Intelligent identification method and system for mineral resources based on multi-source remote sensing data

Abstract

This invention discloses a method and system for intelligent identification of mineral resources based on multi-source remote sensing data, relating to the field of mineral resource management technology. In the identification platform, a basic geographic model within a target geographic area is constructed based on big data. A model network is set up corresponding to the basic geographic model. The basic geographic model is improved based on multi-source remote sensing data and stored in the unit model sections of the corresponding range. Geological data is labeled based on the target geographic model, and mineral information is matched against the target geographic surface model based on a mineral information matching rule base. This invention, through the model network, enables the pre-construction of models within a local area of ​​the basic geographic model, followed by replacement of local models within the basic geographic model, avoiding reconstruction of the entire basic geographic model, reducing server load. Furthermore, it allows for simultaneous reconstruction and replacement of multiple layout models, improving the efficiency and stability of basic geographic model updates.

Description

Technical Field

[0001] This invention relates to the field of mineral resource management technology, specifically to a method and system for intelligent identification of mineral resources based on multi-source remote sensing data. Background Technology

[0002] With the continuous iteration of technologies such as satellite remote sensing (optical, radar, hyperspectral, thermal infrared), airborne remote sensing (UAV aerial survey), and low-altitude remote sensing, the spatial resolution, spectral resolution, and temporal resolution of multi-source remote sensing data have been steadily improved. It has the capability to conduct comprehensive, non-contact, and large-scale exploration of the Earth's surface and shallow geological bodies. It can accurately capture key information closely related to mineral mineralization, such as lithological characteristics, geological structures, alteration zone distribution, and mineralization anomalies, providing rich and high-quality data support for mineral resource identification and geological model construction.

[0003] However, traditional update methods for geological models based on multi-source remote sensing data suffer from numerous problems, severely hindering the refined and efficient development of mineral resource exploration. On one hand, traditional geological model updates cannot achieve parallel processing of multiple local models. Even if multiple local areas are unrelated, updates still require full-scale model reconstruction, leading to excessive server load, significant waste of computing resources, and substantial increases in update time costs. On the other hand, the traditional model strictly follows a fixed sequence of "model reconstruction—verification—replacement," lacking a flexible and efficient processing mechanism. This not only affects the overall efficiency of geological model updates and construction but also reduces model update stability, thus delaying mineral resource identification and failing to meet the core industry needs for refined and efficient mineral resource exploration. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for intelligent identification of mineral resources based on multi-source remote sensing data, so as to solve the problems in the background technology.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for intelligent identification of mineral resources based on multi-source remote sensing data, comprising the following steps: In the identification platform, a basic geographic model is constructed based on big data within the target geographic area. A model network is set up corresponding to the basic geographic model. The model network includes multiple unit model sections, and the multiple unit model sections are interconnected. The basic geographic model is improved based on multi-source remote sensing data and stored in the corresponding range of unit model sections. The geological model is stored based on the unit model sections. The target geographic model is obtained by replacing the basic geographic model with the target unit model sections. Geological data is labeled based on the target geographic model, and mineral information of the target geographic model is matched based on the mineral information matching rule base to complete the identification of mineral resource information within the target geographic area.

[0006] In a preferred embodiment, the step of constructing a basic geographic model within the target geographic area based on big data in the identification platform, and setting up a model network corresponding to the basic geographic model, includes: Geological data within the target geographic area is obtained from big data based on the identification platform, and a basic geographic model within the target geographic area is constructed based on the geological data. A model network is configured corresponding to the basic geographic model, and multiple unit model sections in the model network are connected to the basic geographic model.

[0007] In a preferred embodiment, the step of configuring a model network corresponding to the basic geographic model, wherein multiple unit model sections in the model network are all connected to the basic geographic model, includes: Multiple blank geographic model outlines are configured for the basic geographic model. These blank geographic model outlines are then spatially overlaid on the outer layer of the basic geographic model. The blank geographic model outline closest to the basic geographic model is used as the update model. The basic geographic model is then divided into multiple unit geographic models, where each unit geographic model includes the unit basic geographic model and its corresponding connection point. The outlines of multiple blank geographic models are divided according to multiple unit geographic models to obtain multiple unit model cross-sections, where each unit model cross-section includes the blank unit geographic model and the corresponding connection point. Connect the corresponding unit model sections in the unit geographic model and the updated model through connection points. Multiple blank geographic model outlines and the updated model are accessed in the order from the outside to the inside, and the unit model sections are connected sequentially through connection points. The corresponding update model sets multiple dynamic domains. Each dynamic domain is set one-to-one with the unit model section in the update model. Each dynamic domain includes multiple dynamic nodes. The connection points of the multiple dynamic nodes and the unit model section are connected. The multiple dynamic nodes are set at the edge of the unit model section to form an enclosure. Adjacent dynamic nodes in the dynamic domain are connected to each other. The dynamic domain is mapped to the corresponding unit geographic model.

