A prospecting method, device, equipment and medium based on multi-source remote sensing technology

Abstract

The present invention discloses a prospecting method, device, equipment and medium based on multi-source remote sensing technology, which relates to the technical field of remote sensing geological survey. The prospecting method based on multi-source remote sensing technology includes: using multi-source remote sensing technology to determine the distribution data of ore-indicating factors in the study area; the ore-indicating factors include: geological data, geophysical data and geochemical element content data of the study area extracted by multi-source remote sensing technology; analyzing the mineral content of known ore points in the study area to determine the ore grade of the known ore points; using the ore grade of the known ore points and the ore-indicating factors of the known ore points as sample data to train a pre-constructed prospecting model to obtain a trained prospecting model; inputting the distribution data of the ore-indicating factors in the study area into the trained prospecting model to predict the mineral content of the study area through the prospecting model and determine the prospecting potential area in the study area.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing geological surveys, and particularly to a prospecting method, device, equipment and medium based on multi-source remote sensing technology. Background Technique

[0002] Geological prospecting methods, geophysical prospecting methods, geochemical prospecting methods, and remote sensing prospecting methods are currently relatively mature prospecting exploration technical methods. Among them, the geological prospecting method is the most basic technology in prospecting methods and is also the most widely used. It starts from studying the geological background of mineralization, comprehensively uses knowledge such as mineralogy, petrology, structural geology, and ore deposit geology to study the geological conditions of mineralization and prospecting geological criteria, so as to search for minerals; the geophysical prospecting method is a method based on the changes in the geophysical field, using physical principles and instrument equipment to obtain information such as the electromagnetic, gravity, and density of ore bodies, and then judging the potential location and distribution law of mineral resources; the geochemical prospecting method is a method that masters the distribution law of geochemical elements based on the theories of mineralogy and geochemistry and then guides prospecting; the remote sensing prospecting method has developed rapidly in recent years and is a method that uses remote sensing technology to obtain surface and underground information to predict and locate mineral resources.

[0003] Traditional geological, geophysical, and geochemical prospecting methods usually rely on intensive ground surveys and drilling means, which are not only costly but also have a greater impact on the environment. Conventional remote sensing prospecting methods mainly focus on obtaining geological information such as lithology, structure, and altered minerals, and the remote sensing extraction of important geological, geophysical, and geochemical information such as geothermal anomalies, gravity and magnetic anomalies, and geochemical element contents has not been included in the consideration scope of comprehensive prospecting. Although it has become a reality to extract ore-indicating factors such as geology (lithology, structure, altered minerals, geothermal), geophysics (gravity anomaly, magnetic anomaly), and geochemistry (geochemical element content) using remote sensing data, how to combine the ore-indicating factors of geology, geophysics, and geochemistry obtained by remote sensing means to complete prospecting exploration remains to be studied in depth. Summary of the Invention

[0004] In view of the above technical status quo, the present invention provides a prospecting method, device, equipment and medium based on multi-source remote sensing technology.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] A prospecting method based on multi-source remote sensing technology, comprising:

[0007] Using multi-source remote sensing technology to determine the distribution data of ore-indicating factors in the study area; wherein, the ore-indicating factors include: the geological data, geophysical data, and geochemical element content data of the study area extracted by multi-source remote sensing technology;

[0008] Analyze the mineral content of the known ore spots in the study area to determine the ore grade of the known ore spots; wherein, the ore grade represents the mineral content of the known ore spots.

[0009] Use the ore grade of the known ore spots and the ore-indicating factors of the known ore spots as sample data to train a pre-constructed ore prospecting model, and obtain a trained ore prospecting model.

[0010] Input the distribution data of the ore-indicating factors in the study area into the trained ore prospecting model, so as to combine the distribution data of the ore-indicating factors through the ore prospecting model to predict the mineral content of the study area and determine the ore prospecting potential area in the study area.

[0011] In an alternative embodiment of the present application, the analyzing the mineral content of the known ore spots in the study area to determine the ore grade of the known ore spots includes:

[0012] Analyze the mineral content of multiple ore samples collected from the known ore spots to obtain the mineral content of each ore sample.

[0013] Calculate the average value of the mineral content of each ore sample to determine the ore grade of the known ore spots.

[0014] In an alternative embodiment of the present application, the geological data includes: altered mineral data, lithology data, structure data, and geothermal data.

[0015] The geological data is obtained by the following method:

[0016] Obtain the ground spectral data of altered minerals in the study area measured by a geophysical spectrometer, and obtain the hyperspectral satellite remote sensing data of the study area collected by a satellite remote sensing platform; according to the ground spectral data of altered minerals and the hyperspectral satellite remote sensing data, obtain the altered mineral data of the study area.

[0017] Obtain the Landsat-8 multispectral remote sensing data, ASTER multispectral remote sensing data, and Sentinel-1 radar remote sensing data of the study area collected by the satellite remote sensing platform; according to the color, shape, and texture characteristics of different lithologies in the visible-shortwave infrared bands of the Landsat-8 multispectral remote sensing data and the ASTER multispectral remote sensing data, obtain the lithology data of the study area.

[0018] According to the color, shape, and texture characteristics of different lithologies in the visible-shortwave infrared bands of the Landsat-8 multispectral remote sensing data and the ASTER multispectral remote sensing data, and the radar remote sensing characteristics of the Sentinel-1 radar remote sensing data, obtain the structure data of the study area.

[0019] Through the single-window algorithm, perform land surface temperature inversion on the thermal infrared bands of the Landsat-8 multispectral remote sensing data and the ASTER multispectral remote sensing data to obtain the geothermal data of the study area.

[0020] In an alternative embodiment of the present application, the geophysical data includes: gravity anomaly data and magnetic anomaly data;

[0021] The geophysical data is obtained by the following method:

[0022] Obtain the distance change data between two satellites in the GRACE satellite; combine the geophysical model and the distance change data to analyze the gravity field change of the study area and determine the gravity anomaly data of the study area;

[0023] Obtain the intensity and direction of the earth's magnetic field measured by the Zhangheng-1 satellite; perform magnetic anomaly analysis on the intensity and direction of the earth's magnetic field to determine the magnetic anomaly data of the study area.

[0024] In an alternative embodiment of the present application, the content of the geochemical elements is obtained by the following method:

[0025] Obtain the content of geochemical elements at the sample points in the study area, and the UAV hyperspectral remote sensing data of the study area collected by the UAV remote sensing platform;

[0026] Based on the UAV hyperspectral remote sensing data of the study area, determine the pixel values of the sample points in different bands;

[0027] Perform correlation analysis on the content of geochemical elements at the sample points and the pixel values of the sample points in different bands to determine the characteristic bands in the UAV hyperspectral remote sensing data;

[0028] Based on the content of geochemical elements at the sample points and the pixel values of the sample points in the characteristic bands, train the pre-constructed geochemical element prediction model to obtain a trained geochemical element prediction model;

[0029] Input the UAV hyperspectral remote sensing data into the trained geochemical element prediction model, so as to predict the content of geochemical elements in the study area based on the UAV hyperspectral remote sensing data through the geochemical element prediction model, and obtain the content of geochemical elements in the study area.

