CN114898109A - Porphyry shallow-formation low-temperature hydrothermal type mineral prediction method and system based on deep learning - Google Patents
Porphyry shallow-formation low-temperature hydrothermal type mineral prediction method and system based on deep learning Download PDFInfo
- Publication number
- CN114898109A CN114898109A CN202210388950.0A CN202210388950A CN114898109A CN 114898109 A CN114898109 A CN 114898109A CN 202210388950 A CN202210388950 A CN 202210388950A CN 114898109 A CN114898109 A CN 114898109A
- Authority
- CN
- China
- Prior art keywords
- data
- prediction
- porphyry
- shallow
- mineral
- 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.)
- Granted
Links
- 229910052500 inorganic mineral Inorganic materials 0.000 title claims abstract description 57
- 239000011707 mineral Substances 0.000 title claims abstract description 57
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000013135 deep learning Methods 0.000 title claims abstract description 26
- 238000012549 training Methods 0.000 claims abstract description 52
- 238000012545 processing Methods 0.000 claims abstract description 21
- 230000007246 mechanism Effects 0.000 claims abstract description 18
- 238000013528 artificial neural network Methods 0.000 claims abstract description 15
- 238000005065 mining Methods 0.000 claims description 37
- 238000012795 verification Methods 0.000 claims description 36
- 238000012360 testing method Methods 0.000 claims description 19
- 230000006870 function Effects 0.000 claims description 17
- 238000005070 sampling Methods 0.000 claims description 11
- 238000003860 storage Methods 0.000 claims description 8
- 229910052934 alunite Inorganic materials 0.000 claims description 6
- 239000010424 alunite Substances 0.000 claims description 6
- 238000002360 preparation method Methods 0.000 claims description 6
- KPZTWMNLAFDTGF-UHFFFAOYSA-D trialuminum;potassium;hexahydroxide;disulfate Chemical compound [OH-].[OH-].[OH-].[OH-].[OH-].[OH-].[Al+3].[Al+3].[Al+3].[K+].[O-]S([O-])(=O)=O.[O-]S([O-])(=O)=O KPZTWMNLAFDTGF-UHFFFAOYSA-D 0.000 claims description 6
- 229910001649 dickite Inorganic materials 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 5
- 229910052903 pyrophyllite Inorganic materials 0.000 claims description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Chemical compound O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 5
- CWYNVVGOOAEACU-UHFFFAOYSA-N Fe2+ Chemical compound [Fe+2] CWYNVVGOOAEACU-UHFFFAOYSA-N 0.000 claims description 4
- 229910001919 chlorite Inorganic materials 0.000 claims description 4
- 229910052619 chlorite group Inorganic materials 0.000 claims description 4
- QBWCMBCROVPCKQ-UHFFFAOYSA-N chlorous acid Chemical compound OCl=O QBWCMBCROVPCKQ-UHFFFAOYSA-N 0.000 claims description 4
- NLYAJNPCOHFWQQ-UHFFFAOYSA-N kaolin Chemical compound O.O.O=[Al]O[Si](=O)O[Si](=O)O[Al]=O NLYAJNPCOHFWQQ-UHFFFAOYSA-N 0.000 claims description 4
- 229910052622 kaolinite Inorganic materials 0.000 claims description 4
- 229910021532 Calcite Inorganic materials 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- GUJOJGAPFQRJSV-UHFFFAOYSA-N dialuminum;dioxosilane;oxygen(2-);hydrate Chemical compound O.[O-2].[O-2].[O-2].[Al+3].[Al+3].O=[Si]=O.O=[Si]=O.O=[Si]=O.O=[Si]=O GUJOJGAPFQRJSV-UHFFFAOYSA-N 0.000 claims description 3
- 239000010459 dolomite Substances 0.000 claims description 3
- 229910000514 dolomite Inorganic materials 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 claims description 3
- 229910052901 montmorillonite Inorganic materials 0.000 claims description 3
- 230000010287 polarization Effects 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000002310 reflectometry Methods 0.000 claims description 3
- 230000003595 spectral effect Effects 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000010801 machine learning Methods 0.000 abstract description 8
- 238000012216 screening Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 13
- 238000004590 computer program Methods 0.000 description 11
- 230000033558 biomineral tissue development Effects 0.000 description 8
- 238000000605 extraction Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 238000005259 measurement Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- KDLHZDBZIXYQEI-UHFFFAOYSA-N Palladium Chemical compound [Pd] KDLHZDBZIXYQEI-UHFFFAOYSA-N 0.000 description 2
- JYIBXUUINYLWLR-UHFFFAOYSA-N aluminum;calcium;potassium;silicon;sodium;trihydrate Chemical compound O.O.O.[Na].[Al].[Si].[K].[Ca] JYIBXUUINYLWLR-UHFFFAOYSA-N 0.000 description 2
- 229910052797 bismuth Inorganic materials 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 229910001603 clinoptilolite Inorganic materials 0.000 description 2
- 229910052802 copper Inorganic materials 0.000 description 2
- 239000010949 copper Substances 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 229910052737 gold Inorganic materials 0.000 description 2
- 239000010931 gold Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 229910052750 molybdenum Inorganic materials 0.000 description 2
- 239000013049 sediment Substances 0.000 description 2
- 229910052709 silver Inorganic materials 0.000 description 2
- 229910052725 zinc Inorganic materials 0.000 description 2
- 239000011701 zinc Substances 0.000 description 2
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 1
- ZOKXTWBITQBERF-UHFFFAOYSA-N Molybdenum Chemical compound [Mo] ZOKXTWBITQBERF-UHFFFAOYSA-N 0.000 description 1
- BUGBHKTXTAQXES-UHFFFAOYSA-N Selenium Chemical compound [Se] BUGBHKTXTAQXES-UHFFFAOYSA-N 0.000 description 1
- BQCADISMDOOEFD-UHFFFAOYSA-N Silver Chemical compound [Ag] BQCADISMDOOEFD-UHFFFAOYSA-N 0.000 description 1
- HCHKCACWOHOZIP-UHFFFAOYSA-N Zinc Chemical compound [Zn] HCHKCACWOHOZIP-UHFFFAOYSA-N 0.000 description 1
- 239000002253 acid Substances 0.000 description 1
- 229910052787 antimony Inorganic materials 0.000 description 1
