, and the locations of suspected defects were initially identified.

[0008] Based on multi-channel raw echo data, a multi-layer dielectric model inversion is performed on each profile. The multi-layer dielectric model includes the dielectric constant and interface depth field of the multi-layer dielectric.

[0009] Based on the multilayer medium model, the propagation path and arrival time of electromagnetic waves in each channel to the suspected lesion location are solved, and the delay time of each channel is obtained.

[0010] The echo data of each channel are time-aligned according to the delay time of each channel, and the echo data of multiple channels are superimposed to obtain the superimposed image of each profile.

[0011] Based on the superimposed images of each profile, the disease is located within the neighborhood of the suspected disease location.

[0012] In some embodiments, the inversion of a multi-layer dielectric model based on multi-channel raw echo data includes: pre-setting a candidate set of dielectric constants; using the candidate set of dielectric constants as an optimization space, for each profile, based on a pre-constructed focusing quality evaluation function, with the goal of achieving the best multi-channel focusing effect, optimizing the dielectric constant of each layer starting from the surface layer to obtain a multi-layer dielectric model.

[0013] In some embodiments, specific spatial locations at different underground depths are used as the dielectric constant optimization objects, and the neighborhood of each specific spatial location is used as the evaluation region. The focus quality evaluation function is a weighted sum of image entropy, coherence aggregation degree, and sidelobe energy ratio within the evaluation region. The dielectric constant that maximizes the value of the focus quality evaluation function is taken as the dielectric constant of that spatial location.

[0014] In some embodiments, the propagation path is determined for each channel through the following process:

[0015] Based on the established multi-layer medium model, the propagation path of electromagnetic waves from the transmission channel to the suspected disease location is initialized. The intersection of this path with the interface of each underground medium layer is the initial refraction point.

[0016] Based on the initial refraction point, define the optimization range of the refraction point on the interface of each medium layer;

[0017] Based on the dielectric constant of each layer of the medium and the length of each path between the layers, the propagation time of the electromagnetic wave in each layer of the medium is calculated, and the total propagation time is obtained by summing them up.

[0018] With the total time as the objective, the position of the refraction point is iteratively optimized within the optimization range of each interface refraction point to obtain the actual propagation path and propagation time of electromagnetic waves from each channel to the suspected disease location.

[0019] In some embodiments, after time-aligning the echo data of each channel according to the delay time of each channel, the angle of arrival of the receiving channel is determined based on the direction of the end of the transmission path from the suspected defect location to the receiving antenna. The polarization synthesis unit vector is determined based on the angle of arrival, and the original signal vector collected by the dual-polarization receiving array is projected and superimposed on the synthesis direction.

[0020] In some embodiments, locating a disease within a neighborhood of a suspected disease location based on the overlay of images from various profiles includes:

[0021] Based on the superimposed imaging images of each profile, a three-dimensional mesh is divided within the neighborhood of the suspected disease location;

[0022] For each grid cell within the neighborhood, calculate the coherent sum across all channels;

[0023] Grids whose coherent sum across all channels is greater than or equal to the disease determination threshold are recorded as disease points.

[0024] In some embodiments, the method further includes: using the obtained diseased area as a candidate target, controlling the phased array to scan at multiple scanning deflection angles; obtaining the local focusing response area of ​​the corresponding candidate target at each scanning angle as the candidate target area; and performing spatial consistency evaluation on the positions of multiple candidate targets to determine the authenticity of the disease.

[0025] A second aspect of the present invention provides a phased array ground-penetrating radar defect location device, comprising:

[0026] The raw data acquisition module is configured to acquire multi-channel raw echo data through broadband scanning and transmission focusing, and to preliminarily identify the location of suspected defects;

[0027] The dielectric model inversion module is configured to perform multi-layer dielectric model inversion on each profile based on multi-channel raw echo data. The multi-layer dielectric model includes the dielectric constant and interface depth field of the multi-layer dielectric.

[0028] The propagation delay calculation module is configured to solve the propagation path and arrival time of electromagnetic waves in each channel to the suspected disease location based on the multilayer medium model, and obtain the delay time of each channel.

[0029] The data alignment and overlay module is configured to time-align the echo data of each channel according to the delay time of each channel, and overlay the echo data of multiple channels to obtain the overlay image of each profile.

[0030] The disease location module is configured to locate the disease within the neighborhood of the suspected disease location based on the superimposed imaging images of each profile.

[0031] A third aspect of the present invention provides an electronic device comprising one or more processors and a memory; wherein the memory stores one or more computer programs, the one or more computer programs comprising instructions that, when executed by the electronic device, cause the electronic device to perform the method.

[0032] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed on an electronic device, cause the electronic device to perform the method.

[0033] A fifth aspect of the present invention provides a computer program product comprising a computer program that, when executed by a processor, implements the method described herein.

[0034] One or more of the above technical solutions can solve the actual propagation path of electromagnetic waves from each channel to the disease by inverting the structural parameters of the underground multi-layer medium and combining them with the suspected disease location. This allows for the determination of the time difference between each channel, which can be used for time alignment of multi-channel data, improving the reliability of superimposed imaging and providing a foundation for the accurate location of deep weak diseases. Attached Figure Description

[0035] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0036] Figure 1 This is an example scenario of one or more embodiments of the present invention;

[0037] Figure 2 This is a flowchart illustrating an example method for locating defects in phased array ground-penetrating radar in an embodiment of the present invention.

[0038] Figure 3 This is a schematic diagram illustrating the path of electromagnetic wave refraction and propagation in a multi-layered underground medium in an embodiment of the present invention.

[0039] Figure 4 This is a schematic diagram illustrating the principle of polarization projection and vector coherence superposition in an embodiment of the present invention;

[0040] Figure 5 This is a program module architecture diagram of an example method for locating defects in phased array ground-penetrating radar in an embodiment of the present invention. Detailed Implementation

[0041] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While some embodiments of this application are shown in the drawings, it should be understood that this application can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this application. It should be understood that the drawings and embodiments of this application are for illustrative purposes only and are not intended to limit the scope of protection of this application.

[0042] In the description of the embodiments of this application, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on".

[0043] To address the shortcomings of existing ground-penetrating radar (GPR) positioning methods in multi-layered underground media scenarios, such as the assumption of a uniform half-space or a straight electromagnetic wave path, which cannot adapt to the layered media and result in large positioning deviations, this paper proposes a new method. This method first collects multi-channel echo data and preliminarily identifies suspected disease locations. Then, it inverts the structural parameters of the multi-layered underground media to construct a precise media model. Next, it calculates the actual propagation path and delay time of electromagnetic waves from each channel to the disease location. Using this delay time, it achieves time alignment and superimposed imaging of multi-channel data. Combined with multi-angle verification to eliminate ghosting interference, and finally, through the coordinated operation of these steps, it achieves accurate and reliable positioning of underground diseases, especially deep and weak diseases.

[0044] Figure 1 This is an example scenario of one or more embodiments of the present invention, which includes a transmitting array, a receiving array, a front-end control module, and a host computer. Using a multi-layered underground medium as the detection target, the transmitting array (M channel) transmits detection signals into the multi-layered underground medium under the phased-array scanning / focusing command of the front-end control module; the receiving array (N channel, supporting VH dual polarization) receives the echo signals reflected by the medium, which are then acquired by an ADC and sent to the front-end control module. The front-end control module transmits the sampled data to the host computer, and the host computer executes the method flow for locating ground-penetrating radar defects.

[0045] Figure 2 A flowchart illustrating an example method for locating defects in phased array ground-penetrating radar is shown, including the following steps:

[0046] S101: Acquire multi-channel raw echo data through broadband scanning and transmission focusing, and preliminarily identify suspected defect locations;

[0047] S102: Based on the multi-channel raw echo data, perform multi-layer dielectric model inversion for each profile. The multi-layer dielectric model includes the dielectric constant and interface depth field of the multi-layer dielectric.

[0048] S103: Based on the multilayer medium model, solve the propagation path and arrival time of electromagnetic waves in each channel to the suspected lesion location, and obtain the delay time of each channel;

[0049] S104: Time-align the echo data of each channel according to the delay time of each channel, and superimpose the echo data of multiple channels to obtain the superimposed image of each profile;

[0050] S105: Based on the superimposed imaging of each profile, locate the disease within the neighborhood of the suspected disease location.

[0051] By inverting the structural parameters of the underground multi-layer medium and combining them with the suspected location of the disease, the actual propagation path of electromagnetic waves from each channel to the disease can be solved, and the time difference between each channel can be obtained. This time difference can be used for time alignment of multi-channel data, which improves the reliability of superimposed imaging and provides a basis for accurate disease location.

[0052] In step S101, the transmitting array is controlled to transmit pulse signals in the manner of full ergodic polling, focused illumination, or phased deflection fan-shaped scanning. The receiving array synchronously acquires waveform data of at least two orthogonal polarization components of each receiving channel, constructs a full matrix vector data volume containing the dimensions of "transmitting channel-receiving channel-time-polarization", and performs preprocessing such as time zero correction, channel amplitude and phase equalization, and direct wave suppression to obtain a calibration data volume.

[0053] Based on the preprocessed calibration data, signal regions that significantly differ from normal underground medium echoes in each receiving channel are extracted. These include strong energy reflection signals, hyperbolic characteristic signals, signals with abrupt amplitude changes, and waveform distortion signals, which are identified as potential fault locations. These signals typically correspond to the interface reflection between underground faults (such as cavities, pipelines, and medium damage) and the normal medium. Hyperbolic characteristic signals can be identified using conventional algorithms such as Hough transform and hyperbolic fitting, primarily corresponding to linear faults like underground pipelines. Strong energy reflection signals and signals with abrupt amplitude changes can be filtered by setting amplitude thresholds and calculating signal gradient changes, primarily corresponding to regional faults such as cavities and medium damage. Waveform distortion signals can be identified by comparing the similarity of the waveforms of normal medium echoes, eliminating normal signals with high waveform consistency and retaining the distorted signals.

[0054] When electromagnetic waves emitted by ground-penetrating radar propagate through underground media, their propagation speed is directly related to the dielectric constant of the medium. The larger the dielectric constant, the slower the propagation speed. The focusing effect is determined by the propagation delay of the electromagnetic waves and the channel alignment accuracy. If the dielectric constant deviates significantly from the actual medium, even if the focusing depth Z is set reasonably, problems such as focusing offset and signal divergence will occur, resulting in a significant reduction in focusing quality. In other words, the accuracy of the dielectric constant directly affects the focusing quality. Based on this, step S102 performs digital domain virtual focusing processing on the multi-channel raw data and inverts the multi-layer medium model by evaluating the focusing quality. Specifically, a candidate set of dielectric constants is taken and used as the optimization space. For each profile, based on the pre-constructed focusing quality evaluation function, with the goal of optimizing the multi-channel focusing effect, the dielectric constant of each layer is optimized starting from the surface layer to obtain the multi-layer medium model. Step S102 specifically includes:

[0055] S1021: Based on the detection scenario (such as underground media such as soil, rock, and concrete), a preset range of candidate dielectric constants (usually 2-30) is set, and candidate values ​​are divided into intervals of 0.1-0.5 to ensure coverage of the common dielectric constant range of the target medium;

[0056] S1022: Construction of the focus quality evaluation function. The evaluation function is used to evaluate the focus effect of a specific underground spatial location in a multi-channel fused image. The depth and the horizontal position at each depth are sampled. Multiple sampling points are obtained at each depth, i.e. multiple specific spatial locations. After substituting different candidate values ​​of dielectric constant into the function, the focus quality evaluation results of each specific spatial location can be obtained. With the goal of optimizing the multi-channel focus effect, the dielectric constant of each specific spatial location is determined.

[0057] For example, for each specific spatial location, its neighborhood is used as the evaluation region for that spatial location. Image entropy, coherence convergence, and sidelobe energy ratio are selected as evaluation indicators. A focus quality evaluation function is constructed using a weighted summation method. The focus effect of the evaluation region is evaluated based on this focus quality evaluation function, and the dielectric constant at which the focus effect is optimal is taken as the dielectric constant of that spatial location. The focus quality evaluation function is as follows:

[0058]

[0059] in, These are non-negative weighting coefficients. Indicates the evaluation area Image entropy within, Indicates the evaluation area Coherent aggregation degree within, Indicates the evaluation area The energy ratio of the side lobes within.

[0060] Image entropy The calculation formula is:

[0061]

[0062]

[0063] in, is the candidate equivalent dielectric constant; z is the depth parameter, and r represents a spatial location at depth z; In the candidate Below Spatial Position The amplitude response obtained by virtual focusing; Indicates the evaluation area Any spatial location within, This is a numerically stable term used to avoid the denominator being zero and... .

[0064] coherence factor The calculation formula is:

[0065]

[0066] in, For channel indexing, K The number of channels participating in the focus; The effective component of the echo vector projection received by the k-th channel; For reference sampling time; For the first k Channel from array element to spatial location r The propagation delay; Channel weights can be determined based on channel amplitude and phase calibration, pattern compensation, propagation attenuation compensation, etc. This is a numerically stable term. The more accurate the delay alignment, the closer the coherence factor is to 1.

[0067] Coherent aggregation degree Take coherence factor exist The average or weighted average.

[0068] Sidelobe energy ratio The calculation formula is:

[0069]

[0070] in For the evaluation area The position of the main peak within the lobe; R is the main lobe radius parameter (used to define the main lobe neighborhood). This is an indicator function; it takes the value 1 if the condition is true, and 0 otherwise. It is a numerically stable term. SLR The smaller the value, the more concentrated the energy is in the main lobe and the lower the leakage of the side lobes.

[0071] For each evaluation region , take The largest dielectric constant is denoted as the dielectric constant at the corresponding spatial location.

[0072] In specific calculations, one can first fix the focusing depth, and for points at different horizontal positions at that focusing depth, substitute each candidate value of the dielectric constant into the above reference formula to calculate, and select the candidate value that makes the function value optimal as the initial dielectric constant for that position; alternatively, one can first fix the dielectric constant, select all depths for virtual focusing, select the score with the highest focusing score as the score of the dielectric constant, obtain the scores of all dielectric constants for virtual focusing, and select the dielectric constant with the highest score as the equivalent dielectric constant.

[0073] By inverting the structural parameters of the underground multilayer medium, the dielectric constant and interface depth field of each layer are obtained, and a multilayer medium model that fits the real underground environment is constructed, which can fundamentally solve the problem of mismatch between the propagation path and the delay model.

[0074] S1024: Using the obtained surface dielectric constant and depth field (i.e., interface spatial shape model), define the detection area of ​​the next layer of medium. Repeat steps S1022-S1023 to obtain the depth field and dielectric constant of the next layer. Repeat this step to sequentially complete the inversion of dielectric constants and depth fields of all detectable underground layers until the echo signal has no effective response, thus obtaining the depth field of the underground multi-layer medium and the corresponding dielectric constant of each layer. The detection area of ​​the next layer of medium starts below the interface of the previous layer and extends to the end of the effective echo signal response, excluding interference from surface medium signals.

[0075] In step S103, when electromagnetic waves propagate in the underground multi-layer medium, their propagation speed is determined by the dielectric constant of each layer of medium, and the propagation path will be refracted due to the existence of the medium interface, and it is not a straight-line propagation. Figure 3 This is a schematic diagram illustrating the propagation path of electromagnetic waves through a multi-layered underground medium in an embodiment of the present invention. The underground medium is divided into a first layer and a second layer. The dielectric constant of the first layer is ε1, and its lower interface is h1(x,y). The dielectric constant of the second layer is ε2, and its lower interface is h2(x,y). For shallow targets, electromagnetic waves propagate directly to the target within the first layer. For deep targets, electromagnetic waves originate from the transmitting elements of the multi-channel phased array, propagate through the first layer to the refraction point on the interface h1(x,y), and then enter the second layer after refraction, finally reaching the deep target. Based on this, combined with the multi-layered medium model, the actual propagation path of electromagnetic waves from each transmitting channel, through refraction at each medium interface, and to the suspected defect location can be calculated first. Then, based on the length of each path segment and the propagation speed in the corresponding medium, the total time from emission to arrival at the suspected defect location can be calculated, thereby determining the delay time corresponding to each channel. This is specifically achieved through the following steps:

[0076] S1031: Obtain the position parameters of each transmitting and receiving channel in the raw echo data. Using the interface depth field in the multilayer medium model as a constraint, for each suspected defect location and each pair of transmitting and receiving channels, based on the principle of optical path minimization (Fermat's principle), solve for the optical path refraction points when electromagnetic waves propagate across layers, and then establish a nonlinear propagation delay model including the refraction path. Specifically, for each channel, the propagation path is determined through the following process: First, based on the constructed multilayer medium model, initialize the propagation path of the electromagnetic wave from the transmitting channel to the suspected defect location. The intersection of this path with the interface of each underground medium layer is the initial refraction point. Then, according to these initial refraction points, decompose the entire propagation path into segmented paths corresponding to each medium layer, and based on the spatial position of each initial refraction point, reasonably define the optimization range of the refraction points on the interface of each medium layer. Based on this, the propagation speed of electromagnetic waves in the corresponding medium is calculated by combining the dielectric constant of each medium layer. Then, according to the length of each path segment and the corresponding propagation speed, the total time of electromagnetic waves from emission to arrival at the suspected lesion location is initially calculated. Finally, with the total time as the target, the position of the refraction point is iteratively optimized within the optimization range of the refraction point on each interface to obtain the actual propagation path and propagation time of electromagnetic waves in each channel to arrive at the suspected lesion location.

[0077] S1032: Based on the propagation time of each channel, obtain the transmission delay time of the electromagnetic waves in each channel. For example, a standardized delay table can be constructed by organizing the data in the format of "channel number - delay time". Specifically, first determine the maximum value of the time it takes for the electromagnetic waves from each channel to reach the corresponding suspected defect point, and then determine the delay time of other channels based on this maximum value.

[0078] It is understandable that the location of the suspected disease can be the centroid of the suspected disease area or the location with the highest brightness in the image.

[0079] Based on the constructed multi-layer medium model, the actual propagation path and arrival time of electromagnetic waves in each channel to the suspected disease location are solved, and the accurate delay time of each channel is obtained. This effectively avoids the problem of estimation error in the depth and horizontal position of deep diseases caused by the path assumption deviation in existing methods, and lays the foundation for subsequent signal processing and localization.

[0080] In step S104, the echo data of each channel is first time-aligned based on the delay time of each channel to ensure that the echo signals of multiple channels are synchronously focused on the suspected lesion location, generating a preliminary focusing response. Then, the multiple channels are superimposed to obtain a superimposed imaging image.

[0081] Before performing multi-channel superposition, the true arrival direction of the signal in each received channel is determined based on the direction vector of the last segment of the transmission path of each channel, thus obtaining the angle of arrival of the received channel. ,like Figure 3As shown. The polarization synthesis unit vector is determined by the angle of arrival, and the dual-polarization receiving vector is projected based on the unit vector, thereby facilitating subsequent multi-channel coherent superposition in a unified polarization direction and suppressing polarization mismatch and vector cancellation caused by refraction.

[0082] Figure 4 This is a schematic diagram illustrating the principle of polarization projection and vector coherent superposition in an embodiment of the present invention. The horizontal axis represents the horizontal polarization component (H), the vertical axis represents the vertical polarization component (V), and the receiving vector... This represents the original signal vector acquired by the dual-polarized receiver array, and its components. , These correspond to the horizontally and vertically polarized echo signal components, respectively. Based on the direction of the final segment of the transmission path from the suspected defect location to the receiving antenna, the angle of arrival of the receiving channel is calculated. The unit vector in the polarization resultant direction is determined by this angle of arrival. This vector is the composite projection weight on the H and V polarization bases. By receiving the vector... unit vector in the direction of composition Projection is performed to obtain the effective components. This operation can extract effective signal components that are consistent with the synthetic polarization direction from the dual-polarization receiving vector, suppress polarization mismatch and vector cancellation caused by refraction of underground media, and help to achieve multi-channel coherent superposition in a unified polarization direction, improve imaging signal-to-noise ratio and focusing accuracy.

[0083] By combining the delay time of each channel, precise time alignment of multi-channel echo data is achieved, thereby completing the superposition of multi-channel echoes and obtaining a superimposed image with a high signal-to-noise ratio. This effectively solves the pain points of low coherent superposition efficiency and widening focusing response range in existing technologies, and significantly improves the reliability of superimposed imaging.

[0084] In step S105, based on the overlaid imaging image, a three-dimensional search is performed within the neighborhood of the suspected lesion location using an objective function with a time delay parameter to achieve precise lesion localization. Specifically, this includes the following steps:

[0085] S1051: Based on a multi-layered media model and combined with the detection area range, a three-dimensional grid covering the entire area to be located is constructed. The XY plane of the three-dimensional grid corresponds to the horizontal plane of the detection area, and the Z-axis depth corresponds to the distribution of each layer of the underground media. The grid accuracy can be adjusted according to the detection accuracy requirements.

[0086] S1052: Combining the anomaly characteristics of the underground medium with the focusing quality requirements, a dedicated disease evaluation function is constructed as a location criterion. This function formula is the coherent summation across all channels of each grid. Within the 3D grid corresponding to the neighborhood of the suspected disease location, for each grid point, the dielectric constant and depth field data corresponding to each grid point are substituted into the aforementioned disease evaluation function to calculate the evaluation function value. Grids with a coherent summation across all channels greater than or equal to the disease determination threshold are recorded as disease locations. Simultaneously, the preliminary spatial coordinates (XY plane position and Z-axis depth) and dielectric constant value of this location are recorded to complete the precise location of the disease.

[0087] The evaluation function is as follows:

[0088]

[0089] in Candidate disease locations / intervals, For receiving channel number, For the first k The effective component of the echo sampled by the channel or the echo vector projected from the k-th channel. The distance from the channel to the location of the lesion is calculated based on the above underground medium model. r The delay in propagation For reference sampling time.

[0090] Setting a disease determination threshold T will satisfy... The grid is recorded as the location of the disease.

[0091] In addition, by combining the preset dielectric constant and depth field anomaly judgment threshold of the detection scene, after identifying the grids on all channels whose coherent sum is greater than or equal to the disease judgment threshold, the following verification is also performed: when the grid simultaneously satisfies the following conditions (the difference between the dielectric constant and the surrounding normal medium is greater than or equal to the set dielectric constant threshold, such as 5, and the depth of the disease area formed by the continuous disease points is greater than or equal to the depth field fluctuation anomaly threshold, such as ≥0.5m), it is considered to be a disease point.

[0092] The method further includes step S106: repeatedly scanning the location of the disease from multiple incident angles and analyzing the consistency of the response of the echoes from each angle at that location to evaluate the authenticity of the disease.

[0093] The diseased area obtained in step S105 is used as a candidate target, and the phased array is controlled to scan at multiple scanning deflection angles; the local focusing response area of ​​the corresponding candidate target at each scanning angle is obtained as the candidate target area; the spatial consistency of multiple candidate target positions is evaluated to determine the authenticity of the disease.

[0094] Specifically, the evaluation indicators for spatial consistency evaluation include spatial location consistency and spatial size consistency. Based on the disease evaluation function and disease judgment threshold in step S1052, the suspected disease area corresponding to the candidate target position at each angle is located to obtain the position, size and depth information of each suspected disease area. When the deviation of the position, size and depth information between each suspected disease area is within the set consistency threshold, each suspected disease area is considered to have spatial consistency and the disease located is a real disease; otherwise, it is regarded as a ghost or interference.

[0095] By accurately locating the disease in the vicinity of the suspected disease location, and then further eliminating ghost images and interference through multi-angle verification, the detection reliability of deep and weak diseases is greatly improved.

[0096] Based on the above method, one or more embodiments of the present invention also provide a phased array ground-penetrating radar defect location device, such as... Figure 3 As shown, the system includes: a raw data acquisition module 201, configured to acquire multi-channel raw echo data through broadband scanning and transmission focusing, and preliminarily identify suspected disease locations; a medium model inversion module 202, configured to perform multi-layer medium model inversion on each profile based on the multi-channel raw echo data, wherein the multi-layer medium model includes the dielectric constant and interface depth field of the multi-layer medium; a propagation delay calculation module 203, configured to calculate the propagation path and arrival time of electromagnetic waves from each channel to the suspected disease location based on the multi-layer medium model, and obtain the delay time of each channel; a data alignment and overlay module 204, configured to time-align the echo data of each channel according to the delay time of each channel, and overlay the multi-channel echo data to obtain an overlay image of each profile; and a disease location positioning module 205, configured to locate the disease within the neighborhood of the suspected disease location based on the overlay image of each profile.

[0097] One or more embodiments of the present invention also provide an electronic device that can be used to implement the methods in the above embodiments. The electronic device includes one or more processors, one or more memories coupled to the processors, and a communication module coupled to the processors.

[0098] The memory in this embodiment of the invention is used to store various types of data to support, for example... Figure 1 The execution of the method shown.

[0099] It is understood that the memory can be volatile memory or non-volatile memory, or it can include both volatile and non-volatile memory. The memory in this embodiment of the invention is capable of storing, for example... Figure 1The computer programs corresponding to each step in the method shown are as follows. The operating system contains various system programs, such as the framework layer, core library layer, and driver layer, used to implement various basic business functions and handle hardware-based tasks. Application programs can contain various other applications.

[0100] As an example, a processor can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., where a general-purpose processor can be a microprocessor or any conventional processor, etc.

[0101] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including functions for executing... Figure 1 The program code for the method shown. In such an embodiment, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by the central processing unit, it performs the various functions defined in the apparatus of this application.

[0102] in, Figure 1 The computer program instructions corresponding to the method shown may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in the process. Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0103] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for locating defects in phased array ground-penetrating radar, characterized in that, Includes the following steps: Multi-channel raw echo data was acquired through broadband scanning and transmission focusing, and the locations of suspected defects were initially identified. Based on multi-channel raw echo data, a multi-layer dielectric model inversion is performed on each profile. The multi-layer dielectric model includes the dielectric constant and interface depth field of the multi-layer dielectric. The inversion of a multi-layer dielectric model based on multi-channel raw echo data includes: pre-setting a candidate set of dielectric constants; using the candidate set of dielectric constants as the optimization space, for each profile, based on a pre-constructed focusing quality evaluation function, with the goal of optimizing the multi-channel focusing effect, optimizing the dielectric constant of each layer starting from the surface layer to obtain the multi-layer dielectric model; Based on the multilayer dielectric model, the propagation path and arrival time of electromagnetic waves from each channel to the suspected lesion location are solved to obtain the delay time of each channel. After time alignment of the echo data of each channel based on the delay time of each channel, the arrival angle of the receiving channel is determined based on the direction of the last segment of the transmission path from the suspected lesion location to the receiving antenna. The polarization synthesis unit vector is determined based on the arrival angle, and the original signal vectors collected by the dual-polarization receiving array are projected and superimposed on the synthesis direction. The echo data from each channel are time-aligned based on the delay time of each channel, and the echo data from multiple channels are superimposed to obtain a superimposed image of each profile; based on the superimposed image of each profile, the disease is located in the neighborhood of the suspected disease location, including: Based on the superimposed imaging images of each profile, a three-dimensional mesh is divided within the neighborhood of the suspected disease location; For each grid cell within the neighborhood, calculate the coherent sum across all channels; Grids whose coherent sum across all channels is greater than or equal to the disease determination threshold are recorded as disease points. Based on the superimposed images of each profile, the disease can be located within the neighborhood of the suspected disease location.

2. The phased array ground-penetrating radar defect location method as described in claim 1, characterized in that, Specific spatial locations at different underground depths are used as the objects for optimizing the dielectric constant. The neighborhood of each specific spatial location is used as the evaluation region. The focus quality evaluation function is a weighted sum of image entropy, coherence aggregation degree, and sidelobe energy ratio within the evaluation region. The dielectric constant that maximizes the value of the focus quality evaluation function is taken as the dielectric constant of that spatial location.

3. The phased array ground penetrating radar disease localization method of claim 1, wherein, For each channel, the propagation path is determined through the following process: Based on the established multi-layer medium model, the propagation path of electromagnetic waves from the transmission channel to the suspected disease location is initialized. The intersection of this path with the interface of each underground medium layer is the initial refraction point. Based on the initial refraction point, define the optimization range of the refraction point on the interface of each medium layer; Based on the dielectric constant of each layer and the length of each path between layers, the propagation time of the electromagnetic wave in each layer is calculated, and the total propagation time is obtained by summing them. With the total time as the objective, the position of the refraction point is iteratively optimized within the optimization range of each interface refraction point to obtain the actual propagation path and propagation time of electromagnetic waves from each channel to the suspected disease location.

4. The phased array ground penetrating radar disease localization method of claim 1, wherein, The method further includes: using the obtained diseased area as a candidate target, controlling the phased array to scan at multiple scanning deflection angles; obtaining the local focusing response area of ​​the corresponding candidate target at each scanning angle as the candidate target area; and performing spatial consistency evaluation on the positions of multiple candidate targets to determine the authenticity of the disease.

5. A phased array ground penetrating radar disease locating device, characterized by, include: The raw data acquisition module is configured to acquire multi-channel raw echo data through broadband scanning and transmission focusing, and to preliminarily identify the location of suspected defects; Based on the superimposed images of various profiles, the localization of diseases within the neighborhood of suspected disease locations includes: Based on the superimposed imaging images of each profile, a three-dimensional mesh is divided within the neighborhood of the suspected disease location; For each grid cell within the neighborhood, calculate the coherent sum across all channels; Grids whose coherent sum across all channels is greater than or equal to the disease determination threshold are recorded as disease points. The dielectric model inversion module is configured to perform multi-layer dielectric model inversion on each profile based on multi-channel raw echo data. The multi-layer dielectric model includes the dielectric constant and interface depth field of the multi-layer dielectric. The propagation delay calculation module is configured to solve the propagation path and arrival time of electromagnetic waves from each channel to the suspected lesion location based on the multilayer medium model, and obtain the delay time of each channel; after time alignment of the echo data of each channel based on the delay time of each channel, the arrival angle of the receiving channel is determined based on the direction of the last segment of the transmission path from the suspected lesion location to the receiving antenna, and the polarization synthesis unit vector is determined based on the arrival angle. The original signal vectors collected by the dual-polarization receiving array are projected and superimposed on the synthesis direction. The data alignment and overlay module is configured to time-align the echo data of each channel according to the delay time of each channel, and overlay the multi-channel echo data to obtain an overlay image of each profile; based on the overlay image of each profile, the module performs disease localization within the neighborhood of the suspected disease location, including: Based on the superimposed imaging images of each profile, a three-dimensional mesh is divided within the neighborhood of the suspected disease location; For each grid cell within the neighborhood, calculate the coherent sum across all channels; Grids whose coherent sum across all channels is greater than or equal to the disease determination threshold are recorded as disease points. The disease location module is configured to locate the disease within the neighborhood of the suspected disease location based on the superimposed imaging images of each profile.

6. An electronic device, the electronic device comprising: one or more processors; and a memory; wherein, The memory stores one or more computer programs, the one or more computer programs including instructions, characterized in that, when the instructions are executed by the electronic device, the electronic device performs the method according to any one of claims 1-4.

7. A computer-readable storage medium having stored therein instructions, the computer-readable storage medium comprising: When the instructions are executed on an electronic device, the electronic device causes the electronic device to perform the method according to any one of claims 1-4.

Images