[0030] In an alternative embodiment of the present application, the sample data includes training sample data and test sample data; the training sample data is used to train the pre-constructed prospecting model, and the test sample data is used to test the accuracy of the prospecting model.

[0031] In an alternative embodiment of the present application, it further includes:

[0032] Determine the acceptance rate of each hidden layer of the prospecting model;

[0033] Train the prospecting model with the sample data to determine the first training loss of the prospecting model;

[0034] When data augmentation is performed on the sample data input to the prospecting model in the m-th hidden layer of the prospecting model, determine the second training loss of the m-th hidden layer;

[0035] According to the second training loss and the first training loss, optimize the acceptance rate of the m-th hidden layer by using the gradient descent method;

[0036] Based on the optimized acceptance rate of the m-th hidden layer, optimize the acceptance rates of other hidden layers of the prospecting model except the m-th hidden layer;

[0037] Where m is a positive integer, and m is less than or equal to the number of hidden layers of the prospecting model.

[0038] Compared with the prior art, the prospecting method based on multi-source remote sensing technology provided by the present invention uses multi-source remote sensing technology to extract the distribution data of different types of ore-indicating factors in the study area, and at the same time combines the ore-indicating factors and ore grades of known ore points to train the pre-constructed prospecting model, so that the prospecting model can learn the relationship between the ore-indicating factors and the ore grades. Furthermore, based on the prospecting model, the mineral content of the distribution data of the ore-indicating factors in the study area collected by multi-source remote sensing technology is predicted to determine the prospecting potential area in the study area. This method realizes the joint prediction and positioning of mineral resources based on ore-indicating factors of geology, geophysics, and geochemistry, which is beneficial to improving the prediction and positioning accuracy of mineral resources.

[0039] The present invention also provides a prospecting device based on multi-source remote sensing technology, including:

[0040] An ore-indicating factor acquisition unit for using multi-source remote sensing technology to determine the distribution data of ore-indicating factors in the study area; wherein, the ore-indicating factors include: geological data, geophysical data, and geochemical element content data of the study area extracted by multi-source remote sensing technology;

[0041] An ore grade determination unit for analyzing the mineral content of known ore points in the study area to determine the ore grade of the known ore points; wherein the ore grade represents the mineral content of the known ore points.

[0042] An ore prospecting model training unit for using the ore grade of the known ore points and the ore indicating factors of the known ore points as sample data to train a pre-constructed ore prospecting model to obtain a trained ore prospecting model.

[0043] An ore prospecting model prediction unit for inputting the distribution data of the ore indicating factors in the study area into the trained ore prospecting model, so as to combine the distribution data of the ore indicating factors through the ore prospecting model to predict the mineral content in the study area and determine the ore prospecting potential area in the study area.

[0044] Compared with the prior art, the beneficial effects of the ore prospecting device based on multi-source remote sensing technology provided by the present invention are the same as those of the ore prospecting method based on multi-source remote sensing technology described in the above technical solution, and will not be elaborated here.

[0045] The present invention also provides an electronic device, including:

[0046] A processor;

[0047] A memory for storing executable instructions of the processor;

[0048] The processor is used to execute the above-mentioned ore prospecting method based on multi-source remote sensing technology by running the instructions in the memory.

[0049] Compared with the prior art, the beneficial effects of the electronic device provided by the present invention are the same as those of the ore prospecting method based on multi-source remote sensing technology described in the above technical solution, and will not be elaborated here.

[0050] The present invention also provides a computer storage medium, in which instructions are stored, and when the instructions are run, the above-mentioned ore prospecting method based on multi-source remote sensing technology is implemented.

[0051] Compared with the prior art, the beneficial effects of the computer storage medium provided by the present invention are the same as those of the ore prospecting method based on multi-source remote sensing technology described in the above technical solution, and will not be elaborated here. Description of the Drawings

[0052] The drawings described herein are used to provide a further understanding of the present invention, and constitute a part of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention. In the drawings:

[0053] Figure 1Flowchart of the ore prospecting method based on multi-source remote sensing technology provided by the embodiments of the present application.

[0054] Figure 2 Schematic diagram of ore-indicating factor extraction provided by the embodiments of the present application.

[0055] Figure 3 Flowchart of the application and training of an ore prospecting model provided by the embodiments of the present application.

[0056] Figure 4 Ore-forming prediction potential map provided by the embodiments of the present application.

[0057] Figure 5 Structure diagram of the ore prospecting device based on multi-source remote sensing technology provided by the embodiments of the present application.

[0058] Figure 6 Schematic diagram of the structure of an electronic device provided by the embodiments of the present application. Detailed implementation manners

[0059] For the convenience of clearly describing the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, terms such as "first" and "second" are used to distinguish identical or similar items with basically the same functions and effects. For example, the first threshold and the second threshold are only used to distinguish different thresholds, and do not limit their sequence. Those skilled in the art can understand that terms such as "first" and "second" do not limit the quantity and execution order, and terms such as "first" and "second" do not necessarily limit being different.

[0060] It should be noted that in the present invention, words such as "exemplary" or "for example" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as "exemplary" or "for example" in the present invention should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Exactly speaking, using words such as "exemplary" or "for example" aims to present related concepts in a specific manner.

[0061] In the present invention, "at least one" means one or more, and "a plurality" means two or more. "And / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone, where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects before and after. "At least one (item)" or its similar expression refers to any combination of these items, including any combination of a single item or multiple items. For example, at least one (item) of a, b, or c can represent: a, b, c, the combination of a and b, the combination of a and c, the combination of b and c, or the combination of a, b, and c, where a, b, and c can be single or multiple.

[0062] Geological prospecting methods, geophysical prospecting methods, geochemical prospecting methods, and remote sensing prospecting methods are relatively mature prospecting and exploration technical methods currently in use. Among them, geological prospecting method is the most fundamental technology in prospecting methods and is also the most widely used. It starts from studying the geological background of ore-forming processes, comprehensively applies knowledge such as mineralogy, petrology, structural geology, and ore deposit geology, studies the ore-forming geological conditions and geological criteria for prospecting, and thus searches for mineral deposits; Geophysical prospecting method is a method based on the changes in the geophysical field, using physical principles and instrument equipment to obtain information such as the electromagnetic, gravity, and density of ore bodies, and then judging the potential location and distribution law of mineral resources; Geochemical prospecting method is a method that, based on the theories of mineralogy and geochemistry, masters the distribution law of geochemical elements and then guides prospecting; Remote sensing prospecting method has developed rapidly in recent years and is a method that uses remote sensing technology to obtain surface and underground information to predict and locate mineral resources.