- 229910052785 arsenic Inorganic materials 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- JCXGWMGPZLAOME-UHFFFAOYSA-N bismuth atom Chemical compound [Bi] JCXGWMGPZLAOME-UHFFFAOYSA-N 0.000 description 1
- 229910052804 chromium Inorganic materials 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 230000009545 invasion Effects 0.000 description 1
- 229910052745 lead Inorganic materials 0.000 description 1
- 229910052748 manganese Inorganic materials 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 238000002844 melting Methods 0.000 description 1
- 230000008018 melting Effects 0.000 description 1
- 229910052753 mercury Inorganic materials 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 150000002739 metals Chemical class 0.000 description 1
- 239000011733 molybdenum Substances 0.000 description 1
- 229910052759 nickel Inorganic materials 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 229910052763 palladium Inorganic materials 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 229910052702 rhenium Inorganic materials 0.000 description 1
- WUAPFZMCVAUBPE-UHFFFAOYSA-N rhenium atom Chemical compound [Re] WUAPFZMCVAUBPE-UHFFFAOYSA-N 0.000 description 1
- 239000011435 rock Substances 0.000 description 1
- 229910052711 selenium Inorganic materials 0.000 description 1
- 239000011669 selenium Substances 0.000 description 1
- 239000004332 silver Substances 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 229910052714 tellurium Inorganic materials 0.000 description 1
- PORWMNRCUJJQNO-UHFFFAOYSA-N tellurium atom Chemical compound [Te] PORWMNRCUJJQNO-UHFFFAOYSA-N 0.000 description 1
- 229910052718 tin Inorganic materials 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 229910052721 tungsten Inorganic materials 0.000 description 1
- 229910052720 vanadium Inorganic materials 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- 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
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
-
- 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
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Business, Economics & Management (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Primary Health Care (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Probability & Statistics with Applications (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Mining & Mineral Resources (AREA)
- Marine Sciences & Fisheries (AREA)
- Animal Husbandry (AREA)
- Agronomy & Crop Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
Abstract
The invention belongs to the technical field of mineral resource prediction, and particularly discloses a porphyry shallow-formation low-temperature hydrothermal type mineral prediction method and system based on deep learning, wherein the method comprises the following steps: preparing a predictive variable, processing data of the predictive variable, selecting a training sample, constructing a model and identifying a target area of the prospecting. According to the scheme, geophysical, geochemical and hyperspectral mineral information data are extracted based on the spatial resolution of 60m, geological elements are not used, the influence of uncertainty of the geological elements is effectively avoided, and the prediction precision of the model is improved. In the field of mineral resource prediction, an attention-free mechanism full-connection neural network is constructed for the first time to perform supervised classification type mineral resource prediction, the acquisition capability of the correlation information among the prediction variables is enhanced, and the screening capability of the effective characteristics of the prediction variables is improved, so that the mineral resource prediction precision is improved, and a technical basis is provided for effectively developing machine learning in mineral resource application.
Description
Technical Field
The invention belongs to the technical field of mineral resource prediction, and particularly relates to a porphyry-shallow-formation low-temperature hydrothermal type mineral resource prediction method and system based on deep learning.
Background
With the exponential growth of geological big data and the rise of artificial intelligence, machine learning methods such as a support vector machine and a convolutional neural network are adopted in the prior art to be applied to mineral resource prediction. Compared with the traditional method, the machine learning has higher prediction precision, and particularly has obvious advantages for large data volume, high dimensionality, complex nonlinear relation among input variables or the input variables with complex statistical distribution characteristics.
The prior art mainly has the following two technical defects:
firstly, the current machine learning method adopts geological, geophysical, geochemical and remote sensing data as prediction variables. However, uncertainty of geological elements (such as fracture and lithology) inevitably affects machine learning, and particularly in severe areas in the west, such as Qinghai-Tibet plateau, basic geological survey is relatively weak. In addition, for remote sensing data, spectral measurement data is used at present, and not extracted mineral information, but altered minerals such as alunite, pyrophyllite, dickite and the like have a good indication effect on exploration of porphyry-shallow low-temperature hydrothermal deposit, and particularly alunite is a high-sulfur-type shallow-low-temperature hydrothermal deposit marker mineral.
Second, there is no spatial resolution for explicitly extracting the prediction data at present. Too little spatial resolution may result in too much data, low value density, slow model operation, and too much spatial resolution may result in reduced classification accuracy.
