A method and system for constructing a mineralization model of a wollastonite mine based on space-time constraints

Abstract

The present application relates to the technical field of ore-forming research of wollastonite deposit, and particularly relates to a method and system for constructing an ore-forming model of wollastonite deposit based on space-time constraints. The method comprises: taking the contact zone and peripheral buffer range of intrusive body and carbonate rock as a target area, and obtaining multi-source basic data in the target area; extracting an initial geographic feature data set from the multi-source basic data; performing time series stability analysis on the observed features of the mineral edge zone corresponding to the wollastonite contact metasomatic mineralization, and determining a target time scale and a spatial data block in combination with the spatial continuity constraint of the wollastonite contact metasomatic mineralization; screening a space-time coupled feature data set based on the target time scale and the spatial data block; performing multiple linear regression analysis on the space-time coupled feature data set, screening a target feature parameter related to the wollastonite mineralization process, and constructing a wollastonite ore-forming model based on the target feature parameter. The present application can improve the geological reliability of the ore-forming features.

Description

Technical Field

[0001] This invention relates to the technical field of wollastonite mineralization research, specifically to a method and system for constructing a wollastonite mineralization model based on spatiotemporal constraints. Background Technology

[0002] Wollastonite is an important non-metallic mineral resource, widely used in ceramics, metallurgy, building materials, and new materials. Wollastonite deposits are mostly formed by contact metasomatism, and their formation process is usually closely related to the contact between acidic intrusive bodies and carbonate rocks. The mineralization process is jointly controlled by various geological factors such as hydrothermal activity, lithological assemblage, and tectonic conditions, exhibiting obvious spatial continuity and temporal stability.

[0003] Current methods for studying and exploring the mineralization of wollastonite deposits mainly rely on field geological surveys, drilling and sampling, and analysis of geochemical and geophysical data. While these methods can reveal mineralization patterns within local areas, they suffer from problems such as long working cycles, high costs, and limited spatial coverage. With the development of remote sensing and geographic information technologies, multi-temporal and multi-scale remote sensing data are gradually being introduced into mineral resource surveys, providing new data sources for extracting regional-scale mineralization information.

[0004] Currently, when constructing metallogenic models for wollastonite deposits using remote sensing data, feature extraction of the acquired geographic data is required. A key issue is identifying which factors are relevant to mineralization during the metallogenic model construction. However, actual metallogenic model construction necessitates feature extraction from the acquired geographic data. Because geographic data contains a large amount of feature data, irrelevant features are introduced, reducing the geological reliability of the mineralization characteristics and consequently decreasing the accuracy of the wollastonite metallogenic model construction. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for constructing a metallogenic model of wollastonite based on spatiotemporal constraints, in order to solve the technical problem that the large amount of feature data in the geographical data in the prior art reduces the geological reliability of the metallogenic features, resulting in low accuracy in the construction of metallogenic models of wollastonite.

[0006] In a first aspect, one embodiment of the present invention provides a method for constructing a mineralization model of wollastonite based on spatiotemporal constraints, the method comprising: The target area is the contact zone between the intrusive body and the carbonate rock and the surrounding buffer zone. Multi-source basic data within the target area are acquired. The multi-source basic data includes multi-temporal optical remote sensing images, digital elevation models and derived topographic parameters, basic geological data and historical mineral data. An initial geographic feature dataset is extracted from multi-source basic data; the initial geographic feature dataset includes spectral statistical features, topographic and structural features, and spatial statistical and textural features. By conducting time series stability analysis on the observation characteristics of the mineralized edge zone corresponding to wollastonite contact metasomatic mineralization, and combining the spatial continuity constraints of wollastonite contact metasomatic mineralization, the target time scale and spatial data blocks matching the target time scale are determined. Based on the target time scale and spatial data blocks, a spatiotemporal coupled feature dataset is selected from the initial geographic feature dataset. The spatiotemporal coupled feature dataset is used to characterize the correspondence between the spatiotemporal coupled feature vector and the mineralization intensity level vector. The spatiotemporal coupled feature vector is used to characterize the geographic feature vectors related to wollastonite mineralization. Multiple linear regression analysis was performed on the spatiotemporal coupling feature dataset to screen target feature parameters related to the wollastonite mineralization process, and a wollastonite mineralization model was constructed based on the target feature parameters.

[0007] In one embodiment, determining the target time scale by performing time-series stability analysis on the observation characteristics of the mineralized edge zone corresponding to wollastonite contact metasomatic mineralization includes: Based on the mineralization impact zone, mineral edge zone information is extracted at each time point. The mineral edge zone information includes the spatial range of the mineral edge zone and the observation characteristics within the transition zone; the spatial range is used to characterize the transition zone between the mineralized area and the non-mineralized area. Based on mineral edge zone information, the number of overlapping graticles and the total number of covered graticles of adjacent time nodes within different time windows are counted using a grid indicator function; the grid indicator function is used to indicate whether each graticle in the target area belongs to the edge zone of wollastonite ore. The spatial overlap rate is calculated based on the number of overlapping rasters and the total number of covering rasters, and the target time scale is determined based on the spatial overlap rate; the spatial overlap rate is used to characterize the ratio of the number of overlapping rasters to the total number of covering rasters.

[0008] In one embodiment, determining spatial data blocks that match the target timescale, in conjunction with the spatial continuity constraints of wollastonite contact metasomatism mineralization, includes: Starting from the stable mineral margin zone within the target time scale, a profile is constructed along the contact metasomatic direction between the intrusive body and the surrounding rock; the contact metasomatic direction is the hydrothermal migration direction of the intrusive body towards the surrounding rock. Establish a local coordinate system with the contact zone direction as the x-axis and the contact zone normal as the y-axis, and divide the spatial units according to different spatial sampling intervals; Calculate the wollastonite zoning index of each spatial unit and the continuous transition index of the target area; Based on the wollastonite zoning index and continuous transition index, spatial data blocks matching the target time scale are identified.

[0009] In one embodiment, calculating the wollastonite zoning index of each spatial unit includes: For any spatial unit, the comprehensive characteristic values ​​of silicification and carbonate rocks are obtained. The comprehensive characteristic value of silicification is used to characterize the intensity and distribution characteristics of silicification within the ore-forming region; the comprehensive characteristic value of carbonate rocks is used to characterize the distribution and reaction degree of carbonate rocks within the ore-forming region. Based on the comprehensive characteristic values ​​of silicification and carbonate rocks, the wollastonite zoning index of the spatial unit is determined.

[0010] In one embodiment, calculating the continuous transition index of the target region includes: Calculate the difference in wollastonite zoning index between adjacent spatial units; Based on a preset threshold, the difference in the zoning index of wollastonite ore is binarized to obtain the labeling information; Based on the marking information, the total number of continuous spatial units in all profiles is determined, and the continuity transition index of the target region is determined based on the total number of continuous spatial units and the total number of profiles.

[0011] In one embodiment, the step of performing multiple linear regression analysis on the spatiotemporal coupling feature dataset to screen target feature parameters related to the wollastonite mineralization process includes: Calculate the variance inflation factor of the spatiotemporal coupling feature vector, and perform initial screening of the spatiotemporal coupling feature vector based on the preset variance inflation factor threshold to obtain the initial screening spatiotemporal coupling feature vector. Determine the regression coefficients corresponding to the spatiotemporal coupling feature vectors after initial screening, and construct a regression function based on the regression coefficients; Calculate the probability that the regression coefficient is 0; Based on preset judgment thresholds and probabilities, target feature parameters related to the wollastonite mineralization process are determined from the spatiotemporal coupling feature vectors after initial screening.

[0012] In one embodiment, the extraction of the initial geographic feature dataset from multi-source basic data includes: Based on multi-temporal optical remote sensing images, spectral statistical features and spatial statistical and textural features are extracted. The spectral statistical features include the mean, variance and band ratio features of each band, and the spatial statistical and textural features include gray-level co-occurrence matrix texture features and local change intensity index. Based on digital elevation models and derived topographic parameters and basic geological data, topographic and structural features are extracted; these features include slope, aspect, relief, and structural density. An initial geographic feature dataset was constructed based on spectral statistical features, topographic and structural features, spatial statistics and texture features.

[0013] In one embodiment, constructing the regression function based on the regression coefficients includes: Based on the mineralization intensity level, a regression function is constructed using each spatiotemporally coupled feature vector and its corresponding regression coefficient.

[0014] Secondly, another embodiment of the present invention provides a system for constructing a mineralization model of wollastonite based on spatiotemporal constraints, the system comprising: The data acquisition module is used to acquire multi-source basic data within the target area, which is the contact zone between the intrusive body and the carbonate rock and the surrounding buffer zone. The multi-source basic data includes multi-temporal optical remote sensing images, digital elevation models and derived topographic parameters, basic geological data, and historical mineral data. An initial geographic feature dataset is extracted from the multi-source basic data. The initial geographic feature dataset includes spectral statistical features, topographic and structural related features, and spatial statistical and textural features. The condition constraint module is used to perform time series stability analysis on the observation characteristics of the mineralized edge zone corresponding to wollastonite contact metasomatic mineralization, and to determine the target time scale and the spatial data block that matches the target time scale by combining the spatial continuity constraint of wollastonite contact metasomatic mineralization. The target feature parameter determination module is used to select spatiotemporally coupled feature datasets from the initial geographic feature dataset based on the target time scale and spatial data blocks. The spatiotemporally coupled feature datasets are used to characterize the correspondence between spatiotemporally coupled feature vectors and mineralization intensity level vectors. The spatiotemporally coupled feature vectors are used to characterize geographic feature vectors related to wollastonite mineralization. Multiple linear regression analysis is performed on the spatiotemporally coupled feature datasets to select target feature parameters related to the wollastonite mineralization process, and a wollastonite mineralization model is constructed based on the target feature parameters.