[0063] Traditional geological, geophysical, and geochemical prospecting methods usually rely on intensive ground surveys and drilling means, which are not only costly but also have a greater impact on the environment. Conventional remote sensing prospecting methods mainly focus on obtaining geological information such as lithology, structure, and altered minerals, and the remote sensing extraction of important geological, geophysical, and geochemical information such as geothermal anomalies, gravity and magnetic anomalies, and geochemical element contents has not been included in the consideration scope of comprehensive prospecting. Although it has become a reality to extract ore-indicating factors such as geology (lithology, structure, altered minerals, geothermal), geophysics (gravity anomaly, magnetic anomaly), and geochemistry (geochemical element content) using remote sensing data, how to combine the ore-indicating factors of geology, geophysics, and geochemistry obtained by remote sensing means to complete prospecting and exploration needs to be further studied.

[0064] In view of the above technical status quo, the embodiments of the present application provide a prospecting method, device, equipment, and medium based on multi-source remote sensing technology, which will be described in detail one by one in the following embodiments.

[0065] The embodiments of the present application first provide a prospecting method based on multi-source remote sensing technology. Please refer to Figure 1 , Figure 1 which is the flowchart of the prospecting method based on multi-source remote sensing technology provided by the embodiments of the present application.

[0066] As Figure 1 shown, the prospecting method based on multi-source remote sensing technology includes the following S101 to S104:

[0067] S101, using multi-source remote sensing technology to determine the distribution data of ore-indicating factors in the study area; wherein, the ore-indicating factors include: geological data, geophysical data, and geochemical element content data of the study area extracted by multi-source remote sensing technology.

[0068] The multi-source remote sensing technology refers to a technology that comprehensively uses remote sensing data from different sensors and platforms to obtain remote sensing information of the same area. In the embodiments of the present application, the remote sensing data collected by the multi-source remote sensing technology includes: remote sensing data collected by satellites, remote sensing data collected by ground object spectrometers, and remote sensing data collected by unmanned aerial vehicles.

[0069] The ore-indicating factors can be understood as geological, geochemical, geophysical and other characteristics or signs that can indicate the existence of ore deposits or potential mineralization areas. In the embodiments of the present application, the ore-indicating factors include: geological data, geophysical data, and chemical element content data.

[0070] Please refer to Figure 2 , Figure 2 which is the schematic diagram of ore-indicating factor extraction provided by the embodiments of the present application.

[0071] Among them, the types of geological data in the ore-indicating factors, as well as the geochemical element content, can be obtained based on the analysis of typical ore deposits in the study area and the collection and summary of geochemical remote sensing data of the mining areas in the study area.

[0072] The geological data includes: altered mineral data, lithological data, structural data, and geothermal data.

[0073] The altered mineral data is used to represent the data of secondary minerals formed by the chemical change of minerals in the original rock due to natural processes such as hydrothermal action and weathering in the study area. The rocks around the ore deposit often undergo alteration, forming a specific alteration zone. For example, altered mineral data such as skarnization, sericitization, and chloritization can indicate the existence of hydrothermal ore deposits.

[0074] Specifically, the altered mineral data can be obtained through the following method:

[0075] Obtain the ground spectral data of the altered minerals in the study area measured by a ground object spectrometer, and obtain the hyperspectral satellite remote sensing data of the study area collected by a satellite remote sensing platform;

[0076] According to the ground spectral data of the altered minerals and the hyperspectral satellite remote sensing data, obtain the altered mineral data of the study area.

[0077] As Figure 2 shown, the ground object spectrometer can adopt a CSD350-type ground object spectrometer to conduct actual spectral measurements on the altered minerals in the study area, and then construct the ground spectral data of the altered minerals in the study area to construct a spectral data set that conforms to the actual environment of the study area for different types of altered minerals in the study area.

[0078] The hyperspectral satellite remote sensing data is measured by the GF-5 hyperspectral satellite or the Ziyuan-1 02D satellite. After obtaining the hyperspectral satellite remote sensing data, conduct altered mineral analysis on the hyperspectral satellite remote sensing data with reference to the ground spectral data of the altered minerals, and then obtain the altered mineral data.

[0079] Lithological data mainly includes the rock types and mineral assemblages in the study area. Different rock types are related to different types of ore deposit characteristics. For example, sulfide deposits are usually related to igneous rocks, and sedimentary deposits are related to sedimentary rocks. In addition, specific mineral assemblages can also indicate the presence of ore deposits. For example, the combination of pyrite, galena, and sphalerite often indicates sulfide deposits.

[0080] Tectonic data, which is used to represent the geological structure of the study area; such as faults, fractures, folds, etc. These geological structures are important channels for the migration and enrichment of ore fluids. Collecting tectonic data related to the geological structure of the study area is conducive to ore fluid exploration in the study area.

[0081] Geothermal data, which is used to indicate the existence of hydrothermal activities in the study area and thus indicate potential ore deposits; many ore deposits, especially metal ore deposits, such as gold, silver, copper, lead, zinc, etc., are formed in hydrothermal activities. Hydrothermal activity refers to the process in which high-temperature fluids (hydrothermal fluids) deep underground rise along geological structures to the surface or near the surface, carrying and precipitating minerals. Hydrothermal activities will cause the temperature on the surface or in the shallow subsurface to rise, forming geothermal anomalies.

[0082] Specifically, the lithological data, tectonic data, and geothermal data can be obtained through the following methods:

[0083] First, for the lithological data and tectonic data.

[0084] The lithological data can be obtained from the Landsat-8 multispectral remote sensing data and ASTER multispectral remote sensing data of the study area collected by a satellite remote sensing platform.

[0085] Specifically, according to the color, shape, and texture characteristics of different lithologies in the visible - shortwave infrared bands of the Landsat - 8 multispectral remote sensing data and the ASTER multispectral remote sensing data, the lithology data of the study area is obtained.

[0086] The tectonic data can be obtained from the Landsat - 8 multispectral remote sensing data, ASTER multispectral remote sensing data, and Sentinel - 1 radar remote sensing data of the study area collected by the satellite remote sensing platform.

[0087] Specifically, according to the color, shape, and texture characteristics of different lithologies in the visible - shortwave infrared bands of the Landsat - 8 multispectral remote sensing data and the ASTER multispectral remote sensing data, and the radar remote sensing characteristics of the Sentinel - 1 radar remote sensing data, the tectonic data of the study area is obtained.