Disclosure of Invention
The invention aims to provide a porphyry-shallow-temperature hydrothermal type mineral resource prediction method and system based on deep learning, which can solve the technical problem of low mineral resource prediction accuracy.
The invention provides a porphyry-shallow-formation low-temperature hydrothermal type mineral resource prediction method based on deep learning, which comprises the following steps of:
s1, acquiring hyperspectral remote sensing data, geochemical data and geophysical data of the target mining area, and sampling all the data into 60m resolution;
s2, normalizing the geochemical data and the geophysical data to remove the influence of data dimension;
s3, selecting a target mining area and an area with an extension of 100m as positive samples, selecting an area with an extension of 400m of the target mining area as a negative sample, judging that the area without the mining area is a negative sample, and dividing the area without the mining area into a training set, a verification set and a test set according to a ratio of 7:2: 1;
s4, constructing a model of the deep fully-connected neural network based on the self-attention mechanism, training the model by using the training set constructed in the previous step, simultaneously performing precision verification on the model by using a verification set and a test set, and completing the training when the verification precision converges to a stable state;
and S5, obtaining prediction result data after training is completed, wherein 1 is obtained to indicate that the ore exists, and 0 is obtained to indicate that the ore does not exist.
Preferably, the target mine area is a shallow low temperature hydrothermal deposit or a porphyry deposit.
Preferably, the S1 specifically includes:
firstly, preprocessing hyperspectral remote sensing data of a target mining area to obtain reflectivity data, eliminating bad wave bands influenced by water vapor, and then performing minimum noise classification transformation (MNF) and Pure Pixel Index (PPI) analysis;
then, mineral data information is extracted and resampled to 60m resolution using a combination of the hybrid tuned matched filter method (MTMF) and the Spectral Angle Method (SAM).
Preferably, the mineral data information includes data information of alunite, pyrophyllite, dickite, kaolinite, chlorite/chlorothalolite, limonite, high-Al sericite, medium-Al sericite, low-Al sericite, dolomite, calcite, montmorillonite.
Preferably, the S1 specifically includes:
optimizing a variation function model for the geochemical data by using GS + software, and performing Kriging interpolation to obtain a resolution of 60m according to an optimized result;
and (3) carrying out polarization, horizontal gradient film, vertical first derivative and upward continuation processing on the geophysical data, and sampling to obtain the spatial resolution of 60 m.
Preferably, the S4 specifically includes:
1) inputting characteristic information of each prediction variable extracted from the hyperspectral remote sensing data, the geochemical data and the geophysical data into a model, and calculating similarity to obtain weight;
2) normalizing the weights using a Softmax function;
3) and carrying out weighted summation on the weight and the corresponding key value to obtain the final attention score.
Preferably, the model comprises a 1-layer self-attention mechanism and a 4-layer fully-connected neural network.
The invention also provides a porphyry shallow-to-low-temperature hydrothermal type mineral prediction system based on deep learning, which is used for realizing the porphyry shallow-to-low-temperature hydrothermal type mineral prediction method based on deep learning and comprises the following steps:
the system comprises a predictive variable preparation module, a data acquisition module and a data analysis module, wherein the predictive variable preparation module is used for acquiring hyperspectral remote sensing data, geochemical data and geophysical data of a target mining area and sampling all groups of data into 60m resolution;
the predictive variable data processing module is used for carrying out normalization processing on the geochemical data and the geophysical data so as to remove the influence of data dimension;
the training sample selection module selects a target mining area and a range with the extension of 100m as positive samples, selects a region with the extension of 400m beyond the target mining area as a negative sample, judges that the region without the mine is the negative sample, and divides the region without the mine into a training set, a verification set and a test set according to the ratio of 7:2: 1;
the model construction module is used for constructing a model of the deep fully-connected neural network based on the self-attention mechanism, training the model by using the training set constructed in the previous step, simultaneously performing precision verification on the model by using the verification set and the test set, and finishing the training when the precision of the verification set is converged to a stable state;
and the ore-finding target area identification module obtains prediction result data after training is finished, outputs the prediction result data by using a torch.max () function, and takes the output result as the corresponding classification with the maximum prediction probability as a two-classification problem, wherein 1 represents that ore exists and 0 represents that no ore exists.
The invention also provides electronic equipment which comprises a memory and a processor, wherein the processor is used for realizing the deep learning-based porphyry shallow-to-low-temperature hydrothermal type mineral prediction method when executing the computer management program stored in the memory.
The invention also provides a computer readable storage medium, on which a computer management program is stored, which when executed by a processor implements the steps of the porphyry shallow-to-low temperature hydrothermal type mineral prediction method based on deep learning.
Compared with the prior art, the porphyry shallow-to-low-temperature hydrothermal type mineral prediction method and system based on deep learning, provided by the invention, comprise the following steps: preparing a predictive variable, processing data of the predictive variable, selecting a training sample, constructing a model and identifying a target area of the prospecting. According to the scheme, geophysical, geochemical and hyperspectral mineral information data are extracted based on the spatial resolution of 60m, geological elements are not used, the influence of uncertainty of the geological elements is effectively avoided, and the prediction precision of the model is improved. In the field of mineral resource prediction, an attention-free mechanism full-connection neural network is constructed for the first time to perform supervised classification type mineral resource prediction, the acquisition capability of the correlation information among the prediction variables is enhanced, and the screening capability of the effective characteristics of the prediction variables is improved, so that the mineral resource prediction precision is improved, and a technical basis is provided for effectively developing machine learning in mineral resource application.