[0008] In a preferred embodiment, the steps of refining the basic geographic model based on multi-source remote sensing data and storing it in the corresponding range of unit model sections, storing the geological model based on the unit model sections, and replacing the basic geographic model with the target unit model sections to obtain the target geographic model include: Acquire multi-source remote sensing data, convert the multi-source remote sensing data into geological data, compare the geological data with the geological data in the basic geographic model, and obtain the difference data; Determine the cell model section where the differential data is located and sequentially infiltrate the data into the updated model according to the outlines of multiple blank geographic models; When the unit model sections containing the differential data are adjacent, if the model range corresponding to the differential data of one unit model section is larger than the model range corresponding to the differential data of another unit model section, exceeding the preset conditions, the unit model section with the larger proportion of the model range corresponding to the differential data is taken as the main section. The dynamic domain corresponding to the main section is used to squeeze the adjacent dynamic domains to obtain a dynamic domain including the differential data, which is then mapped to the unit geographic model corresponding to the basic geographic model. The corresponding range is divided by the data points corresponding to the unit geographic model to obtain the main unit geographic model. Based on the multi-source remote sensing data corresponding to the differential data, a model is constructed in the main cross-section. The constructed main cross-section is disconnected from the adjacent unit model cross-sections, the main unit geographic model is disconnected from the adjacent unit geographic models, the main cross-section and the main unit geographic model are interchanged, and the connection between the main cross-section and the adjacent unit geographic models is established to obtain a new basic geographic model as the target geographic model. The main unit geographic model is connected to the adjacent unit model cross-sections to obtain a new updated model. Set access permissions for the model network to protect the underlying geographic model when unauthorized networks access the model network.

[0009] In a preferred embodiment, the step of setting access permissions for the model network to protect the underlying geographic model when unauthorized networks access the model network includes: When authorized network access to the model network exists, access is sequentially transferred from any connection point on the model network to the connection point of the basic geographic model. When an unauthorized network accesses the model network, the connection between the updated model's connection point and the connection points in other blank geographic model outlines is disconnected. At the same time, the connection points in other blank geographic model outlines are cross-connected. The unauthorized network can cross-access the connection points in the other blank geographic model outlines, thus preventing it from accessing the basic geographic model and completing the protection of the basic geographic model.

[0010] In a preferred embodiment, the steps of labeling geological data based on a target geographic model and matching mineral information of the target geographic model based on a mineral information matching rule base to complete the identification of mineral resource information within the target geographic area include: Establish association rules between geological data and geographic surface models based on the target geographic model, and mark the core feature information of geological data in the target geographic model; The geological data in the target geographic model is matched according to the mineral information matching rule base to obtain the matching level, corresponding mineral type, and mineralization potential characteristics, which are then marked in the target geographic model. Based on the target geographic model, the mineral resource information is integrated to form a mineral resource information identification result set, and the key mineralization areas are determined in the target geographic model.

[0011] This invention also provides a mineral resource intelligent identification system based on multi-source remote sensing data, comprising: The configuration module is used to construct a basic geographic model within the target geographic area based on big data in the identification platform. A model network is set up for the basic geographic model. The model network includes multiple unit model sections, which are interconnected. The data acquisition and construction module, connected to the settings module, is used to improve the basic geographic model based on multi-source remote sensing data and store it in the corresponding range of unit model sections. The geological model is stored based on the unit model sections. The target geographic model is obtained by replacing the basic geographic model through the target unit model sections. The identification module, connected to the data acquisition and construction module, is used to label geological data based on the target geographic model and match mineral information of the target geographic model based on the mineral information matching rule base, thereby completing the identification of mineral resource information within the target geographic area.

[0012] In a preferred embodiment, the setting module includes: The basic building blocks are used to acquire geological data within the target geographic area based on the identification platform, and to construct a basic geographic model within the target geographic area based on the geological data. The model building unit is used to configure the model network corresponding to the basic geographic model. Multiple unit model sections in the model network are connected to the basic geographic model.

[0013] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention enables the pre-construction of local models within a basic geographic model using a model network, and then replaces local models within the basic geographic model, avoiding reconstruction of the entire basic geographic model, reducing server load. Furthermore, it allows for the simultaneous reconstruction and replacement of multiple layout models, improving the efficiency and stability of basic geographic model updates. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0015] Figure 1 This is a flowchart of the method of the present invention.

[0016] Figure 2 This is a system block diagram of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1, please refer to Figure 1 As shown in this embodiment, the intelligent identification method for mineral resources based on multi-source remote sensing data includes the following steps: S1. Construct a basic geographic model within the target geographic area based on big data in the identification platform, and set up a model network corresponding to the basic geographic model. The model network includes multiple unit model sections, and the multiple unit model sections are interconnected. S2. Based on multi-source remote sensing data, the basic geographic model is improved and stored in the corresponding range of unit model sections. The geological model is stored based on the unit model sections. The target geographic model is obtained by replacing the basic geographic model with the target unit model sections. S3. Mark geological data based on the target geographic model, match mineral information of the target geographic model based on the mineral information matching rule base, and complete the identification of mineral resource information within the target geographic area.

[0019] As described in steps S1-S3 above, the model network enables the pre-construction of models within a local area of ​​the basic geographic model, and then the local models in the basic geographic model are replaced, avoiding reconstruction of the entire basic geographic model, reducing the server's operating load. Furthermore, multiple layout models can be reconstructed and replaced simultaneously, improving the efficiency and stability of updating the basic geographic model.

[0020] In one embodiment, step S1, which involves constructing a basic geographic model within the target geographic area based on big data in the identification platform and setting up a model network corresponding to the basic geographic model, includes: S11. Obtain geological data within the target geographic area based on the identification platform, and construct a basic geographic model within the target geographic area based on the geological data; S12. Configure a model network corresponding to the basic geographic model. Multiple unit model sections in the model network are connected to the basic geographic model. In one embodiment, step S12, where the corresponding basic geographic model is configured with a model network, and multiple unit model sections in the model network are all connected to the basic geographic model, includes: S121. Configure multiple blank geographic model outlines corresponding to the basic geographic model, and stack the multiple blank geographic model outlines in the access outer layer of the basic geographic model (the basic geographic model and the multiple blank geographic model outlines are stored in a large access space. According to the access settings of the storage space, the multiple blank geographic model outlines are placed in order before the access address of the space where the basic geographic model is located). The blank geographic model outline closest to the basic geographic model is used as the update model. The basic geographic model is divided into ranges to obtain multiple unit geographic models. The unit geographic model includes the unit basic geographic model and the corresponding connection point. S122. Divide the outlines of multiple blank geographic models according to multiple unit geographic models to obtain multiple unit model cross-sections, wherein the unit model cross-section includes the blank unit geographic model and the corresponding connection point. S123. Connect the corresponding unit model sections in the unit geographic model and the updated model through connection points. Multiple blank geographic model outlines and the updated model are connected sequentially through connection points in the order of access from the outside to the inside. S124. Multiple dynamic domains are set for the corresponding update model. The dynamic domains are set one-to-one with the unit model sections in the update model. The dynamic domains include multiple dynamic nodes (virtual machines). The connection points of the multiple dynamic nodes and the unit model sections are connected. The multiple dynamic nodes are set at the edges of the unit model sections to form an enclosure. Adjacent dynamic nodes in the dynamic domains are connected to each other. The dynamic domains are mapped to the corresponding unit geographic models.

[0021] As described in steps S11 and S12 above, before identifying mineral resources, it is necessary to determine the geological range to be identified. The range is determined on the geographic plane, and then the underground depth range is determined below the range, which is the underground space range. Then, based on big data, the existing geological data corresponding to the target geographic range is determined. For example, the existing geological data can be obtained through the Internet, including publicly available topographic and geomorphological geographic data, basic geographic vector data, remote sensing geographic feature data, geological structure geographic data, and geographic spatiotemporal correlation data of the region. A basic geographic model can be constructed based on existing geological data, although this data may not be accurate at present. Its purpose is to first build a basic geographic model based on existing data, so that the model can be quickly replaced and updated later. At the same time, in order to protect the final basic geographic model, a model net is set up for the basic geographic model. The model net is a blank geographic model with the same outline as the basic geographic model. Then, in order to update the basic geographic model in blocks, the basic geographic model is divided into multiple unit geographic models. The unit geographic model has a corresponding connection point. This connection point is a server used to store the model and connect it with the cross-sections of other unit models, and can also connect with the connection points of adjacent unit geographic models. The blank geographic model outline is divided according to the division method of the basic geographic model. Here, the blank geographic model outline is a three-dimensional spatial model with the same outline as the basic geographic model outline. This three-dimensional spatial model is divided into multiple blank geographic model outlines according to multiple unit geographic models, resulting in multiple unit model cross-sections. Connect the unit geographic model to the corresponding unit model section of the updated model, and set access permissions (authorized network) for the corresponding model network. This allows the range that needs to be updated to be provided to the unit model section, resulting in a specific unit model section corresponding to the range. The corresponding unit geographic model of the base geographic model can then be replaced through this unit model section. This replacement involves disconnecting the connection points of the range from the connection points of the adjacent unit geographic models, and replacing the original unit geographic model with the connection points of the unit model section.The unit geographic model and its cross-sections here are not fixed ranges. The range of the unit geographic model can be adaptively modified according to the range that needs to be continuously updated. Therefore, multiple dynamic domains are set for the updated model. In the updated model, each unit model cross-section has a dynamic domain, which is composed of multiple dynamic nodes. The dynamic nodes surround the edge of the unit model cross-section, which can limit the range of the unit model cross-section. When there are changes in the future, the edge of the unit model cross-section is re-divided by the dynamic nodes, and the corresponding connection points are notified. Since it is completely divided, the dynamic nodes corresponding to adjacent unit model cross-sections are close together. In this way, the range of the unit geographic model can be enlarged and reduced by "squeezing" or "yielding" to adjacent dynamic nodes. Then, the dynamic nodes adjacent to the dynamic domain are connected, and the resulting dynamic domain is mapped to the unit geographic model. This enables the unit geographic model to be updated and selected adaptively in the future, realizing the updating and selection of the basic geographic model layout, avoiding comprehensive and large-scale model modifications, and making the model update more efficient and stable.

[0022] In one embodiment, step S2, which involves refining the basic geographic model based on multi-source remote sensing data and storing it in a corresponding range of unit model sections, storing a geological model based on the unit model sections, and replacing the basic geographic model with the target unit model sections to obtain the target geographic model, includes: S21. Acquire multi-source remote sensing data, convert the multi-source remote sensing data into geological data, compare the geological data with the geological data in the basic geographic model, and obtain the difference data. S22. Determine the cell model section where the differential data is located and sequentially infiltrate the data into the updated model according to the outlines of multiple blank geographic models; S23. When the unit model sections containing the difference data are adjacent, if the model range corresponding to the difference data of one unit model section is larger than the model range corresponding to the difference data of another unit model section, exceeding the preset conditions, the unit model section with the larger proportion of the model range corresponding to the difference data is taken as the main section. The dynamic domain corresponding to the main section is used to squeeze the adjacent dynamic domains (the squeezing process is: the dynamic nodes that are adjacent and attached are moved to each other. Here, the dynamic nodes of the main section and the dynamic nodes of the adjacent unit model section are moved towards the adjacent unit model section, "squeezing" the model range of the adjacent unit model section), to obtain the dynamic domain including the difference data, and map it to the unit geographic model corresponding to the basic geographic model. The corresponding range is divided by the data points corresponding to the unit geographic model to obtain the main unit geographic model. S24. Based on the multi-source remote sensing data corresponding to the difference data, build a model in the main cross section. After the main cross section is built, disconnect it from the adjacent unit model cross section. Disconnect the main unit geographic model from the adjacent unit geographic model. Interchange the main cross section with the main unit geographic model. Establish a connection between the main cross section and the adjacent unit geographic model to obtain a new basic geographic model as the target geographic model. Connect the main unit geographic model with the adjacent unit model cross section to obtain a new updated model. S25. Set access permissions for the model network to protect the basic geographic model when there is unauthorized network access to the model network.

[0023] In one embodiment, step S25, which sets access permissions for the model network to protect the underlying geographic model when unauthorized networks access the model network, includes: S251. When there is an authorized network access model network, access can be transferred sequentially to the connection point of the basic geographic model through any connection point on the model network (since multiple unit model sections are sequentially hierarchically located before access to the basic geographic model, access can be transferred sequentially through the connections between the connection points on the model network to the access point on the basic geographic model). S252. When an unauthorized network accesses the model network, disconnect the connection points of the updated model from the connection points in other blank geographic model outlines, and cross-connect the connection points in other blank geographic model outlines. The unauthorized network will cross-access the connection points in the other blank geographic model outlines and will not be able to access the basic geographic model, thus completing the protection of the basic geographic model.

[0024] As described in steps S21-S23 above, the current multi-source remote sensing data is then acquired, specifically including multispectral remote sensing (typical data sources are: Sentinel-2, Landsat 8 / 9, GF series), hyperspectral remote sensing (typical data sources are: ASTER, PRISMA, GF-5, HyMap), thermal infrared remote sensing (typical data sources are: SDGSAT-1, ASTERTIR, LandsatTIRS), synthetic aperture radar (SAR) (typical data sources are: Sentinel-1, ALOSPALSAR), lidar (LiDAR) (typical data sources are: airborne / UAV LiDAR), and auxiliary data (typical data sources are: gravity, magnetic, geochemical, geological maps, drilling data). The multi-source remote sensing data covers the entire target geographic area. It is then converted into a geological structure and basic geographic model based on the current multi-source remote sensing data for improvement. (The purpose of this comparison includes selection, supplementation and reconstruction. When it is consistent with the existing basic geographic model, reconstruction is not required. When it is inconsistent with the existing basic geographic model, reconstruction is required based on the currently collected data. The reconstructed model is stored in any unit model section. Areas without geological data in the existing basic geographic model need to be supplemented.) The method for converting multi-source remote sensing data into geological structures is as follows: using the multi-source remote sensing feature set extracted in the first stage as input, combined with prior geological knowledge (such as the geological laws of metallogenic belts and the correspondence between lithology and spectral density), the remote sensing features are transformed into geological structures through qualitative interpretation, quantitative inversion, and spatial verification. Through joint matching of spectral features, texture features, and topographic features, different lithologies (such as granite, limestone, and basalt) have unique spectral absorption characteristics and remote sensing color textures, and correspond to specific topographic features (such as granite often forming low hills and mountains, and limestone often forming karst landforms). The inversion method is as follows: first, the lithology endmembers are determined by spectral unmixing of hyperspectral endmembers, then, combined with optical textures and LiDAR topographic features, random forest / support vector machine is used for lithology classification, and finally, geological prior constraint correction is performed by combining regional geological maps to output the spatial distribution boundaries, distribution direction, and contact relationships of lithological strata. Structural trace inversion involves identifying structural features (fractures / folds / joints) in remote sensing data as linear / circular characteristics, abrupt topographic changes, and anomalous scattering characteristics (e.g., fault zones appear as continuous linear textures, and folds appear as arc-shaped tonal variations). The process includes automatic extraction and manual interpretation of linear / circular features from optical / SAR data (e.g., Hough transform for linear structures); identification of structural deformation zones (hidden faults) using InSAR interferometry; identification of structural landforms (e.g., fault scarps, fold ridges) using LiDAR topographic data; and classification of structures by grade (primary / secondary faults), orientation (strike / dip / dip), and spatial relationships (intersection / parallelism). The output includes the spatial location, geometric parameters, and combination relationships of structural traces, forming a structural framework. Mineralization alteration structure inversion is a direct indicator of mineral formation. Alteration minerals (such as kaolinite, hematite, and sericite) have unique hyperspectral absorption characteristics and exhibit specific band ratio anomalies in optical data. Inversion method: Using hyperspectral data as the core, alteration anomaly areas are extracted through the spectral indices of alteration minerals (such as HCI clay index and Fe2O3 iron index). Anomaly verification is performed by combining the optical band ratio method. Then, spatial cluster analysis is used to determine the distribution range, zoning characteristics (inner / middle / outer zone), and spatial correlation with structure / lithology of the alteration zone, outputting the spatial distribution and grade of mineralization alteration structure. The inversion of concealed geological structures involves: targeting concealed rock masses / concealed faults in vegetation-covered / thickly topsoil areas, and using comprehensive anomaly inversion from multiple sources (such as thermal inertia anomalies from thermal infrared, coherence anomalies from SAR, and weak alteration anomalies from hyperspectral data); the inversion method is to integrate the anomaly features of each data source using a multi-source feature fusion algorithm (such as DS evidence theory and Bayesian fusion), combined with gravity / magnetic auxiliary data as constraints, to output the spatial location, burial depth, and distribution range of the concealed geological structures.The above approach combines geological prior knowledge constraints with cross-validation using multi-source data. The geological prior knowledge constraints involve incorporating regional geological maps, metallogenic regularities, and known mineral deposit data to correct errors in remote sensing inversion (such as excluding linear features of artificial features from being misidentified as faults). The multi-source cross-validation involves verifying the spatial correlation between alteration anomalies extracted by hyperspectral imaging and tectonic structures using SAR data, and verifying their topographic distribution patterns using LiDAR data, to ensure the accuracy and reliability of geological structure inversion.The above method enables the rapid construction of geological models within the target geographic area, significantly improving the speed of mineral identification analysis. After obtaining multi-source remote sensing data, the data is converted into geological structures using the aforementioned conversion method. Then, the geological data is compared with the geological data in the basic geographic model to obtain differential data. This differential data is the geological data that differs from the basic geographic model. The unit model sections containing the differential data are determined, and data is sequentially infiltrated into the updated model according to multiple blank geographic model outlines. This infiltration involves sequentially transmitting data through the outer unit model sections, but not storing it in the outer unit model sections; instead, it is stored in the updated model. The geological data with differential data is then constructed within the updated model. When dealing with models in certain localized areas, if the model range corresponding to the difference data in one unit model section is larger than the model range corresponding to the difference data in another unit model section (exceeding a preset condition), the unit model section with the larger proportion of the difference data's model range is designated as the primary section. The dynamic domain corresponding to the primary section is then used to compress the adjacent dynamic domains. For example, if the difference data is distributed across unit model sections A and B, with 70% in section A and 30% in section B (a difference of 40%), exceeding the preset condition of 30%, then section A is designated as the primary section, and the dynamic domain corresponding to the primary section is used to compress the adjacent dynamic domains. Dynamic domain compression means that only one dynamic domain is needed to encompass the range containing the difference data. Adjacent dynamic domains are attached, so the movement only requires attachment, moving towards the dynamic nodes of adjacent dynamic domains to shrink the range they encompass. If this doesn't exceed preset conditions, then the dynamic domains of the two unit model sections need to automatically shrink the model range corresponding to the difference data within their respective unit model sections. The dynamic domain will subsequently only include the model range corresponding to the difference data, discarding the rest. The dynamic domain movement architecture involves setting multiple model points corresponding to the model range within the unit model section that updates the model. These model points are communication nodes, set within the empty space where the model resides. During this process, dynamic nodes (virtual machines) can move between model points to define the model range space where differential data is located. The same model point is also set in the basic geographic model, and the model points in the basic geographic model and the model points in the updated model are positionally corresponding and connected. Therefore, the position of the model point in the updated model where the dynamic node is located can be sensed by the model point in the basic geographic model. This enables the mapping of the model range on the basic geographic model between multiple dynamic nodes in the subsequent dynamic domain. The dynamic domain is mapped to the corresponding unit geographic model of the basic geographic model. The range is divided according to the data points corresponding to the unit geographic model to obtain the main unit geographic model. This main unit geographic model is consistent with the dynamic domain that is reduced to only include differential data.Models are built on the main cross-section based on the multi-source remote sensing data corresponding to the differential data. The constructed main cross-section is then disconnected from the adjacent unit model cross-sections. The updated model at this point needs to be protected. The main unit geographic model is disconnected from the adjacent unit geographic models, and the main cross-section and the main unit geographic model are interchanged. This interchange involves swapping the corresponding data points, narrowing the model range corresponding to the differential data. This is to perform lightweight interchange here, improving the efficiency and stability of the interchange. It is not necessary to rebuild the entire basic geographic model; only a local interchange is needed. Multiple local models can be built and interchanged simultaneously, improving the construction efficiency of the basic geographic model. A connection is established between the main cross-section and the adjacent unit geographic models to obtain a new basic geographic model, completing the interchange operation. The main unit geographic model is connected to the adjacent unit model cross-sections to obtain a new updated model. Here, the connection points are reorganized through dynamic domains (reorganization of cloud servers). When there is authorized network access to the model network, access is sequentially transferred to the connection point of the basic geographic model through any connection point on the model network. Data protection can be provided for the updated model and the basic geographic model through multiple unit model cross-sections. When unauthorized network access to the model network exists, disconnecting the connection points of the updated model from the connection points in other blank geographic model outlines isolates the updated model from the outer blank geographic model outlines. Simultaneously, cross-connecting the connection points in other blank geographic model outlines allows unauthorized networks to cross-access through these cross-connected blank geographic model outlines, thus trapping external unauthorized network access within the cross-connected blank geographic model outlines and preventing access to the basic geographic model, thereby protecting it. Furthermore, if the updated model subsequently becomes corrupted, a blank geographic model outline outside the updated model can be sequentially used to replace it as the new updated model. This model network has multiple functions: it allows for the pre-construction of models within a local area of ​​the basic geographic model, followed by replacement of local models within the basic geographic model, avoiding reconstruction of the entire basic geographic model and reducing server load; secondly, it allows for the simultaneous reconstruction and replacement of multiple layout models, improving the efficiency and stability of basic geographic model updates.

[0025] In one embodiment, step S3, which involves labeling geological data based on a target geographic model and matching mineral information from the target geographic model using a mineral information matching rule base to identify mineral resource information within the target geographic area, includes: S31. Establish association rules between geological data and geographic surface model based on the target geographic model, and mark the core feature information of geological data in the target geographic model; S32. Match the geological data in the target geographic model according to the mineral information matching rule base to obtain the matching level, corresponding mineral type, and mineralization potential characteristics, and mark them in the target geographic model. Based on the target geographic model, integrate the mineral resource information to form a mineral resource information identification result set, and determine the key mineralization area in the target geographic model.

[0026] As described in steps S31 and S32 above, establish the association rules between geological data and geographic surface models. Based on the surface code, spatial location, and geological framework characteristics of the geographic surface model, mark the core feature information of the geological data (such as the copper element anomaly value of a certain surface, the lithology being granite, the structural type being fault, etc.) one by one in the corresponding geographic surface model unit. Integrate the marked geographic surface models to form a target geographic surface model with geological data markings. Each surface unit carries complete and standardized geological feature data, and the marking results are bound to the surface code to achieve a one-to-one correspondence between model units and geological data. Generate a geological data marking ledger to record the source, marking location, feature information, marking time, etc. of the marked data to complete the traceability of the marking process. This study analyzes the relationships between known mineral types, mineralization regularities, mineral characteristics, and geological conditions both domestically and internationally, establishing a basic mineral information database. This database includes mineral types (metallic, non-metallic, energy minerals, etc.), metallogenic geological indicators (lithology, structure, geochemical anomalies, etc.), mineral resource characteristics (grade, reserve size, occurrence state, etc.), and the geographical and geological matching relationships of known mineral deposits. Based on mineralization regularities, a mineral information matching rule database is constructed, clarifying the geological characteristic matching thresholds corresponding to different mineral types (e.g., the copper element anomaly threshold, lithological matching requirements, structural matching conditions, etc. for a specific copper deposit), and setting multiple feature weights (e.g., the weight of metallogenic structural features is higher than that of ordinary lithological features). The matching database is dynamically maintained and updated, promptly supplementing it with new mineralization theories, new exploration results, and new geological data to optimize matching rules and weights, thereby improving the adaptability and accuracy of the database. Using a basic geographic model with geological data labels as the matching subject, the mineral information matching rule library is invoked to initiate mineral information matching at the area unit level. The geological label data of each area unit is compared with the mineral mineralization geological indicators in the mineral information matching rule library (geological data comparison). According to a multi-feature weighted matching algorithm, the matching similarity between the target geographic area model area unit and each mineral type in the mineral information matching rule library is calculated. If the similarity reaches a preset matching threshold, the area unit is determined to have the mineralization potential of the corresponding mineral type. The matching results are graded, and based on the matching similarity, area units are divided into three levels: high matching degree, medium matching degree, and low matching degree (high matching degree indicates areas with significant mineralization potential, medium matching degree indicates areas with questionable mineralization potential, and low matching degree indicates areas with no obvious mineralization potential). If a single area unit matches multiple mineral types, cross-validation is performed based on mineralization patterns to eliminate unreasonable cross-type matching results and retain matching conclusions that conform to geological mineralization logic.The mineral information matching results from all target geographic models are integrated, and the matching level, corresponding mineral type, and metallogenic potential characteristics are marked in the spatial model of the target geographic model to form a spatial distribution map of mineral resources identification. Core identification information of mineral resources in the target geographic area is extracted, including: possible mineral types and distribution ranges in the area to be identified, metallogenic potential areas (high / medium / low) for each mineral type, key geological indicators of metallogenesis, and geographic area characteristics of mineral resource occurrence, etc., as mineral resource information, and a standardized mineral resource information identification result set is formed. The identification results are spatially integrated and statistically analyzed, and the metallogenic potential of mineral resources is statistically analyzed according to the administrative division, geological division, and geographic area unit of the target geographic area, and key exploration areas are identified as key metallogenic areas.

[0027] Example 2, please refer to Figure 2 As shown in this embodiment, the intelligent mineral resource identification system based on multi-source remote sensing data includes: The configuration module is used to construct a basic geographic model within the target geographic area based on big data in the identification platform. A model network is set up for the basic geographic model. The model network includes multiple unit model sections, which are interconnected. The data acquisition and construction module, connected to the settings module, is used to improve the basic geographic model based on multi-source remote sensing data and store it in the corresponding range of unit model sections. The geological model is stored based on the unit model sections. The target geographic model is obtained by replacing the basic geographic model through the target unit model sections. The identification module, connected to the data acquisition and construction module, is used to label geological data based on the target geographic model and match mineral information of the target geographic model based on the mineral information matching rule base, thereby completing the identification of mineral resource information within the target geographic area.

[0028] In one embodiment, the setting module includes: The basic building blocks are used to acquire geological data within the target geographic area based on the identification platform, and to construct a basic geographic model within the target geographic area based on the geological data. The model building unit is used to configure the model network corresponding to the basic geographic model. Multiple unit model sections in the model network are connected to the basic geographic model. It should be noted that model networks can pre-build models within a local area of ​​the basic geographic model, and then replace local models within the basic geographic model, avoiding reconstruction of the entire basic geographic model, reducing server load. Secondly, multiple layout models can be rebuilt and replaced simultaneously, improving the efficiency of basic geographic model updates. It also exhibits good stability in model building and updating.

[0029] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for intelligent identification of mineral resources based on multi-source remote sensing data, characterized in that, Includes the following steps: In the identification platform, a basic geographic model is constructed based on big data within the target geographic area. A model network is set up corresponding to the basic geographic model. The model network includes multiple unit model sections, and the multiple unit model sections are interconnected. The basic geographic model is improved based on multi-source remote sensing data and stored in the corresponding range of unit model sections. The geological model is stored based on the unit model sections. The target geographic model is obtained by replacing the basic geographic model with the target unit model sections. Geological data is labeled based on the target geographic model, and mineral information of the target geographic model is matched based on the mineral information matching rule base to complete the identification of mineral resource information within the target geographic area.

2. The intelligent identification method for mineral resources based on multi-source remote sensing data according to claim 1, characterized in that, The steps of constructing a basic geographic model within the target geographic area based on big data in the identification platform, and setting up a model network corresponding to the basic geographic model, include: Geological data within the target geographic area is obtained from big data based on the identification platform, and a basic geographic model within the target geographic area is constructed based on the geological data. A model network is configured corresponding to the basic geographic model, and multiple unit model sections in the model network are connected to the basic geographic model.

3. The intelligent identification method for mineral resources based on multi-source remote sensing data according to claim 2, characterized in that, The step of configuring a model network corresponding to the basic geographic model, wherein multiple unit model sections in the model network are connected to the basic geographic model, includes: Multiple blank geographic model outlines are configured for the basic geographic model. These blank geographic model outlines are then spatially overlaid on the outer layer of the basic geographic model. The blank geographic model outline closest to the basic geographic model is used as the update model. The basic geographic model is then divided into multiple unit geographic models, where each unit geographic model includes the unit basic geographic model and its corresponding connection point. The outlines of multiple blank geographic models are divided according to multiple unit geographic models to obtain multiple unit model cross-sections, where each unit model cross-section includes the blank unit geographic model and the corresponding connection point. Connect the corresponding unit model sections in the unit geographic model and the updated model through connection points. Multiple blank geographic model outlines and the updated model are accessed in the order from the outside to the inside, and the unit model sections are connected sequentially through connection points. The corresponding update model sets multiple dynamic domains. Each dynamic domain is set one-to-one with the unit model section in the update model. Each dynamic domain includes multiple dynamic nodes. The connection points of the multiple dynamic nodes and the unit model section are connected. The multiple dynamic nodes are set at the edge of the unit model section to form an enclosure. Adjacent dynamic nodes in the dynamic domain are connected to each other. The dynamic domain is mapped to the corresponding unit geographic model.

4. The intelligent identification method for mineral resources based on multi-source remote sensing data according to claim 1, characterized in that, The steps of refining the basic geographic model based on multi-source remote sensing data and storing it in the corresponding range of unit model sections, storing the geological model based on the unit model sections, and replacing the basic geographic model with the target unit model sections to obtain the target geographic model include: Acquire multi-source remote sensing data, convert the multi-source remote sensing data into geological data, compare the geological data with the geological data in the basic geographic model, and obtain the difference data; Determine the cell model section where the differential data is located and sequentially infiltrate the data into the updated model according to the outlines of multiple blank geographic models; When the unit model sections containing the differential data are adjacent, if the model range corresponding to the differential data of one unit model section is larger than the model range corresponding to the differential data of another unit model section, exceeding the preset conditions, the unit model section with the larger proportion of the model range corresponding to the differential data is taken as the main section. The dynamic domain corresponding to the main section is used to squeeze the adjacent dynamic domains to obtain a dynamic domain including the differential data, which is then mapped to the unit geographic model corresponding to the basic geographic model. The corresponding range is divided by the data points corresponding to the unit geographic model to obtain the main unit geographic model. Based on the multi-source remote sensing data corresponding to the differential data, a model is constructed in the main cross-section. The constructed main cross-section is disconnected from the adjacent unit model cross-sections, the main unit geographic model is disconnected from the adjacent unit geographic models, the main cross-section and the main unit geographic model are interchanged, and the connection between the main cross-section and the adjacent unit geographic models is established to obtain a new basic geographic model as the target geographic model. The main unit geographic model is connected to the adjacent unit model cross-sections to obtain a new updated model. Set access permissions for the model network to protect the underlying geographic model when unauthorized networks access the model network.

5. The intelligent identification method for mineral resources based on multi-source remote sensing data according to claim 4, characterized in that, The step of setting access permissions for the model network to protect the basic geographic model when unauthorized networks access the model network includes: When authorized network access to the model network exists, access is sequentially transferred from any connection point on the model network to the connection point of the basic geographic model. When an unauthorized network accesses the model network, the connection between the updated model's connection point and the connection points in other blank geographic model outlines is disconnected. At the same time, the connection points in other blank geographic model outlines are cross-connected. The unauthorized network can cross-access the connection points in the other blank geographic model outlines, thus preventing it from accessing the basic geographic model and completing the protection of the basic geographic model.

6. The intelligent identification method for mineral resources based on multi-source remote sensing data according to claim 1, characterized in that, The steps of labeling geological data based on the target geographic model and matching mineral information of the target geographic model based on a mineral information matching rule base to complete the identification of mineral resource information within the target geographic area include: Establish association rules between geological data and geographic surface models based on the target geographic model, and mark the core feature information of geological data in the target geographic model; The geological data in the target geographic model is matched according to the mineral information matching rule base to obtain the matching level, corresponding mineral type, and mineralization potential characteristics, which are then marked in the target geographic model. Based on the target geographic model, the mineral resource information is integrated to form a mineral resource information identification result set, and the key mineralization areas are determined in the target geographic model.

7. A mineral resource intelligent identification system based on multi-source remote sensing data, used to implement the mineral resource intelligent identification method based on multi-source remote sensing data as described in any one of claims 1-6, characterized in that, include: The configuration module is used to construct a basic geographic model within the target geographic area based on big data in the identification platform. A model network is set up for the basic geographic model. The model network includes multiple unit model sections, which are interconnected. The data acquisition and construction module, connected to the settings module, is used to improve the basic geographic model based on multi-source remote sensing data and store it in the corresponding range of unit model sections. The geological model is stored based on the unit model sections. The target geographic model is obtained by replacing the basic geographic model through the target unit model sections. The identification module, connected to the data acquisition and construction module, is used to label geological data based on the target geographic model and match mineral information of the target geographic model based on the mineral information matching rule base, thereby completing the identification of mineral resource information within the target geographic area.

8. The intelligent identification system for mineral resources based on multi-source remote sensing data according to claim 7, characterized in that, The settings module includes: The basic building blocks are used to acquire geological data within the target geographic area based on the identification platform, and to construct a basic geographic model within the target geographic area based on the geological data. The model building unit is used to configure the model network corresponding to the basic geographic model. Multiple unit model sections in the model network are connected to the basic geographic model.

Images