[0088] Furthermore, the geothermal data can be obtained by using the single - window algorithm to perform land surface temperature inversion on the thermal infrared bands of the Landsat - 8 multispectral remote sensing data and the ASTER multispectral remote sensing data, thereby obtaining the geothermal data of the study area.

[0089] Geophysical data includes: gravity anomaly data and magnetic anomaly data. Among them, the gravity anomaly data can indicate underground density changes to facilitate the identification of ore deposits or mineralized areas; the magnetic anomaly data can indicate the distribution of underground magnetic substances to facilitate the identification of magnetic ore deposits and geological structures.

[0090] As Figure 2 shown, for the gravity anomaly data, it can be obtained based on GRACE satellite data. The GRACE satellite consists of two satellites flying along the same orbit. The distance change between the two satellites can be accurately measured by a microwave rangefinder. The change in distance reflects the inhomogeneity of the Earth's gravity field. By combining the distance change data with a geophysical model, the change in the Earth's gravity field can be deduced. Based on this principle, the gravity anomaly data of the study area can be deduced by identifying the distance change information of the GRACE satellite.

[0091] For the magnetic anomaly data, it can be obtained based on the intensity and direction of the Earth's magnetic field. The intensity and direction of the Earth's magnetic field can be obtained from the Zhangheng - 1 satellite data collected by the Zhangheng - 1 satellite equipped with a high - precision magnetometer. Then, after correcting the intensity and direction of the Earth's magnetic field by combining solar activity data and an atmospheric model, the magnetic anomaly data can be identified.

[0092] The geochemical element content data refers to the element content data of geochemical elements such as lead and zinc in the research area. In the actual application process, the geochemical element content can intuitively reflect the existence of ore deposits. For example, analyzing the anomalies of trace elements such as selenium, tellurium, and rhenium can indicate the existence of copper deposits; analyzing the sulfur isotope ratio can indicate the hydrothermal source of ore deposits.

[0093] Specifically, the geochemical element content can be obtained through the following methods:

[0094] Obtain the geochemical element content of the sample points in the research area and the unmanned aerial vehicle (UAV) hyperspectral remote sensing data of the research area collected by the UAV remote sensing platform.

[0095] Based on the UAV hyperspectral remote sensing data of the research area, determine the pixel values of the sample points in different bands.

[0096] Conduct a correlation analysis on the geochemical element content of the sample points and the pixel values of the sample points in different bands to determine the characteristic bands in the UAV hyperspectral remote sensing data.

[0097] Based on the geochemical element content of the sample points and the pixel values of the sample points in the characteristic bands, train a pre-constructed geochemical element prediction model to obtain a trained geochemical element prediction model.

[0098] Input the UAV hyperspectral remote sensing data into the trained geochemical element prediction model, so as to predict the geochemical element content of the research area based on the UAV hyperspectral remote sensing data through the geochemical element prediction model, and obtain the geochemical element content of the research area.

[0099] In the actual application process, hyperspectral satellites usually provide a spatial resolution of dozens of meters to hundreds of meters, and the remote sensing observation data of satellites is easily affected by cloud cover. For UAVs, due to their low flight altitude, UAVs can capture minute details on the ground and can provide a relatively high spatial resolution, usually between a few centimeters and a few meters. Therefore, in order to predict the geochemical element content of the research area and improve the prediction accuracy of chemical element content, the embodiments of the present application use UAVs to collect UAV hyperspectral remote sensing data, and determine the pixel values of the UAV hyperspectral remote sensing data of the sample points in different bands from the UAV hyperspectral remote sensing data collected by the UAVs.

[0100] As Figure 2 shown, the geochemical element content of the sample points in the research area can be obtained based on the pre-collected geochemical remote sensing data of the research area.

[0101] Further, each pixel of the UAV hyperspectral remote sensing data corresponds to a small area on the ground. Each pixel contains a series of numerical values, which are the pixel values. These numerical values represent the reflectance or radiation intensity of the small area in different bands. In the embodiments of the present application, these numerical values are related to the content of geochemical elements in the small area.

[0102] In the actual application process, the relationship between the pixel value and the content of geochemical elements is mainly established through the spectral characteristics of the earth surface substances. Different chemical elements and compounds have different absorption and reflection characteristics in a specific wavelength range, and these absorption and reflection characteristics can be reflected through the pixel value.

[0103] Further, a correlation analysis is performed on the content of geochemical elements at the sample point and the pixel values of the sample point in different bands. The purpose is to find the characteristic bands that can reflect the content of geochemical elements at the sample point from the pixel values of different bands of the sample point. In the actual application process, statistical methods (such as Pearson correlation coefficient, Spearman rank correlation coefficient, etc.) can be used for calculation.

[0104] The geochemical element prediction model can be understood as a machine learning model. Specifically, the geochemical element prediction model can be an AlexNet convolutional neural network model (The adaptive layer selection, AdaLASE), and the present application does not limit this.

[0105] S102, analyze the mineral content of the known ore points in the study area to determine the ore grade of the known ore points; wherein, the ore grade represents the mineral content of the known ore points.

[0106] The ore grade refers to the content of minerals or metals contained in the ore, usually expressed in percentage (%) or grams per ton (g / t).

[0107] Specifically, the above S102 includes:

[0108] Analyze the mineral content of multiple ore samples collected from the known ore points to obtain the mineral content of each ore sample; calculate the average value of the mineral content of each ore sample to determine the ore grade of the known ore points.

[0109] In the actual application process, the analysis of the ore samples can be realized by methods such as wet chemical analysis, atomic absorption spectrometry, magnetic testing, radioactive testing, etc., and the present application does not limit this.

[0110] In another alternative embodiment of the present application, when the data of the known ore points in the research area is relatively complete, the method of searching for data can also be used to analyze the mineral content of the known ore points in the research area.

[0111] S103. Use the ore grade of the known ore point and the ore indication factor of the known ore point as sample data to train a pre-constructed prospecting model to obtain a trained prospecting model.

[0112] The ore grade is used to directly reflect the actual content of useful minerals or metals in the ore, while the ore indication factor is used to reflect the characteristics of the ore point in different dimensions. Therefore, from one perspective, the ore indication factor can explain the formation reason of useful minerals or metals in the known ore point.

[0113] In view of this, the present application uses the ore grade of the known ore point and the ore indication factor of the known ore point as sample data to train a pre-constructed prospecting model, so that the prospecting model can learn to have the prospecting factor based on the ore point and predict the ore grade of the ore point.