Drawings
FIG. 1 is a flow chart of a porphyry-shallow-phase low-temperature hydrothermal mineral resource prediction method based on deep learning provided by the invention;
FIG. 2 is a schematic diagram of a hardware structure of a possible electronic device provided in the present invention;
FIG. 3 is a schematic diagram of a hardware structure of a possible computer-readable storage medium provided by the present invention;
fig. 4 is a schematic diagram of the division of the training area and the prediction area of the porphyry shallow-to-low-temperature hydrothermal mineral prediction method based on deep learning provided by the invention.
Detailed Description
The following detailed description of the present invention is provided in conjunction with the accompanying drawings, but it should be understood that the scope of the present invention is not limited to the specific embodiments.
Throughout the specification and claims, unless explicitly stated otherwise, the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element or component but not the exclusion of any other element or component.
As shown in fig. 1 and 4, a method for predicting shallow-to-low-temperature hydrothermal type mineral deposits of porphyry based on deep learning according to a preferred embodiment of the present invention comprises the following steps:
s1, acquiring hyperspectral remote sensing data, geochemical data and geophysical data of the target mining area, and sampling all the data into 60m resolution;
s2, normalizing the geochemical data and the geophysical data to remove the influence of data dimension;
s3, selecting a target mining area and an area with an extension of 100m as positive samples, selecting an area with an extension of 400m of the target mining area as a negative sample, judging that the area without the mining area is a negative sample, and dividing the area without the mining area into a training set, a verification set and a test set according to a ratio of 7:2: 1;
s4, constructing a model of the deep fully-connected neural network based on the self-attention mechanism, training the model by using the training set constructed in the previous step, simultaneously performing precision verification on the model by using a verification set and a test set, and completing the training when the precision of the verification set is converged to a stable state;
and S5, obtaining prediction result data after training, outputting the prediction result data by using a torch.max () function, wherein the output result is the corresponding classification with the maximum prediction probability, and obtaining 1 representing the existence of the ore and 0 representing the absence of the ore as a binary classification problem.
The porphyry mineralization system is controlled by the same structure-magma-thermodynamic system in a broad sense, develops around a multi-stage invaded medium-acid rock mass, is dependent on space and time, and is formed by a series of symbiotic ore deposit type combinations related to shallow and ultra-shallow invasion and terrestrial volcano-subplanaria. Statistical studies show that the porphyry mineralization system provides about 75% of copper, 50% of molybdenum, almost all rhenium and more than 20% of gold for the world at present, is one of the main sources of metals such as silver, palladium, tellurium, selenium, bismuth, lead, zinc and the like, and is one of the most important exploration targets of the research of the mineral deposit academia and the industry at present. A porphyry-shallow low-temperature hydrothermal mineralization system is a mineralization symbiotic combination type in which a porphyry mineralization system in the porphyry mineralization system and a shallow low-temperature hydrothermal mineralization system at the top or the side part of the porphyry mineralization system are symbiotic with each other.
In one particular implementation scenario:
first, predictor variables are prepared. The predictive variables include hyperspectral remote sensing data, geochemical data and geophysical data. The three kinds of data are respectively extracted to obtain data information with the resolution of 60m, so that the subsequent data processing is facilitated.
1. Hyperspectral remote sensing data: preprocessing hyperspectral data of a test area to obtain reflectivity data, eliminating bad wave bands affected by water vapor and the like, and then performing minimum noise classification transformation (MNF) and Pure Pixel Index (PPI) analysis. 12 minerals such as alunite, pyrophyllite, dickite, kaolinite, chlorite/clinoptilolite, limonite, high-Al sericite, medium-Al sericite, low-Al sericite and the like are extracted by a mode of combining a mixed tuning matched filter method (MTMF) and a Spectrum Angle Method (SAM). Resampled to 60m resolution.
2. Geochemical data: in other words, for the measured data of the water system sediments of more than 1:5 ten thousand, the GS + software is used for optimizing the variation function model, and the Kriging interpolation is carried out to obtain the resolution of 60m according to the optimization result.
3. Geophysical data: the method is characterized in that the method carries out the pole melting, horizontal gradient film, vertical first derivative and upward continuation processing on the high-precision magnetic layer data which is superior to 1:5 ten thousand, and the spatial resolution of 60m is obtained by sampling.
And secondly, processing the predictive variable data. And carrying out normalization (min-max method) processing on the geochemical data and the geophysical data to remove the influence of data dimension. The extraction result of the hyperspectral remote sensing data is binary of 0 (none) and 1 (existence), and normalization processing is not needed.
Geochemical data: namely water system sediment measurement data, namely 15 elements of Ag, As, Au, Bi, Cr, Cu, Hg, Mn, Mo, Ni, Pb, Sb, Sn, W and Zn.
Geophysical data: and (3) carrying out high-precision magnetic measurement on data, and carrying out polarization, horizontal gradient film, vertical first-order derivative and upward continuation processing.