[0015] In one embodiment, the condition constraint module includes: The calculation module is used to extract mineral edge zone information at various time points based on the mineralization influence zone. The mineral edge zone information includes the spatial range of the mineral edge zone and the observation characteristics within the transition zone. The spatial range is used to characterize the transition zone between the mineralized and unmineralized areas. Based on the mineral edge zone information, the module uses a raster indicator function to count the number of overlapping rasters and the total number of covered rasters at adjacent time points within different time windows. The raster indicator function is used to indicate whether each raster in the target area belongs to the edge zone of wollastonite. Based on the number of overlapping rasters and the total number of covered rasters, the module calculates the spatial overlap rate and determines the target time scale based on the spatial overlap rate. The spatial overlap rate is used to characterize the ratio of the number of overlapping rasters to the total number of covered rasters.

[0016] Thirdly, in another embodiment of the present invention, an electronic device is provided, including a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method described in the first aspect.

[0017] Fourthly, in another embodiment of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0018] The present invention has the following beneficial effects: In constructing the mineralization model of wollastonite deposits, this invention targets the contact zone between the intrusive body and carbonate rocks, as well as the surrounding buffer zone, and acquires multi-source basic data within this target area. An initial geographic feature dataset is extracted from the multi-source basic data. Time-series stability analysis is performed on the observed characteristics of the mineralized edge zone corresponding to wollastonite contact metasomatic mineralization. Combined with the spatial continuity constraint of wollastonite contact metasomatic mineralization, a target time scale and spatial data blocks matching the target time scale are determined. Based on the target time scale and spatial data blocks, a spatiotemporal coupled feature dataset is selected from the initial geographic feature dataset. Multiple linear regression analysis is performed on the spatiotemporal coupled feature dataset to select target feature parameters related to the wollastonite mineralization process. Based on these target feature parameters, a wollastonite mineralization model is constructed. This invention introduces the spatial continuity constraint of the wollastonite contact metasomatic mineralization process and time-series stability analysis of observed features in the construction stage of the mineralization target feature parameters, avoiding the feature selection distortion problem caused by purely data-driven methods in mineralization analysis. Under the constraint of time scale, the size of spatial data blocks is determined so that the time scale and spatial scale are mutually constrained, thereby ensuring that the constructed feature dataset can truly reflect the zonal structure characteristics of wollastonite. Through multiple linear regression, the correlation between spatiotemporal coupling characteristics and mineral changes is statistically analyzed, so as to effectively screen the target feature parameters that are significantly related to the wollastonite mineralization process and improve the accuracy of the wollastonite mineralization model construction. Attached Figure Description

[0019] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating a method for constructing a mineralization model of wollastonite based on spatiotemporal constraints, as provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of a system for constructing a mineralization model of wollastonite based on spatiotemporal constraints, provided in an embodiment of the present invention. Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0021] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method and system for constructing a spatiotemporally constrained wollastonite mineralization model according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0023] The following description, in conjunction with the accompanying drawings, details the specific scheme of the method for constructing a spatiotemporally constrained mineralization model of wollastonite ore provided by the present invention.

[0024] This invention proposes a method for constructing a mineralization model of wollastonite based on spatiotemporal constraints. Please refer to [link to relevant documentation]. Figure 1 The diagram illustrates a schematic flowchart of a method for constructing a spatiotemporally constrained mineralization model of wollastonite, according to an embodiment of the present invention. The method includes the following steps: Step S1: Taking the contact zone between the intrusive body and the carbonate rock and the surrounding buffer zone as the target area, acquire multi-source basic data within the target area; the multi-source basic data includes multi-temporal optical remote sensing images, digital elevation models and derived topographic parameters, basic geological data and historical mineral data.

[0025] The contact zone between the intrusive body and the carbonate rock refers to the core area of ​​wollastonite mineralization. It is the interface between acidic intrusive bodies (such as granite) and carbonate rocks (such as limestone). Hydrothermal metasomatism occurs intensely here, making it a key location for the formation of wollastonite ore bodies.

[0026] The outer buffer zone refers to the extended area at a certain distance around the contact zone, which serves as the radiation range of the mineralization influence. It is used to avoid omitting potential mineralization-related areas around the contact zone and to ensure the completeness of the analysis scope.

[0027] The target area refers to the contact zone between the intrusive body and the carbonate rock and the surrounding buffer zone. It is the spatial boundary for data acquisition, feature extraction and analysis in the entire process of constructing the mineralization model.

[0028] Multi-temporal optical remote sensing images refer to remote sensing data covering different seasons and years in optical bands (visible light, near infrared, etc.). They are digital images that record surface reflection or radiation information, possessing temporal continuity and spatial consistency, and can avoid short-term surface interference.

[0029] A digital elevation model (DEM) is a three-dimensional digital topographic model constructed using a limited number of surface elevation sampling points, storing the elevation of various locations on the surface. Derived topographic parameters are indicators derived from the DEM through spatial analysis algorithms, including slope, aspect, and undulation, which are used to reflect the geomorphic background of the mineralized area.

[0030] Basic geological data refers to fundamental and objective data that characterize the geological background of the study area. It may include stratigraphic and lithological distribution maps, intrusive body boundary information, structural line information, etc., and mostly comes from long-term geological surveys and exploration results. It is used to clarify the core space of mineralization and the direction of analysis.

[0031] Historical mineral data refers to targeted data that records the mineralization, distribution, and resource changes of wollastonite at different times. It includes records of ore body distribution, resource quantity changes, and mineralization intensity level data (0-no mineralization, 1-weak mineralization, 2-industrial ore body) at different times, and serves as a reference reflecting the actual mineralization situation.

[0032] It should be noted that, in the embodiments of the present invention, by defining the target area as the contact zone between the intrusive body and the carbonate rock and the surrounding buffer zone, it can be ensured that the analysis focuses on the mineralization core and the potential impact area; furthermore, by clearly acquiring four types of core multi-source basic data, including multi-temporal optical remote sensing images, DEM and derived topographic parameters, basic geological data and historical mineral data, a data system of remote sensing-topography-geology-mineral change can be formed, providing data for subsequent feature extraction, spatiotemporal constraints and mineralization model construction.

[0033] Step S2: Extract the initial geographic feature dataset from the multi-source basic data; the initial geographic feature dataset includes spectral statistical features, topographic and structural features, and spatial statistical and textural features.

[0034] The initial geographic feature dataset refers to a set of features extracted from multi-source basic data that covers multi-dimensional mineralization-related information.

[0035] Spectral statistical features refer to features extracted from multi-temporal optical remote sensing images, including the mean, variance, and band ratio of each band, as well as mineral sensitivity indices such as silicification correlation index and carbonate rock correlation index, which can reflect the mineral composition and lithological characteristics of the Earth's surface.

[0036] Topographic and structural features refer to features extracted from DEM and basic geological data, including topographic parameters such as slope, aspect, and undulation derived from DEM, as well as structural density and distance from major structural lines calculated based on structural line data, which are used to characterize the geomorphic background and structural control conditions of the mineralization area.

[0037] Spatial statistics and texture features refer to features extracted from multi-temporal optical remote sensing images. They mainly include gray level co-occurrence matrix (GLCM) texture features and local variation intensity indicators, which can reflect the spatial heterogeneity of lithological contact zones and alteration zones and can help identify the spatial distribution patterns related to mineralization.

[0038] Furthermore, the extraction of the initial geographic feature dataset from multi-source basic data includes: Based on multi-temporal optical remote sensing images, spectral statistical features and spatial statistical and textural features are extracted. The spectral statistical features include the mean, variance and band ratio features of each band, while the spatial statistical and textural features include gray-level co-occurrence matrix texture features and local change intensity indicators.

[0039] Among them, the band mean refers to the average value of the gray values ​​of all pixels in a certain band of a remote sensing image, which is used to reflect the overall intensity of surface reflection or radiation corresponding to that band.

[0040] Variance is used to describe the dispersion of pixel gray values ​​within a certain band of a remote sensing image, and can reflect the uniformity or difference of land cover.

[0041] Band ratio characteristics can be obtained by calculating the ratio of gray values ​​of different bands in remote sensing images. This can enhance the spectral differences of minerals and lithology, and facilitate the identification of mineral assemblages related to mineralization.

[0042] GLCM texture features can be obtained by statistically analyzing the co-occurrence relationship of gray values ​​of adjacent pixels in an image. They can characterize the texture of an image, such as its coarseness, density, and uniformity, and can reflect the spatial distribution structure of lithology.

[0043] The local variation intensity index is used to quantify the variation range of pixel gray values ​​in a local area of ​​remote sensing image. It can highlight spatially abrupt changes in areas such as lithological contact zones and alteration zones, and help locate key areas related to mineralization.

[0044] Based on digital elevation models and derived topographic parameters and basic geological data, topographic and structural features are extracted; these features include slope, aspect, undulation, and structural density.

[0045] Among them, derived topographic parameters refer to topographic feature indicators derived from DEM data through spatial analysis algorithms. They are an extension of the application of DEM data and can accurately reflect the geomorphic details of the mineralized area.