[0114] The prospecting model can be understood as a machine learning model. Machine learning (a multi-disciplinary field that involves multiple disciplines such as probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory) is specifically used to study how a computer reorganizes its existing knowledge structure to continuously improve its performance. Machine learning generally includes technologies such as artificial neural networks, belief networks, reinforcement learning, transfer learning, and inductive learning. And machine learning belongs to a branch of artificial intelligence (Artificial Intelligence, AI) technology.

[0115] In the embodiment of the present application, the prospecting model is an AlexNet convolutional neural network model. The AlexNet convolutional neural network model is a classic convolutional neural network (Convolutional Neural Network, CNN) model. The AlexNet convolutional neural network model consists of five convolutional layers, two fully connected hidden layers, and one fully connected output layer, and has powerful recognition ability. At the same time, this model uses regularization technology to improve the accuracy of the model and reduce overfitting. And it uses ReLU instead of sigmoid as its activation function, which greatly speeds up the convergence and reduces the time-consuming of model training.

[0116] After obtaining the sample data through the above S101 to S103, the sample data can be input into the pre-constructed prospecting model, so that the prospecting model can learn the relationship between the ore indication factor and the ore grade, and make the prospecting model have the ability to predict the ore grade based on the ore indication factor.

[0117] In an alternative embodiment of the present application, the sample data includes: training sample data and test sample data; the training sample data is used to train a pre-constructed prospecting model, and the test sample data is used to test the accuracy of the prospecting model.

[0118] Specifically, 70% of the sample data can be used as the training sample data, and 30% of the sample data can be used as the test sample data.

[0119] Please refer to Figure 3 , Figure 3 which is a flowchart of the application and training of a prospecting model provided by an embodiment of the present application.

[0120] In the process of optimizing the prospecting model with the training sample data, the prospecting model will iteratively process the training sample set multiple times, and adjust the parameters of the prospecting model through an optimization algorithm (such as gradient descent) to minimize the loss.

[0121] Furthermore, the test sample data is mainly used to optimize the model to avoid overfitting.

[0122] In a neural network model, data augmentation is often an important part of model training. However, currently, data augmentation often stays at the data input layer of the model, and does not consider whether applying data augmentation to the hidden layer of the neural network will improve the performance of the model. At the same time, since there are often multiple hidden layers in a neural network, it is difficult to determine at which hidden layer data augmentation can achieve the best effect.

[0123] In order to improve the data processing ability of the prospecting model, improve the accuracy and speed of data processing of the prospecting model, in the process of training the prospecting model in an embodiment of the present application, the adaptive layer selection method (AdaLASE) is combined to optimize the prospecting model.

[0124] Specifically, the process of optimizing the prospecting model includes:

[0125] Determine the acceptance rate of each hidden layer of the prospecting model;

[0126] Train the prospecting model with the sample data to determine the first training loss of the prospecting model;

[0127] When performing data augmentation on the sample data input to the prospecting model at the m-th hidden layer of the prospecting model, determine the second training loss of the m-th hidden layer;

[0128] Optimize the acceptance rate of the m-th hidden layer by using the gradient descent method according to the second training loss and the first training loss;

[0129] Based on the optimized acceptance rate of the m-th hidden layer, optimize the acceptance rates of the other hidden layers except the m-th hidden layer of the prospecting model; where m is a positive integer, and m is less than or equal to the number of hidden layers of the prospecting model.

[0130] The acceptance rate of the hidden layer is used to guide the model to dynamically select which hidden layers to perform data augmentation or other operations during training, thereby improving the generalization ability and performance of the model.

[0131] In the actual application process, the sum of the acceptance rates of each hidden layer is 1, and at the initial iterative optimization, the initial acceptance rates of each hidden layer are the same. For example, when the prospecting model includes 5 hidden layers, the initial acceptance rate of each hidden layer is 0.2.

[0132] The above optimization process of the prospecting model is an iterative process. At the beginning of the iteration, the acceptance rate of the i-th hidden layer is the initial acceptance rate , and the training loss of each hidden layer can be expressed by the following formula (1):

[0133] (1);

[0134] where represents the total loss of the hidden layer of the prospecting model; K represents the number of hidden layers of the prospecting model; represents the training loss of the i-th hidden layer of the prospecting model.

[0135] In the actual application process, when the m-th hidden layer is selected as the most suitable hidden layer, after the training data to be input into the prospecting model propagates forward to the m-th hidden layer, the m-th hidden layer performs data augmentation, and finally the augmented data is propagated forward to the output layer to obtain the data result output by the output layer.

[0136] However, in order to obtain the most suitable hidden layer for data augmentation, it is necessary to update the acceptance rates of each layer iteratively and determine the most suitable hidden layer according to the acceptance rate.

[0137] Specifically, the optimization of the acceptance rate of the m-th hidden layer by using the gradient descent method can be achieved by the following formula (2):

[0138] (2);

[0139] where represents the optimized acceptance rate of the m-th hidden layer; represents the acceptance rate of the m-th hidden layer before optimization; represents the step size; represents the first training loss of the ore prospecting model; represents the second training loss of the m-th hidden layer; .

[0140] S104. Input the distribution data of the ore-indicating factors in the study area into the trained ore prospecting model, so as to combine the distribution data of the ore-indicating factors through the ore prospecting model to predict the mineral content in the study area and determine the ore prospecting potential areas in the study area.

[0141] Specifically, after obtaining the ore prospecting model, the distribution data of the ore-indicating factors in the study area can be input into the ore prospecting model, and then the ore prospecting model combines the ore-indicating factors of each ore point in the study area to analyze the ore grades of each ore point in the study area, and then determine the ore prospecting potential areas in the study area.

[0142] That is, after the ore prospecting model predicts the ore grades of each ore point, based on the mineral content expressed by the ore grades of each ore point, the ore prospecting potential areas in the study area are defined.

[0143] Please refer to Figure 4 , Figure 4 which is the metallogenic prediction potential map provided by the embodiment of the present application.

[0144] As Figure 4 shown, Figure 4 the predicted potential map shown includes the ore grades of each ore point in the study area. The mineral content expressed by the ore grades is different, and the corresponding colors in the predicted potential map are also different.

[0145] In the actual application process, after obtaining the metallogenic prediction potential map, relevant staff can calibrate the potential of the study area according to the metallogenic potential map. For example, the study area can be classified into high-potential metallogenic areas, medium-potential metallogenic areas, and low-potential metallogenic areas according to the potential map, so as to facilitate relevant staff to carry out exploration work.

[0146] In summary, the prospecting method based on multi-source remote sensing technology provided by the embodiments of the present application utilizes multi-source remote sensing technology to extract the distribution data of ore-indicating factors of different types in the study area. At the same time, in combination with the ore-indicating factors and ore grades of known ore points, the pre-constructed prospecting model is trained, enabling the prospecting model to learn the relationship between ore-indicating factors and ore grades. Furthermore, based on the prospecting model, the distribution data of ore-indicating factors of the study area collected by multi-source remote sensing technology is used to predict the mineral content, and the prospecting potential areas in the study area are determined. This method realizes the joint prediction and positioning of mineral resources based on ore-indicating factors of geology, geophysics, and geochemistry, which is beneficial to improving the prediction and positioning accuracy of mineral resources.

[0147] The embodiments of the present application also provide a prospecting device based on multi-source remote sensing technology. Please refer to Figure 5 , Figure 5 which is the structural diagram of the prospecting device based on multi-source remote sensing technology provided by the embodiments of the present application.

[0148] As Figure 5 shown, the prospecting device based on multi-source remote sensing technology includes:

[0149] An ore-indicating factor acquisition unit 501, configured to use multi-source remote sensing technology to determine the distribution data of ore-indicating factors in the study area; wherein, the ore-indicating factors include: geological data, geophysical data, and geochemical element content data of the study area extracted by multi-source remote sensing technology;

[0150] An ore grade determination unit 502, configured to perform mineral content analysis on the known ore points in the study area to determine the ore grades of the known ore points; wherein, the ore grade represents the mineral content of the known ore points;

[0151] A prospecting model training unit 503, configured to use the ore grades of the known ore points and the ore-indicating factors of the known ore points as sample data to train a pre-constructed prospecting model to obtain a trained prospecting model;

[0152] A prospecting model prediction unit 504, configured to input the distribution data of ore-indicating factors of the study area into the trained prospecting model, and through the prospecting model, in combination with the distribution data of ore-indicating factors, perform mineral content prediction on the study area to determine the prospecting potential areas in the study area.

[0153] In an alternative embodiment of the present application, the performing mineral content analysis on the known ore points in the study area to determine the ore grades of the known ore points includes:

[0154] Performing mineral content analysis on multiple ore samples collected from the known ore points to obtain the mineral content of each ore sample;

[0155] Calculate the average value of the mineral content of each of the ore samples to determine the ore grade of the known ore point.

[0156] In an alternative embodiment of the present application, the geological data includes: altered mineral data, lithological data, tectonic data, and geothermal data;

[0157] The geological data is obtained by the following method:

[0158] Obtain the altered mineral ground spectral data of the study area measured by a geophysical spectrometer, and obtain the hyperspectral satellite remote sensing data of the study area collected by a satellite remote sensing platform; according to the altered mineral ground spectral data and the hyperspectral satellite remote sensing data, obtain the altered mineral data of the study area;

[0159] Obtain the Landsat-8 multispectral remote sensing data, ASTER multispectral remote sensing data, and Sentinel-1 radar remote sensing data of the study area collected by the satellite remote sensing platform; according to the color, shape, and texture characteristics of different lithologies in the visible-shortwave infrared bands of the Landsat-8 multispectral remote sensing data and the ASTER multispectral remote sensing data, obtain the lithological data of the study area;

[0160] According to the color, shape, and texture characteristics of different lithologies in the visible-shortwave infrared bands of the Landsat-8 multispectral remote sensing data and the ASTER multispectral remote sensing data, and the radar remote sensing characteristics of the Sentinel-1 radar remote sensing data, obtain the tectonic data of the study area;

[0161] Through the single-window algorithm, perform land surface temperature inversion on the thermal infrared bands of the Landsat-8 multispectral remote sensing data and the ASTER multispectral remote sensing data to obtain the geothermal data of the study area.

[0162] In an alternative embodiment of the present application, the geophysical data includes: gravity anomaly data and magnetic anomaly data;

[0163] The geophysical data is obtained by the following method:

[0164] Obtain the distance change data between two satellites in the GRACE satellite; combine the geophysical model and the distance change data to analyze the gravity field change of the study area to determine the gravity anomaly data of the study area;

[0165] Obtain the intensity and direction of the earth's magnetic field measured by the Zhangheng-1 satellite; perform magnetic anomaly analysis on the intensity and direction of the earth's magnetic field to determine the magnetic anomaly data of the study area.

[0166] In an alternative embodiment of the present application, the content of the geochemical elements is obtained by the following method:

[0167] Obtain the content of the geochemical elements of the sample points in the study area, and the high - spectral remote - sensing data of the unmanned aerial vehicle (UAV) of the study area collected by the UAV remote - sensing platform;

[0168] Based on the UAV high - spectral remote - sensing data of the study area, determine the pixel values of the sample points in different bands;

[0169] Conduct a correlation analysis on the content of the geochemical elements of the sample points and the pixel values of the sample points in different bands to determine the characteristic bands in the UAV high - spectral remote - sensing data;

[0170] Based on the content of the geochemical elements of the sample points and the pixel values of the sample points in the characteristic bands, train a pre - constructed geochemical - element prediction model to obtain a trained geochemical - element prediction model;

[0171] Input the UAV high - spectral remote - sensing data into the trained geochemical - element prediction model, so as to predict the content of the geochemical elements of the study area based on the UAV high - spectral remote - sensing data through the geochemical - element prediction model, and obtain the content of the geochemical elements of the study area.

[0172] In an alternative embodiment of the present application, the sample data includes training sample data and test sample data; the training sample data is used to train the pre - constructed ore - prospecting model, and the test sample data is used to test the accuracy of the ore - prospecting model.

[0173] In an alternative embodiment of the present application, the device is further configured to:

[0174] Determine the acceptance rate of each hidden layer of the ore - prospecting model;

[0175] Train the ore - prospecting model with the sample data to determine the first training loss of the ore - prospecting model;

[0176] When data augmentation is performed on the sample data input into the ore - prospecting model in the m - th hidden layer of the ore - prospecting model, determine the second training loss of the m - th hidden layer;

[0177] According to the second training loss and the first training loss, optimize the acceptance rate of the m - th hidden layer by using the gradient - descent method;

[0178] Based on the optimized acceptance rate of the m - th hidden layer, optimize the acceptance rates of the other hidden layers except the m - th hidden layer of the ore - prospecting model;

[0179] Wherein, m is a positive integer, and m is less than or equal to the number of hidden layers of the prospecting model.

[0180] The above device embodiment provided in this embodiment and the method embodiment of the present application belong to the same inventive concept. For technical details not described in detail in this embodiment, reference may be made to the specific processing content of the prospecting method based on multi-source remote sensing technology provided in the above embodiments of the present application, which will not be elaborated herein.

[0181] This application embodiment also provides an electronic device, such as Figure 6 shown. Figure 6 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

[0182] Such as Figure 6 shown, the electronic device includes:

[0183] A processor 210;

[0184] A memory 200 for storing executable instructions of the processor 210;

[0185] The processor 210 is configured to execute the prospecting method based on multi-source remote sensing technology disclosed in any of the above embodiments by running the instructions in the memory 200.

[0186] The processor 210, the memory 200, the communication interface 220, the input device 230, and the output device 240 are interconnected through a bus. Among them:

[0187] The bus may include a path for transmitting information between various components of the computer system.

[0188] The processor 210 may be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the program of the present invention. It may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0189] The processor 210 may include a main processor, and may also include a baseband chip, a modem, etc.

[0190] The program for implementing the technical solution of the present invention is stored in the memory 200, and the operating system and other key services can also be stored. Specifically, the program can include program code, and the program code includes computer operation instructions. More specifically, the memory 200 can include a read-only memory (ROM), other types of static storage devices that can store static information and instructions, a random access memory (RAM), other types of dynamic storage devices that can store information and instructions, a disk memory, a flash memory, and so on.

[0191] The input device 230 can include devices for receiving data and information input by the user, such as a keyboard, a mouse, a camera, a scanner, a touch screen, etc.

[0192] The output device 240 can include devices for allowing information to be output to the user, such as a display screen, a printer, a speaker, etc.

[0193] The communication interface 220 can include devices of any transceiver type for communicating with other devices or communication networks, such as Ethernet, a radio access network (RAN), a wireless local area network (WLAN), etc.

[0194] The processor 210 executes the program stored in the memory 200 and calls other devices, and can be used to implement each step of any one of the prospecting methods based on multi-source remote sensing technology provided in the above embodiments of the present application.

[0195] In addition to the above methods and devices, the embodiments of the present application can also be a computer program product, which includes computer program instructions. When the computer program instructions are run by a processor, the processor executes the steps in the prospecting methods based on multi-source remote sensing technology in various embodiments of the present application.

[0196] The computer program product can be written in any combination of one or more programming languages for the program code to execute the operations of the embodiments of the present application. The programming languages include object-oriented programming languages, such as Java, C++, etc., and also include conventional procedural programming languages, such as the "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, executed as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0197] In addition, the embodiments of the present application can also be a storage medium, on which a computer program is stored, and the computer program is executed by a processor for the steps in the prospecting methods based on multi-source remote sensing technology in various embodiments of the present application.

[0198] For the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should understand that this application is not limited by the described action sequence, because according to this application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0199] It should be noted that the embodiments in this specification are all described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. For the similarities and commonalities among the embodiments, reference can be made to each other. For the device embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and reference can be made to the relevant parts of the method embodiments for the related content.

[0200] The steps in the methods of the embodiments of this application can be adjusted, combined, and deleted according to actual needs. The technical features recorded in each embodiment can be replaced or combined.

[0201] The modules and sub-modules in the devices and terminals in the embodiments of this application can be combined, divided, and deleted according to actual needs.

[0202] In several embodiments provided by this application, it should be understood that the disclosed terminals, devices, and methods can be implemented in other ways. For example, the terminal embodiments described above are only illustrative. For example, the division of modules or sub-modules is only a logical function division. In actual implementation, there can be other division methods. For example, multiple sub-modules or modules can be combined or integrated into another module, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of devices or modules can be in electrical, mechanical, or other forms.

[0203] The modules or sub-modules described as separate components may or may not be physically separated. The components as modules or sub-modules may or may not be physical modules or sub-modules, that is, they can be located in one place, or they can be distributed to multiple network modules or sub-modules. Some or all of the modules or sub-modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0204] In addition, in each embodiment of the present application, each functional module or sub-module can be integrated into a processing module, or each module or sub-module can exist physically alone, or two or more modules or sub-modules can be integrated into one module. The above-mentioned integrated module or sub-module can be implemented in the form of hardware, or in the form of a software functional module or sub-module.

[0205] Those skilled in the art can further realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Skilled professionals can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0206] The steps of the method or algorithm described in combination with the embodiments disclosed herein can be directly implemented by hardware, a software unit executed by a processor, or a combination of the two. The software unit can be placed in a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, register, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the technical field.

[0207] Finally, it should also be noted that in this document, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element.

[0208] The foregoing description of the disclosed embodiments enables those skilled in the art to practice or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A prospecting method based on multi-source remote sensing technology, characterized in that: include: Using multi-source remote sensing technology to determine the distribution data of the ore-indicating factors in the study area; wherein the ore-indicating factors include: geological data, geophysical data and geochemical element content data of the study area extracted by multi-source remote sensing technology; The geophysical data include: gravity anomaly data and magnetic anomaly data; The geophysical data are obtained by the following method: Obtaining distance change data between two satellites in the GRACE satellites; combining the geophysical model and the distance change data, analyzing the gravity field change of the study area, and determining the gravity anomaly data of the study area; Obtaining the strength and direction of the Earth's magnetic field measured by the Zhangheng-1 satellite; performing magnetic anomaly analysis on the strength and direction of the Earth's magnetic field to determine magnetic anomaly data for the study area; The geochemical element contents are obtained by the following method: Obtaining the geochemical element contents of the sample points in the study area, and the drone hyperspectral remote sensing data of the study area collected by the drone remote sensing platform; Based on the UAV hyperspectral remote sensing data of the study area, determining the pixel values ​​of the sample points in different bands; Performing correlation analysis on the geochemical element content of the sample point and the pixel value of the sample point in different bands to determine the characteristic bands in the UAV hyperspectral remote sensing data; Based on the geochemical element content of the sample point and the pixel value of the sample point in the characteristic band, a pre-constructed geochemical element prediction model is trained to obtain a trained geochemical element prediction model; Inputting the UAV hyperspectral remote sensing data into the trained geochemical element prediction model, so as to predict the geochemical element content of the study area based on the UAV hyperspectral remote sensing data through the geochemical element prediction model, and obtain the geochemical element content of the study area; Conducting mineral content analysis on known mineral points in the study area to determine the ore grade of the known mineral points; wherein the ore grade represents the mineral content of the known mineral points; Using the ore grade of the known mineral point and the ore-indicating factor of the known mineral point as sample data, training a pre-constructed mineral prospecting model to obtain a trained mineral prospecting model; Inputting the distribution data of the ore-indicating factors of the study area into the trained ore prospecting model, so as to predict the mineral content of the study area by combining the ore prospecting model with the distribution data of the ore-indicating factors, obtain the ore grade of each ore point predicted by the ore prospecting model, and determine the ore prospecting potential area in the study area based on the ore grade of each ore point; The method further comprises: Determine the acceptance rate of each hidden layer of the prospecting model; train the prospecting model through the sample data to determine the first training loss of the prospecting model; when the mth hidden layer of the prospecting model performs data enhancement on the sample data input into the prospecting model, determine the second training loss of the mth hidden layer; optimize the acceptance rate of the mth hidden layer using the gradient descent method according to the second training loss and the first training loss; based on the optimized acceptance rate of the mth hidden layer, optimize the acceptance rates of other hidden layers of the prospecting model except the mth hidden layer; wherein m is a positive integer, and m is less than or equal to the number of hidden layers of the prospecting model; The second training loss of the mth hidden layer is determined by the following formula: ； in, represents the total loss of the hidden layer of the prospecting model; K represents the number of hidden layers of the prospecting model; represents the second training loss of the mth hidden layer of the prospecting model; represents the acceptance rate of the mth hidden layer before optimization; The acceptance rate of the optimized mth hidden layer is determined by the following formula: ； in, represents the acceptance rate of the optimized mth hidden layer; represents the step length; represents the first training loss of the prospecting model; .

2. The prospecting method based on multi-source remote sensing technology according to claim 1, characterized in that: The method of analyzing the mineral content of the known mineral points in the study area to determine the ore grade of the known mineral points includes: Performing mineral content analysis on a plurality of ore samples collected from the known mining sites to obtain the mineral content of each of the ore samples; The mineral content of each of the ore samples is averaged to determine the ore grade of the known ore point.

3. The prospecting method based on multi-source remote sensing technology according to claim 1, characterized in that: The geological data include: altered mineral data, lithology data, structural data and geothermal data; The geological data are obtained by the following method: Obtaining ground spectrum data of altered minerals in the study area measured by a ground object spectrometer, and obtaining high-spectral satellite remote sensing data of the study area collected by a satellite remote sensing platform; obtaining altered mineral data in the study area based on the ground spectrum data of altered minerals and the high-spectral satellite remote sensing data; Acquire Landsat-8 multispectral remote sensing data, ASTER multispectral remote sensing data and Sentinel-1 radar remote sensing data of the study area collected by the satellite remote sensing platform; acquire lithology data of the study area according to the color, shape and texture characteristics of different lithologies in the visible light-shortwave infrared band of the Landsat-8 multispectral remote sensing data and the ASTER multispectral remote sensing data; Acquire structural data of the study area based on the color, shape and texture characteristics of different lithologies in the visible light-shortwave infrared band of the Landsat-8 multispectral remote sensing data and the ASTER multispectral remote sensing data, and the radar remote sensing characteristics of the Sentinel-1 radar remote sensing data; The surface temperature is inverted on the thermal infrared bands of the Landsat-8 multispectral remote sensing data and the ASTER multispectral remote sensing data through a single window algorithm to obtain the geothermal data of the study area.

4. The prospecting method based on multi-source remote sensing technology according to claim 1, characterized in that: The sample data includes training sample data and test sample data; the training sample data is used to train the pre-built prospecting model, and the test sample data is used to verify the accuracy of the prospecting model.

5. A prospecting device based on multi-source remote sensing technology, characterized in that: include: The ore-indicating factor acquisition unit is used to determine the distribution data of the ore-indicating factors of the study area by using multi-source remote sensing technology; wherein the ore-indicating factors include: geological data, geophysical data and geochemical element content data of the study area extracted by multi-source remote sensing technology; The geophysical data include: gravity anomaly data and magnetic anomaly data; The geophysical data are obtained by the following method: Obtaining distance change data between two satellites in the GRACE satellites; combining the geophysical model and the distance change data, analyzing the gravity field change of the study area, and determining the gravity anomaly data of the study area; Obtaining the strength and direction of the Earth's magnetic field measured by the Zhangheng-1 satellite; performing magnetic anomaly analysis on the strength and direction of the Earth's magnetic field to determine magnetic anomaly data for the study area; The geochemical element contents are obtained by the following method: Obtaining the geochemical element contents of the sample points in the study area, and the drone hyperspectral remote sensing data of the study area collected by the drone remote sensing platform; Based on the UAV hyperspectral remote sensing data of the study area, determining the pixel values ​​of the sample points in different bands; Performing correlation analysis on the geochemical element content of the sample point and the pixel value of the sample point in different bands to determine the characteristic bands in the UAV hyperspectral remote sensing data; Based on the geochemical element content of the sample point and the pixel value of the sample point in the characteristic band, a pre-constructed geochemical element prediction model is trained to obtain a trained geochemical element prediction model; Inputting the UAV hyperspectral remote sensing data into the trained geochemical element prediction model, so as to predict the geochemical element content of the study area based on the UAV hyperspectral remote sensing data through the geochemical element prediction model, and obtain the geochemical element content of the study area; An ore grade determination unit, used to analyze the mineral content of the known mineral points in the study area to determine the ore grade of the known mineral points; wherein the ore grade represents the mineral content of the known mineral points; A prospecting model training unit is used to train a pre-built prospecting model using the ore grade of the known mineral point and the ore-indicating factor of the known mineral point as sample data to obtain a trained prospecting model; A prospecting model prediction unit is used to input the distribution data of the ore-indicating factors of the study area into the trained prospecting model, so as to predict the mineral content of the study area by combining the prospecting model with the distribution data of the ore-indicating factors, obtain the ore grade of each ore point predicted by the prospecting model, and determine the prospecting potential area in the study area based on the ore grade of each ore point; the method also includes: The prospecting device is also used to determine the acceptance rate of each hidden layer of the prospecting model; train the prospecting model through the sample data to determine the first training loss of the prospecting model; when the mth hidden layer of the prospecting model performs data enhancement on the sample data input into the prospecting model, determine the second training loss of the mth hidden layer; optimize the acceptance rate of the mth hidden layer using the gradient descent method according to the second training loss and the first training loss; based on the optimized acceptance rate of the mth hidden layer, optimize the acceptance rates of other hidden layers of the prospecting model other than the mth hidden layer; wherein m is a positive integer, and m is less than or equal to the number of hidden layers of the prospecting model; The second training loss of the mth hidden layer is determined by the following formula: ； in, represents the total loss of the hidden layer of the prospecting model; K represents the number of hidden layers of the prospecting model; represents the second training loss of the mth hidden layer of the prospecting model; represents the acceptance rate of the mth hidden layer before optimization; The acceptance rate of the optimized mth hidden layer is determined by the following formula: ； in, represents the acceptance rate of the optimized mth hidden layer; represents the step length; represents the first training loss of the prospecting model; .

6. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is used to execute the prospecting method based on multi-source remote sensing technology as described in any one of claims 1 to 4 by running instructions in the memory.

7. A computer storage medium, characterized in that: The computer storage medium stores instructions, and when the instructions are executed, the prospecting method based on multi-source remote sensing technology as described in any one of claims 1 to 4 is executed.

Images