Hyperspectral remote sensing data: alunite, pyrophyllite, dickite, kaolinite, chlorite/clinoptilolite, limonite, high Al sericite, medium Al sericite, low Al sericite, dolomite, calcite and montmorillonite, which are 12 kinds of minerals.
And thirdly, selecting a training sample. Selecting a known mining area and a range of 100m of extension as a positive sample; the known mine area extends beyond 400m, and areas considered mine-free based on expert knowledge are negative examples. The method is divided into a training set, a verification set and a test set according to the ratio of 7:2: 1.
And fourthly, constructing a model. And constructing a model of the deep fully-connected neural network based on the self-attention mechanism, training the model by using the training set constructed in the previous step, performing precision verification on the model by using the verification set and the test set, and finishing the training when the precision of the verification set is converged to a stable state.
The Self-attention (Self-attention) mechanism was proposed by the Google machine translation team in 2017 (Vaswani et al, 2017), the basic idea is to calculate the weights of different parts and then perform weighted summation, and the final effect is to pay different degrees of attention to the different parts. The self-attention mechanism adopts parallel learning, and the relation between each feature and other features is not extracted in sequence, so that the calculation time is shortened, and the long-distance dependence problem is solved. For input a i e.R (i ═ 1,2, …, n) is mapped to three different spaces through linearity, a query matrix Q (query), a key matrix K (Keys) and a value matrix V (value) are respectively obtained, and a is obtained through training i Weight matrix w linearly mapped to Q, K, V q 、w k 、w v 。
The calculation of the attention score is mainly divided into three steps:
1) a is to i Query and each key (including a) i ) Calculating similarity to obtain weight, wherein common similarity functions comprise dot products, perceptrons and the like;
2) normalizing the weights using a Softmax function;
3) and carrying out weighted summation on the weight and the corresponding key Value to obtain the final attention score. The mathematical formula of the attention operation is (Vaswani et al, 2017):
in the formula d k Is the vector latitude of K and,
the weights are mainly scaled to prevent the calculated dot product result from being too large when the vector dimension is too high.
Wherein the input a of the self-attention mechanism i That is, the characteristic information of each prediction variable extracted from the hyperspectral remote sensing data, the geochemical data and the geophysical data is output and calculated for each prediction variable characteristic informationThe latter attention score, the higher the score the more important.
The model of the deep fully-connected neural network based on the self-attention mechanism consists of a 1-layer self-attention mechanism and a 4-layer fully-connected neural network. The machine learning model is applied to mineral prediction for the first time.
And fifthly, identifying the target area of the prospecting. Based on the prediction result data obtained by the previous model operation, a value of 0 or 1 is obtained, wherein 1 is mineral and 0 is non-mineral. And (4) combining geological data by experts to screen and identify the target area of the ore exploration.
The scheme of the embodiment has the following two innovation points:
1. the extraction spatial resolution of the predictor variable is 60 m;
2. the prediction variable uses mineral information extracted by hyperspectral remote sensing data with the resolution of better than 30m, geophysical and geochemical data with the scale of 1:5 ten thousand, and geology (such as fracture, lithology, altered zone and the like) is not used.
The embodiment of the invention also provides a porphyry shallow-to-low-temperature hydrothermal type mineral prediction system based on deep learning, which is used for realizing the steps of the porphyry shallow-to-low-temperature hydrothermal type mineral prediction method based on deep learning, and comprises the following steps:
the system comprises a predictive variable preparation module, a data acquisition module and a data analysis module, wherein the predictive variable preparation module is used for acquiring hyperspectral remote sensing data, geochemical data and geophysical data of a target mining area and sampling all groups of data into 60m resolution;
the predictive variable data processing module is used for carrying out normalization processing on the geochemical data and the geophysical data so as to remove the influence of data dimension;
the training sample selection module selects a target mining area and a range with the extension of 100m as positive samples, selects a region with the extension of 400m beyond the target mining area as a negative sample, judges that the region without the mine is the negative sample, and divides the region without the mine into a training set, a verification set and a test set according to the ratio of 7:2: 1;
the model construction module is used for constructing a model of the deep fully-connected neural network based on the self-attention mechanism, training the model by using the training set constructed in the previous step, simultaneously performing precision verification on the model by using the verification set and the test set, and finishing the training when the verification precision converges to a stable state;
and the ore-finding target area identification module obtains prediction result data after training is finished, obtains 1 indicating that ore exists and obtains 0 indicating that no ore exists.
Fig. 2 is a schematic diagram of an embodiment of an electronic device according to an embodiment of the invention. As shown in fig. 2, an embodiment of the present invention provides an electronic device, which includes a memory 1310, a processor 1320, and a computer program 1311 stored in the memory 1310 and executable on the processor 1320, where the processor 1320 executes the computer program 1311 to implement the following steps: s1, acquiring hyperspectral remote sensing data, geochemical data and geophysical data of the target mining area, and sampling all the data into 60m resolution;
s2, normalizing the geochemical data and the geophysical data to remove the influence of data dimension;
s3, selecting a target mining area and an area with an extension of 100m as positive samples, selecting an area with an extension of 400m of the target mining area as a negative sample, judging that the area without the mining area is a negative sample, and dividing the area without the mining area into a training set, a verification set and a test set according to a ratio of 7:2: 1;
s4, constructing a model of the deep fully-connected neural network based on the self-attention mechanism, training the model by using the training set constructed in the previous step, simultaneously performing precision verification on the model by using a verification set and a test set, and finishing training when the precision of the verification set is converged to a stable state;
and S5, obtaining prediction result data after training, outputting the prediction result data by using a torch.max () function, wherein the output result is the corresponding classification with the maximum prediction probability, and obtaining 1 representing the existence of the ore and 0 representing the absence of the ore as a binary classification problem.