[0046] Slope refers to the degree of inclination of the earth's surface, which can reflect the steepness of the terrain and affect the hydrothermal migration path and the occurrence conditions of ore bodies.

[0047] Slope aspect refers to the direction in which the earth's surface tilts. It is related to environmental factors such as sunlight and precipitation, and indirectly affects hydrothermal activity during the mineralization process.

[0048] Undulation refers to the degree of topographic relief within a local area, which can reflect the degree of geomorphic fragmentation in a mineralized area and provide a basis for structural analysis.

[0049] Tectonic density refers to a characteristic index calculated based on tectonic line information in basic geological data. It refers to the number or length of geological structures (such as faults and folds) per unit area, which can reflect the intensity of regional tectonic activity and is an important controlling factor for mineralization.

[0050] It should be noted that, in the embodiments of the present invention, DEM and derived topographic parameters and basic geological data are used as joint data sources to extract four types of key features (slope, aspect, undulation, and structural density). Among them, slope, aspect, and undulation reflect the geomorphic background of the mineralized area, and structural density reflects the intensity of regional tectonic activity. The two types of information together constitute topographic and structural related features.

[0051] An initial geographic feature dataset was constructed based on spectral statistical features, topographic and structural features, spatial statistics and texture features.

[0052] The initial geographic feature dataset refers to a collection that integrates spectral statistical features, topographic and structural features, spatial statistics and texture features. It is a collection of multi-dimensional information on minerals, landforms, structures and spatial distribution involved in the mineralization process, extracted from multi-source basic data using existing methods. This lays the foundation for subsequent spatiotemporal constraint screening and target feature extraction.

[0053] It should be noted that, in the embodiments of the present invention, by integrating three types of core features, an initial geographic feature dataset covering multi-dimensional information on mineralization is formed, thereby realizing the transformation from multi-source basic data to mineralization-related features.

[0054] Step S3: By conducting time series stability analysis on the observation characteristics of the mineralized edge zone corresponding to wollastonite contact metasomatic mineralization, and combining the spatial continuity constraints of wollastonite contact metasomatic mineralization, the target time scale and the spatial data block matching the target time scale are determined.

[0055] Among them, wollastonite contact metasomatic mineralization refers to the main mineralization type of wollastonite deposits, which are formed at the contact interface between acidic intrusive bodies and carbonate rocks. It is jointly controlled by multiple factors such as hydrothermal activity, lithological combination and tectonic conditions, and is the basis for mineralization constraints.

[0056] The observational characteristics of the mineral edge zone refer to a set of multi-dimensional features extracted from the transition zone between the wollastonite mineralized and unmineralized areas (i.e., the mineral edge zone) based on multi-temporal optical remote sensing images, digital elevation models and derived topographic parameters, basic geological data and historical mineral data, which can reflect the geological and mineral-related attributes of the area.

[0057] Time series stability analysis refers to the process of analyzing the observation characteristics of the marginal zone of wollastonite deposits, based on multi-temporal optical remote sensing images and historical mineral data. It involves quantifying the spatial consistency of observation characteristics within different time windows, eliminating short-term interferences caused by surface cover (vegetation, snow cover, soil moisture) and imaging conditions, and selecting the best observation time window that can truly reflect the static geological structure of the ore body.

[0058] For example, time series stability analysis can be based on the geological nature of wollastonite deposits, which are formed during geological history and whose physical location and morphology remain static within the remote sensing observation window. Based on multi-temporal optical remote sensing images and combined with the known distribution range of ore bodies in historical mineral data, the observation characteristics (such as spectral statistical characteristics and spatial texture characteristics) of the mineral edge zone at each time node are extracted. By converting the mineral edge zone at different time nodes into a binary mask raster, the Cumulative Spatial Overlap Rate (CT) of the edge zones of adjacent nodes within different time windows is calculated. The consistency and anti-interference ability of the observation characteristics of the mineral edge zone in the time series are analyzed, thereby eliminating short-term disturbances caused by surface cover and imaging conditions such as vegetation, snow cover, and soil moisture, and selecting the best observation time window that can truly reflect the static geological structure of the ore body.

[0059] Spatial continuity constraint refers to the data screening criteria based on the geological pattern of wollastonite mineralization distributed along the intrusive-wall rock contact zone. It is necessary to ensure that the feature extraction conforms to the hydrothermal migration and mineralization zoning pattern of the intrusive-wall rock, and to avoid spatial analysis deviating from the mineralization mechanism.

[0060] For example, the target time scale refers to the effective time window selected through time series stability analysis, that is, the time range with the highest spatial continuity (overlap rate) of the mineral edge zone, which can avoid short-term interference and provide stable time dimension support for subsequent feature extraction.

[0061] Spatial data blocks refer to the spatial analysis range that matches the target time scale. They are determined by spatial continuity constraints, and their size conforms to the mineralization zoning pattern of wollastonite, ensuring that the spatial dimension features are extracted corresponding to the mineralization area.

[0062] It should be noted that, in the embodiments of the present invention, the optimal spatiotemporal range for mineralization analysis is determined through the dual constraints of mineralization mechanism transformation. That is, the geological characteristics of temporal stability and spatial continuity of wollastonite contact metasomatic mineralization are transformed into constraints for data screening. The time series stability constraint locks in the target time scale that can reflect the real mineralization structure, and the spatial continuity constraint determines the spatial data blocks that match the time scale and fit the mineralization zoning pattern. Finally, the mutual constraints of time and spatial scales are realized, laying the foundation for subsequent screening of anti-interference features and construction of spatiotemporally coupled datasets, and ensuring the geological credibility of mineralization features from the source.

[0063] Furthermore, the determination of the target time scale by performing time series stability analysis on the observation characteristics of the mineralized edge zone corresponding to wollastonite contact metasomatic mineralization includes: Based on the mineralization impact zone, mineral edge zone information is extracted at each time point. The mineral edge zone information includes the spatial range of the mineral edge zone and the observation characteristics within the transition zone; the spatial range is used to characterize the transition zone between the mineralized area and the non-mineralized area.

[0064] Among them, the mineralization influence zone refers to the area around the known ore body that may have weak mineralization or mineralization-related geological conditions due to mineralization processes (such as hydrothermal activity, element migration, etc.). It is the reference range for extracting the mineral edge zone, and its boundary is determined based on the results of historical mineral exploration and geological surveys.

[0065] Time nodes refer to the specific time points at which multi-temporal optical remote sensing images and geographic data are acquired (covering different seasons and years). These time nodes need to be organized according to a unified time axis to form continuous time windows or fixed time period aggregation windows, providing a basis for analyzing the temporal stability of mineral edge zones.

[0066] Mineral marginal zone information refers to a comprehensive set of information extracted from multiple sources of data, including multi-temporal optical remote sensing images, DEM and derived topographic parameters, basic geological data, and historical mineral data, based on the contact metasomatic mineralization mechanism of wollastonite. It can include the spatial range determined by the distribution of known ore bodies in historical mineral data and combined with basic geological data such as intrusive body boundaries, as well as the multi-dimensional observation features extracted within this spatial range, including spectral statistical features, spatial texture features, and topographic and structural features.

[0067] It should be noted that the observed features of the mineral edge zone are a subset of the initial geographic feature dataset. They refer to the geographic features within the spatial range of the mineral edge zone (the transition zone between mineralized and unmineralized areas). The feature categories are consistent with the initial geographic feature dataset, but the spatial range is constrained to the mineral edge zone.

[0068] For example, the process of acquiring mineral boundary zone information at each time point includes: first, based on the known ore body distribution in historical mineral data, and combined with the contact zone between intrusive bodies and carbonate rocks and the surrounding buffer zone clearly defined in basic geological data, a fixed spatial range of the transition zone between mineralized and unmineralized areas is delineated; then, for each time point, observation features such as spectral statistical features, topographic and structural features, and spatial texture features are extracted within this spatial range using existing methods, based on multi-temporal optical remote sensing images, DEMs, and derived topographic parameters; finally, isolated noise points are removed through 3×3 raster neighborhood analysis, and historical mineral data is compared and corrected with basic geological data to form complete mineral boundary zone information with fixed spatial range and dimensional observation features for each time point. Furthermore, during the extraction process, data consistency within the same spatial coordinate system must be ensured to avoid spatial deviations between time points.

[0069] Mineralized areas refer to regions where wollastonite mineralization (including weak mineralization and industrial ore bodies) has been confirmed. Their boundaries are defined based on historical mineral data (such as ore body distribution maps and resource records) and geological exploration results, and they are the core objects of metallogenic model analysis.

[0070] The non-mineralized zone refers to the area within the study area that is not affected by wollastonite mineralization or does not have the geological conditions for mineralization (such as being far from the intrusive body-carbonate rock contact zone, or lacking tectonic fault control, etc.), forming a clear spatial boundary with the mineralized zone.

[0071] The transition zone refers to the essential attribute of the mineral edge zone. It is the spatial transition area between the mineralized area and the non-mineralized area. Its spatial location and morphology can reflect the extensional boundary of the wollastonite ore body and have stability within the effective time scale. It is controlled by the mineralization mechanism and is not affected by short-term surface disturbances such as vegetation and snow cover.

[0072] It should be noted that, in the embodiments of the present invention, the mineralization-affected area is used as a reference range. For each data acquisition time point, information on the mineral edge zone that can characterize the spatial transition relationship between the mineralized area and the non-mineralized area is extracted. This provides data for subsequent analysis of the temporal continuity of the mineral edge zone and screening of effective time scales, thereby avoiding the impact of short-term surface disturbances on the extraction of mineralization features.

[0073] Based on mineral edge zone information, the number of overlapping graticules and the total number of covered graticules of adjacent time nodes within different time windows are counted using a raster indicator function. The raster indicator function is used to indicate whether each graticule in the target area belongs to the edge zone of wollastonite.

[0074] Among them, the grid indicator function refers to a binary function used to identify the attributes of spatial grids. It can determine whether each grid in the target area belongs to the edge zone of wollastonite deposits. The output result is only 1 or 0 (1 represents that the grid belongs to the edge zone, and 0 represents that it does not). It is stored in the form of a binary mask and is a tool for quantifying the spatial positional relationship of mineral edge zones.

[0075] A time window refers to a multi-temporal data set organized along a unified time axis. It is divided into continuous time windows (such as N-year sliding windows) and fixed time period aggregation windows (such as dry season and low vegetation period windows). It is used to centrally analyze the spatial continuity of mineral edge zones within a specific time range and is the core unit for screening effective time scales.

[0076] Adjacent time nodes refer to two consecutive data acquisition time points within a time window (such as the acquisition time of remote sensing images in different seasons or years). It is necessary to ensure that the data at the same spatial location are consistent, and they are the comparison objects for calculating the spatial overlap rate of the edge zone.

[0077] The mineral edge zone refers to the spatial transition zone between mineralized and unmineralized areas, reflecting the extensional boundary of wollastonite ore bodies. Its spatial location and morphology are stable within an effective time scale and can be extracted from remote sensing data using artificial or neural network methods.

[0078] The number of overlapping graticles refers to the total number of graticles that are simultaneously identified as belonging to the edge zone (with graticle indicator function values ​​of 1) in adjacent time nodes of the mineral edge zone. It is a quantitative indicator for measuring the spatial continuity of the edge zone. The more overlapping graticles there are, the better the continuity of the edge zone.

[0079] The total number of covered rasters refers to the total number of rasters that are determined to belong to the edge zone at least at one time point in the mineral edge zone of adjacent time points (at least one of the raster indicator function values ​​is 1). It is used to normalize the number of overlapping rasters and calculate the CT value.

[0080] It should be noted that, in the embodiments of the present invention, each grid is identified as belonging to the edge zone of wollastonite deposits by a grid indicator function. Then, the number of overlapping grids and the total number of covered grids in the edge zone of adjacent time nodes within different time windows are counted. Finally, the effective time scale with high continuity can be screened out by the ratio of the two (spatial overlap rate CT value), and non-mineralization changes caused by short-term disturbances such as vegetation and snow cover are eliminated, providing a time constraint basis for subsequent mineralization feature extraction.

[0081] The spatial overlap rate is calculated based on the number of overlapping rasters and the total number of covering rasters, and the target time scale is determined based on the spatial overlap rate; the spatial overlap rate is used to characterize the ratio of the number of overlapping rasters to the total number of covering rasters.

[0082] Among them, the spatial overlap rate (CT value) is calculated by the ratio of the number of overlapping grids to the total number of covered grids (range 0~1). The closer the ratio is to 1, the more stable the spatial location of the mineral edge zone at adjacent time points and the better the continuity. It is a quantitative indicator for screening effective time scales.

[0083] The target time scale refers to the optimal time window selected based on the spatial overlap rate, that is, the time range with the highest spatial overlap rate. This time scale can avoid short-term surface disturbances such as vegetation, snow cover, and soil moisture, and truly reflect the long-term stability of the wollastonite mineralization structure.

[0084] It should be noted that, in the embodiments of the present invention, based on the grid statistics of the mineral edge zone, the spatial overlap rate is obtained by calculating the ratio of the number of overlapping grids to the total number of covered grids, and then the time window with the highest spatial overlap rate is selected as the target time scale. The spatial overlap rate directly quantifies the temporal continuity of the mineral edge zone, ensuring that the target time scale can eliminate short-term interference and conform to the temporal stability characteristics of wollastonite mineralization, providing reliable time dimension support for subsequent spatial data block matching and feature screening.

[0085] Furthermore, the determination of spatial data blocks matching the target timescale, based on the spatial continuity constraints of wollastonite contact metasomatism mineralization, includes: Starting from the stable mineral margin zone within the target time scale, a profile is constructed along the contact metasomatic direction between the intrusive body and the surrounding rock; the contact metasomatic direction is the hydrothermal migration direction from the intrusive body to the surrounding rock.

[0086] Among them, the stable mineral edge zone refers to the transition zone between mineralized and non-mineralized areas that maintains consistent spatial location and morphology within the target time scale. It is a spatial marker that reflects the true mineralization structure and provides a reliable starting point for subsequent spatial analysis.

[0087] The contact metasomatic direction between the intrusive body and the surrounding rock refers to the key spatial direction of wollastonite contact metasomatic mineralization. Specifically, it refers to the extension direction from the acidic intrusive body to the carbonate surrounding rock, which is consistent with the hydrothermal migration path and is the main extension direction of mineralization zoning.

[0088] A profile refers to a longitudinal or transverse analytical section constructed along the contact metasomatic direction between the intrusive body and the surrounding rock. It is used to capture element migration, mineral phase transformation, and mineralization zoning characteristics near the contact zone, replacing traditional arbitrary two-dimensional grid analysis.

[0089] The contact metasomatic direction refers to the direction of hydrothermal migration from the intrusive body to the surrounding rock. It is the path of hydrothermal fluid carrying minerals from the intrusive body to the surrounding rock and undergoing metasomatic reactions during the mineralization process, which determines the spatial distribution pattern of mineralization zoning.

[0090] Intrusive bodies refer to acidic rock masses (such as granite) that intrude into surrounding rocks such as carbonate rocks. They are the hydrothermal source and material source for wollastonite mineralization, and their contact interface with the surrounding rocks is the core mineralization area.

[0091] The surrounding rock refers to the rock (e.g., carbonate rock) around the intrusive body. It is the carrier for the formation of wollastonite deposits and forms wollastonite ore bodies after undergoing a metasomatic reaction with the hydrothermal fluids released by the intrusive body.

[0092] The direction of hydrothermal migration refers to the direction in which hydrothermal fluids move and diffuse from the intrusive body to the surrounding rock. During this process, the hydrothermal fluids carry minerals such as silica, which react with the carbonate rocks in the surrounding rock to eventually form wollastonite. This direction is the focus of spatial analysis.

[0093] It should be noted that, in the embodiments of the present invention, a stable mineral edge zone within the target time scale is used as a reliable spatial starting point, and a targeted profile is constructed along the hydrothermal migration direction (contact metasomatism direction) of the intrusive body towards the surrounding rock. This profile can conform to the spatial law of wollastonite contact metasomatism mineralization, providing a spatial analysis framework for subsequent division of spatial units, calculation of zoning index (B value), and determination of spatial data block size, ensuring that spatial analysis is highly compatible with the mineralization mechanism.

[0094] Establish a local coordinate system with the contact zone direction as the x-axis and the contact zone normal as the y-axis, and divide the space into units according to different spatial sampling intervals.

[0095] The local coordinate system refers to a dedicated coordinate system established to adapt to the spatial distribution pattern of wollastonite contact metasomatic mineralization, rather than a general geographic coordinate system. It is used to focus on the contact zone between the intrusive body and the surrounding rock and the mineralization-related areas, simplifying the spatial analysis dimensions and ensuring that subsequent operations such as spatial unit division and zoning index calculation conform to the geological mineralization mechanism.

[0096] The contact zone refers to the boundary area between acidic intrusive bodies and carbonate rocks. It is the core site for wollastonite contact metasomatism mineralization. Hydrothermal activity, element migration, and mineral phase transformation all occur in this area, and its spatial morphology and orientation directly determine the distribution characteristics of mineralization zones.

[0097] The orientation of the contact zone refers to the direction in which the contact zone extends on the horizontal plane (such as east-west or north-south). It is the basis for constructing the x-axis of the local coordinate system. Choosing this direction as the x-axis allows it to be parallel to the extension trajectory of the ore-forming zone, making it easier to obtain the geological feature changes along the zoning direction.

[0098] The contact zone normal refers to the direction perpendicular to the contact zone and pointing towards the dominant direction of hydrothermal migration, that is, the direction from the intrusive body to the surrounding rock. As the y-axis of the local coordinate system, it can reflect the spatial progression of element migration and metasomatism during the mineralization process.

[0099] Spatial sampling interval refers to the distance (e.g., Δd) between two adjacent spatial sampling points (or adjacent spatial units) when dividing spatial units along the hydrothermal migration direction (perpendicular to the normal direction of the intrusive-wall rock contact zone) during the construction of a wollastonite contact metasomatic profile.

[0100] For example, multiple different Δd values ​​can be set first (such as 10 meters, 50 meters, 100 meters, etc., with specific candidate values ​​calibrated according to the metallogenic zoning width of the target area and the spatial resolution of the remote sensing image), and then the continuous transition index of the zoning index (B value) can be used. (Value) to filter for the optimal interval.

[0101] By comparing the continuity index corresponding to different candidate Δd ( (value), filter out The sampling interval with the highest value (i.e. the best continuous transition between bands) is taken as the optimal spatial sampling interval that matches the target time scale, and spatial data blocks are finally divided based on this interval.

[0102] A spatial unit refers to the smallest analytical unit (e.g., ...) that is divided within a profile constructed based on a local coordinate system according to a preset spatial sampling interval. , representing the i-th unit on the k-th metasomatic profile, is the basic spatial carrier for calculating the wollastonite zoning index (B value) and summarizing geographical features.

[0103] It should be noted that, in the embodiments of the present invention, in order to conform to the spatial laws of contact metasomatic mineralization, a local coordinate system is established with the contact zone orientation as the x-axis and the contact zone normal (hydrothermal migration direction) as the y-axis. Then, the mineralization-related profile is divided into several spatial units according to different spatial sampling intervals. Finally, by selecting the optimal sampling interval (matching the continuous transition of mineralization zoning), a data foundation is provided for subsequent zoning index calculation and spatiotemporal coupling feature extraction, ensuring that spatial analysis is highly compatible with the mineralization mechanism.

[0104] Calculate the wollastonite zoning index for each spatial unit and the continuous transition index for the target region.

[0105] Here, a spatial unit refers to the smallest analytical unit (denoted as Δd) divided according to different spatial sampling intervals (Δd) in a mineralization-related profile constructed based on a local coordinate system. , representing the i-th unit on the k-th metasomatic profile, is the basic spatial carrier for calculating the wollastonite zoning index and summarizing geographical features. Its division conforms to the mineralization spatial law of intrusive body-wall rock contact metasomatism.

[0106] The wollastonite zoning index (B value) is an indicator used to quantify the relative equilibrium between silicification and carbonate rock characteristics within a spatial unit. Its value is fixed in the range of [-1, +1]. A value closer to 0 indicates that the ratio of silicon (Si) to calcium (Ca) is more consistent with the chemical reaction conditions for wollastonite formation. It can reflect the core characteristics of mineralization zones.

[0107] For example, the zoning index of wollastonite can be expressed as: ; In the formula, In the section element The comprehensive eigenvalues ​​related to silicification at the location, In the section element At the same location, the composite characteristic value is related to the characteristics of carbonate rocks (Ca). To prevent the use of small positive numbers with a denominator of zero (such as 0.001).

[0108] The target area refers to the mineralization analysis area constructed along the contact metasomatic direction of the intrusive body and the surrounding rock (the hydrothermal migration direction), including the stable mineralized edge zone and the gradually expanding spatial range. It is the area for assessing the continuous transition of mineralization zones.

[0109] The continuous transition index is a quantitative indicator used to determine whether the wollastonite mineralization zones within a target area are continuously distributed. It is obtained by calculating whether the difference between the zoning indices (B values) of adjacent spatial units is less than a preset threshold (0.5 times the standard deviation of the zoning index). The value is 0 or 1. The higher the cumulative mean, the better the continuity of the mineralization zones. It can be used to determine the optimal spatial data block size.

[0110] For example, the continuous transition index can be expressed as: ; ; In the formula, M represents the number of cross-sections. Indicates the first In the cross section, the set of indices determined to be intermediate zones is based on... Compare with a preset threshold to obtain , The value can be 0 or 1. This preset threshold is determined based on the statistical distribution characteristics of the zoning index within the region. The preset threshold can be 0.5 times the standard deviation of the zoning index. When the difference is greater than the threshold, When it is 0, less than the threshold The value is 1.

[0111] It should be noted that, in order to avoid If there is an order-of-magnitude issue, a maximum-minimum normalization function can be used. After normalization, its mapping interval is [0,1].

[0112] When expanding spatial units along the cross-sectional direction, the maximum expansion distance Max(D) (5km) is set, or expansion is stopped when the rate of change of the zoning index is less than a preset threshold of 0.3.

[0113] Based on the wollastonite zoning index and continuous transition index, spatial data blocks matching the target time scale are identified.

[0114] For example, the target time scale refers to the effective time window selected through time series stability analysis, that is, the time range with the highest spatial overlap rate (CT value) of the mineral edge zone, which can avoid short-term surface disturbances and ensure the long-term stability of the mineralization structure.

[0115] Spatial data blocks refer to the spatial analysis range that matches the target time scale. Their size is determined by spatial continuity constraints and conforms to the zonal distribution pattern of wollastonite along the intrusive-surrounding rock contact zone. They are the basic spatial units for subsequent feature statistical summarization.

[0116] It should be noted that, in the embodiments of the present invention, the mineralization potential of spatial units is characterized by calculating the wollastonite zoning index (B value) based on the target time scale, and by using the continuous transition index (…). The spatial continuity of mineralization zones was verified by the value; finally, the spatial range with the best zonation index balance and continuous transition was selected as the spatial data block that matches the target time scale, so as to realize the mutual constraint between the time scale and the spatial scale, and ensure that the subsequent feature extraction fits the mineralization mechanism, laying the spatial foundation for the construction of a spatiotemporally coupled dataset.

[0117] Furthermore, the calculation of the wollastonite zoning index for each spatial unit includes: For any spatial unit, the comprehensive characteristic values ​​of silicification and carbonate rocks are obtained. The comprehensive characteristic value of silicification is used to characterize the intensity and distribution characteristics of silicification in the mineralization area. The comprehensive characteristic value of carbonate rocks is used to characterize the distribution and reaction degree of carbonate rocks in the mineralization area.

[0118] Among them, the comprehensive characteristic value of silicification ( This refers to an index that quantifies the intensity and spatial distribution of silicification within a mineralized area, calculated based on the thermal infrared bands of remote sensing images (such as ASTER data). The band ratio can effectively enhance the spectral response of siliceous rocks and silicified zones, reflecting the strength and coverage of silicification.

[0119] Comprehensive characteristic value of carbonate rocks ( This refers to an index characterizing the distribution range and reaction degree of carbonate rocks within a mineralization area, calculated based on shortwave infrared bands of remote sensing images (such as ASTER data). (Band combination) can highlight the spectral differences between carbonate rocks and other lithologies, reflecting the degree to which they participate in hydrothermal metasomatic reactions.

[0120] The mineralization area refers to the core area within the target area (the contact zone between the intrusive body and the carbonate rock and the surrounding buffer zone) that has mineralization potential. It is the main area where silicification and carbonate rock metasomatism occur.

[0121] Silicification refers to the geological process during mineralization in which hydrothermal fluids carry siliceous minerals and react with the surrounding rocks, enriching the surrounding rocks with siliceous materials. It is a preliminary process for the formation of wollastonite deposits, and its intensity and distribution directly affect the formation and scale of the ore body.

[0122] Carbonate rocks refer to the core host rock type of mineralization (such as limestone), which are the material carriers for the formation of wollastonite deposits and undergo metasomatic reactions with hydrothermal fluids released by intrusive bodies. This eventually forms a wollastonite ore body.

[0123] The degree of reaction refers to the thoroughness of the metasomatic reaction between carbonate rocks and minerals such as silica in hydrothermal fluids. The higher the degree of reaction, the easier it is to form industrial-grade wollastonite ore bodies, which is an important basis for judging mineralization potential.

[0124] It should be noted that, in the embodiments of the present invention, for each spatial unit divided by the cross-section, two types of characteristic values ​​are obtained: a comprehensive characteristic value of silicification that characterizes the intensity and distribution of silicification. ), and the comprehensive characteristic values ​​of carbonate rocks that characterize the distribution and degree of metasomatic reaction of carbonate rocks ( The two types of characteristic values ​​support the quantitative calculation of the zoning index, providing data for subsequent judgment of the continuity of mineralization zoning and determination of the size of spatial data blocks, ensuring that spatial analysis conforms to the mineralization mechanism of wollastonite contact metasomatism.

[0125] Based on the comprehensive characteristic values ​​of silicification and carbonate rocks, the wollastonite zoning index of the spatial unit is determined.

[0126] Among them, the wollastonite zoning index (B value) refers to a quantitative index calculated by combining the comprehensive characteristic values ​​of silicification and carbonate rocks. It is used to quantify the relative balance between silicon (Si) and calcium (Ca) within a spatial unit. The value range is fixed in [-1, +1]. The closer the value is to 0, the more it indicates that the remote sensing characteristic response intensity of silicification and carbonate rocks is in a relatively balanced state, indicating a transitional area with strong contact metasomatism.

[0127] It should be noted that, in the embodiments of the present invention, spatial units are used as the basic analysis carriers. The comprehensive characteristic values ​​of silicification and carbonate rocks within the unit are extracted by remote sensing data, and then substituted into the formula to calculate the zoning index (B value). Finally, the silicon-calcium balance state is quantified by the index, which provides a basis for judging the continuous transition of mineralization zoning and determining the optimal size of spatial data blocks, ensuring that spatial analysis is highly compatible with the wollastonite contact metasomatic mineralization mechanism.

[0128] Furthermore, the continuous transition index of the calculated target region includes: Calculate the difference in wollastonite zoning index between adjacent spatial units.

[0129] In this context, adjacent spatial units refer to the smallest analytical units (denoted as ) that are spatially connected after being divided according to the same spatial sampling interval (Δd) in an ore-forming profile constructed based on a local coordinate system (with the contact zone orientation as the x-axis and the contact zone normal as the y-axis). and , representing two adjacent units on the k-th metasomatic section. Its division conforms to the mineralization spatial pattern of intrusive-wall rock contact metasomatism, and adjacent units jointly reflect the geological feature changes along the hydrothermal migration direction.

[0130] The difference refers to the absolute difference in the wollastonite zoning index between two adjacent spatial units (denoted as ). ( ), used to determine whether the mineralization zone is in a continuous transition state, and is an intermediate parameter for calculating the continuous transition index.

[0131] Based on a preset threshold, the difference in the zoning index of wollastonite is binarized to obtain the labeling information.

[0132] The preset threshold refers to the quantitative standard used to determine whether the difference in the zoning index of wollastonite ore meets the characteristics of continuous transition. It is determined based on the statistical distribution characteristics of the zoning index in the study area. For example, the preset threshold can be 0.5 times the standard deviation of the zoning index, which is the basis for the determination of binarization transformation.

[0133] Binarization refers to the process of converting continuous zone index differences (GB values) into discrete values ​​containing only 0 or 1. For example, the rule is to take 1 when the GB value is less than a preset threshold and 0 when it is greater than the threshold. It is a quantization operation that simplifies data and highlights continuous transition characteristics.

[0134] The label information refers to the discrete values ​​(0 or 1) obtained after binarization. 1 represents a continuous transition in the mineralization zoning of adjacent spatial units, while 0 represents a jump in zoning. It is used to calculate the continuous transition index. The basic data for (value).

[0135] For example, in an embodiment of the present invention, 0.5 times the standard deviation of the zoning index is used as a preset threshold to perform binarization transformation on the difference in the zoning index of wollastonite ore between adjacent spatial units, and the continuity of the zoning is transformed into intuitive label information (0 or 1). This operation realizes the discretization processing of continuous data, provides standardized data for subsequent calculation of continuous transition index and judgment of the spatial continuity of mineralization zoning, and ensures that the spatial scale screening fits the mineralization mechanism.

[0136] Based on the marking information, the total number of continuous spatial units in all profiles is determined, and the continuity transition index of the target region is determined based on the total number of continuous spatial units and the total number of profiles.

[0137] The total number of continuous spatial units refers to the sum of the number of adjacent spatial units with a marker information of 1 in all profiles, that is, the total number of spatial unit combinations that are in a continuous transition state along the hydrothermal migration direction, which can reflect the continuous range of mineralization zones.

[0138] The total number of profiles (M) refers to the total number of targeted profiles constructed along the contact metasomatic direction between the intrusive body and the surrounding rock. It is a parameter for balancing the continuous transition characteristics of different profiles and calculating the overall indicators of the region.

[0139] Continuous transition index ( The value (value) refers to a quantitative index that measures the spatial continuity of wollastonite mineralization zones within a target area. It is calculated by the total number of continuous spatial units, the total number of profiles, and the effective index set within the profile. The closer the value is to 1, the more continuous the mineralization zone is along the hydrothermal migration direction, with no obvious jumps, and it fits the mineralization mechanism.

[0140] It should be noted that, in the embodiments of the present invention, based on the binarized marker information, the total number of spatial units in a continuous transition state in all profiles is counted, and then combined with the total number of profiles and the set of intermediate zone indexes within the profiles, the continuous transition index of the target area is calculated by means of the average value. This index quantifies the spatial continuity of mineralization zones, providing a basis for subsequently determining the size of spatial data blocks that match the target time scale, and ensuring that the spatial scale screening conforms to the zoning pattern of wollastonite contact metasomatism mineralization.

[0141] Step S4: Based on the target time scale and spatial data blocks, select the spatiotemporal coupling feature dataset from the initial geographic feature dataset; the spatiotemporal coupling feature dataset is used to characterize the correspondence between the spatiotemporal coupling feature vector and the mineralization intensity level vector; the spatiotemporal coupling feature vector is used to characterize the geographic feature vector related to wollastonite mineralization.

[0142] Among them, the spatiotemporal coupled feature dataset refers to the structured dataset formed by statistically summarizing the features in the initial geographic feature dataset under the dual constraints of the target time scale and spatial data blocks. It is used to establish the correspondence between the spatiotemporal coupled feature vector and the mineralization intensity level vector.

[0143] Spatiotemporal coupled feature vectors refer to the vector form after integrating geographical features (such as silicification index, tectonic density, slope, etc.) related to wollastonite mineralization within a spatial data block. They centrally represent the comprehensive state of mineralization-related geographical features within a specific spatiotemporal range.

[0144] The mineralization intensity level vector refers to a vector composed of mineralization intensity levels (divided into 0-no mineralization, 1-weak mineralization, and 2-industrial ore bodies according to mineralization potential) in historical mineral data. It serves as the dependent variable (Y) in multiple linear regression analysis to verify the correlation between the spatiotemporal coupling feature vector and the mineralization process.

[0145] Geographic feature vectors refer to vectors composed of one or more geographic features (such as spectral features, topographic parameters, tectonic features, etc.) related to wollastonite mineralization. They are the basic data form that characterizes the geological, topographic, and spectral attributes of the mineralization area. Spatiotemporal coupled feature vectors are their integrated form under specific spatiotemporal constraints.

[0146] It should be noted that, in the embodiments of the present invention, the selected target time scale and the matching spatial data block are used as dual constraints to extract and integrate geographical features related to wollastonite mineralization from the massive initial geographical feature dataset, forming a spatiotemporal coupled feature dataset. This dataset establishes the correspondence between the spatiotemporal coupled feature vector representing the mineralization-related geographical features and the mineralization intensity level vector representing the actual mineralization situation, providing structured and highly reliable data support for subsequent screening of target mineralization feature parameters and construction of mineralization models through multiple linear regression analysis.

[0147] Step S5: Perform multiple linear regression analysis on the spatiotemporal coupling feature dataset, screen out target feature parameters related to the wollastonite mineralization process, and construct a wollastonite mineralization model based on the target feature parameters.

[0148] Among them, multiple linear regression analysis refers to a statistical analysis method that uses the spatiotemporal coupling feature vector as the independent variable (X) and the mineralization intensity level as the dependent variable (Y) to construct a regression model. The regression coefficients are solved by the least squares method to quantify the correlation strength between each feature and the mineralization process. It is a tool for screening target feature parameters.

[0149] The mineralization process of wollastonite refers to the formation process of wollastonite deposits, which is mainly characterized by contact metasomatic mineralization. It is primarily related to the contact between acidic intrusive bodies and carbonate rocks, and is controlled by geological factors such as hydrothermal activity, lithological assemblage, and tectonic conditions. It features temporal stability (long-term stability of the ore body location) and spatial continuity (zonal distribution along the contact zone).

[0150] For example, the target feature parameters refer to key features that are significantly related to the wollastonite mineralization process, selected through multiple linear regression analysis and significance testing (t-test, p-value < 0.05), such as silicification index, carbonate rock index, tectonic density, and slope. These features not only conform to statistical laws but also have clear geological interpretability, and can truly reflect the mineralization controlling factors.

[0151] For example, a wollastonite mineralization model refers to a model that reflects the contact metasomatic mineralization mechanism of wollastonite, constructed based on target characteristic parameters and using existing methods (such as rule-based induction, weighted superposition analysis, and expert knowledge-driven mineralization favorability evaluation). It is used to reconstruct mineralization regularities and provide reliable data support for delineating mineral exploration target areas and deploying mineral exploration.

[0152] It should be noted that, in the embodiments of the present invention, the spatiotemporal coupling feature dataset is used as the basis for analysis. The correlation strength between geographical features and mineralization processes is quantified through multiple linear regression analysis. The target feature parameters with both statistical reliability and geological significance are screened out by combining significance test. Then, these key parameters are integrated using existing mineralization analysis methods. Finally, a mineralization model that can accurately reflect the contact metasomatic mineralization mechanism of wollastonite is constructed. This can solve the problems of unclear mechanism and low prediction accuracy of traditional methods and provide scientific support for mineral resource exploration.

[0153] Furthermore, the multivariate linear regression analysis performed on the spatiotemporal coupling feature dataset to screen target feature parameters related to the wollastonite mineralization process includes: Calculate the variance inflation factor of the spatiotemporal coupling feature vector, and perform initial screening of the spatiotemporal coupling feature vector based on the preset variance inflation factor threshold to obtain the initial screening spatiotemporal coupling feature vector.

[0154] It should be noted that the variance inflation factor (VIF) is a core indicator that measures the strength of multicollinearity among spatiotemporally coupled feature vectors. Its value is ≥1. The larger the value, the stronger the multicollinearity among features, which will lead to distortion of regression coefficients.

[0155] For example, the preset variance inflation factor threshold refers to a quantitative standard used to screen low collinearity features. If VIF ≤ 10, features exceeding this threshold are judged as high collinearity features and need to be removed.

[0156] Initial screening refers to the feature preprocessing step before multiple linear regression analysis. It is used to remove highly collinear features, optimize the dataset structure, avoid the interference of multicollinearity on the regression analysis results, and lay the foundation for accurate calculation of regression coefficients in the subsequent process.

[0157] The spatiotemporal coupling feature vectors after initial screening refer to the set of feature vectors that retain VIF≤10 (or retain them after compression by Lasso regression) after the variance inflation factor test. This set eliminates redundant high collinear features and retains only mutually independent and non-overlapping mineralization-related features.

[0158] It should be noted that, in the embodiments of the present invention, by calculating the variance inflation factor (VIF) of the spatiotemporal coupled feature vector, the features are initially screened with a VIF≤10 as a preset threshold to remove highly collinear features (or Lasso regression is used to compress irrelevant features), and finally a low-redundancy, highly independent feature vector is obtained after initial screening. This process solves the problem of regression model distortion caused by information overlap between features, optimizes the dataset structure, and provides reliable input data for subsequent calculation of regression coefficients and screening of key mineralization features using the least squares method.

[0159] Determine the regression coefficients corresponding to the spatiotemporal coupling feature vectors after initial screening, and construct a regression function based on the regression coefficients.

[0160] Among them, the regression coefficient ( In a multiple linear regression model, the parameter denoted as is a parameter that measures the strength and direction of the influence of each spatiotemporal coupling feature vector after initial screening on the dependent variable (mineral change). It is obtained by solving the least squares method. The larger its absolute value, the stronger the explanatory power of the corresponding feature on the mineralization process; the sign reflects the positive / negative correlation between the feature and the mineralization intensity, and is a quantitative basis for judging the correlation between the feature and mineralization.

[0161] The regression function refers to the function based on the spatiotemporal coupling feature vector (independent variable X) and regression coefficients after initial screening. ) and constant term ( The mathematical model is constructed to quantify the linear relationship between geographical features and mineral resource changes.

[0162] For example, the regression function can be expressed as: ; in, This represents the change in mineral resources (mineralization intensity level vector). to These are the spatiotemporal coupling feature vectors after each initial screening. For constant terms, to These are the regression coefficients for the corresponding feature vectors.

[0163] It should be noted that, in the embodiments of the present invention, the regression coefficients corresponding to each feature are first obtained by using the least squares method on the initial screening spatiotemporally coupled feature vectors after spatiotemporal constraints and high collinearity elimination. Then, by integrating the regression coefficients, eigenvectors, and constant terms, a regression function is constructed to quantify the linear relationship between geographical features and mineral resource changes. This function forms the basis for subsequent t-tests to screen key mineralization features (determining significant features with p-values ​​< 0.05), providing support for accurately identifying target feature parameters closely related to the wollastonite mineralization process.

[0164] Calculate the probability that the regression coefficient is 0.

[0165] For example, regression coefficients The statistical probability of a value equal to 0 (i.e., the p-value) is a quantitative indicator calculated through a t-test. It is used to determine whether the regression coefficient is significantly different from 0 and to reflect the reliability of the correlation between the corresponding feature and the mineralization process.

[0166] It should be noted that calculating the probability (p-value) that the regression coefficient is equal to 0 is essentially to verify the statistical significance of the coefficient through a t-test. If the p-value is less than the preset threshold, it means that the regression coefficient is significantly different from 0, and the corresponding feature has a reliable correlation with the wollastonite mineralization process; otherwise, it means that the feature has a weak correlation with mineralization and should be eliminated, ultimately supporting the accurate screening of key mineralization feature parameters.

[0167] Based on preset judgment thresholds and probabilities, target feature parameters related to the wollastonite mineralization process are determined from the spatiotemporal coupling feature vectors after initial screening.

[0168] Among them, the preset judgment threshold is used as a quantitative standard to determine whether the feature is significantly related to the mineralization process. For example, the preset judgment threshold can be 0.05, which is the basis for judging the target feature parameters.

[0169] For example, the statistical probability (i.e., p-value) that the regression coefficient is significantly different from 0 is calculated through a t-test, reflecting the reliability of the association between the corresponding spatiotemporal coupling feature vector and the mineralization intensity level (dependent variable Y).

[0170] The mineralization process of wollastonite refers to the mineralization process centered on the contact metasomatism between acidic intrusive bodies and carbonate rocks, and controlled by factors such as hydrothermal activity, lithological combination, and tectonic conditions. It has the characteristics of temporal stability and spatial continuity.

[0171] Target feature parameters refer to key features (such as silicification index, carbonate rock index, tectonic density, etc.) that have a significant statistical correlation with the wollastonite mineralization process after significant screening. They have both geological interpretability and statistical reliability and are the core input for constructing mineralization models.

[0172] It should be noted that, in the embodiments of the present invention, a preset significance level threshold (p value < 0.05) is used as the judgment criterion, and the probability (p value) calculated by the t-test is combined to perform a secondary screening of the spatiotemporal coupling feature vectors after the initial screening; only features with p values ​​less than the threshold and significantly related to the mineralization process are retained, and the target feature parameters are finally determined. This process realizes dual control of geological constraints and statistical verification, ensuring that the screened features are both consistent with the mineralization mechanism and have reliable statistical correlation, providing high-quality data support for the subsequent construction of mineralization models.

[0173] Furthermore, the construction of the regression function based on the regression coefficients includes: Based on the mineralization intensity level, a regression function is constructed using each spatiotemporally coupled feature vector and its corresponding regression coefficient.

[0174] It should be noted that, in the embodiments of this invention, the actual mineralization intensity level, representing mineralization, is used as the dependent variable, and the mineralization-related geographical features (spatiotemporally coupled feature vectors) under spatiotemporal constraints are used as the independent variables. A regression function quantifying the linear correlation between the two is constructed by combining the regression coefficients corresponding to each feature obtained through the least squares method. This function forms the basis for subsequently extracting key mineralization feature parameters through t-tests (screening for significant features with p-values ​​< 0.05), providing a mathematical tool for constructing a geologically reliable and statistically dependable wollastonite mineralization model.

[0175] In summary, this invention, in constructing the mineralization model of wollastonite deposits, targets the contact zone between the intrusive body and carbonate rocks, as well as the surrounding buffer zone, and acquires multi-source basic data within this target area. An initial geographic feature dataset is extracted from the multi-source basic data. Time-series stability analysis is performed on the observed characteristics of the mineralized edge zone corresponding to wollastonite contact metasomatic mineralization, combined with the spatial continuity constraint of wollastonite contact metasomatic mineralization, to determine the target time scale and spatial data blocks matching the target time scale. Based on the target time scale and spatial data blocks, a spatiotemporal coupled feature dataset is selected from the initial geographic feature dataset. Multiple linear regression analysis is performed on the spatiotemporal coupled feature dataset to select target feature parameters related to the wollastonite mineralization process, and a wollastonite mineralization model is constructed based on these target feature parameters. This invention, by introducing the spatial continuity constraint of the wollastonite contact metasomatic mineralization process and time-series stability analysis of observed features in the construction stage of the mineralization target feature parameters, avoids the feature selection distortion problem caused by purely data-driven methods in mineralization analysis. Under the constraint of time scale, the size of spatial data blocks is determined so that the time scale and spatial scale are mutually constrained, thereby ensuring that the constructed feature dataset can truly reflect the zonal structure characteristics of wollastonite. Through multiple linear regression, the correlation between spatiotemporal coupling characteristics and mineral changes is statistically analyzed, so as to effectively screen the target feature parameters that are significantly related to the wollastonite mineralization process and improve the accuracy of the wollastonite mineralization model construction.

[0176] This invention proposes a spatiotemporal constraint-based metallogenic model construction system for wollastonite deposits. Please refer to [link to relevant documentation]. Figure 2 The diagram illustrates a structural schematic of a spatiotemporally constrained wollastonite mineralization model construction system 200 according to an embodiment of the present invention. The system includes: Data acquisition module 201 is used to acquire multi-source basic data within the target area, which is the contact zone between the intrusive body and the carbonate rock and the surrounding buffer zone. The multi-source basic data includes multi-temporal optical remote sensing images, digital elevation models and derived topographic parameters, basic geological data and historical mineral data. An initial geographic feature dataset is extracted from the multi-source basic data. The initial geographic feature dataset includes spectral statistical features, topographic and structural related features, and spatial statistical and textural features. The condition constraint module 202 is used to perform time series stability analysis on the observation characteristics of the mineral edge zone corresponding to wollastonite contact metasomatic mineralization, and combine the spatial continuity constraint of wollastonite contact metasomatic mineralization to determine the target time scale and the spatial data block that matches the target time scale. The target feature parameter determination module 203 is used to select a spatiotemporal coupled feature dataset from the initial geographic feature dataset based on the target time scale and spatial data blocks. The spatiotemporal coupled feature dataset is used to characterize the correspondence between the spatiotemporal coupled feature vector and the mineralization intensity level vector. The spatiotemporal coupled feature vector is used to characterize the geographic feature vector related to wollastonite mineralization. Multiple linear regression analysis is performed on the spatiotemporal coupled feature dataset to select target feature parameters related to the wollastonite mineralization process, and a wollastonite mineralization model is constructed based on the target feature parameters.

[0177] In one embodiment, the calculation module is used to extract mineral edge zone information at each time point based on the mineralization influence zone, whereby the mineral edge zone is used to characterize the transition zone between the mineralized and unmineralized areas; to count the number of overlapping graticles and the total number of covered graticles of mineral edge zones at adjacent time points within different time windows using a raster indicator function; the raster indicator function is used to indicate whether each graticle in the target area belongs to the edge zone of wollastonite ore; to calculate the spatial overlap rate based on the number of overlapping graticles and the total number of covered graticles, and to determine the target time scale based on the spatial overlap rate; the spatial overlap rate is used to characterize the ratio of the number of overlapping graticles to the total number of covered graticles.

[0178] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the system for constructing a wollastonite mineralization model based on spatiotemporal constraints and the embodiment of a method for constructing a wollastonite mineralization model based on spatiotemporal constraints provided in the above embodiments belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0179] This invention also provides an electronic device. Please refer to [link to relevant documentation]. Figure 3 The electronic device may include a processor 301, a memory 302, and a program 3021 stored in the memory 302 and capable of running on the processor 301.

[0180] When program 3021 is executed by processor 301, it can achieve the following: Figure 1 Any steps in the corresponding method embodiments and the achievement of the same beneficial effects will not be repeated here.

[0181] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by hardware related to program instructions, and the program can be stored in a readable medium.

[0182] This invention also provides a readable storage medium storing a computer program, which, when executed by a processor, can perform the above-described functions. Figure 1 Any step in the corresponding method embodiment can achieve the same technical effect, and will not be repeated here to avoid repetition.

[0183] The computer-readable storage medium of this invention can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. For example, a computer-readable storage medium can be an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0184] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0185] The program code contained on the storage medium can be transmitted using any suitable medium, including but not limited to wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0186] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or terminal. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0187] This invention also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned steps to implement the spatiotemporal constraint-based wollastonite mineralization model construction method provided in the above embodiments.

[0188] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0189] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for constructing a mineralization model of a wollastonite deposit based on a space-time constraint, characterized in that, The method includes: The target area is the contact zone between the intrusive body and the carbonate rock and the surrounding buffer zone. Multi-source basic data within the target area are acquired. The multi-source basic data includes multi-temporal optical remote sensing images, digital elevation models and derived topographic parameters, basic geological data and historical mineral data. An initial geographic feature dataset is extracted from the multi-source basic data; the initial geographic feature dataset includes spectral statistical features, topographic and structural related features, and spatial statistical and textural features. By performing time series stability analysis on the observation characteristics of the mineralized edge zone corresponding to wollastonite contact metasomatic mineralization, and combining the spatial continuity constraints of the wollastonite contact metasomatic mineralization, the target time scale and the spatial data block matching the target time scale are determined. Based on the target time scale and the spatial data block, a spatiotemporal coupled feature dataset is selected from the initial geographic feature dataset; the spatiotemporal coupled feature dataset is used to characterize the correspondence between the spatiotemporal coupled feature vector and the mineralization intensity level vector; the spatiotemporal coupled feature vector is used to characterize the geographic feature vector related to wollastonite mineralization; Multiple linear regression analysis was performed on the spatiotemporal coupling feature dataset to screen target feature parameters related to the wollastonite mineralization process, and a wollastonite mineralization model was constructed based on the target feature parameters. The step of performing multiple linear regression analysis on the spatiotemporal coupling feature dataset to screen target feature parameters related to the wollastonite mineralization process includes: Calculate the variance inflation factor of the spatiotemporal coupling feature vector, and perform initial screening on the spatiotemporal coupling feature vector based on a preset variance inflation factor threshold to obtain the initial screening spatiotemporal coupling feature vector. Determine the regression coefficients corresponding to the spatiotemporal coupling feature vectors after the initial screening, and construct a regression function based on the regression coefficients; Calculate the probability that the regression coefficient is 0; Based on the preset judgment threshold and the probability, target feature parameters related to the wollastonite mineralization process are determined from the spatiotemporal coupling feature vector after the initial screening. The method involves performing time-series stability analysis on the observational characteristics of the mineralized edge zone corresponding to wollastonite contact metasomatism to determine the target time scale, including: Based on the mineralization impact zone, mineral edge zone information is extracted at each time point. The mineral edge zone information includes the spatial range of the mineral edge zone and the observation characteristics within the transition zone. The spatial range is used to characterize the transition zone between the mineralized area and the non-mineralized area. Based on the mineral edge zone information, the number of overlapping grids and the total number of covered grids of mineral edge zones at adjacent time nodes within different time windows are counted using a grid indicator function; the grid indicator function is used to indicate whether each grid in the target area belongs to the edge zone of wollastonite ore. Based on the number of overlapping graticles and the total number of covered graticles, the spatial overlap rate is calculated, and the target time scale is determined based on the spatial overlap rate; the spatial overlap rate is used to characterize the ratio of the number of overlapping graticles to the total number of covered graticles.

2. The method for constructing a mineralization model of wollastonite based on spatiotemporal constraints according to claim 1, characterized in that, The determination of spatial data blocks matching the target timescale, based on the spatial continuity constraints of wollastonite contact metasomatism mineralization, includes: Starting from a stable mineral margin zone within the target timescale, a profile is constructed along the contact metasomatic direction between the intrusive body and the surrounding rock; the contact metasomatic direction is the hydrothermal migration direction of the intrusive body towards the surrounding rock. Establish a local coordinate system with the contact zone direction as the x-axis and the contact zone normal as the y-axis, and divide the spatial units according to different spatial sampling intervals; Calculate the wollastonite zoning index of each of the spatial units and the continuity transition index of the target region; Based on the wollastonite zoning index and the continuous transition index, spatial data blocks matching the target time scale are determined.

3. The method for constructing a mineralization model of wollastonite based on spatiotemporal constraints according to claim 2, characterized in that, The calculation of the wollastonite zoning index for each of the spatial units includes: For any spatial unit, a comprehensive characteristic value of silicification and a comprehensive characteristic value of carbonate rocks are obtained; the comprehensive characteristic value of silicification is used to characterize the intensity and distribution characteristics of silicification within the ore-forming region; the comprehensive characteristic value of carbonate rocks is used to characterize the distribution and reaction degree of carbonate rocks within the ore-forming region. Based on the comprehensive characteristic values ​​of silicification and carbonate rocks, the wollastonite zoning index of the spatial unit is determined.

4. The method for constructing a mineralization model of wollastonite based on spatiotemporal constraints according to claim 3, characterized in that, The calculation of the continuous transition index of the target region includes: Calculate the difference in wollastonite zoning index between adjacent spatial units; Based on a preset threshold, the difference in the zoning index of the wollastonite ore is binarized to obtain the marking information; Based on the marking information, the total number of continuous spatial units in all profiles is determined, and the continuity transition index of the target region is determined based on the total number of continuous spatial units and the total number of profiles.

5. The method for constructing a mineralization model of wollastonite based on spatiotemporal constraints according to claim 1, characterized in that, The extraction of the initial geographic feature dataset from the multi-source basic data includes: Based on the multi-temporal optical remote sensing images, the spectral statistical features and the spatial statistical and texture features are extracted; the spectral statistical features include the mean, variance and band ratio features of each band, and the spatial statistical and texture features include gray-level co-occurrence matrix texture features and local change intensity index. Based on the digital elevation model and derived topographic parameters, and the basic geological data, the topographic and structural features are extracted; the topographic and structural features include slope, aspect, relief, and structural density. An initial geographic feature dataset is constructed based on the spectral statistical features, the topographic and structural features, and the spatial statistics and texture features.

6. The method for constructing a mineralization model of wollastonite based on spatiotemporal constraints according to claim 1, characterized in that, The construction of the regression function based on the regression coefficients includes: Based on the mineralization intensity level, a regression function is constructed using each of the spatiotemporal coupled feature vectors and their corresponding regression coefficients.

7. A system for constructing a mineralization model of wollastonite based on spatiotemporal constraints, characterized in that, The system includes: The data acquisition module is used to acquire multi-source basic data within the target area, which includes the contact zone between the intrusive body and the carbonate rock and the surrounding buffer zone. The multi-source basic data includes multi-temporal optical remote sensing images, digital elevation models and derived topographic parameters, basic geological data, and historical mineral data. An initial geographic feature dataset is extracted from the multi-source basic data. The initial geographic feature dataset includes spectral statistical features, topographic and structural correlation features, and spatial statistical and textural features. The condition constraint module is used to perform time series stability analysis on the observation characteristics of the mineralized edge zone corresponding to wollastonite contact metasomatic mineralization, and combine the spatial continuity constraint of the wollastonite contact metasomatic mineralization to determine the target time scale and the spatial data block that matches the target time scale. The target feature parameter determination module is used to filter out a spatiotemporal coupled feature dataset from the initial geographic feature dataset based on the target time scale and the spatial data block; the spatiotemporal coupled feature dataset is used to characterize the correspondence between the spatiotemporal coupled feature vector and the mineralization intensity level vector; the spatiotemporal coupled feature vector is used to characterize the geographic feature vector related to wollastonite mineralization; multiple linear regression analysis is performed on the spatiotemporal coupled feature dataset to filter out target feature parameters related to the wollastonite mineralization process, and a wollastonite mineralization model is constructed based on the target feature parameters; The target feature parameter determination module is further configured to calculate the variance inflation factor of the spatiotemporal coupling feature vector, perform initial screening of the spatiotemporal coupling feature vector based on a preset variance inflation factor threshold, obtain the initial-screened spatiotemporal coupling feature vector; determine the regression coefficients corresponding to the initial-screened spatiotemporal coupling feature vectors, and construct a regression function based on the regression coefficients; calculate the probability that the regression coefficient is 0; and determine the target feature parameters related to the wollastonite mineralization process from the initial-screened spatiotemporal coupling feature vectors according to a preset judgment threshold and the probability. The condition constraint module is further configured to extract mineral edge zone information for each time node based on the mineralization influence zone. The mineral edge zone information includes the spatial range of the mineral edge zone and the observation features within the transition zone. The spatial range is used to characterize the transition zone between the mineralized and non-mineralized areas. Based on the mineral edge zone information, the number of overlapping graticles and the total number of covered graticles for adjacent time nodes within different time windows are statistically analyzed using a raster indicator function. The raster indicator function is used to indicate whether each graticle in the target area belongs to the edge zone of wollastonite. Based on the number of overlapping graticles and the total number of covered graticles, the spatial overlap rate is calculated, and the target time scale is determined based on the spatial overlap rate. The spatial overlap rate is used to characterize the ratio of the number of overlapping graticles to the total number of covered graticles.

Images