CN116224265A - Ground penetrating radar data inversion method and device, computer equipment and storage medium - Google Patents
Ground penetrating radar data inversion method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN116224265A CN116224265A CN202211641636.5A CN202211641636A CN116224265A CN 116224265 A CN116224265 A CN 116224265A CN 202211641636 A CN202211641636 A CN 202211641636A CN 116224265 A CN116224265 A CN 116224265A
- Authority
- CN
- China
- Prior art keywords
- training
- dielectric constant
- model
- ground penetrating
- network model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
- G01S7/418—Theoretical aspects
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V3/00—Electric or magnetic prospecting or detecting; Measuring magnetic field characteristics of the earth, e.g. declination, deviation
- G01V3/12—Electric or magnetic prospecting or detecting; Measuring magnetic field characteristics of the earth, e.g. declination, deviation operating with electromagnetic waves
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Radar, Positioning & Navigation (AREA)
- Computer Networks & Wireless Communication (AREA)
- Environmental & Geological Engineering (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geophysics (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The application provides a ground penetrating radar data inversion method, a device, computer equipment and a storage medium, comprising the following steps: receiving ground penetrating radar data; invoking a preset one-stage network model to perform inversion operation on the ground penetrating radar data to obtain a first dielectric constant model; and calling a preset two-stage network model to perform inversion operation on the ground penetrating radar data and the first dielectric constant model to obtain a second dielectric constant model. The method and the device ensure the accuracy of the first dielectric constant model, modify the wrong structural characteristics of the first dielectric constant model, obtain the technical effect of the second dielectric constant model which is more accurate, greatly improve the operation efficiency of inversion of the ground penetrating radar data and reduce the consumption of calculation resources.
Description
Technical Field
The application relates to the technical field of artificial intelligence, in particular to a ground penetrating radar data inversion method, a ground penetrating radar data inversion device, computer equipment and a storage medium.
Background
The ground penetrating radar data inversion is: solving the pathological problem that the ground penetrating radar data and the dielectric constant model with at least one relative dielectric constant spatially correspond to each other, wherein the pathological problem refers to the problem that an output result is very sensitive to input data, namely, a tiny error in the input data can cause great change of the output result, the condition index of the problem is generally measured by a condition number, and the condition number is larger, and the pathological degree of the problem is heavier.
However, current ground penetrating radar data inversion typically employs the full waveform inversion (FWI, full waveform inversion) method of conventional ground penetrating radars for qualitative and quantitative reconstruction of solutions to the formation structure images. It directly uses the entire received waveform to match the emulated GPR data. It then reconstructs the dielectric profile of the structure by minimizing the mismatch between the two sets of data.
However, the inventor has found that conventional FWI generally uses an iterative manner to reduce the error between the analog data and the ground penetrating radar data, and thus, it takes a lot of computing resources to complete one FWI job; moreover, the FWI method is difficult to extract the characteristic value with a sensitive relation with the dielectric constant model from the ground penetrating radar data, so that the inversion accuracy of the FWI method is low.
Disclosure of Invention
The application provides a ground penetrating radar data inversion method, a device, computer equipment and a storage medium, which are used for solving the problems that a large amount of calculation resources are required to be consumed by a traditional FWI method, and a characteristic value with a sensitive relation with a dielectric constant model is difficult to extract from the ground penetrating radar data by the FWI method, so that the inversion precision of the FWI method is low.
In a first aspect, the present application provides a ground penetrating radar data inversion method, including:
receiving ground penetrating radar data; the ground penetrating radar data are waveform data of electromagnetic waves for carrying out nondestructive detection on the shallow earth surface of the target area;
invoking a preset one-stage network model to perform inversion operation on the ground penetrating radar data to obtain a first dielectric constant model; wherein the first permittivity model is a factor determining a propagation speed of the electromagnetic wave in a stratum structure of the target area; the first dielectric constant model has at least one relative dielectric constant; the relative permittivity is a physical parameter that characterizes the dielectric properties of the formation structure;
invoking a preset two-stage network model to perform inversion operation on the ground penetrating radar data and the first dielectric constant model to obtain a second dielectric constant model; the second dielectric constant model comprises physical parameters after the relative dielectric constant in the first dielectric constant model is corrected according to the ground penetrating radar data.
In the above scheme, before the invoking the preset one-stage network model to perform inversion operation on the ground penetrating radar data, the method further includes:
Acquiring a first training sample;
and training through a first initial network model preset by the first training sample to obtain the one-stage network model.
In the above scheme, before the invoking the preset two-stage network model to perform inversion operation on the ground penetrating radar data and the first dielectric constant model, the method further includes:
acquiring a second training sample;
and training a preset second initial network model through the second training sample and the first-stage network model to obtain the second-stage network model.
In the above aspect, before the obtaining the first training sample, the method further includes:
acquiring a stratum structure, embedding irregular blocks in the stratum structure to convert the stratum structure into a training dielectric constant model, and performing forward modeling according to the training dielectric constant model to obtain training ground penetrating radar data;
summarizing one training dielectric constant model and the training ground penetrating radar data to form training data;
and summarizing a plurality of training data to obtain a training set.
In the above scheme, the acquiring the first training sample includes:
m training data are obtained from the training set, and the M training data are summarized to obtain the first training sample; wherein M is a positive integer, M is more than or equal to 1;
The obtaining a second training sample includes:
acquiring N training data from the training set, and summarizing the N training data to obtain the second training sample; wherein N is a positive integer, and N is more than or equal to 1.
In the above solution, training by using a first initial network model preset by the first training sample to obtain the first-stage network model includes:
the ground penetrating radar data of training data in the first training sample are used as first input information of the first initial network model, and the first initial network model is operated to carry out inversion operation on the first input information to obtain first output information;
the training dielectric constant model of training data in the first training sample is used as first reference information of the first initial network model, and a first loss value is generated according to the first output information and the first reference information through a preset first loss function; wherein the first loss value characterizes a degree of difference between the first output information and the first reference information;
iterating the first initial network model according to the first loss value through a preset optimizing model to adjust the weight of a hidden layer in the first initial network model, so that the first loss value between the first output information generated by the first initial network model and the first reference information is in a preset first threshold interval, and setting the first initial network model after iteration as the first-stage network model.