Please refer to fig. 3, which is a schematic diagram of an embodiment of a computer-readable storage medium according to the present invention. As shown in fig. 3, the present embodiment provides a computer-readable storage medium 1400, on which a computer program 1411 is stored, which computer program 1411, when executed by a processor, implements the steps of: s1, acquiring hyperspectral remote sensing data, geochemical data and geophysical data of a target mining area, and sampling all the data into 60m resolution;
s2, normalizing the geochemical data and the geophysical data to remove the influence of data dimension;
s3, selecting a target mining area and an area with an extension of 100m as positive samples, selecting an area with an extension of 400m of the target mining area as a negative sample, judging that the area without the mining area is a negative sample, and dividing the area without the mining area into a training set, a verification set and a test set according to a ratio of 7:2: 1;
s4, constructing a model of the deep fully-connected neural network based on the self-attention mechanism, training the model by using the training set constructed in the previous step, simultaneously performing precision verification on the model by using a verification set and a test set, and completing the training when the precision of the verification set is converged to a stable state;
and S5, obtaining prediction result data after training, outputting the prediction result data by using a torch.max () function, wherein the output result is the corresponding classification with the maximum prediction probability, and obtaining 1 representing the existence of the ore and 0 representing the absence of the ore as a binary classification problem.
Has the beneficial effects that:
1. by adding the hyperspectral remote sensing mineral information extraction data and not using geological elements, the influence of uncertainty of the geological elements is effectively avoided, and the prediction precision of the model is improved.
2. For the hyperspectral data with spatial resolution better than 30m and the geochemical and geophysical measured data with a scale of 1:5 ten thousand, the extraction spatial resolution of the predictive variable is reasonably determined to be 60 m. The results show that the target area of the prospecting predicted by the model accounts for 2.96 percent of the total area of the research area, and 100 percent of potential ore deposits are covered. At present, the target area of the ore exploration defined by machine learning accounts for 7-25% of the total area of the research area, and 77-100% of potential ore deposits are covered. The technology effectively improves the prediction precision and reduces the potential target area for finding the ore.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing descriptions of specific exemplary embodiments of the present invention have been presented for purposes of illustration and description. It is not intended to limit the invention to the precise form disclosed, and obviously many modifications and variations are possible in light of the above teaching. The exemplary embodiments were chosen and described in order to explain certain principles of the invention and its practical application to enable one skilled in the art to make and use various exemplary embodiments of the invention and various alternatives and modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the claims and their equivalents.
Claims (10)
1. A porphyry shallow-to-low-temperature hydrothermal type mineral prediction method based on deep learning is characterized by comprising the following steps of:
s1, acquiring hyperspectral remote sensing data, geochemical data and geophysical data of the target mining area, and sampling all the data into 60m resolution;
s2, normalizing the geochemical data and the geophysical data to remove the influence of data dimension;
s3, selecting a target mining area and an area with an extension of 100m as positive samples, selecting an area with an extension of 400m of the target mining area as a negative sample, judging that the area without the mining area is a negative sample, and dividing the area without the mining area into a training set, a verification set and a test set according to a ratio of 7:2: 1;
s4, constructing a model of the deep fully-connected neural network based on the self-attention mechanism, training the model by using the training set constructed in the previous step, simultaneously performing precision verification on the model by using a verification set and a test set, and completing the training when the precision of the verification set is converged to a stable state;
and S5, obtaining prediction result data after training is finished, outputting the prediction result data by using a torch.max () function, wherein the output result is the corresponding classification with the maximum prediction probability, and the output result is used as a two-classification problem to obtain 1 representing that the ore exists and obtain 0 representing that the ore does not exist.
2. The deep learning-based porphyry shallow-to-low-temperature hydrothermal type mineral prediction method according to claim 1, wherein the target mine area is a shallow-to-low-temperature hydrothermal type mineral deposit or a porphyry type mineral deposit.
3. The method for predicting shallow-to-low-temperature hydrothermal mineral deposits of porphyry based on deep learning of claim 1, wherein the S1 specifically comprises:
firstly, preprocessing hyperspectral remote sensing data of a target mining area to obtain reflectivity data, eliminating bad wave bands influenced by water vapor, and then performing minimum noise classification transformation (MNF) and Pure Pixel Index (PPI) analysis;
then, mineral data information is extracted and resampled to 60m resolution using a combination of the hybrid tuned matched filter method (MTMF) and the Spectral Angle Method (SAM).
4. The method of claim 3, wherein the mineral data information includes data information of alunite, pyrophyllite, dickite, kaolinite, chlorite/brevifibrite, limonite, high-Al sericite, medium-Al sericite, low-Al sericite, dolomite, calcite, and montmorillonite.
5. The method for predicting shallow-to-low-temperature hydrothermal mineral deposits of porphyry based on deep learning of claim 1, wherein the S1 specifically comprises:
optimizing a variation function model for the geochemical data by using GS + software, and performing Kriging interpolation to obtain a resolution of 60m according to an optimized result;
and (3) carrying out polarization, horizontal gradient film, vertical first derivative and upward continuation processing on the geophysical data, and sampling to obtain the spatial resolution of 60 m.
6. The method for predicting shallow-to-low-temperature hydrothermal mineral deposits of porphyry based on deep learning of claim 1, wherein the S4 specifically comprises:
1) inputting characteristic information of each prediction variable extracted from the hyperspectral remote sensing data, the geochemical data and the geophysical data into a model, and calculating similarity to obtain weight;
2) normalizing the weights using a Softmax function;
3) and carrying out weighted summation on the weight and the corresponding key value to obtain the final attention score.
7. The deep learning-based porphyry shallow-temperature hydrothermal type mineral prediction method as claimed in claim 1, wherein the model comprises a 1-layer self-attention mechanism and a 4-layer fully-connected neural network.
8. A deep learning based prediction system for shallow-to-low-temperature hydrothermal mineral production of porphyry, which is used for implementing the deep learning based prediction method for shallow-to-low-temperature hydrothermal mineral production of porphyry according to any one of claims 1-7, and comprises the following steps:
the system comprises a predictive variable preparation module, a data acquisition module and a data processing module, wherein the predictive variable preparation module is used for acquiring hyperspectral remote sensing data, geochemical data and geophysical data of a target mining area and sampling all groups of data into 60m resolution;
the predictive variable data processing module is used for carrying out normalization processing on the geochemical data and the geophysical data so as to remove the influence of data dimension;
the training sample selection module is used for selecting a target mining area and positive samples within the range of 100m of extension, negative samples beyond 400m of extension of the target mining area, judging that the area without the mine is a negative sample, and dividing the area without the mine into a training set, a verification set and a test set according to the ratio of 7:2: 1;
the model construction module is used for constructing a model of the deep fully-connected neural network based on the self-attention mechanism, training the model by using the training set constructed in the previous step, simultaneously performing precision verification on the model by using the verification set and the test set, and finishing the training when the precision of the verification set is converged to a stable state;
and the ore-finding target area identification module is used for obtaining prediction result data after training is finished, outputting the prediction result data by using a torch.max () function, and taking the output result as the corresponding classification with the maximum prediction probability as a two-classification problem to obtain 1 representing that ore exists and 0 representing that no ore exists.
9. An electronic device, comprising a memory and a processor, wherein the processor is configured to execute a computer management program stored in the memory to implement the method for predicting low-temperature hydrothermal mineral deposit based on deep learning porphyry according to any one of claims 1 to 7.
10. A computer-readable storage medium, having stored thereon a computer management-like program which, when executed by a processor, implements the steps of the deep learning-based porphyry shallow-to-low-temperature hydrothermal mineral prediction method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210388950.0A CN114898109B (en) | 2022-04-14 | 2022-04-14 | Zeolite type mineral product prediction method and system based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210388950.0A CN114898109B (en) | 2022-04-14 | 2022-04-14 | Zeolite type mineral product prediction method and system based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114898109A true CN114898109A (en) | 2022-08-12 |
| CN114898109B CN114898109B (en) | 2023-05-02 |
Family
ID=82717271
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210388950.0A Active CN114898109B (en) | 2022-04-14 | 2022-04-14 | Zeolite type mineral product prediction method and system based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114898109B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115879648A (en) * | 2023-02-21 | 2023-03-31 | 中国地质科学院 | Machine learning-based ternary deep mineralization prediction method and system |
| CN116307123A (en) * | 2023-02-23 | 2023-06-23 | 中国地质大学（武汉） | Knowledge graph driving-based mineral resource prediction method and storage medium |
| CN117934019A (en) * | 2024-03-22 | 2024-04-26 | 南京海关工业产品检测中心 | Copper concentrate sample tracing method and system based on deep learning |
| CN118211614A (en) * | 2024-05-22 | 2024-06-18 | 中南大学 | Method for constructing three-dimensional ore-forming prediction deep learning model based on attention mechanism |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6430547B1 (en) * | 1999-09-22 | 2002-08-06 | International Business Machines Corporation | Method and system for integrating spatial analysis and data mining analysis to ascertain relationships between collected samples and geology with remotely sensed data |
| CN108021853A (en) * | 2016-11-01 | 2018-05-11 | 核工业北京地质研究院 | A kind of remote sensing images recognition methods suitable for Granite Region vein |
| CN111859811A (en) * | 2020-07-30 | 2020-10-30 | 中国地质科学院矿产资源研究所 | Ore formation prediction method and system based on fuzzy logic and machine learning |
| CN112766321A (en) * | 2020-12-31 | 2021-05-07 | 中国地质调查局成都地质调查中心 | Geological feature detection and identification method and system based on deep learning |
| CN112906940A (en) * | 2021-01-19 | 2021-06-04 | 中国地质调查局发展研究中心 | Sniffing intelligent ore finding prediction method based on deep learning |
| CN113723255A (en) * | 2021-08-24 | 2021-11-30 | 中国地质大学（武汉） | Hyperspectral image classification method and storage medium |
-
2022
- 2022-04-14 CN CN202210388950.0A patent/CN114898109B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6430547B1 (en) * | 1999-09-22 | 2002-08-06 | International Business Machines Corporation | Method and system for integrating spatial analysis and data mining analysis to ascertain relationships between collected samples and geology with remotely sensed data |
| CN108021853A (en) * | 2016-11-01 | 2018-05-11 | 核工业北京地质研究院 | A kind of remote sensing images recognition methods suitable for Granite Region vein |
| CN111859811A (en) * | 2020-07-30 | 2020-10-30 | 中国地质科学院矿产资源研究所 | Ore formation prediction method and system based on fuzzy logic and machine learning |
| CN112766321A (en) * | 2020-12-31 | 2021-05-07 | 中国地质调查局成都地质调查中心 | Geological feature detection and identification method and system based on deep learning |
| CN112906940A (en) * | 2021-01-19 | 2021-06-04 | 中国地质调查局发展研究中心 | Sniffing intelligent ore finding prediction method based on deep learning |
| CN113723255A (en) * | 2021-08-24 | 2021-11-30 | 中国地质大学（武汉） | Hyperspectral image classification method and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 常睿春等: "蚀变矿物信息与构造在高光谱遥感找矿中的应用初探", 《国土资源科技管理》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115879648A (en) * | 2023-02-21 | 2023-03-31 | 中国地质科学院 | Machine learning-based ternary deep mineralization prediction method and system |
| CN115879648B (en) * | 2023-02-21 | 2023-06-06 | 中国地质科学院 | Ternary deep mining prediction method and system based on machine learning |
| CN116307123A (en) * | 2023-02-23 | 2023-06-23 | 中国地质大学（武汉） | Knowledge graph driving-based mineral resource prediction method and storage medium |
| CN116307123B (en) * | 2023-02-23 | 2023-11-14 | 中国地质大学（武汉） | Knowledge graph driving-based mineral resource prediction method and storage medium |
| CN117934019A (en) * | 2024-03-22 | 2024-04-26 | 南京海关工业产品检测中心 | Copper concentrate sample tracing method and system based on deep learning |
| CN117934019B (en) * | 2024-03-22 | 2024-05-28 | 南京海关工业产品检测中心 | Copper concentrate sample tracing method and system based on deep learning |
| CN118211614A (en) * | 2024-05-22 | 2024-06-18 | 中南大学 | Method for constructing three-dimensional ore-forming prediction deep learning model based on attention mechanism |
| CN118211614B (en) * | 2024-05-22 | 2024-08-02 | 中南大学 | Method for constructing three-dimensional ore-forming prediction deep learning model based on attention mechanism |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114898109B (en) | 2023-05-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114898109B (en) | Zeolite type mineral product prediction method and system based on deep learning | |
| CN110728224B (en) | Remote sensing image classification method based on attention mechanism depth Contourlet network | |
| Parsa | A data augmentation approach to XGboost-based mineral potential mapping: an example of carbonate-hosted ZnPb mineral systems of Western Iran | |
| WO2021139069A1 (en) | General target detection method for adaptive attention guidance mechanism | |
| Yin et al. | Ensemble learning models with a Bayesian optimization algorithm for mineral prospectivity mapping | |
| CN110442666B (en) | Mineral resource prediction method and system based on neural network model | |
| CN116665067B (en) | Ore finding target area optimization system and method based on graph neural network | |
| CN113344045B (en) | Method for improving SAR ship classification precision by combining HOG characteristics | |
| CN114399073A (en) | Ocean surface temperature field prediction method based on deep learning | |
| CN114997501A (en) | Deep learning mineral resource classification prediction method and system based on sample unbalance | |
| CN114139819A (en) | Geochemical variable space prediction method based on geostatistical weighted random forest | |
| CN113435254A (en) | Sentinel second image-based farmland deep learning extraction method | |
| CN115951430A (en) | Geomagnetic index forecasting method based on LSTM | |
| Wei et al. | Seven-day sea surface temperature prediction using a 3DConv-LSTM model | |
| Pan et al. | Geological mapping via convolutional neural network based on remote sensing and geochemical survey data in vegetation coverage areas | |
| Liu et al. | A deep-learning-based mineral prospectivity modeling framework and workflow in prediction of porphyry–epithermal mineralization in the Duolong ore District, Tibet | |
| CN117710744A (en) | Hyperspectral image classification method based on band selection feature fusion multi-attention network | |
| Farzamian et al. | A weighted fuzzy aggregation GIS model in the integration of geophysical data with geochemical and geological data for Pb–Zn exploration in Takab area, NW Iran | |
| Bougrain et al. | Knowledge recovery for continental-scale mineral exploration by neural networks | |
| Li et al. | Prospectivity and Uncertainty Analysis of Tungsten Polymetallogenic Mineral Resources in the Nanling Metallogenic Belt, South China: A Comparative Study of AdaBoost, GBDT, and XgBoost Algorithms | |
| CN116994071A (en) | Multispectral laser radar point cloud classification method based on self-adaptive spectrum residual error | |
| Hittawe et al. | Time-series weather prediction in the Red sea using ensemble transformers | |
| CN111523258B (en) | Microseism effective signal first arrival pickup method and system based on MS-Net network | |
| CN114063063A (en) | Geological disaster monitoring method based on synthetic aperture radar and point-like sensor | |
| Bai et al. | Seamount age prediction machine learning model based on multiple geophysical observables: methods and applications in the Pacific Ocean |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |