CN116861955A - Method for inverting submarine topography by machine learning based on topography unit partition - Google Patents
Method for inverting submarine topography by machine learning based on topography unit partition Download PDFInfo
- Publication number
- CN116861955A CN116861955A CN202310763911.9A CN202310763911A CN116861955A CN 116861955 A CN116861955 A CN 116861955A CN 202310763911 A CN202310763911 A CN 202310763911A CN 116861955 A CN116861955 A CN 116861955A
- Authority
- CN
- China
- Prior art keywords
- water depth
- gravity
- machine learning
- topography
- anomaly
- 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
- 238000012876 topography Methods 0.000 title claims abstract description 42
- 238000010801 machine learning Methods 0.000 title claims abstract description 38
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000005192 partition Methods 0.000 title claims abstract description 21
- 230000005484 gravity Effects 0.000 claims abstract description 108
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 64
- 238000012549 training Methods 0.000 claims abstract description 48
- 238000010219 correlation analysis Methods 0.000 claims abstract description 6
- 230000005856 abnormality Effects 0.000 claims description 27
- 238000000638 solvent extraction Methods 0.000 claims description 9
- 238000011160 research Methods 0.000 claims description 7
- 238000003062 neural network model Methods 0.000 claims description 6
- 238000012417 linear regression Methods 0.000 claims description 4
- 238000005259 measurement Methods 0.000 claims description 4
- 230000006870 function Effects 0.000 claims description 3
- 239000013535 sea water Substances 0.000 claims description 3
- 238000001228 spectrum Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 abstract description 4
- 230000000694 effects Effects 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000002922 simulated annealing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method for inverting submarine topography by machine learning based on topography unit partition, which belongs to the technical field of geophysics and is used for inverting the submarine topography, and comprises the steps of forming a characteristic data set required by a machine learning model by gravity anomaly, shortwave gravity anomaly, vertical gravity gradient anomaly and residual vertical gravity gradient anomaly, wherein the characteristic data set forms a training set, and a grid characteristic data set forms a prediction set; and carrying out overall correlation analysis on the training set, adjusting various parameters of the model, inputting all the training sets into a machine learning model for training, inputting the prediction set into the trained machine learning model to obtain a full-area water depth value, and fusing the regional prediction results to obtain the final water depth of the overall area. The invention solves the problem that the traditional inversion method ignores the action of gravity information and the non-linear term of the sea bottom topography by utilizing the strong non-linear mapping capability of machine learning, thereby improving the inversion precision of the region with large topography fluctuation and rapid change.
Description
Technical Field
The invention discloses a method for inverting submarine topography by machine learning based on topography unit partition, and belongs to the technical field of geophysics.
Background
The traditional submarine topography measurement mainly relies on the shipborne sonar measuring water depth technology to acquire submarine topography data, and has the problems of low efficiency, high cost and the like. In the traditional research method, most methods only consider the linear mapping relation between gravity anomaly and sea bottom topography, and neglect the influence of nonlinear terms on the inversion result, so that the accuracy of the inversion result is limited. In addition, the vertical gravity gradient abnormality is used as the extension of gravity abnormality data, the sensitivity to the high-frequency part of the submarine topography exceeds the gravity itself, the influence of the second order quantity in Park theory is considered by utilizing the vertical gravity gradient abnormality in the prior art, the submarine topography is inverted by adopting a simulated annealing method, and the precision is improved by 22%. The method considers the effect of nonlinear terms, and proves that the use of the vertical gravity gradient anomaly can reflect more terrain detail information, so that the inversion accuracy is improved. In the inversion of seafloor terrain, the effect of higher order nonlinear term effects between the seafloor gravitational field and the terrain is not negligible.
Machine learning has strong nonlinear mapping capability, has great potential in the field of earth science, but the relationship between different seabed terrains and gravitational fields is uncertain, especially in areas with large relief of terrains, such as sea-time, sea-time and the like, the relationship between different terrains and gravity is more complex, and the adoption of a single model is insufficient to reflect the complex relationship, especially in the case of less training data, the inversion accuracy is reduced.
Disclosure of Invention
The invention aims to provide a method for inverting submarine topography by using machine learning based on topography unit partition, which solves the problem of low accuracy of inverting submarine topography by using machine learning under complex topography in the prior art.
A method of inverting a seafloor terrain using machine learning based on terrain unit partitioning, comprising:
s1, separating short-wave gravity anomaly from gravity anomaly;
s2, separating residual vertical gravity gradient abnormality from the vertical gravity gradient abnormality;
s3, combining the gravity anomaly, the shortwave gravity anomaly, the vertical gravity gradient anomaly and the residual vertical gravity gradient anomaly into a characteristic data set required by a machine learning model, wherein the characteristic data set forms a training set, the gridded characteristic data set forms a prediction set, and the machine learning model uses a BP neural network model;
s4, performing correlation analysis on the training set and the water depth data, measuring the correlation between the training set and the water depth data by adopting a Pearson correlation coefficient, and determining the weight proportion of the training set in the training model according to the correlation between the training set and the water depth data; inputting all training sets into a machine learning model for training, and inputting a prediction set into the trained machine learning model to obtain a full-area water depth value;
s5, gridding the water depth values of the whole region to obtain a submarine topography model of the whole region;
judging the topography of the region by a full-region submarine topography model, partitioning the topography of the research region according to three topography features of the sea, the sea ditch and the sea basin, respectively performing steps S1, S2 and S3 on ship water depth data and gravity data of each subarea to obtain a characteristic data set of each subarea, taking the characteristic data set of the ship water depth control point position as a partition training set, inputting the partition training set into a machine learning model for training, inputting the partition prediction set into the machine learning model to obtain a partition prediction result, and fusing the partition prediction result to obtain the final water depth of the whole region.
S1 comprises the following steps:
short wave gravity Δg of measured single beam point res Using control point j n The water depth is calculated by the grid plate formula:
in the formula ,indicated at control point j n Short wave component on the upper part; g is the gravitational constant; Δρ is the optimal density difference constant between the sea water and the subsea ocean shell; d represents a reference water depth, and the maximum water depth of the ship-borne sounding data is taken; />The water depth value of the control point is;
and calculating the correlation relation and root mean square error of the predicted water depth and the actually measured water depth corresponding to different density difference constants by adopting an iteration method, wherein the density difference constant is the optimal value when the root mean square error is minimum and the correlation coefficient is maximum.
S2 comprises the following steps:
the gravity data and the water depth are regarded as two different signals, and the coherence of the two signals is as follows:
wherein ,is a cross spectrum coherence function; g (k) and H (k) respectively represent the Fourier transform of the gravity signal and the Fourier transform of the terrain signal; g * (k)、H * (k) Respectively representing complex conjugates of G (k) and B (k);
performing coherence analysis on the water depth data after trend removal and the vertical gravity gradient abnormality, and selectingLinear regression is carried out on the data in the wave band range larger than 0.5, and the abnormal wave band vertical gravity gradient-wave band water depth proportion factor is obtainedMultiplying the obtained scale factors with the ship water depth to obtain a reference vertical gravity gradient abnormality, subtracting the reference vertical gravity gradient abnormality from the vertical gravity gradient abnormality to obtain a residual vertical gravity gradient abnormality of the depth measuring point, and interpolating to obtain a vertical gravity gradient abnormality field of the research area.
After the short wave gravity anomaly is obtained, subtracting the short wave gravity anomaly from the gravity anomaly of the actually measured single beam point position to obtain the long wave gravity anomaly of each point, gridding the long wave gravity anomaly, and subtracting the gridded long wave gravity anomaly from the gridded gravity anomaly to obtain the gridded short wave gravity anomaly.
Compared with the prior art, the invention has the following beneficial effects: the invention provides a method for inverting submarine topography according to a topography unit by combining gravity anomaly and vertical gravity gradient anomaly based on a machine learning method, and solves the problem that the traditional inversion method ignores the effects of gravity information and a submarine topography nonlinear item by utilizing strong nonlinear mapping capability of machine learning, thereby improving inversion precision of a region with large topography fluctuation and rapid change.
Drawings
FIG. 1 is a technical flow chart of the present invention;
FIG. 2 is a block diagram of a BP neural network used in the present invention;
fig. 3 is a flow chart for acquiring a feature dataset.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions in the present invention will be clearly and completely described below, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
A method of inverting a seafloor terrain using machine learning based on terrain unit partitioning, comprising:
s1, separating short-wave gravity anomaly from gravity anomaly;
s2, separating residual vertical gravity gradient abnormality from the vertical gravity gradient abnormality;
s3, combining the gravity anomaly, the shortwave gravity anomaly, the vertical gravity gradient anomaly and the residual vertical gravity gradient anomaly into a characteristic data set required by a machine learning model, wherein the characteristic data set forms a training set, the gridded characteristic data set forms a prediction set, and the machine learning model uses a BP neural network model;
s4, performing correlation analysis on the training set and the water depth data, measuring the correlation between the training set and the water depth data by adopting a Pearson correlation coefficient, and determining the weight proportion of the training set in the training model according to the correlation between the training set and the water depth data; inputting all training sets into a machine learning model for training, and inputting a prediction set into the trained machine learning model to obtain a full-area water depth value;
s5, gridding the water depth values of the whole region to obtain a submarine topography model of the whole region;
judging the topography of the region by a full-region submarine topography model, partitioning the topography of the research region according to three topography features of the sea, the sea ditch and the sea basin, respectively performing steps S1, S2 and S3 on ship water depth data and gravity data of each subarea to obtain a characteristic data set of each subarea, taking the characteristic data set of the ship water depth control point position as a partition training set, inputting the partition training set into a machine learning model for training, inputting the partition prediction set into the machine learning model to obtain a partition prediction result, and fusing the partition prediction result to obtain the final water depth of the whole region.
S1 comprises the following steps:
short wave gravity Δg of measured single beam point res Using control point j n The water depth is calculated by the grid plate formula:
in the formula ,indicated at control point j n Short wave component on the upper part; g is the gravitational constant; Δρ is the optimal density difference constant between the sea water and the subsea ocean shell; d represents a reference water depth, and the maximum water depth of the ship-borne sounding data is taken; />The water depth value of the control point is;
and calculating the correlation relation and root mean square error of the predicted water depth and the actually measured water depth corresponding to different density difference constants by adopting an iteration method, wherein the density difference constant is the optimal value when the root mean square error is minimum and the correlation coefficient is maximum.
S2 comprises the following steps:
the gravity data and the water depth are regarded as two different signals, and the coherence of the two signals is as follows:
wherein ,is a cross spectrum coherence function; g (k) and H (k) respectively represent the Fourier transform of the gravity signal and the Fourier transform of the terrain signal; g * (k)、H * (k) Respectively representing complex conjugates of G (k) and B (k);
performing coherence analysis on the water depth data after trend removal and the vertical gravity gradient abnormality, and selectingPerforming linear regression on data in a wave band range larger than 0.5, obtaining a wave band vertical gravity gradient abnormality-wave band water depth scaling factor, multiplying the obtained scaling factor by the ship water depth to obtain a reference vertical gravity gradient abnormality, subtracting the reference vertical gravity gradient abnormality from the vertical gravity gradient abnormality to obtain a residual vertical gravity gradient abnormality of a depth measurement point, and interpolating to obtain a vertical gravity gradient abnormality field of the research area.
After the short wave gravity anomaly is obtained, subtracting the short wave gravity anomaly from the gravity anomaly of the actually measured single beam point position to obtain the long wave gravity anomaly of each point, gridding the long wave gravity anomaly, and subtracting the gridded long wave gravity anomaly from the gridded gravity anomaly to obtain the gridded short wave gravity anomaly.
The technical process of the invention is as shown in figure 1, firstly, gravity anomaly, vertical gravity gradient anomaly and ship water depth measurement data are respectively valued along the same grid point and are put in the same document to form a data set, then the data set is divided into a training set and a prediction set, wherein the training set carries out overall correlation analysis, is guided into a neural network model for training, and is combined with the training set to form a pre-estimated water depth model after the training is completed; meanwhile, the training sets are partitioned through the terrain units to form a plurality of training sets, partition correlation analysis is carried out, then the training sets are respectively led into the neural network model for training, a multi-region water depth model is formed after multi-model training, and a full-region submarine terrain model is formed through fusion. The neural network model in the embodiment is specifically a BP neural network, that is, the machine learning model, as shown in fig. 2, four kinds of data are input into an input layer, and are processed by an hidden layer to form an output layer, and finally the predicted water depth is output. The process of obtaining the characteristic data set is as shown in fig. 3, the gravity anomaly and the ship survey water depth data form a ship survey point short wave gravity anomaly through a grid plate formula, meanwhile, the ship survey water depth data and the vertical gravity gradient anomaly are subjected to trending, coherence analysis and linear regression to obtain a scale factor, the scale factor and the ship survey water depth data are multiplied to obtain a ship survey point reference vertical gravity gradient anomaly, the ship survey point reference vertical gravity gradient anomaly is subtracted from the vertical gravity gradient anomaly to obtain a ship survey point residual vertical gravity gradient anomaly, and the ship survey point short wave gravity anomaly is fused to form the characteristic data set. Error characteristics of inversion results of a submarine topography model obtained by inversion of a gravity geological method by a traditional method, an undivided BP model and a partitioned BP-S model are counted, and error statistics results of three model inversion results and 14000 single-beam nuclear detection points are shown in a table 1.
Table 1 checking error statistics in different models;
| model | Root mean square error/m | Average relative error |
| BP neural network | 89 | 2.07％ |
| The method of the invention | 47 | 1.45％ |
| Gravity geology method | 91 | 2.82％ |
The method comprises the steps of carrying out a first treatment on the surface of the It can be seen in table 1 that the method of the present invention has advantages in both root mean square error and average error.
The above embodiments are only for illustrating the technical aspects of the present invention, not for limiting the same, and although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may be modified or some or all of the technical features may be replaced with other technical solutions, which do not depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (4)
1. A method of inverting a seafloor terrain using machine learning based on terrain cell partitioning, comprising:
s1, separating short-wave gravity anomaly from gravity anomaly;
s2, separating residual vertical gravity gradient abnormality from the vertical gravity gradient abnormality;
s3, combining the gravity anomaly, the shortwave gravity anomaly, the vertical gravity gradient anomaly and the residual vertical gravity gradient anomaly into a characteristic data set required by a machine learning model, wherein the characteristic data set forms a training set, the gridded characteristic data set forms a prediction set, and the machine learning model uses a BP neural network model;
s4, performing correlation analysis on the training set and the water depth data, measuring the correlation between the training set and the water depth data by adopting a Pearson correlation coefficient, and determining the weight proportion of the training set in the training model according to the correlation between the training set and the water depth data; inputting all training sets into a machine learning model for training, and inputting a prediction set into the trained machine learning model to obtain a full-area water depth value;
s5, gridding the water depth values of the whole region to obtain a submarine topography model of the whole region;
judging the topography of the region by a full-region submarine topography model, partitioning the topography of the research region according to three topography features of the sea, the sea ditch and the sea basin, respectively performing steps S1, S2 and S3 on ship water depth data and gravity data of each subarea to obtain a characteristic data set of each subarea, taking the characteristic data set of the ship water depth control point position as a partition training set, inputting the partition training set into a machine learning model for training, inputting the partition prediction set into the machine learning model to obtain a partition prediction result, and fusing the partition prediction result to obtain the final water depth of the whole region.
2. A method of inverting a seafloor terrain using machine learning based on terrain cell partitioning as claimed in claim 1 wherein S1 comprises:
short wave gravity Δg of measured single beam point res Using control point j n The water depth is calculated by the grid plate formula:
in the formula ,indicated at control point j n Short wave component on the upper part; g is the gravitational constant; Δρ is the optimal density difference constant between the sea water and the subsea ocean shell; d represents a reference water depth, and the maximum water depth of the ship-borne sounding data is taken; />The water depth value of the control point is;
and calculating the correlation relation and root mean square error of the predicted water depth and the actually measured water depth corresponding to different density difference constants by adopting an iteration method, wherein the density difference constant is the optimal value when the root mean square error is minimum and the correlation coefficient is maximum.
3. A method of inverting a seafloor terrain using machine learning based on terrain cell partitioning as claimed in claim 1 wherein S2 comprises:
the gravity data and the water depth are regarded as two different signals, and the coherence of the two signals is as follows:
wherein ,is a cross spectrum coherence function; g (k) and H (k) respectively represent the Fourier transform of the gravity signal and the Fourier transform of the terrain signal; g * (k)、H * (k) Respectively representing complex conjugates of G (k) and H (k);
performing coherence analysis on the water depth data after trend removal and the vertical gravity gradient abnormality, and selectingPerforming linear regression on data in a wave band range greater than 0.5, and obtainingObtaining a wave band vertical gravity gradient abnormality-wave band water depth scaling factor, multiplying the obtained scaling factor with the ship water depth to obtain a reference vertical gravity gradient abnormality, subtracting the reference vertical gravity gradient abnormality from the vertical gravity gradient abnormality to obtain a residual vertical gravity gradient abnormality of a depth measurement point, and interpolating to obtain a vertical gravity gradient abnormality field of the research area.
4. The method for inverting submarine topography by machine learning based on topography unit partitioning according to claim 1, wherein after obtaining short wave gravity anomalies, subtracting the short wave gravity anomalies from the gravity anomalies of the measured single beam point location to obtain long wave gravity anomalies of each point, gridding the long wave gravity anomalies, and subtracting the gridded long wave gravity anomalies from the gridded gravity anomalies to obtain gridded short wave gravity anomalies.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310763911.9A CN116861955A (en) | 2023-06-27 | 2023-06-27 | Method for inverting submarine topography by machine learning based on topography unit partition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310763911.9A CN116861955A (en) | 2023-06-27 | 2023-06-27 | Method for inverting submarine topography by machine learning based on topography unit partition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116861955A true CN116861955A (en) | 2023-10-10 |
Family
ID=88227809
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310763911.9A Pending CN116861955A (en) | 2023-06-27 | 2023-06-27 | Method for inverting submarine topography by machine learning based on topography unit partition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116861955A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117113857A (en) * | 2023-10-23 | 2023-11-24 | 自然资源部第一海洋研究所 | Full-connection depth neural network model and method for inverting submarine topography |
-
2023
- 2023-06-27 CN CN202310763911.9A patent/CN116861955A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117113857A (en) * | 2023-10-23 | 2023-11-24 | 自然资源部第一海洋研究所 | Full-connection depth neural network model and method for inverting submarine topography |
| CN117113857B (en) * | 2023-10-23 | 2024-01-30 | 自然资源部第一海洋研究所 | Full-connection depth neural network model and method for inverting submarine topography |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Vardy | Deriving shallow‐water sediment properties using post‐stack acoustic impedance inversion | |
| Pyrcz et al. | Stratigraphic rule-based reservoir modeling | |
| Fedi et al. | Multiscale analysis of potential fields by a ridge consistency criterion: the reconstruction of the Bishop basement | |
| Yuan et al. | A probabilistic shear wave velocity model of the crust in the central West Australian craton constrained by transdimensional inversion of ambient noise dispersion | |
| CN116861955A (en) | Method for inverting submarine topography by machine learning based on topography unit partition | |
| CN109521469B (en) | Regularization inversion method for elastic parameters of submarine sediments | |
| Mulia et al. | Initial tsunami source estimation by inversion with an intelligent selection of model parameters and time delays | |
| Jun et al. | Trans-dimensional Markov chain Monte Carlo inversion of sound speed and temperature: Application to Yellow Sea multichannel seismic data | |
| CN111399074A (en) | Gravity and gravity gradient modulus combined three-dimensional inversion method | |
| Titus et al. | Conditioning surface-based geological models to well data using artificial neural networks | |
| CN112861322B (en) | Quantitative analysis method and system for submarine stepped landform evolution | |
| CN111273349B (en) | Transverse wave velocity extraction method and processing terminal for seabed shallow sediment layer | |
| CN112100906B (en) | Data-driven large-scale density modeling method, computing device and storage medium | |
| Florio et al. | A fractional vertical derivative technique for regional-residual separation | |
| CN105893329A (en) | Monthly-scale-based tide level data consistency correction method | |
| Zhang et al. | 3D Airborne Electromagnetic Data Inversion with Flight Altitude Correction | |
| CN113221651B (en) | Submarine sediment classification method using acoustic propagation data and unsupervised machine learning | |
| Bai et al. | A wavelet transformation approach for multi-source gravity fusion: Applications and uncertainty tests | |
| Mouzong Pemi et al. | Depth and lineament maps derived from North Cameroon gravity data computed by artificial neural network | |
| CN114482995A (en) | Fine determination method for argillaceous content of fine-grain sediment | |
| Daniel OB et al. | Application of 3D Euler deconvolution and 2D inverse modelling to basin depth estimation, the case of the Keta basin, Ghana | |
| WO2012060888A1 (en) | System and method for providing a physical property model | |
| Chen et al. | The improved Kriging interpolation algorithm for local underwater terrain based on fractal compensation | |
| Cox et al. | Two dimensional numerical models of the California electromagnetic coastal anomaly | |
| Wang et al. | The High-Speed Inversion of Rayleigh Wave and Its Microtremor Application Analysis |
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 |