In the above scheme, training the preset second initial network model to obtain the two-stage network model through the second training sample and the one-stage network model includes:
taking the training ground penetrating radar data of the training data in the second training sample as second input information, operating the first-stage network model to perform inversion operation on the second input information to obtain first-stage output information, and operating the second initial network model to perform inversion operation on the first-stage output information to obtain second output information;
the training dielectric constant model of training data in the second training sample is used as second reference information of the second initial network model, and a second loss value is generated according to the second output information and the second reference information through a preset second loss function; wherein the second loss value characterizes a degree of difference between the second output information and the second reference information;
and iterating the second initial network model according to the second loss value through a preset optimizing model to adjust the weight of a hidden layer in the second initial network model, so that the second loss value between second output information generated by the second initial network model and the second reference information is in a preset second threshold interval, and setting the iterated second initial network model as the two-stage network model.
In a second aspect, the present application provides a ground penetrating radar data inversion apparatus, comprising:
the input module is used for receiving ground penetrating radar data; the ground penetrating radar data are waveform data of electromagnetic waves for carrying out nondestructive detection on the shallow earth surface of the target area;
the first inversion module is used for calling a preset one-stage network model to perform inversion operation on the ground penetrating radar data to obtain a first dielectric constant model; wherein the first permittivity model is a factor determining a propagation speed of the electromagnetic wave in a stratum structure of the target area; the first dielectric constant model has at least one relative dielectric constant; the relative permittivity is a physical parameter that characterizes the dielectric properties of the formation structure;
the second inversion module is used for calling a preset two-stage network model to perform inversion operation on the ground penetrating radar data and the first dielectric constant model to obtain a second dielectric constant model; the second dielectric constant model comprises physical parameters after the relative dielectric constant in the first dielectric constant model is corrected according to the ground penetrating radar data.
In a third aspect, the present application provides a computer device comprising: a processor and a memory communicatively coupled to the processor;
The memory stores computer-executable instructions;
the processor executes the computer-executable instructions stored in the memory to implement the ground penetrating radar data inversion method as described in the claims.
In a fourth aspect, the present application provides a computer readable storage medium having stored therein computer executable instructions that when executed by a processor are configured to implement the above-described ground penetrating radar data inversion method.
In a fifth aspect, the present application provides a computer program product comprising a computer program which, when executed by a processor, implements the above-described ground penetrating radar data inversion method.
According to the ground penetrating radar data inversion method, the device, the computer equipment and the storage medium, inversion operation is carried out on the ground penetrating radar data by calling a preset one-stage network model, and the first dielectric constant model is obtained according to the characteristic value by extracting the characteristic value with a sensitive relation between the ground penetrating radar data and the dielectric constant model so as to ensure the accuracy of the first dielectric constant model.
And carrying out inversion operation on the ground penetrating radar data and the first dielectric constant model by calling a preset two-stage network model so as to modify the structural characteristics of errors in the first dielectric constant model and obtain the technical effect of a more accurate second dielectric constant model, thereby obtaining the second dielectric constant model based on the first dielectric constant model.
And inversion operation is carried out on the ground penetrating radar data by using a network model, so that the operation efficiency of the inversion of the ground penetrating radar data is greatly improved, and the consumption of calculation resources is reduced.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description, serve to explain the principles of the application.
Fig. 1 is a schematic view of an application scenario provided in an embodiment of the present application;
fig. 2 is a flowchart of an embodiment 1 of a ground penetrating radar data inversion method provided in an embodiment of the present application;
FIG. 3 is an image of a second dielectric constant model in a method for inversion of ground penetrating radar data according to an embodiment of the present application;
fig. 4 is a flowchart of embodiment 2 of a ground penetrating radar data inversion method provided in an embodiment of the present application;
fig. 5 is an image of the training dielectric constant model in embodiment 2 of a ground penetrating radar data inversion method according to the embodiment of the present application;
fig. 6 is an image of the training ground penetrating radar data in embodiment 2 of a ground penetrating radar data inversion method according to the embodiment of the present application;
fig. 7 is a schematic structural diagram of a first initial network model in embodiment 2 of a ground penetrating radar data inversion method according to an embodiment of the present application;
FIG. 8 is a graph of a relationship between a first loss value and a number of iterations of a first initial network model during training;
fig. 9 is a schematic structural diagram of a second initial network model in embodiment 2 of a ground penetrating radar data inversion method according to an embodiment of the present application;
FIG. 10 is a graph of the relationship between the second loss value and the number of iterations of the second initial network model during training;
FIG. 11 is a schematic diagram of a program module of a ground penetrating radar data inversion device provided by the invention;
fig. 12 is a schematic diagram of a hardware structure of a computer device in the computer device according to the present invention.
Specific embodiments thereof have been shown by way of example in the drawings and will herein be described in more detail. These drawings and the written description are not intended to limit the scope of the inventive concepts in any way, but to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with some aspects of the present application as detailed in the accompanying claims.
The application scene is as follows:
ground penetrating radar (Ground Penetrating radar. Gpr) is a shallow earth surface geophysical prospecting technique for nondestructive detection using electromagnetic waves, and has been widely used in many fields including glacial, archaeological and geotechnical engineering fields. GPR is capable of converting electromagnetic information in a subsurface medium into information related to characteristics of the geological medium (e.g., position, shape, and dielectric properties), which is important in the geological exploration and analysis of shallow surfaces.
The focus of the existing GPR inversion method is to generally infer the position, size and dimension of the detection target body according to the observed GPR data. Conventional Full Waveform Inversion (FWI) of GPR is considered a solution to qualitatively and quantitatively reconstructing images of the formation structure. It directly uses the entire received waveform to match the emulated GPR data. It then reconstructs the dielectric profile of the structure by minimizing the mismatch between the two sets of data. FWI originates in the field of seismic exploration and is thereafter rapidly used to process radar data. However, since the actual formation structure always has irregular geometric features and complex distribution patterns, the received subsurface ground penetrating radar data is typically interleaved and clutter with discontinuous and distorted echoes. In addition, the characteristics of radar signals are masked by a plurality of reflections caused by abnormal bodies in the underground medium, and the characteristics of signals with disordered shapes are often present in images. On the other hand, conventional FWI generally uses an iterative manner to reduce the error between the analog data and the ground penetrating radar data, and thus, it takes a lot of computing resources to complete one FWI job. Therefore, the conventional FWI not only consumes a large amount of computing resources, but also has a great room for improvement in the accuracy of the inversion model.
In recent years, deep Neural Networks (DNNs) have been rapidly developed in seismic denoising, signal processing, geophysical inversion. DNNs automatically learn advanced features through training data and are then able to estimate the nonlinear mapping between the input image data and the various data fields. With the rapid development of the deep learning technology, the breadth and depth of intelligent inversion work in the field of geology are also expanding. Convolutional Neural Networks (CNNs) are utilized to predict high resolution impedance. Li et al have realized super-resolution speed image prediction using a multitasking approach. And the two-dimensional inversion imaging of GPR and earthquake is realized through end-to-end learning mode resolution.
To date, DNN-based GPR inversion work has had some progress, but there is still significant room for development. A difficulty in reconstructing the complex structure of electrical properties of subsurface media is extracting valid features from complex GPR data and preserving spatial alignment between input and output.
The present application proposes to use an incremental learning approach to predict subsurface dielectric constant models. The one-stage network extracts the initial dielectric constant model from the GPR data in an end-to-end fashion. And then a two-stage network model with two channels is built, the initial dielectric constant model is regarded as prior information, and inversion prediction is carried out by combining GPR data as input. The one-stage network can extract signal features of GPR from adjacent tracks, and the two-stage network can effectively modify wrong structural features in the dielectric constant model. The method combines the idea of incremental learning, and provides a new algorithm for a learning mechanism in a GPR inversion task.
Specifically, referring to fig. 1, the present application proposes a server 2 running a ground penetrating radar data inversion method, by receiving ground penetrating radar data from a ground penetrating radar device 3; the ground penetrating radar data are waveform data of electromagnetic waves for carrying out nondestructive detection on the shallow surface of the target area; invoking a preset one-stage network model to perform inversion operation on the ground penetrating radar data to obtain a first dielectric constant model; the first dielectric constant model is a factor for determining the propagation speed of electromagnetic waves in a stratum structure of a target area; the first dielectric constant model has at least one relative dielectric constant; the relative permittivity is a physical parameter that characterizes the dielectric properties of the formation structure;
and invoking a preset two-stage network model to perform inversion operation on the ground penetrating radar data and the first dielectric constant model to obtain a second dielectric constant model; the second dielectric constant model comprises physical parameters obtained by correcting the relative dielectric constant in the first dielectric constant model according to the ground penetrating radar data.
Thus, the present application uses an incremental learning based approach to achieve GPR inversion. GPR inversion is to solve a pathological problem of space correspondence between ground penetrating radar data and a dielectric constant model, and the GPR inversion task is realized by learning new knowledge from the GPR data for a plurality of times. The inversion method used by the method not only can effectively improve the GPR inversion precision, but also has the capabilities of suppressing false anomalies and recovering deep stratum structures.
According to the method, accurate analysis of ground penetrating radar data is achieved through the first-stage network model and the second-stage network model, and the second dielectric constant model with the relative dielectric constant capable of accurately representing the dielectric property of the stratum structure is obtained, wherein the dielectric property refers to geological information showing storage and loss of electrostatic energy by the stratum structure under the action of an electric field.
Therefore, according to the second dielectric constant model, analysis and extraction of the characteristic value of the dielectric property are carried out on the ground penetrating radar data, so that geological information is obtained, and accurate analysis and extraction of the characteristic value of the ground penetrating radar data are realized.
The following describes the technical solutions of the present application and how the technical solutions of the present application solve the prior art problems in detail with specific embodiments. The following embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments. Embodiments of the present application will be described below with reference to the accompanying drawings.
Example 1:
referring to fig. 2, the present application provides a ground penetrating radar data inversion method, which includes:
s101: receiving ground penetrating radar data; the ground penetrating radar data are waveform data of electromagnetic waves for performing nondestructive detection on the shallow surface of the target area.
In this step, a ground penetrating radar (Ground Penetrating radar. Gpr) is a geophysical method for detecting the characteristics and distribution of substances inside a medium by transmitting and receiving high-frequency electromagnetic waves through an antenna. Ground penetrating radar data refers to GPR data, which is a shallow earth surface geophysical exploration technology using electromagnetic waves for nondestructive detection, and is widely applied to various fields including glacial, archaeological and geotechnical engineering fields. GPR data is capable of converting electromagnetic information in subsurface media into information related to characteristics of the geological media (e.g., location, shape, and dielectric properties), which is important in the geological exploration and analysis of shallow surfaces.
S102: invoking a preset one-stage network model to perform inversion operation on the ground penetrating radar data to obtain a first dielectric constant model; the first dielectric constant model is a factor for determining the propagation speed of electromagnetic waves in a stratum structure of a target area; the first dielectric constant model has at least one relative dielectric constant; the relative permittivity is a physical parameter that characterizes the dielectric properties of the formation structure.
In the embodiment, inversion operation is performed on the ground penetrating radar data by calling a one-stage network model to obtain a first dielectric constant model, so that the stratum structure of the target area is predicted according to the ground penetrating radar data, and the first dielectric constant model representing the stratum structure is obtained. The first-stage network model extracts a first dielectric constant model from the ground penetrating radar data in an end-to-end mode.
The relative permittivity (relative permittivity) is a physical parameter that characterizes the dielectric properties of a dielectric material. Dielectric properties refer to properties that exhibit storage and loss of electrostatic energy under the action of an electric field, and are generally expressed in terms of dielectric constant and dielectric loss.
Optionally, the one-stage network model can extract signal features of the GPR from the adjacent tracks, and further obtain the first dielectric constant model through the signal features.
In this embodiment, the first-stage network model extracts the eigenvalue of the dielectric property in the ground penetrating radar data, performs inversion operation according to the eigenvalue to obtain at least one relative dielectric constant, and generates the first dielectric constant model according to the at least one relative dielectric constant, so as to ensure the accuracy of the first node constant model.
S103: invoking a preset two-stage network model to perform inversion operation on the ground penetrating radar data and the first dielectric constant model to obtain a second dielectric constant model; the second dielectric constant model comprises physical parameters obtained by correcting the relative dielectric constant in the first dielectric constant model according to the ground penetrating radar data.
In the embodiment, the first dielectric constant model is regarded as prior information by constructing a two-channel two-stage network model, and the first dielectric constant model and the ground penetrating radar data are input as the two-stage network model, so that the two-stage network model can adjust the first dielectric constant model according to the ground penetrating radar data (namely, GPR data represent the original waveform of the shallow surface of the target area) to realize inversion prediction, wherein the two-stage network model can effectively modify the wrong structural characteristics in the first dielectric constant model, and obtain the technical effect of a more accurate second dielectric constant model. The method combines the idea of incremental learning, and provides a new algorithm for a learning mechanism in the ground penetrating radar data inversion task. The second dielectric constant model is shown in fig. 3, wherein the vertical axis on the left side in fig. 3 is the depth of the formation structure, the horizontal axis is the position of the target region, and the vertical axis on the right side is the relative dielectric constant of the second dielectric constant model.
Specifically, the subsurface dielectric constant model is predicted by using an incremental learning manner. The first dielectric constant model is extracted from the GPR data by a one-stage network model in an end-to-end manner. And then a two-stage network model with two channels is built, the first dielectric constant model is regarded as prior information, and inversion prediction is carried out by combining GPR data as input. The one-stage network model can extract signal characteristics of GPR from adjacent tracks, and the two-stage network model can effectively modify wrong structural characteristics in the dielectric constant model. The method combines the idea of incremental learning, and provides a new algorithm for a learning mechanism in the ground penetrating radar data inversion task.
The method and the device realize inversion of the ground penetrating radar data by using a mode based on incremental learning. The ground penetrating radar data inversion is to solve the pathological problem of the space correspondence between the ground penetrating radar data and the dielectric constant model, and the ground penetrating radar data inversion task is realized by learning new knowledge from the GPR data for a plurality of times. The inversion method used by the method not only can effectively improve the inversion precision of the data of the ground penetrating radar, but also has the capabilities of suppressing false anomalies and recovering deep stratum structures.
Therefore, constraint of the inversion process is achieved by using the one-stage network model prediction results as prior information of the two-stage network model. Compared with the traditional FWI algorithm, the method has the advantages that inversion accuracy is effectively improved, and proper calculation efficiency can be ensured. Compared with the ground penetrating radar data inversion algorithm based on deep learning at the present stage, the method has the advantages that the prediction result is better matched with the model parameters, and the structural distribution of the deep layer can be well reflected.
In this embodiment, the first-stage network model characterizes the characteristic value of the dielectric property in the stratum structure by extracting the ground penetrating radar data, and creates the first dielectric constant model according to the characteristic value, so as to ensure the accuracy of the first node constant model.
The two-stage network-free model is used for further modifying the relative dielectric constants in the first dielectric constant model by comparing the first dielectric constant model and detecting all characteristic values in radar data to obtain a second dielectric constant model, so that the relative dielectric constants in the second dielectric constant model are more matched with the ground penetrating radar data, and the accuracy of each relative dielectric constant in the second dielectric constant model is further ensured.
Meanwhile, in the embodiment, the first-stage network model and the second-stage network model are called to directly operate the ground penetrating radar data, and the error between the simulation data and the ground penetrating radar data is reduced in a mode of multiple iterations as in the FWI method in the prior art, so that the situation that a large amount of computing resources are consumed when the FWI work is completed once occurs, and compared with the prior art, the consumption of the computing resources is greatly reduced.
Example 2:
referring to fig. 4, the present application provides a ground penetrating radar data inversion method, which includes:
s201: receiving ground penetrating radar data; the ground penetrating radar data are waveform data of electromagnetic waves for performing nondestructive detection on the shallow surface of the target area.
This step is the same as S101 in embodiment 1, and thus will not be described here.
S202: acquiring a stratum structure, embedding irregular blocks in the stratum structure to convert the stratum structure into a training dielectric constant model, and performing forward modeling according to the training dielectric constant model to obtain training ground penetrating radar data;
summarizing a training dielectric constant model and training ground penetrating radar data to form training data;
and summarizing a plurality of training data to obtain a training set.
In this example, by generating data similar to the actual ground penetrating radar data scene characteristics as training data, the capability of the network for solving the actual problem is effectively improved, as shown in fig. 5, a training dielectric constant model is generated randomly based on exploration conditions and a mathematical method, the training dielectric constant model comprises a randomly simulated undulating stratum structure, dielectric parameters from top to bottom become larger gradually, a random equivalent medium technology is used for describing the non-uniformity of stratum medium, an irregular block is simulated to be embedded into the stratum by combining with an abrasive grain technology, and the medium parameter range in the model is generated randomly in a reasonable interval.
Local random feature anomalies are considered. The application adopts a non-uniformity mixed function random medium modeling method to simulate the horizon structure, and the function is expressed by the following formula:
wherein r represents a fuzzy factor, a, b and c represent autocorrelation lengths in x, y and z directions, and various training dielectric constant models in different forms can be constructed by selecting local disturbance radii (a, b and c) and local disturbance intensities (r), so that diversity of the training dielectric constant models is realized.
After obtaining the training dielectric constant model, forward modeling of the training ground penetrating radar data (GPR) was performed using FDTD algorithm, as shown in fig. 6. The dielectric constant model size is 4.5m×4m, the cell grid size is 0.025m×0.025m, the sampling time interval is 0.0201ns, and the main frequency of the transmitting antenna is 650MHz. The sample set used in this application contains 1000 pairs of ground penetrating radar data and a dielectric constant model.
It should be noted that, the random equivalent medium technology is to simulate a deposit structure, embed the deposit structure into a random fluctuating stratum structure to obtain a geological model containing the deposit structure, convert the geological model into a resistivity model based on the existing geological data, obtain a sample data set according to the resistivity model, wherein the sample data set comprises the random simulated fluctuating stratum structure data, deposit structure data and corresponding resistivity data, and train an initial resistivity model reconstruction network based on the sample data set to obtain a resistivity model reconstruction network, and the resistivity model reconstruction network is used for performing deep learning on the first electromagnetic inversion data to obtain the second electromagnetic inversion data.
Abrasive grain techniques are computer algorithms that generate a three-dimensional array based on matlab or python, and from that three-dimensional array, an irregular shape.
The FDTD algorithm is a Finite-difference-time-Domain (FDTD) method, which is a common method in the field of electromagnetic field calculation. The model basis for the time-domain finite difference method is Maxwell's equations (Maxwell's equations) that are the most basic in electrodynamics. After the FDTD method is proposed, along with the development of computing technology, particularly electronic computer technology, the FDTD method has been developed to a great extent, and has been widely used in fields of electromagnetism, electronics, optics, and the like.
S203: and acquiring a first training sample, and training through a first initial network model preset by the first training sample to obtain a one-stage network model.
In a preferred embodiment, obtaining the first training sample comprises:
m training data are obtained from the training set, and the M training data are summarized to obtain a first training sample; wherein M is a positive integer, M is more than or equal to 1.
In this example, through the preset number M, M training data are obtained from the training set and summarized to obtain the first training sample, so that the controllability of the number of training data in the sample is ensured.
In a preferred embodiment, training is performed by a first initial network model preset by a first training sample to obtain a first-stage network model, including:
training ground penetrating radar data of training data in a first training sample are used as first input information of a first initial network model, and the first initial network model is operated to carry out inversion operation on the first input information to obtain first output information;
the training dielectric constant model of training data in the first training sample is used as first reference information of a first initial network model, and a first loss value is generated according to first output information and first reference information through a preset first loss function; wherein the first loss value characterizes a degree of difference between the first output information and the first reference information;
Iterating the first initial network model according to the first loss value through a preset optimizing model to adjust the weight of a hidden layer in the first initial network model, so that the first loss value between the first output information generated by the first initial network model and the first reference information is in a preset first threshold interval, and setting the iterated first initial network model as a one-stage network model.
In this example, a deep neural network is used as a first initial network model, as shown in fig. 7, which uses a downsampled codec structure. The network performs four downsampling operations, which are implemented by a convolutional layer with a "stride" parameter of 2. In the network, 4 groups of characteristic diagrams with different scales are shared, and the size ratios are respectively 8:4:2:1. the decoding layer is similar to the encoding layer, deconvolution is performed before 4 repeated structures form each repeated structure, the number of characteristic channels after each deconvolution is halved, and the size of the characteristic map is doubled. After deconvolution, the deconvolution results are spliced with the feature maps of the corresponding steps of the encoded portion. The convolution kernel of the last layer is a 1x1 convolution kernel, which converts the 64-channel feature map into a specific class number of results. The input of the one-stage network model is single channel (ground penetrating radar data), the input of the two-stage network model is double channel (the output of the one-stage network model and the ground penetrating radar data), and the outputs of the two networks are both dielectric constant models.
The first initial network model uses a mean square error MSE as a loss function of the deep neural network, uses an Adam optimizer as an optimization model to train the first initial network model, and adopts an attenuation learning rate in the training process. After training data are input, training times, loss function values and learning rate are recorded, and network parameters are stored in a specific file. And integrating the one-stage network model prediction model with the GPR data, and then putting the integrated GPR data into the training of the two-stage network model for prediction.
The expression of the first loss function is:
where loss_1 is a first loss value, y is first output information, r is first reference information, and 1/n represents a mean value. Training data is input, an Adam optimizer is used for training, and an attenuation learning rate is adopted in the training process. As shown in fig. 8, the vertical axis of the graph is a first loss value loss_1, and the horizontal axis is an iteration number Epoch of the first initial network model, where fig. 8 includes a training set curve formed by a first loss value of a training set in an iteration process in a first training sample, and a verification set curve formed by a first loss value of a verification set in the iteration process in the first training sample.
The learning rate is determined by the initial learning rate eta 0 The three parameters of the decay period T and the decay rate alpha are characterized, and the real-time learning rate expression in the training process is as follows
η i ＝α i η 0
Where i is the current learning rate decay times. The training times, the loss function value and the learning rate are recorded in the text file, and the network parameters are stored in the specific file.
It should be noted that the deep neural network is a technology in the field of machine learning (ML, machineLearning). The advantage of multiple layers is that complex functions can be represented with fewer parameters. In supervised learning, the problem with previous multi-layer neural networks is the tendency to trap local extremum points. If the training samples are sufficient to cover future samples, the learned multi-layer weights can be used well to predict new test samples.
S204: acquiring a second training sample; and training a preset second initial network model through the second training sample and the first-stage network model to obtain a second-stage network model.
In a preferred embodiment, obtaining the second training sample comprises:
acquiring N training data from the training set, and summarizing the N training data to obtain a second training sample; wherein N is a positive integer, and N is more than or equal to 1.
In this example, through the preset number N, N training data are obtained from the training set and summarized to obtain the first training sample, so that the controllability of the number of the training data in the sample is ensured.
Preferably, training the preset second initial network model through the second training sample and the first-stage network model to obtain a second-stage network model, including:
taking the training ground penetrating radar data of the training data in the second training sample as second input information, operating a first-stage network model to perform inversion operation on the second input information to obtain first-stage output information, and operating a second initial network model to perform inversion operation on the first-stage output information to obtain second output information;
the training dielectric constant model of training data in the second training sample is used as second reference information of a second initial network model, and a second loss value is generated according to second output information and second reference information through a preset second loss function; wherein the second loss value characterizes a degree of difference between the second output information and the second reference information;
iterating the second initial network model according to the second loss value through a preset optimizing model to adjust the weight of a hidden layer in the second initial network model, so that the second loss value between second output information generated by the second initial network model and second reference information is in a preset second threshold interval, and setting the iterated second initial network model as a two-stage network model.
In this example, a deep neural network is used as a second initial network model, as shown in fig. 9, which uses a downsampled codec structure. The network performs four downsampling operations, which are implemented by a convolutional layer with a "stride" parameter of 2. In the network, 4 groups of characteristic diagrams with different scales are shared, and the size ratios are respectively 8:4:2:1. the decoding layer is similar to the encoding layer, deconvolution is performed before 4 repeated structures form each repeated structure, the number of characteristic channels after each deconvolution is halved, and the size of the characteristic map is doubled. After deconvolution, the deconvolution results are spliced with the feature maps of the corresponding steps of the encoded portion. The convolution kernel of the last layer is a 1x1 convolution kernel, which converts the 64-channel feature map into a specific class number of results. The input of the one-stage network model is single channel (ground penetrating radar data), the input of the two-stage network model is double channel (the output of the one-stage network model and the ground penetrating radar data), and the outputs of the two networks are both dielectric constant models.
The second initial network model uses a mean square error MSE as a loss function of the deep neural network, uses an Adam optimizer as an optimization model to train the second initial network model, and adopts an attenuation learning rate in the training process. After training data are input, training times, loss function values and learning rate are recorded, and network parameters are stored in a specific file. And integrating the one-stage network model prediction model with the GPR data, and then putting the integrated GPR data into the training of the two-stage network model for prediction.
The expression of the second loss function is:
where loss_2 is a second loss value, x is second output information, r is second reference information, and 1/n represents a mean value. Training data is input, an Adam optimizer is used for training, and an attenuation learning rate is adopted in the training process. As shown in fig. 10, the vertical axis is the second loss value loss_2, and the horizontal axis is the number of iterations Epoch of the second initial network model.
The learning rate is determined by the initial learning rate eta 0 The three parameters of the decay period T and the decay rate alpha are characterized, and the real-time learning rate expression in the training process is as follows
η i ＝α i η 0
Where i is the current learning rate decay times. The training times, the loss function value and the learning rate are recorded in the text file, and the network parameters are stored in the specific file.
It should be noted that the deep neural network is a technology in the field of machine learning (ML, machineLearning). The advantage of multiple layers is that complex functions can be represented with fewer parameters. In supervised learning, the problem with previous multi-layer neural networks is the tendency to trap local extremum points. If the training samples are sufficient to cover future samples, the learned multi-layer weights can be used well to predict new test samples.
S205: invoking a preset one-stage network model to perform inversion operation on the ground penetrating radar data to obtain a first dielectric constant model; the first dielectric constant model is a factor for determining the propagation speed of electromagnetic waves in a stratum structure of a target area; the first dielectric constant model has at least one relative dielectric constant; the relative permittivity is a physical parameter that characterizes the dielectric properties of the formation structure.
The step is the same as S102 in embodiment 1, so that the description is omitted here.
S206: invoking a preset two-stage network model to perform inversion operation on the ground penetrating radar data and the first dielectric constant model to obtain a second dielectric constant model; the second dielectric constant model comprises physical parameters obtained by correcting the relative dielectric constant in the first dielectric constant model according to the ground penetrating radar data.
This step is the same as S103 in embodiment 1, and thus will not be described here.
An experimental result based on the technical scheme comprises:
in evaluating the inversion algorithm used in the present invention, the predicted results of the one-stage network model and the two-stage network model are compared. Compared with a one-stage network prediction result, the two-stage network has higher precision for detecting deep horizons and abnormal targets. The GPR inversion method used by the invention uses the idea of incremental learning as a constraint, and the dielectric constant predicted by a one-stage network is required to be used as the constraint, so that the prediction precision of the two-stage network on the medium model is improved. From the indexes of MSE, PSNR and SSIM, the two-stage network prediction model is obviously improved compared with a one-stage network.
The quantitative comparison of the test results is shown in the following table:
| Algorithm | MSE↓ | PSNR↑ | SSIM↑ |
| One-stage network | 0.1719 | 55.7793 | 0.9957 |
| Two-stage network | 0.0859 | 58.7929 | 0.9979 |
Wherein, MSE: mean-square error (MSE) is a measure reflecting the degree of difference between an estimated quantity and an estimated quantity. Let t be an estimate of the overall parameter θ determined from the subsamples, (θ -t) 2, the mathematical expectation, called the mean square error of the estimate t.
PSNR: peak signal-to-noise ratio (PSNR), is an engineering term that represents the ratio of the maximum possible power of a signal to the destructive noise power affecting its accuracy of representation.
SSIM (Structural Similarity), structural similarity, is an indicator for measuring the similarity of two images. The index was first proposed by the image and video engineering laboratory (Laboratory for Image and Video Engineering) at the university of texas austin. Two images used in SSIM, one being an uncompressed undistorted image and the other being a distorted image.
Example 3:
referring to fig. 11, the present application provides a ground penetrating radar data inversion device 1, including:
an input module 11 for receiving ground penetrating radar data; the ground penetrating radar data are waveform data of electromagnetic waves for carrying out nondestructive detection on the shallow surface of the target area;
The first inversion module 15 is used for calling a preset one-stage network model to perform inversion operation on the ground penetrating radar data to obtain a first dielectric constant model; the first dielectric constant model is a factor for determining the propagation speed of electromagnetic waves in a stratum structure of a target area; the first dielectric constant model has at least one relative dielectric constant; the relative permittivity is a physical parameter that characterizes the dielectric properties of the formation structure;
the second inversion module 16 is configured to invoke a preset two-stage network model to perform inversion operation on the ground penetrating radar data and the first dielectric constant model, so as to obtain a second dielectric constant model; the second dielectric constant model comprises physical parameters obtained by correcting the relative dielectric constant in the first dielectric constant model according to the ground penetrating radar data.
Optionally, the ground penetrating radar data inversion device 1 further includes:
the sample construction module 12 is used for acquiring a stratum structure, embedding irregular blocks in the stratum structure to convert the stratum structure into a training dielectric constant model, and performing forward modeling according to the training dielectric constant model to obtain training ground penetrating radar data; summarizing a training dielectric constant model and training ground penetrating radar data to form training data; and summarizing a plurality of training data to obtain a training set.
The first training module 13 is configured to obtain a first training sample, and perform training through a first initial network model preset by the first training sample to obtain a first-stage network model.
A second training module 14 for acquiring a second training sample; and training a preset second initial network model through the second training sample and the first-stage network model to obtain a second-stage network model.
Example 4:
to achieve the above object, the present application further provides a computer device 4, including: a processor 42 and a memory 41 communicatively connected to the processor 42; the memory stores computer-executable instructions;
the processor executes computer-executable instructions stored in the memory 41 to implement the above-mentioned ground penetrating radar data inversion method, where the components of the ground penetrating radar data inversion apparatus may be distributed in different computer devices, and the computer device 4 may be a smart phone, a tablet computer, a notebook computer, a desktop computer, a rack-mounted server, a blade server, a tower server, or a rack-mounted server (including a stand-alone server, or a server cluster formed by a plurality of application servers) that execute a program, and so on. The computer device of the present embodiment includes at least, but is not limited to: a memory 41, a processor 42, which may be communicatively connected to each other via a system bus, as shown in fig. 12. It should be noted that fig. 12 only shows a computer device with components-but it should be understood that not all of the illustrated components are required to be implemented and that more or fewer components may be implemented instead. In the present embodiment, the memory 41 (i.e., a readable storage medium) includes a flash memory, a hard disk, a multimedia card, a card memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, and the like. In some embodiments, the memory 41 may be an internal storage unit of a computer device, such as a hard disk or a memory of the computer device. In other embodiments, the memory 41 may also be an external storage device of a computer device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like. Of course, the memory 41 may also include both internal storage units of the computer device and external storage devices. In this embodiment, the memory 41 is generally used to store an operating system installed in a computer device and various application software, such as program codes of the ground penetrating radar data inversion apparatus of the third embodiment. In addition, the memory 41 may also be used to temporarily store various types of data that have been output or are to be output. Processor 42 may be a central processing unit (Central Processing Unit, CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments. The processor 42 is typically used to control the overall operation of the computer device. In this embodiment, the processor 42 is configured to execute the program code stored in the memory 41 or process data, for example, execute the ground penetrating radar data inversion device, so as to implement the ground penetrating radar data inversion method of the above embodiment.
The integrated modules, which are implemented in the form of software functional modules, may be stored in a computer readable storage medium. The software functional modules described above are stored in a storage medium and include instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or processor to perform some steps of the methods of the various embodiments of the present application. It should be appreciated that the processor may be a central processing unit (Central Processing Unit, CPU for short), other general purpose processors, digital signal processor (Digital Signal Processor, DSP for short), application specific integrated circuit (Application Specific Integrated Circuit, ASIC for short), etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the present application may be embodied directly in a hardware processor for execution, or in a combination of hardware and software modules in a processor for execution. The memory may comprise a high-speed RAM memory, and may further comprise a non-volatile memory NVM, such as at least one magnetic disk memory, and may also be a U-disk, a removable hard disk, a read-only memory, a magnetic disk or optical disk, etc.
To achieve the above object, the present application further provides a computer readable storage medium such as a flash memory, a hard disk, a multimedia card, a card memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a Read Only Memory (ROM), an Electrically Erasable Programmable Read Only Memory (EEPROM), a Programmable Read Only Memory (PROM), a magnetic memory, a magnetic disk, an optical disk, a server, an App application store, etc., on which computer-executable instructions are stored, which when executed by the processor 42, perform the corresponding functions. The computer readable storage medium of the present embodiment is used to store computer-executable instructions for implementing the ground penetrating radar data inversion method, which when executed by the processor 42 implement the ground penetrating radar data inversion method of the above embodiment.
The storage medium may be implemented by any type or combination of volatile or nonvolatile memory devices such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disk. A storage media may be any available media that can be accessed by a general purpose or special purpose computer.
An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an application specific integrated circuit (Application Specific Integrated Circuits, ASIC for short). It is also possible that the processor and the storage medium reside as discrete components in an electronic device or a master device.
The application provides a computer program product, comprising a computer program which realizes the ground penetrating radar data inversion method when being executed by a processor.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the application following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the application pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.
It is to be understood that the present application is not limited to the precise arrangements and instrumentalities shown in the drawings, which have been described above, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the application is limited only by the appended claims.
Claims (10)
1. The ground penetrating radar data inversion method is characterized by comprising the following steps of:
receiving ground penetrating radar data; the ground penetrating radar data are waveform data of electromagnetic waves for carrying out nondestructive detection on the shallow earth surface of the target area;
invoking a preset one-stage network model to perform inversion operation on the ground penetrating radar data to obtain a first dielectric constant model; wherein the first permittivity model is a factor determining a propagation speed of the electromagnetic wave in a stratum structure of the target area; the first dielectric constant model has at least one relative dielectric constant; the relative permittivity is a physical parameter that characterizes the dielectric properties of the formation structure;
Invoking a preset two-stage network model to perform inversion operation on the ground penetrating radar data and the first dielectric constant model to obtain a second dielectric constant model; the second dielectric constant model comprises physical parameters after the relative dielectric constant in the first dielectric constant model is corrected according to the ground penetrating radar data.
2. The method of claim 1, wherein before the invoking the preset one-stage network model to perform the inversion operation on the ground penetrating radar data, the method further comprises:
acquiring a first training sample;
and training through a first initial network model preset by the first training sample to obtain the one-stage network model.
3. The method of inversion of ground penetrating radar data according to claim 2, wherein before the invoking the preset two-stage network model to perform inversion operation on the ground penetrating radar data and the first dielectric constant model, the method further comprises:
acquiring a second training sample;
and training a preset second initial network model through the second training sample and the first-stage network model to obtain the second-stage network model.
4. A ground penetrating radar data inversion method according to claim 3 wherein prior to said obtaining a first training sample, the method further comprises:
acquiring a stratum structure, embedding irregular blocks in the stratum structure to convert the stratum structure into a training dielectric constant model, and performing forward modeling according to the training dielectric constant model to obtain training ground penetrating radar data;
summarizing one training dielectric constant model and the training ground penetrating radar data to form training data;
and summarizing a plurality of training data to obtain a training set.
5. A ground penetrating radar data inversion method according to claim 3 wherein said obtaining a first training sample comprises:
m training data are obtained from the training set, and the M training data are summarized to obtain the first training sample; wherein M is a positive integer, M is more than or equal to 1;
the obtaining a second training sample includes:
acquiring N training data from the training set, and summarizing the N training data to obtain the second training sample; wherein N is a positive integer, and N is more than or equal to 1.
6. The ground penetrating radar data inversion method according to claim 2, wherein the training by the first initial network model preset by the first training sample to obtain the one-stage network model includes:
The ground penetrating radar data of training data in the first training sample are used as first input information of the first initial network model, and the first initial network model is operated to carry out inversion operation on the first input information to obtain first output information;
the training dielectric constant model of training data in the first training sample is used as first reference information of the first initial network model, and a first loss value is generated according to the first output information and the first reference information through a preset first loss function; wherein the first loss value characterizes a degree of difference between the first output information and the first reference information;
iterating the first initial network model according to the first loss value through a preset optimizing model to adjust the weight of a hidden layer in the first initial network model, so that the first loss value between the first output information generated by the first initial network model and the first reference information is in a preset first threshold interval, and setting the first initial network model after iteration as the first-stage network model.
7. A ground penetrating radar data inversion method according to claim 3 wherein training a preset second initial network model through said second training sample and said one-stage network model to obtain said two-stage network model comprises:
Taking the training ground penetrating radar data of the training data in the second training sample as second input information, operating the first-stage network model to perform inversion operation on the second input information to obtain first-stage output information, and operating the second initial network model to perform inversion operation on the first-stage output information to obtain second output information;
the training dielectric constant model of training data in the second training sample is used as second reference information of the second initial network model, and a second loss value is generated according to the second output information and the second reference information through a preset second loss function; wherein the second loss value characterizes a degree of difference between the second output information and the second reference information;
and iterating the second initial network model according to the second loss value through a preset optimizing model to adjust the weight of a hidden layer in the second initial network model, so that the second loss value between second output information generated by the second initial network model and the second reference information is in a preset second threshold interval, and setting the iterated second initial network model as the two-stage network model.
8. A ground penetrating radar data inversion apparatus, comprising:
the input module is used for receiving ground penetrating radar data; the ground penetrating radar data are waveform data of electromagnetic waves for carrying out nondestructive detection on the shallow earth surface of the target area;
the first inversion module is used for calling a preset one-stage network model to perform inversion operation on the ground penetrating radar data to obtain a first dielectric constant model; wherein the first permittivity model is a factor determining a propagation speed of the electromagnetic wave in a stratum structure of the target area; the first dielectric constant model has at least one relative dielectric constant; the relative permittivity is a physical parameter that characterizes the dielectric properties of the formation structure;
the second inversion module is used for calling a preset two-stage network model to perform inversion operation on the ground penetrating radar data and the first dielectric constant model to obtain a second dielectric constant model; the second dielectric constant model comprises physical parameters after the relative dielectric constant in the first dielectric constant model is corrected according to the ground penetrating radar data.
9. A computer device, comprising: a processor and a memory communicatively coupled to the processor;
The memory stores computer-executable instructions;
the processor executes computer-executable instructions stored by the memory to implement the ground penetrating radar data inversion method of any one of claims 1 to 7.
10. A computer readable storage medium having stored therein computer executable instructions which when executed by a processor are for implementing the ground penetrating radar data inversion method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211641636.5A CN116224265A (en) | 2022-12-20 | 2022-12-20 | Ground penetrating radar data inversion method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211641636.5A CN116224265A (en) | 2022-12-20 | 2022-12-20 | Ground penetrating radar data inversion method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116224265A true CN116224265A (en) | 2023-06-06 |
Family
ID=86568736
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211641636.5A Pending CN116224265A (en) | 2022-12-20 | 2022-12-20 | Ground penetrating radar data inversion method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116224265A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116990772A (en) * | 2023-09-26 | 2023-11-03 | 北京大学 | Ground penetrating radar double-parameter real-time inversion method based on multi-scale convolution network |
-
2022
- 2022-12-20 CN CN202211641636.5A patent/CN116224265A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116990772A (en) * | 2023-09-26 | 2023-11-03 | 北京大学 | Ground penetrating radar double-parameter real-time inversion method based on multi-scale convolution network |
| CN116990772B (en) * | 2023-09-26 | 2024-01-02 | 北京大学 | Ground penetrating radar double-parameter real-time inversion method based on multi-scale convolution network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Giannakis et al. | A machine learning-based fast-forward solver for ground penetrating radar with application to full-waveform inversion | |
| CN106772583B (en) | A kind of earthquake diffracted wave separation method and device | |
| Bi et al. | Deep relative geologic time: A deep learning method for simultaneously interpreting 3‐D seismic horizons and faults | |
| Yablokov et al. | An artificial neural network approach for the inversion of surface wave dispersion curves | |
| Wang et al. | Data-driven S-wave velocity prediction method via a deep-learning-based deep convolutional gated recurrent unit fusion network | |
| Masroor et al. | A multiple-input deep residual convolutional neural network for reservoir permeability prediction | |
| CN112731522A (en) | Intelligent recognition method, device and equipment for seismic stratum and storage medium | |
| WO2022232572A1 (en) | Method and system for high resolution least-squares reverse time migration | |
| Song et al. | Delineating facies spatial distribution by integrating ensemble data assimilationand indicator geostatistics with level‐set transformation | |
| CN116224265A (en) | Ground penetrating radar data inversion method and device, computer equipment and storage medium | |
| CN112444850B (en) | Seismic data velocity modeling method, storage medium and computing device | |
| Jeong et al. | Application of conditional generative model for sonic log estimation considering measurement uncertainty | |
| Naprstek et al. | Convolutional neural networks applied to the interpretation of lineaments in aeromagnetic data | |
| CN115238565A (en) | Resistivity model reconstruction network training method, electromagnetic inversion method and device | |
| Patsia et al. | Background removal, velocity estimation, and reverse-time migration: a complete GPR processing pipeline based on machine learning | |
| Bürkle et al. | Deep physics-aware stochastic seismic inversion | |
| Zhou et al. | Stochastic structure-constrained image-guided inversion of geophysical data | |
| Bao et al. | Soil moisture estimation using tomographic ground penetrating radar in a MCMC–Bayesian framework | |
| US12013508B2 (en) | Method and system for determining seismic processing parameters using machine learning | |
| CN115469376A (en) | Tunnel advanced geological forecast comprehensive detection method and system based on non-homologous data | |
| Xu et al. | The use of 3D convolutional autoencoder in fault and fracture network characterization | |
| CN110794469B (en) | Gravity inversion method based on minimum geological feature unit constraint | |
| Iqbal et al. | Analysis of 2D and 3D GPR data interpretation using continuous wavelet transforms: Case study from an archaeological test site | |
| Menzel | Constrained indicator data resampling—A parameter constrained irregular resampling method for scattered point data | |
| Cheng et al. | Transient Electromagnetic Data Inversion: A Machine Learning Approach with CNN-LightGBM Model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |