A Method and System for Predicting Nickel Ore Target Areas Based on Hyperspectral Remote Sensing Data Analysis

Abstract

This invention provides a method and system for predicting nickel ore target areas based on hyperspectral remote sensing data analysis, belonging to the field of mineral resource exploration technology. First, hyperspectral remote sensing data of the area to be predicted is acquired. Then, alteration feature spatial clustering construction processing is performed on the hyperspectral remote sensing data to form alteration feature spatial clusters, identifying the spatial distribution of various geological alteration types and binding spectral features. Subsequently, sequential spatial evolution processing of mineralized alteration combinations is performed to obtain a sequential evolution body of mineralized alteration. Next, target area indicator feature anchoring and shaping construction processing is performed on the sequential evolution body of mineralized alteration to form a target area indicator spatial shaping body, locating key indicator features of the nickel ore target area. Finally, spatial reorganization and delineation processing of the nickel ore target area is carried out to generate the nickel ore target area prediction result. This invention fully utilizes the spectral and spatial information of hyperspectral remote sensing data, effectively improving the accuracy and reliability of nickel ore target area prediction.

Description

Technical Field

[0001] This invention relates to the field of mineral resource exploration technology, and more specifically, to a method and system for predicting nickel ore target areas based on hyperspectral remote sensing data analysis. Background Technology

[0002] In the field of mineral resource exploration, nickel ore, as an important metallic mineral, requires efficient and accurate exploration for national economic development and resource security. Traditional nickel ore exploration methods mainly rely on geological surveys, geophysical exploration, and geochemical exploration. Geological surveys involve field investigations and analysis of geological information such as rocks and strata to infer potential nickel ore areas. However, these methods are limited by manpower and material resources, have a limited scope, and lack in-depth understanding of deep geological conditions. Geophysical exploration uses physical methods, such as gravity, magnetics, and electrical methods, to detect differences in the physical properties of underground geological bodies to locate nickel ore. However, different geological bodies may have similar physical properties, leading to multiple interpretations and difficulty in guaranteeing accuracy. Geochemical exploration delineates nickel ore target areas by analyzing the chemical element content in media such as soil, rocks, and stream sediments. However, this method is easily affected by surface environmental factors, such as weathering and transportation, which alter element distribution and affect the reliability of exploration results.

[0003] In recent years, hyperspectral remote sensing technology has been increasingly applied to the field of mineral resource exploration. Hyperspectral remote sensing data possesses rich spectral information, capable of reflecting subtle differences in the spectral characteristics of ground features, thus providing new tools for mineral exploration. However, most existing mineral exploration methods based on hyperspectral remote sensing data simply utilize spectral information for classification or identification, failing to fully consider the geospatial information contained within the hyperspectral data and the complex relationship between geological alteration and mineralization. This makes it difficult to accurately and comprehensively predict nickel ore target areas, and thus cannot meet the needs of modern nickel ore exploration. Summary of the Invention

[0004] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide a method for predicting nickel ore target areas based on hyperspectral remote sensing data analysis, the method comprising:

[0005] The hyperspectral remote sensing data volume of the region to be predicted is obtained. The hyperspectral remote sensing data volume includes the full-band spectral response characteristics and the geospatial distribution characteristics of each geographic detection point in the region to be predicted. The full-band spectral response characteristics are the continuous response characteristics of spectral reflectance and spectral absorption formed by the geographic detection points within the hyperspectral remote sensing detection band. The geospatial distribution characteristics are the geographic coordinates of each geographic detection point and the spatial adjacency and spatial extension relationships between the points.

[0006] The hyperspectral remote sensing data volume is subjected to alteration feature spatial clustering construction processing to form an alteration feature spatial cluster body. The alteration feature spatial cluster body is a continuous cluster distribution structure formed by various geological alteration types based on geographic spatial distribution characteristics, and the continuous cluster distribution structure is bound to the full-band spectral response characteristics of the corresponding geological alteration type.

[0007] The alteration feature spatial cluster is subjected to sequential spatial evolution processing of mineralization alteration combination to form mineralization alteration sequential evolution body. The mineralization alteration sequential evolution body is a spatial sequential extension structure formed by geological alteration combination adapted to nickel ore mineralization characteristics in the order of mineralization alteration occurrence. The spatial sequential extension structure is bound to the full-band spectral response feature evolution content of the corresponding alteration combination.

[0008] The target area indicator feature anchoring and shaping construction process is performed on the mineralization alteration sequence evolution body to form a target area indicator spatial shaping body. The target area indicator spatial shaping body is a shaped spatial distribution structure constructed with the core alteration feature of nickel ore mineralization as the anchoring center, and the shaped spatial distribution structure is bound to multi-level target area indicator alteration features.

[0009] The target area indicator space shaping body is subjected to nickel ore target area spatial straightening and delineation processing to form nickel ore target area prediction results.

[0010] Furthermore, embodiments of the present invention also provide a nickel ore target area prediction system based on hyperspectral remote sensing data analysis, comprising:

[0011] A processor; a machine-readable storage medium for storing machine-executable instructions of the processor; wherein the processor is configured to execute the above-described nickel ore target area prediction method based on hyperspectral remote sensing data analysis by executing the machine-executable instructions.

[0012] In another aspect, embodiments of the present invention also provide a computer program product, the computer program product including machine-executable instructions, the machine-executable instructions being stored in a computer-readable storage medium, a processor of a computer device reading the machine-executable instructions from the computer-readable storage medium, the processor executing the machine-executable instructions, causing the computer device to execute the above-described method for predicting nickel ore target areas based on hyperspectral remote sensing data analysis.

[0013] Based on the above, by acquiring hyperspectral remote sensing data volumes containing full-band spectral response characteristics and geospatial distribution characteristics, the spectral information and spatial distribution information of ground objects are fully considered. The alteration feature spatial clustering construction process is performed on the hyperspectral remote sensing data volume to form an alteration feature spatial cluster, which can accurately identify the spatial distribution of various geological alteration types and bind their corresponding spectral response features. The alteration feature spatial cluster is then subjected to the sequential spatial evolution processing of mineralization alteration combination to form a mineralization alteration sequential evolution body, which simulates the spatial evolution process of geological alteration combination adapted to nickel mineralization characteristics in mineralization order and binds the evolution content of spectral response features. The mineralization alteration sequential evolution body is then subjected to the target area indicator feature anchoring and shaping construction process to form a target area indicator spatial shaping body. With the core alteration feature of nickel mineralization as the anchoring center, multi-level target area indicator alteration features are bound. Finally, the target area indicator spatial shaping body is subjected to the nickel ore target area spatial consolidation and delineation processing to form the nickel ore target area prediction result, which greatly improves the accuracy and reliability of nickel ore target area prediction and reduces exploration costs and risks. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the execution flow of the nickel ore target area prediction method based on hyperspectral remote sensing data analysis provided in an embodiment of the present invention.

[0015] Figure 2 This is a schematic diagram of exemplary hardware and software components of a nickel ore target area prediction system based on hyperspectral remote sensing data analysis provided in an embodiment of the present invention. Detailed Implementation

[0016] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart illustrating a method for predicting nickel ore target areas based on hyperspectral remote sensing data analysis, provided in one embodiment of the present invention. A detailed description follows.

[0017] Step S110: Obtain hyperspectral remote sensing data volume of the area to be predicted. The hyperspectral remote sensing data volume includes the full-band spectral response characteristics and geospatial distribution characteristics of each geographic detection point in the area to be predicted. The full-band spectral response characteristics are the continuous response characteristics of spectral reflection and spectral absorption formed by the geographic detection points within the hyperspectral remote sensing detection band. The geospatial distribution characteristics are the geographic coordinates of each geographic detection point and the spatial adjacency and spatial extension relationships between the points.

[0018] In this embodiment, the hyperspectral remote sensing data volume is acquired through hyperspectral remote sensing satellites or airborne remote sensing platforms. It contains geographic observation point data distributed within the area to be predicted according to a set spatial resolution. Each geographic observation point corresponds to a set of full-band spectral response characteristics and geospatial distribution characteristics. The full-band spectral response characteristics are continuous response curves of spectral reflectance and spectral absorptivity as a function of wavelength within the band range from the starting wavelength to the ending wavelength, formed according to the set spectral resolution. This is represented as a one-dimensional array of length N, where each array element corresponds to a spectral reflectance value at a specific wavelength, and N is the total number of bands. The geospatial distribution characteristics include the latitude and longitude coordinates of each geographic observation point, a spatial adjacency matrix calculated using the Delaunay triangulation algorithm, and a spatial extension direction vector determined based on the minimum spanning tree algorithm. For example, the area to be predicted is divided into grids of a set size, with the center of each grid serving as a geographic detection point. There are a total of M points. The geospatial distribution characteristics of each point include longitude, latitude, and the index number and spatial distance value of the surrounding points that are spatially adjacent to that point.

[0019] Step S120: Perform alteration feature spatial clustering construction processing on the hyperspectral remote sensing data volume to form an alteration feature spatial cluster body. The alteration feature spatial cluster body is a continuous cluster distribution structure formed by various geological alteration types based on geographic spatial distribution characteristics, and the continuous cluster distribution structure is bound to the full-band spectral response characteristics of the corresponding geological alteration type.

[0020] In this embodiment, the spatial clustering construction process of alteration features combines spectral features with geospatial features to construct the spatial distribution morphology of geological alteration types related to nickel ore formation. This process first extracts and binds spectral and geospatial features from the hyperspectral remote sensing data volume, then compares and matches them with a standard spectral library to identify the geographic detection points corresponding to different geological alteration types. Finally, it analyzes the spatial distribution relationships of these points to construct a continuous clustered distribution structure.

[0021] Step S121: Extract the full-band spectral response features from the hyperspectral remote sensing data volume, and bind the full-band spectral response features of each geographic detection point to the geographic coordinates of the corresponding geographic detection point point one by one to form a spectral feature geographic point binding set. Each geographic coordinate in the spectral feature geographic point binding set uniquely corresponds to a set of full-band spectral response features.

[0022] In this embodiment, full-band spectral response feature data and geographic coordinate data are first separated from the hyperspectral remote sensing data volume. The full-band spectral response feature data is a two-dimensional array, with row indices corresponding to geographic detection point numbers and column indices corresponding to spectral band numbers. The array element value is the reflectance of that point in the corresponding band. The geographic coordinate data is also a two-dimensional array, with row indices corresponding to geographic detection point numbers and two columns corresponding to longitude and latitude values, respectively. By matching the two arrays by row index, the full-band spectral response feature array of each geographic detection point is combined with its corresponding longitude and latitude values ​​into a tuple. All tuples constitute the spectral feature geographic point binding set. For example, for a geographic detection point numbered P, its full-band spectral response feature is an array of length N, and its geographic coordinates are (longitude value, latitude value). This is combined into a tuple of ([reflectance1, reflectance2, ..., reflectanceN], longitude value, latitude value) and stored in the spectral feature geographic point binding set.

[0023] Step S122: Retrieve the standard spectral response library of geological alteration types related to nickel ore formation, extract the standard full-band spectral response features corresponding to various geological alteration types in the standard spectral response library, and perform band synchronization regularization on all standard full-band spectral response features according to the band division rules of hyperspectral remote sensing to form a regularized standard spectral feature set.

[0024] In this embodiment, the geological alteration types related to nickel ore formation include serpentinization, chloritization, epidote alteration, and carbonatization, among others. A standard spectral response library is stored in a local database. Standard full-band spectral response features for each alteration type are extracted from this library within the band from the starting wavelength to the ending wavelength. The spectral resolution of the original standard spectrum may differ from that of the hyperspectral remote sensing data, requiring interpolation processing based on the spectral resolution of the hyperspectral remote sensing data. A linear interpolation algorithm is used to calculate the reflectance values ​​corresponding to each wavelength at a set spectral resolution interval based on the reflectance values ​​of known wavelength points in the standard spectrum, ensuring that the length of the standard full-band spectral response features is consistent with the detected spectrum, both consisting of N data points. The normalized standard spectral features for each alteration type are associated and stored with the corresponding alteration type name, forming a normalized standard spectral feature set, where each element has the structure (alteration type name, [standard reflectance 1, standard reflectance 2, ..., standard reflectance N]).

[0025] Step S123: Perform band-by-band feature comparison between the single-class standard full-band spectral response features in the normalized standard spectral feature set and the single-group detection full-band spectral response features in the spectral feature geographic location binding set, and select spectral segments that have feature matching with the standard full-band spectral response features from the detection full-band spectral response features to form a single-location spectral feature matching set.

[0026] In this embodiment, for each geographic detection point in the spectral feature geographic point binding set, it is sequentially compared with each type of geological alteration standard spectrum in the normalized standard spectral feature set. The comparison is carried out from the band dimension, analyzing the characteristic morphology and detail differences between the detected spectrum and the standard spectrum in each band, and selecting matching spectral fragments.

[0027] Step S1231: Extract the standard full-band spectral response features of a single type of geological alteration from the regularized standard spectral feature set, and divide it into multiple continuous band units according to the band division rules of hyperspectral remote sensing. Each band unit retains complete spectral reflectance and spectral absorption response features to form a set of single-type standard spectral band units.

[0028] In this embodiment, taking serpentinization alteration as an example, its standard full-band spectral response characteristics are an array of length N. Based on the natural segmentation of the bands detected by hyperspectral remote sensing, the region from the starting wavelength to the first boundary wavelength is divided into a first band region, and the region from the first boundary wavelength to the ending wavelength is divided into a second band region. Within the first band region, several band units are obtained by dividing the region at a set wavelength interval; similarly, several band units are obtained by dividing the region at another set wavelength interval. Each band unit contains spectral reflectance data for all wavelengths within that band, forming a single-class standard spectral band unit set, where each element is (band number, [reflectance value 1, reflectance value 2, ..., reflectance value K]), where K is the number of wavelength points contained in that band unit.

[0029] Step S1232: Extract the full-band spectral response features of a single geographic detection point in the spectral feature geographic point binding set, and split it into multiple band units corresponding to the standard spectrum according to the same band division rules to form a single-point detection spectral band unit set, thereby realizing the synchronous splitting of band units of the detection spectrum and the standard spectrum.

[0030] In this embodiment, for the geographic detection point P bound to the spectral feature geographic point location, its full-band spectral response feature array is extracted. Following the same band division rules as in step S1231—that is, the first band region is divided according to a set wavelength interval, and the second band region is divided according to another set wavelength interval—the detection spectrum is split into multiple band units with the same number of standard spectral band units and corresponding band ranges. Each band unit also contains reflectance data within the corresponding wavelength range, forming a single-point detection spectral band unit set. Its structure is consistent with that of a single-class standard spectral band unit set, facilitating subsequent unit-by-unit comparison.

[0031] Step S1233: Extract a single standard band unit from the single-class standard spectral band unit set and a single detection band unit corresponding to the single-point detection spectral band unit set, analyze the changing trends of spectral reflectance and spectral absorption within the two band units, compare the characteristic morphological matching of the two, and form a single band unit feature comparison result.

[0032] In this embodiment, taking band unit B in the standard spectral band unit set and band unit B' in the detector spectral band unit set as examples, the reflectance data within the two band units are first analyzed for trends. The first derivative of the reflectance data within each band unit is calculated, and the rising segment, falling segment, and inflection point position of the spectral curve are determined by the positive and negative changes of the first derivative. Then, the number and relative positions of the rising and falling segments of the standard band unit B and the detector band unit B', as well as the number and distribution of inflection points, are compared. If the number of rising and falling segments is the same, the relative position deviation is within a set range, and the deviation of the number and distribution of inflection points is also within a set range, then the two are determined to have a matching feature morphology; otherwise, they are determined to be mismatched, forming a single-band unit feature comparison result.

[0033] Step S1234: For the feature comparison results of single-band units with consistent spectral change trends, calculate the spectral angle of the detected band unit and the corresponding standard band unit within the same band range. If the spectral angle is less than the set threshold, it is determined to be a feature detail match, and the matching band unit detail confirmation content is formed.

[0034] In this embodiment, after step S1233 determines that the feature morphology matches, the spectral angle is calculated. The spectral angle is calculated as follows: the reflectance data of the standard band unit B and the probe band unit B' are considered as two vectors, and the angle between these two vectors is calculated. First, the two vectors are normalized so that the magnitude of the vectors is 1. Then, the dot product of the two normalized vectors is calculated, and the spectral angle is obtained by using the inverse cosine function. If the calculated spectral angle is less than a set threshold, it indicates that the two band units have a high degree of matching in spectral detail, forming a matching band unit detail confirmation content, and the number and spectral angle value of the band unit are recorded; otherwise, even if the morphology matches, the details do not match, and the band unit is not recorded.

[0035] Step S1235: Compare all standard band units in the single-class standard spectral band unit set with the corresponding detection band units in the single-point detection spectral band unit set unit by unit, and collect all band units that form characteristic morphological matching and detail matching to form a single-point matching band unit set.

[0036] In this embodiment, according to the band unit numbering order, steps S1233 and S1234 are performed sequentially on each standard band unit in the single-class standard spectral band unit set and the corresponding detection band unit in the single-point detection spectral band unit set. All band unit numbers that simultaneously satisfy feature morphology matching and feature detail matching are aggregated to form a single-point matching band unit set. This single-point matching band unit set contains the numbering information of all detection band units that match the standard spectrum of the current alteration type.

[0037] Step S1236: Extract the band numbers of all matching band units in the single-point matching band unit set, and connect each matching band unit continuously according to the order of the band numbers to form a continuous spectral segment corresponding to a single geographic detection point and a single type of standard spectrum, thus forming a single-point single-type spectral matching segment.

[0038] In this embodiment, the numbers of all matching band units are extracted from the single-point matching band unit set and sorted in ascending order of band number. Since the band units are divided according to wavelength order, the numbering order corresponds to the wavelength sequence. The sorted band units are then connected sequentially, that is, the reflectance value of the last wavelength point of the previous band unit is continuously arranged with the reflectance value of the first wavelength point of the next band unit to form a continuous spectral segment. This spectral segment is the single-point single-type spectral matching segment corresponding to a single geographic detection point and the current alteration type.

[0039] Step S1237: Count the number of band units contained in the single-point single-class spectral matching segment, and confirm that the spectral matching segment is a continuous band distribution according to the continuous distribution state of the band units, thus forming the validity confirmation content of the single-point single-class spectral matching segment.

[0040] In this embodiment, the number of band units contained in a single-point, single-type spectral matching segment is counted. If the proportion of this number to the total number of band units is greater than a set proportion threshold, and the numbering of these band units is continuous, that is, there are no unmatched band units in the middle, then the spectral matching segment is confirmed to be a continuous band distribution pattern and has validity, forming the validity confirmation content of the single-point, single-type spectral matching segment; otherwise, the spectral matching segment is determined to be discontinuous and has low validity.

[0041] Step S1238: For the full-band spectral response characteristics of a single geographic detection point, the above-mentioned band-by-band comparison and spectral matching fragment extraction are completed with the standard full-band spectral response characteristics of all geological alteration types in the normalized standard spectral feature library to form a set of multi-class spectral matching fragments for a single point.

[0042] In this embodiment, for a single geographic detection point, following the process from steps S1231 to S1237, the spectral data is sequentially compared with the standard spectral data of each type of geological alteration in the normalized standard spectral feature library. For each alteration type, a corresponding spectral matching fragment (if present) is extracted, and its validity confirmation is recorded. All spectral matching fragments and validity confirmation information corresponding to all alteration types are summarized to form a set of multi-type spectral matching fragments for a single point.

[0043] Step S1239: Extract all spectral matching segments with continuous band distribution from the single-point multi-class spectral matching segment set, and bind each spectral matching segment with the corresponding geological alteration type to form a single-point spectral feature matching initial set.

[0044] In this embodiment, spectral matching segments with continuous band distribution patterns are selected from the set of multi-type spectral matching segments at a single location. Each of the above spectral matching segments is bound to its corresponding geological alteration type name to form a tuple structure of (alteration type name, spectral matching segment, validity confirmation content). All tuples constitute the initial set of spectral feature matching at a single location.

[0045] Step S12310: Organize the initial set of single-point spectral feature matching, determine the geological alteration type and corresponding continuous spectral matching fragments corresponding to a single geographic detection point, remove non-continuous spectral fragments, and form a single-point spectral feature matching set.

[0046] In this embodiment, the tuples in the initial set of single-point spectral feature matching are checked. If the spectral matching fragment corresponding to a certain geological alteration type is determined to be a continuous band distribution after validity verification, the tuple is retained. If there are discontinuous spectral fragment contents, i.e., the validity verification content is discontinuous, the tuple is removed from the initial set. After processing, the geological alteration type and the corresponding continuous spectral matching fragment corresponding to a single geographic detection point are obtained, forming a single-point spectral feature matching set.

[0047] Step S124: For the spectral feature geographic point binding set, the full-band spectral response features of all geographic detection points are sequentially compared and spectral fragments are screened. All geographic detection points that form a single-point spectral feature matching set are collected. The corresponding geographic detection points are classified according to the matched geological alteration type to form an alteration type geographic point classification set.

[0048] In this embodiment, each geographic detection point in the spectral feature geographic point binding set is traversed, and the operation of step S123 is performed for each point to obtain a single-point spectral feature matching set for each point. For geographic detection points whose single-point spectral feature matching sets are not empty, their geographic coordinates and corresponding alteration type information are recorded. Then, the above-mentioned geographic detection points are classified according to the alteration type name, and the geographic coordinates of all geographic detection points belonging to the same alteration type are grouped together to form an alteration type geographic point classification set, where each element is (alteration type name, [geographic coordinate 1, geographic coordinate 2, ..., geographic coordinate Q]), and Q is the number of geographic detection points corresponding to this alteration type.

[0049] Step S125: Extract the geospatial distribution characteristics of all geographic detection points corresponding to a single geological alteration type in the alteration type geographic point classification set, analyze the spatial adjacency relationship between each geographic detection point, identify geographic detection point groups with continuous spatial adjacency relationships, and form a single-type alteration continuous point group set.

[0050] In this embodiment, taking the serpentinization alteration type in the geographical point classification set of alteration types as an example, the geospatial distribution characteristics of all corresponding geographical detection points are extracted, including the geographical coordinates and spatial adjacency matrix of each point. Based on the spatial adjacency matrix, the connected component analysis method in graph theory is used to group spatially adjacent geographical detection points into a group. Specifically, each geographical detection point is regarded as a node in the graph. If two points are spatially adjacent (determined by the spatial adjacency matrix), an edge is established between the corresponding nodes. Through depth-first search or breadth-first search algorithms, all connected components in the graph are found, and each connected component is a group of geographical detection points with continuous spatial adjacency. All the above point groups are aggregated to form a single-type alteration continuous point group set.

[0051] Step S126: Extract the geographic coordinates of all geographic detection points in a single geographic detection point group within the single-type alteration continuous point group set, and connect the geographic coordinates of each point continuously according to the spatial extension direction of the geographic detection points to form a spatial continuous distribution line of single-type geological alteration, and construct a single-type alteration spatial line set.

[0052] In this embodiment, it is achieved through the following sub-steps:

[0053] For example, step S1261: extract the geographic coordinates of all geographic detection points in a single geographic detection point group within the single-type alteration continuous point group set, and preliminarily arrange all geographic coordinates according to their spatial distribution to form a preliminary arrangement set of coordinates for a single point group.

[0054] In this embodiment, a set of geographic detection points is selected from the set of continuous points of single-type alteration, and the geographic coordinates (longitude and latitude values) of all geographic detection points in the set are extracted. The coordinates of the points are then preliminarily drawn on a two-dimensional plane to form a preliminary set of coordinates for a single point group, which visually displays the spatial distribution of the points.

[0055] Step S1262: Analyze the spatial distribution trend of the concentrated geographic detection points in the preliminary arrangement of the single point group coordinates, identify the main spatial direction of the point distribution, determine the core spatial extension direction of the geographic detection point group, and form the core extension direction of the single point group.

[0056] In this embodiment, principal component analysis is used to analyze the geographic coordinates of the initial distribution set of single-point locations. The geographic coordinates are converted into Cartesian coordinates (X, Y), forming a two-dimensional data matrix. The covariance matrix of this matrix is ​​calculated, and then the eigenvalues ​​and eigenvectors of the covariance matrix are solved. The direction of the eigenvector with the largest eigenvalue is the main spatial direction of the point distribution. This direction is determined as the core extension direction of the single-point group and represented by a unit vector.

[0057] Step S1263: Select the geographic detection point located at the starting point of the core space extension direction from the initial arrangement of the single point group coordinates as the starting point, extract the geographic coordinates of the starting point and the spatial adjacency relationship with the surrounding points, and form the starting point feature of the single point group line.

[0058] In this embodiment, based on the core spatial extension direction vector, the projection value of each geographic coordinate point in the preliminary coordinate arrangement set of the single-point group is calculated in that direction. The point with the smallest projection value is determined to be the geographic detection point located at the starting point of the core spatial extension direction, and is taken as the starting point. The geographic coordinates (longitude value S, latitude value S) of this starting point are extracted, along with the indices of all points adjacent to this starting point obtained through the spatial adjacency matrix, forming the starting point feature of the single-point group line.

[0059] Step S1264: Based on the starting point features of the single-point group lines, select adjacent geographic detection points in sequence according to the core spatial extension direction, extract the geographic coordinates of the adjacent points, connect the geographic coordinates of the adjacent points with straight lines to form the initial spatial connection line segments, and construct the initial line segment set of the single-point group.

[0060] In this embodiment, starting from the starting point, adjacent geographic detection points are searched along the core spatial extension direction. The direction of the line connecting the starting point and each adjacent point is calculated. The adjacent point with the smallest angle to the core spatial extension direction is selected as the next connection point. The geographic coordinates of this point are extracted, and a straight line is used to connect the starting point and this connection point to form a spatial connection line segment. Then, using this connection point as the new starting point, the above process is repeated until no adjacent points that meet the conditions can be found, forming an initial sequence of spatial connection line segments and constructing an initial line segment set for a single point group.

[0061] Step S1265: Continue to connect subsequent geographic detection points along the core spatial extension direction and the secondary extension direction in sequence, and include all geographic detection points with spatial adjacency in the connection range to form a preliminary spatial connection network covering the entire point group, and construct the initial connection network of a single point group.

[0062] In this embodiment, in addition to the core spatial extension direction, there are also secondary extension directions. These secondary extension directions are obtained by solving for the eigenvector corresponding to the second largest eigenvalue of the covariance matrix. Based on the initial set of line segments, for unconnected geographic detection points, it is checked whether they have a spatial adjacency relationship with already connected points. If so, they are connected to already connected points along the secondary extension direction or other possible directions, forming more spatial connection segments. Ultimately, this forms a preliminary spatial connection network covering the entire point group, i.e., the initial connection network for a single point group.

[0063] Step S1266: In the initial connection network of the single point group, calculate the length of each connection segment and the number of points within a certain neighborhood radius of the two ends of the connection segment, remove connection segments whose length is less than a set length threshold and whose number of points in the neighborhood is less than a set number threshold and their associated points, and retain the segments that have not been removed as the core connection branch set of the single point group.

[0064] In this embodiment, for each connecting segment in the initial connection network of a single-point group, its length is calculated, which is the distance between the geographic coordinates of the two endpoints of the segment. Simultaneously, a neighborhood radius is set with the two endpoints of the segment as the center, and the number of geographic detection points contained within this neighborhood is counted. If the length of a connecting segment is less than a set length threshold, and the number of points in the neighborhood of its two endpoints is also less than a set number threshold, then the segment is determined to be a secondary connection, removed from the network, and the isolated points associated with that segment are also removed. The remaining connecting segments constitute the core connection branch set of the single-point group.

[0065] Step S1267: Extract all coordinate points of a single core connection branch in the single-point group core connection branch set, and use the B-spline curve algorithm to fit the coordinate points to generate a smooth spatial line of the single-point group.

[0066] In this embodiment, a core connection branch is selected from the set of core connection branches of a single point group. This core connection branch consists of a series of continuous connection segments, containing multiple geographic coordinate points. These coordinate points are arranged in connection order and used as control points for the B-spline curve. By selecting an appropriate curve order and node vector, the points on the curve are calculated using the B-spline curve formula to generate a smooth spatial line. This spatial line can well fit the distribution trend of the original coordinate points, forming a smooth spatial line for the single point group.

[0067] Step S1268: For cases where there are multiple groups of geographic detection points under the same geological alteration type, construct corresponding smooth spatial lines for each group of geographic detection points to form a set of spatial lines for multiple groups of points.

[0068] In this embodiment, when the same geological alteration type corresponds to multiple geographic detection point groups, steps S1261 to S1267 are performed on each point group to generate a smooth spatial line for each point group. All these smooth spatial lines are then aggregated to form a multi-point group spatial line set.

[0069] Step S1269: Analyze the spatial positional relationship between multiple smooth spatial lines in the multi-point spatial line group. If there are coordinate points that connect the lines, connect the connecting coordinate points to integrate multiple lines into longer continuous spatial distribution lines, forming a single-type etched integrated spatial line.

[0070] In this embodiment, the shortest distance between any two smooth spatial lines in the multi-point spatial line set is calculated. If the shortest distance is less than a set connection threshold, it is determined that there are coordinate points where the two lines are spatially connected. The two closest coordinate points are found and used as connection points. The two lines are connected at the connection points with a straight line or curve to form a longer continuous spatial distribution line. This process is repeated until all connectable lines are integrated to form a single-type etched integrated spatial line.

[0071] Step S12610: Collect all single-type alteration integrated spatial lines and smooth spatial lines that do not form a connection under the same geological alteration type, construct a single-type alteration spatial line set, and determine the starting point, ending point and core spatial extension direction of each spatially continuous distribution line.

[0072] In this embodiment, the single-type alteration integrated spatial lines and the smooth spatial lines that do not form a connection are summarized. Each line records its starting point coordinates, ending point coordinates, and the core spatial extension direction determined by a method similar to step S1262. All these lines together constitute a set of single-type alteration spatial lines.

[0073] Step S127: Extract the full-band spectral response features of all geographic detection points on a single spatially continuous distribution line in the single-type alteration spatial line set, arrange the full-band spectral response features of each point continuously according to the spatial extension direction to form the continuous arrangement content of the spectral features of a single line, and bind it to the corresponding spatially continuous distribution line to form an alteration line spectral binding set.

[0074] In this embodiment, for a single spatially continuous line within a single type of alteration spatial line set, the full-band spectral response feature array of each geographic detection point is extracted sequentially according to the connection order of the points along the line. This array is then continuously arranged along the spatial extension direction (from the starting point to the ending point) to form a two-dimensional array, where rows correspond to geographic detection points, columns correspond to spectral bands, and array elements are reflectance values. This two-dimensional array, as the continuous arrangement of spectral features, is bound to the starting point, ending point, and core extension direction of the corresponding spatially continuous line, forming a tuple structure of (spatial line information, continuous arrangement of spectral features). All tuples constitute the spectral binding set of the alteration line.

[0075] Step S128: Integrate all spatially continuous distribution lines and their bound spectral features under the same geological alteration type in the spectral binding set of the alteration lines, and combine the spatial positional relationship between each line to construct an overall cluster distribution structure of a single type of geological alteration, forming a single-type alteration cluster structure set.

[0076] In this embodiment, for the same geological alteration type, all spatially continuous lines belonging to that type and their corresponding continuous spectral features are collected from the alteration line spectral binding set. The spatial relationships of these lines are analyzed, including distances and intersection angles between them. Based on the spatial distribution characteristics of the lines, a region growing algorithm is used to group lines that are close to each other and extend in the same direction into a cluster. A minimum bounding polygon is constructed for each cluster as its outline boundary, forming the overall cluster distribution structure of a single type of geological alteration. All cluster distribution structures constitute a single-type alteration cluster structure set.

[0077] Step S129: Arrange the cluster distribution structures of all geological alteration types in the single-type alteration cluster structure set according to the actual geographic spatial distribution relationship in the area to be predicted, retain the actual spatial location relationship between various types of cluster distribution structures, and form a global alteration cluster arrangement set.

[0078] In this embodiment, all cluster distribution structures within a single type of alteration cluster structure set are placed according to their actual geographical coordinates within the area to be predicted. During the arrangement process, the relative spatial relationships between each cluster distribution structure are maintained, such as distance and orientation, to ensure that the true distribution of geological alteration within the area to be predicted is reflected. All arranged cluster distribution structures form a global alteration cluster distribution set.

[0079] Step S1210: The cluster distribution structure of all the clusters in the global alteration cluster arrangement concentration is fully integrated with the full-band spectral response characteristics and geographic spatial distribution characteristics of the corresponding geological alteration types to form an alteration feature spatial cluster body. The alteration feature spatial cluster body includes the cluster distribution structure of all nickel ore mineralization-related geological alteration types in the area to be predicted and the corresponding bound feature content.

[0080] In this embodiment, the distribution structure of each cluster in the global alteration cluster distribution set is fused with the full-band spectral response characteristics (associated through the spectral binding set of alteration lines) and geospatial distribution characteristics (including the geographic coordinate range and extension direction of the cluster structure) of its corresponding geological alteration type. The fusion method involves constructing a three-dimensional data structure, where the spatial dimension represents geographic coordinates and the attribute dimension includes information such as alteration type and spectral characteristics. Through this fusion, an alteration feature spatial cluster body is formed, which completely contains the spatial distribution morphology and corresponding feature information of all nickel ore mineralization-related geological alteration types within the area to be predicted.

[0081] Step S130: Perform sequential spatial evolution processing on the alteration feature spatial cluster to form a sequential evolution body of mineralized alteration. The sequential evolution body of mineralized alteration is a spatial sequential extension structure formed by geological alteration combination adapted to the mineralization characteristics of nickel ore according to the order of mineralization alteration. The spatial sequential extension structure is bound to the full-band spectral response feature evolution content of the corresponding alteration combination.

[0082] In this embodiment, the sequential spatial evolution processing of mineralization alteration assemblages aims to identify alteration assemblages related to nickel mineralization within spatial clusters of alteration features, and to construct their spatial sequential extension structures according to the chronological order of alteration during mineralization, while also associating the evolution of spectral features. This process involves steps such as spatial arrangement analysis of alteration clusters, matching of mineralization sequence rules, and construction of spatial sequential structures.

[0083] Step S131: Extract the cluster distribution structure of various geological alterations and their corresponding geographic spatial distribution range in the alteration feature spatial cluster body, and arrange all cluster distribution structures as a whole according to the geographic spatial orientation of the area to be predicted to form an alteration cluster spatial orientation arrangement set.

[0084] In this embodiment, all cluster distribution structures for each geological alteration type are extracted from the alteration feature spatial cluster body, along with the geographic spatial distribution range of each cluster distribution structure (such as the latitude and longitude coordinates of the smallest bounding rectangle). Based on the geographic spatial orientation of the area to be predicted (such as the direction of longitude or latitude), all cluster distribution structures are located and arranged on a two-dimensional plane according to their geographic spatial distribution range, forming an alteration cluster spatial orientation arrangement set, which intuitively displays the spatial distribution pattern of each cluster structure.

[0085] Step S132: Analyze the positional connection relationship of different geological alteration cluster distribution structures in the geographic space of the spatial orientation arrangement of the alteration clusters, identify the combinations of geological alteration cluster distribution structures that are adjacent or directly connected in spatial distribution, and form a set of alteration cluster spatial connection combinations.

[0086] In this embodiment, for any two different geological alteration cluster distribution structures with concentrated spatial orientations, the intersection of their geographic spatial distribution ranges is calculated. If the intersection is not empty, or the shortest distance between the two ranges is less than a set connection distance threshold, it is determined that the two cluster distribution structures are spatially adjacent or directly connected. The two cluster structures are combined into a set, and their alteration type and spatial position relationship are recorded. All such combinations form an alteration cluster spatial connection combination set.

[0087] Step S133: Retrieve the sequence rules of geological alteration during nickel ore formation, extract the order of occurrence of various geological alterations and the evolutionary relationship between alterations in the sequence rules, extract the core geological alteration types and combination forms corresponding to each stage of nickel ore formation, and form a set of ore-forming alteration sequence rules.

[0088] In this embodiment, the geological alteration sequence rules for the nickel ore mineralization process are retrieved from a geological knowledge base. These rules detail the types of geological alteration occurring at different stages of nickel ore mineralization (e.g., early, middle, and late stages), as well as the chronological order and evolutionary relationships between these alterations. For example, serpentinization may occur in the early stage, chloritization may occur in the middle stage, and carbonatization may occur in the late stage. From these rules, the core alteration types for each mineralization stage and their typical combinations, such as the sequence of (serpentinization → chloritization → carbonatization), are extracted to form a set of mineralization alteration sequence rules.

[0089] Step S134: Compare the alteration types and their spatial arrangement order of a single alteration combination in the alteration cluster spatial connection combination set with the main sequence of alteration occurrence defined in the mineralization alteration sequence rule set. If the type and arrangement order of the alteration combination is a continuous subsequence of the main sequence, it is determined to be a match. All matching alteration combinations are grouped into the initial selection set of mineralization-related alteration combinations.

[0090] In this embodiment, it is achieved through the following sub-steps:

[0091] Step S1341: Extract all geological alteration types contained in a single alteration combination in the alteration cluster spatial connection combination set, and sort the alteration types according to the geometric center point coordinates of various alteration cluster distribution structures along a preset spatial reference direction to form a sequential structure of single-combination alteration spatial distribution.

[0092] In this embodiment, for an alteration combination within the spatially connected alteration cluster set, there are alteration types A and B. The geometric center coordinates (longitude A, latitude A) and (longitude B, latitude B) of the cluster distribution structures for alteration types A and B are calculated respectively. The preset spatial reference direction is due north. The coordinate difference between the two center points along this direction is calculated, and the alteration types are sorted in descending or ascending order of coordinate values ​​to form a sequential structure for the spatial distribution of a single alteration combination, such as (alteration type A, alteration type B) or (alteration type B, alteration type A).

[0093] Step S1342: Extract the main sequence of geological alteration corresponding to the core stage of nickel ore mineralization from the mineralization alteration sequence rule set, determine the order of occurrence and type combination requirements of various geological alterations in the main sequence, and form the main sequence set of mineralization core alteration.

[0094] In this embodiment, the main alteration sequence corresponding to the core stage of nickel ore formation (such as the mineralization stage) is extracted from the set of alteration sequence rules. For example, (serpentinization → chloritization → epidote alteration → carbonatization). The order of occurrence of various alterations in this main sequence and the combination requirements of adjacent alteration types are clearly defined. For example, serpentinization must be followed by chloritization, and chloritization can be followed by epidote alteration, etc., forming the set of core alteration sequence rules for ore formation.

[0095] Step S1343: Perform overall feature matching between the single-combination alteration spatial distribution sequential structure and the main sequential set of mineralization core alteration, analyze the overall matching of the combination and arrangement order of geological alteration types in the two, and form the overall sequential matching result of the single combination.

[0096] In this embodiment, the alteration type sequence in the spatial distribution sequence of single-combination alteration is compared with the main sequence sequence in the main sequence set of mineralization core alteration. If the single-combination sequence is a continuous subsequence of the main sequence sequence, for example, if the main sequence is (A→B→C→D) and the single-combination sequence is (B→C), then the overall matching is determined to be valid, and the overall sequence matching result of the single combination is considered a match; otherwise, it is considered a mismatch.

[0097] Step S1344: For the overall sequential matching results of the single combination that form overall matching, extract the spatial extension direction of the cluster distribution structure of each geological alteration type in the alteration spatial distribution sequence of the single combination, analyze the matching of the spatial extension direction with the alteration evolution direction of the main sequence set of the mineralization core alteration, and form the spatial evolution direction matching result of the single combination.

[0098] In this embodiment, after step S1343 determines an overall match, the core spatial extension direction of each alteration type cluster distribution structure in the single-combination alteration spatial distribution sequence is extracted (determined through step S1262). The above direction vectors are arranged in sequential order to obtain a spatial extension direction sequence. The alteration evolution direction of the main sequence set of mineralized core alteration is a preset direction vector (such as along the structural zone). The average angle between the spatial extension direction sequence and the alteration evolution direction is calculated. If the average angle is less than a set angle threshold, the spatial evolution direction is determined to be matched, and the single-combination spatial evolution direction matching result is considered a match; otherwise, it is considered a mismatch.

[0099] Step S1345: Extract the full-band spectral response characteristics of various geological alteration types in the sequential structure of the single-combination alteration spatial distribution, analyze the evolutionary correlation between each characteristic, compare the matching of the evolutionary correlation with the spectral evolution characteristics of the concentrated alteration of the mineralization alteration sequence, and form a single-combination spectral evolution characteristic matching result.

[0100] In this embodiment, the full-band spectral response feature arrays of each alteration type in the spatial distribution sequence of a single alteration combination are extracted. The correlation coefficients of spectral features of adjacent alteration types are calculated, and the position and intensity changes of spectral reflection peaks and absorption valleys are analyzed to determine the evolutionary correlation between spectral features. The ore-forming alteration sequence rule set defines the spectral evolution features of alteration, such as the gradual increase or decrease of reflectance in a specific band from early to late stages. The spectral evolution correlation of a single combination is compared with the spectral evolution features in the rule set. If the trends are consistent, the spectral evolution features are considered to match, and the result of the single combination spectral evolution feature matching is considered a match; otherwise, it is considered a mismatch.

[0101] Step S1346: For the single-combination overall sequential matching results that do not form an overall match, extract the geological alteration type fragments in the alteration spatial distribution sequence of the single combination that have a local match with the main sequence set of alteration of the mineralization core, analyze whether the geological alteration type fragment is an alteration combination of the core stage of nickel mineralization, and form a single-combination local sequential matching result.

[0102] In this embodiment, for single combinations determined to be mismatched overall in step S1343, a sliding window method is used to search for segments that locally match the main sequence set of alteration of the mineralization core within the spatial distribution sequence structure of the single combination alteration. For example, if the main sequence is (A→B→C→D) and the single combination sequence is (A→C→D), then (C→D) may be a locally matching segment. The analysis determines whether this local segment belongs to the alteration combination of the nickel ore mineralization core stage. If so, the single combination local sequence matching result is considered a match; otherwise, it is considered a mismatch.

[0103] Step S1347: For the local sequential matching results of the single combination that form local matching, extract the other geological alteration types in the alteration combination, analyze whether the corresponding alteration type is an associated alteration type of nickel mineralization alteration, and whether it has spatial distribution and evolution correlation with the core alteration type, and form a single combination associated alteration matching result.

[0104] In this embodiment, for single combinations that form local matches, the remaining alteration types besides the locally matched segments are extracted. A geological knowledge base is queried to determine whether these alteration types are known associated alteration types in nickel ore mineralization alteration. Simultaneously, the spatial distribution relationship (e.g., whether they are adjacent) and evolutionary correlation relationship (e.g., whether there is a transition in spectral characteristics) between these associated alteration types and the core alteration type (alteration types in the locally matched segments) are analyzed. If all conditions are met, the associated alteration matching result of the single combination is considered a match; otherwise, it is considered a mismatch.

[0105] Step S1348: Collect alteration combinations that form overall matching and alteration combinations that form local matching and have associated alteration matching to form a candidate set of mineralization-related alteration combinations.

[0106] In this embodiment, all alteration combinations that are fully matched in step S1343 and matched in both steps S1344 and S1345, as well as alteration combinations that are partially matched in step S1346 and matched by associated alteration in step S1347, are collected and combined to form a candidate set of mineralization-related alteration combinations.

[0107] Step S1349: Extract the cluster distribution structure corresponding to each alteration combination in the candidate set of mineralization-related alteration combinations, calculate the ratio of the total length of the continuous distribution lines in each cluster distribution structure to the area enclosed by the lines. If the ratio is lower than a set threshold, the alteration cluster distribution structure is determined to be a combination with poor spatial extensibility and is removed to form a mineralization-related alteration combination screening set.

[0108] In this embodiment, for each alteration combination in the candidate set of mineralization-related alteration combinations, all clustered distribution structures contained therein are extracted. For each clustered distribution structure, the total length of all continuous distribution lines and the area enclosed by the clustered structure are calculated (using the minimum bounding polygon). The ratio of total length to area is calculated. If this ratio is lower than a set threshold, it indicates that the spatial extension of the clustered structure is poor, and it may be a non-mineralization-related scattered alteration, thus it is removed from the candidate set. The remaining alteration combinations form the mineralization-related alteration combination screening set.

[0109] Step S13410: Use the mineralization-related alteration combination screening set as the initial selection set of mineralization-related alteration combinations to determine all geological alteration combinations in the area to be predicted that match the mineralization characteristics of nickel ore.

[0110] In this embodiment, the mineralization-related alteration combination screening set is directly used as the initial selection set of mineralization-related alteration combinations. This initial selection set of mineralization-related alteration combinations includes all geological alteration combinations in the area to be predicted that have been screened and matched multiple times and have been determined to match the mineralization characteristics of nickel ore.

[0111] Step S135: Extract the cluster distribution structure of various geological alterations within a single mineralization-related alteration combination in the preliminary selection set of the mineralization-related alteration combination, and arrange the cluster distribution structure sequentially according to the order of alteration occurrence in nickel ore formation to form a single combination alteration cluster sequential arrangement structure.

[0112] In this embodiment, for a mineralization-related alteration assemblage initially selected, the cluster distribution structures of various geological alterations within the assemblage are sorted according to the alteration occurrence sequence defined in the mineralization alteration sequence rule set. For example, if the assemblage includes serpentinization, chloritization, and carbonatization, and these are arranged in the rule sequence as serpentinization → chloritization → carbonatization, their cluster distribution structures are arranged in this order to form a single-assembly alteration cluster sequential arrangement structure.

[0113] Step S136: Extract the geographic spatial coordinates of the cluster distribution structure of geological alteration at each stage in the single-combination alteration cluster sequential arrangement structure, and continuously connect the cluster distribution structures of the preceding alteration and the subsequent alteration according to the spatial adjacency relationship to form a spatial sequential extension structure of the mineralization alteration combination, and construct a single-combination spatial sequential extension set.

[0114] In this embodiment, the geometric center coordinates of the alteration cluster distribution structure at each stage in the sequential arrangement structure of a single-assembly alteration cluster are extracted. For adjacent preceding and subsequent alterations, the edge points of their cluster structures are calculated, and the two closest edge points are found as connection points. These two connection points are connected with a straight line or curve to achieve continuous connection between the preceding and subsequent alteration cluster structures. All alteration cluster structures are connected sequentially to form a spatial sequential extension structure of the mineralization alteration assemblies. Each mineralization-related alteration assembly corresponds to one of the above structures, and all structures constitute a single-assembly spatial sequential extension set.

[0115] Step S137: Extract the full-band spectral response features of geological alteration at each stage in the single-combination spatial sequential extension concentrated mineralization alteration combination spatial sequential extension structure, arrange the full-band spectral response features of each stage continuously according to the alteration sequential evolution direction to form the spectral feature sequential evolution arrangement content, and bind it to the corresponding spatial sequential extension structure.

[0116] In this embodiment, full-band spectral response feature arrays corresponding to the alteration cluster distribution structures at each stage are extracted from the spatially sequential extension structure of the single-combination spatially sequential extension set (associated through the spectral binding set of alteration lines). Following the direction of alteration sequential evolution (i.e., from early to late stages), the aforementioned spectral feature arrays are arranged continuously to form a three-dimensional array, where the first dimension represents the alteration stage, the second dimension represents the geographic detection point, and the third dimension represents the spectral band. This three-dimensional array is used as the content of the spectral feature sequential evolution arrangement and bound to the corresponding spatially sequential extension structure, forming a tuple of (spatially sequential extension structure, spectral feature sequential evolution arrangement content).

[0117] Step S138: Extract the alteration cluster distribution structure and the bound spectral feature sequential evolution arrangement content corresponding to the core stage of nickel ore mineralization from the single combination spatial sequential extension set, focus on extracting the feature content of the core stage, form the core mineralization stage alteration feature set, and bind it to the spatial sequential extension structure of the corresponding mineralization alteration combination.

[0118] In this embodiment, the alteration type corresponding to the core stage of nickel ore formation (such as the mineralization stage) is determined based on the ore-forming alteration sequence rule set. Within the spatial sequential extension structure of the single-assembly spatial sequential extension set, the cluster distribution structure corresponding to the alteration type of this core stage is found, and the portion related to the core stage in its bound spectral feature sequential evolution arrangement is extracted. This portion of features is then extracted in detail, such as calculating the average spectrum and extracting the reflectance of characteristic bands, to form the core ore-forming stage alteration feature set, which is then bound to the corresponding spatial sequential extension structure.

[0119] Step S139: Arrange the spatial sequential extension structure of all mineralization-related alteration combinations in the preliminary selection set of mineralization-related alteration combinations, the sequential evolution arrangement of the bound spectral features, and the core mineralization stage alteration feature set according to the actual geographic spatial distribution of the area to be predicted, to form a global mineralization alteration sequential arrangement set.

[0120] In this embodiment, the spatial sequential extension structure of each mineralization-related alteration assembly is initially selected and arranged according to its actual geographic coordinates within the region to be predicted. Simultaneously, the bound spectral feature sequential evolution arrangement and the core mineralization stage alteration feature set are also associated with their corresponding spatial locations. All of these elements together constitute a global mineralization alteration sequential arrangement set, demonstrating the spatial sequential distribution of mineralization-related alteration assemblies within the region to be predicted.

[0121] Step S1310: The spatial sequential extension structure of all mineralization alteration combinations and the various bound features in the global mineralization alteration sequential arrangement are fully integrated to form a mineralization alteration sequential evolution body. The mineralization alteration sequential evolution body contains the dual features of the sequential evolution and spatial distribution of all mineralization-related alteration combinations in the area to be predicted.

[0122] In this embodiment, spatial data fusion technology is employed to integrate the spatial sequential extension structure of all mineralization alteration assemblages, the sequential evolution of spectral features, and the alteration feature set of the core mineralization stage from the global mineralization alteration sequential distribution set. The fused mineralization alteration sequential evolution is a comprehensive data structure that includes both the spatial sequential extension morphology of each mineralization-related alteration assemblage and the changes in spectral features with sequential evolution, achieving a unification of both sequential evolution and spatial distribution characteristics.

[0123] Step S140: Perform target area indicator feature anchoring and shaping construction processing on the ore-forming alteration sequential evolution body to form a target area indicator spatial shaping body. The target area indicator spatial shaping body is a shaped spatial distribution structure constructed with the core alteration features of nickel ore formation as the anchoring center, and the shaped spatial distribution structure is bound to multi-level target area indicator alteration features.

[0124] In this embodiment, the target area indicator feature anchoring and shaping construction process is based on the sequential evolution of mineralization alteration, using core alteration features as anchor points to construct a shaped spatial structure with a specific geometric shape, and binding multi-level target area indicator features.

[0125] Step S141: Extract the spatial sequential extension structure of each mineralization alteration combination in the mineralization alteration sequential evolution body, identify the alteration cluster distribution structure corresponding to the core stage of nickel mineralization in each mineralization alteration combination, determine the geographic spatial center coordinates of the alteration cluster distribution structure, and use it as the core anchor point for target area indication to form a core anchor point set.

[0126] In this embodiment, the spatial sequential extension structure of each mineralization alteration assemblage in the mineralization alteration sequence evolution body is traversed. Based on the alteration feature set of the core mineralization stage, the alteration cluster distribution structure corresponding to the core nickel mineralization stage is identified. The geometric center coordinates of this cluster distribution structure are calculated (by averaging the coordinates of all boundary points), and these coordinates are used as the core anchoring points for target area indication. The core anchoring points of all mineralization alteration assemblages constitute the core anchoring point set.

[0127] Step S142: Extract the spatial sequential extension structure of a single mineralization alteration combination in the mineralization alteration sequential evolution body, identify all preceding alteration cluster distribution structures before the core alteration cluster distribution structure corresponding to the core anchoring point, and all subsequent alteration cluster distribution structures after it, extract the geographic spatial range covered by the circumscribed polygon of the alteration cluster distribution structure, and form the initial alteration combination spatial association range.

[0128] In this embodiment, it is achieved through the following sub-steps:

[0129] Step S1421: Extract the spatial sequential extension structure of a single mineralized alteration combination in the mineralized alteration sequential evolution body, extract the core alteration cluster distribution structure corresponding to the core anchoring point, obtain the complete geographic spatial distribution coordinates of the cluster distribution structure, and form a core alteration cluster spatial feature set.

[0130] In this embodiment, a spatial sequential extension structure of a mineralized alteration assemblage is selected from the sequential evolution of mineralized alteration. Based on the core anchoring points, the corresponding core alteration cluster distribution structure is found. The geographic coordinates of all boundary points of this cluster structure, as well as the coordinates of the geographic detection points contained within it, are extracted to form a spatial feature set of the core alteration cluster.

[0131] Step S1422: Based on the alteration sequence defined in the spatial sequential extension structure, extract a geological alteration cluster distribution structure that precedes the core alteration cluster distribution structure in the sequence as a first-level preceding alteration cluster distribution structure, obtain its geographic spatial distribution coordinates, and form a first-level preceding alteration spatial feature set.

[0132] In this embodiment, within the alteration sequence of the spatially sequentially extending structure, the core alteration cluster distribution structure has a defined position. Following the sequence order, the alteration cluster distribution structure preceding it is the first-order preceding alteration cluster distribution structure. The geographical coordinates of the boundary points and internal points of this structure are extracted to form the first-order preceding alteration spatial feature set.

[0133] Step S1423: Based on the alteration sequence defined in the spatial sequential extension structure, extract a geological alteration cluster distribution structure located after the core alteration cluster distribution structure in the sequence as a first-level subsequent alteration cluster distribution structure, obtain its geographic spatial distribution coordinates, and form a first-level subsequent alteration spatial feature set.

[0134] In this embodiment, similar to step S1422, the alteration cluster distribution structure located one position after the core alteration cluster distribution structure in the spatial sequential extension structure is extracted as the first-level subsequent alteration cluster distribution structure, and its geographical coordinates are obtained to form a first-level subsequent alteration spatial feature set.

[0135] Step S1424: Continue to extract the multi-level pre-sequence geological alteration cluster distribution structure corresponding to the first-level pre-sequence alteration spatial feature set along the pre-sequence evolution direction until an alteration type that is not directly related to nickel mineralization is extracted. Record the geographic spatial distribution coordinates of each level of pre-sequence alteration and the spatial distance from the core alteration to form a multi-level pre-sequence alteration spatial feature set.

[0136] In this embodiment, starting from the primary preceding alteration cluster distribution structure, alteration cluster distribution structures are extracted along the preceding direction of the alteration sequence (i.e., earlier alteration stages). For each extracted structure, its geospatial coordinates are recorded, and the spatial distance (e.g., the distance between the geometric centers) between this structure and the core alteration cluster distribution structure is calculated. Extraction stops when the extracted alteration type is marked as having no direct association with nickel mineralization in the mineralization alteration sequence rule set. All extracted preceding alteration structures and their distance information form a multi-level preceding alteration spatial feature set.

[0137] Step S1425: Continue to extract the multi-level geological alteration cluster distribution structure corresponding to the first-level alteration spatial feature set along the subsequent evolution direction until an alteration type that is not directly related to nickel mineralization is extracted. Record the geographic spatial distribution coordinates of each level of alteration and the spatial distance from the core alteration to form a multi-level alteration spatial feature set.

[0138] In this embodiment, similar to step S1424, the alteration cluster distribution structure is extracted along the subsequent direction of the alteration sequence (i.e., the later alteration stage), and the coordinates and distances are recorded until an alteration type with no direct association is encountered, forming a multi-level subsequent alteration spatial feature set.

[0139] Step S1426: Integrate the geographic spatial distribution coordinates of all alteration cluster distribution structures in the core alteration cluster spatial feature set, the multi-level preceding alteration spatial feature set, and the multi-level subsequent alteration spatial feature set, and draw the overall spatial distribution outline of all alteration cluster distribution structures to form a global spatial outline map of alteration combination.

[0140] In this embodiment, the geospatial distribution coordinates of all clustered distribution structures of core alteration, multi-level pre-alteration alteration, and multi-level post-alteration alteration are integrated into the same coordinate system. By drawing the boundaries of each clustered structure, a global spatial outline map of the alteration combination is formed, which intuitively shows the overall spatial distribution range of the mineralization alteration combination.

[0141] Step S1427: Taking the geographic spatial center of the core alteration cluster spatial feature set as the origin, and combining the spatial distance between the pre- and post-alteration alteration cluster distribution structures at each level and the origin, delineate the minimum geographic spatial range that can include all mineralization-related alteration cluster distribution structures, and form an initial outline of the alteration combination spatial association range.

[0142] In this embodiment, the geometric center of the core alteration cluster distribution structure is taken as the origin, and the boundary point farthest from the origin among all mineralization-related alteration cluster distribution structures is identified. A circular or rectangular area is drawn with the origin as the center and this farthest distance as the radius, serving as the minimum geographic spatial range that can encompass all mineralization-related alteration cluster distribution structures, thus forming the initial outline of the spatial association range of alteration assemblies.

[0143] Step S1428: Extract the geographic coordinate boundary of the initial outline of the spatial association range of the alteration combination, and expand the boundary of the initial outline according to the spatial extension edge of all mineralization-related alteration cluster distribution structures to ensure that the whole of all mineralization-related alteration cluster distribution structures is within the expanded range, thus forming the extended outline of the spatial association range of the alteration combination.

[0144] In this embodiment, it is checked whether the initially drawn outline completely includes all mineralization-related alteration cluster distribution structures. If some structures exceed the initially drawn outline, the outline boundary is extended along the exceeding direction, with the extension distance being the maximum distance of the exceeding portion. After extension, it is ensured that the entirety of all mineralization-related alteration cluster distribution structures is within the outline range, forming an extended outline of the alteration combination spatial association range.

[0145] Step S1429: Extract all alteration cluster distribution structures within the initial outline of the alteration combination spatial association range. If an unrelated alteration cluster distribution structure that does not conform to the definition type of the mineralization alteration sequential rule set is identified, remove the vertex closest to the geometric center of the unrelated alteration cluster distribution structure from the vertices of the circumscribed polygon, and regenerate the convex polygon boundary to form the alteration combination spatial association range adjustment outline.

[0146] In this embodiment, within the initial outline of the alteration combination spatial association range, it is checked whether there are any unrelated alteration cluster distribution structures outside the defined type of the mineralization alteration sequential rule set. If they exist, the geometric center of the unrelated structure is found, the distance from each vertex of the circumscribed polygon of the initial outline to the center is calculated, the nearest vertex is removed, and then the convex polygon boundary is regenerated based on the remaining vertices to obtain the adjusted outline of the alteration combination spatial association range, so as to eliminate the influence of unrelated alteration.

[0147] Step S14210: Extract the geographic spatial distribution range and corresponding geographic coordinate boundary of the adjusted contour of the alteration combination spatial association range to form the initial alteration combination spatial association range, and determine all mineralization-related alteration cluster distribution structures included in the geographic spatial distribution range.

[0148] In this embodiment, the geographic coordinate boundary of the adjusted contour of the alteration assemblages spatial association range is extracted, including the latitude and longitude coordinates of all points on the boundary. The geographic space enclosed by this boundary is the initial alteration assemblages spatial association range. Simultaneously, the type and quantity of all mineralization-related alteration cluster distribution structures contained within this range are recorded.

[0149] Step S143: Using the core anchoring point as the center, and based on the outer polygon of the initial alteration combination spatial association range, fit according to the preset geometric template corresponding to the typical nickel ore mineralization spatial morphology to generate a single combination shaped spatial distribution structure.

[0150] In this embodiment, the preset typical nickel ore mineralization spatial morphology geometric templates include elliptical, annular, and strip-like shapes. Based on the actual spatial distribution characteristics of the mineralization alteration assemblages, the best-matching geometric template is selected. Using the core anchoring point as the center of the geometric template, the circumscribed polygon of the initial alteration assemblage spatial association range is fitted to the geometric template. The size and orientation of the template are adjusted to ensure optimal coverage of the circumscribed polygon. The fitted geometric template is the single-assembly shaped spatial distribution structure.

[0151] Step S144: Extract the geological alteration cluster distribution structure and full-band spectral response features corresponding to each geometric part in the single-combination shaping spatial distribution structure, and bind various feature contents to specific geometric parts of the shaping space point by point to form a shaping space feature part binding set.

[0152] In this embodiment, the single-combination shaped spatial distribution structure is divided into different geometric parts, such as the central area, the transition area, the edge area, etc. For each geometric part, the geological alteration cluster-shaped distribution structure and its full-band spectral response characteristics contained in this part are extracted. The above characteristic content is associated with the corresponding geometric part coordinate range to achieve point-by-point binding, that is, each coordinate point corresponds to a specific alteration type and spectral characteristics, forming a shaped spatial characteristic part binding set.

[0153] Step S145: In the shaped spatial characteristic part binding set, mark the alteration characteristics corresponding to the core anchor point positions as the first-level anchor characteristics. With the core anchor point positions as the center, divide multiple concentric annular zones according to a preset distance radius, and mark the alteration characteristics located in different annular zones as multi-level anchor characteristics in sequence, forming a multi-level target area indication anchor characteristic set.

[0154] In this embodiment, first mark the alteration characteristics at the positions where the core anchor point positions are located as the first-level anchor characteristics, which are the most direct target area indication characteristics. Then, with the core anchor point positions as the center, set multiple distance radii R1, R2, R3, etc. (R1 < R2 < R3...), and divide multiple concentric annular zones. The one within the radius R1 is the first-level annular zone, the one between R1 and R2 is the second-level annular zone, and so on. Mark the alteration characteristics located in the first-level annular zone as the second-level anchor characteristics, those in the second-level annular zone as the third-level anchor characteristics, and so on, forming a multi-level target area indication anchor characteristic set.

[0155] Step S146: Extract the geographical spatial coordinates of each level of anchor characteristics in the multi-level target area indication anchor characteristic set, construct the spatial association link between the first-level anchor characteristics and each level of anchor characteristics, and bind all the spatial association links to the corresponding shaped spatial distribution structure, forming a shaped spatial characteristic association link set.

[0156] In this embodiment, extract the geographical spatial coordinate points of each level of anchor characteristics in the multi-level target area indication anchor characteristic set. For each non-first-level anchor characteristic point, find the first-level anchor characteristic point closest to it, and connect these two points with a line segment to form a spatial association link. All the above links constitute a spatial association link set, and bind it to the corresponding shaped spatial distribution structure to form a shaped spatial characteristic association link set, showing the spatial connection between each level of anchor characteristics and the core anchor characteristics.

[0157] Step S147: Arrange the shaped spatial distribution structures corresponding to all the ore-forming alteration combinations in the ore-forming alteration sequential evolution body, as well as the bound shaped spatial characteristic part binding set, multi-level target area indication anchor characteristic set, and shaped spatial characteristic association link set, according to the actual geographical spatial position of the area to be predicted, forming a global target area indication shaped arrangement set.

[0158] In this embodiment, the spatial distribution structure of the morphological alteration assemblage of each mineralization alteration combination in the sequential evolution of mineralization alteration is placed according to the actual geographic coordinates of its core anchoring point. At the same time, the bound set of morphological spatial feature parts, the set of multi-level target area indicator anchoring features, and the set of morphological spatial feature association links are also mapped to their respective geographic locations to form a global target area indicator morphological arrangement set, which shows the spatial distribution of all target area indicator morphological structures in the area to be predicted.

[0159] Step S148: Extract the boundary features of each shaped spatial distribution structure in the global target area indicator shaping arrangement set, and combine them with the corresponding multi-level target area indicator anchoring feature set to perform feature marking on the boundary of the shaped space so that the boundary marking content is consistent with the distribution range of the anchoring features, forming a shaped space feature boundary set.

[0160] In this embodiment, the boundary coordinates of each shaped spatial distribution structure are extracted, and the multi-level target area indicator anchoring features corresponding to each point on the boundary are analyzed. Based on the level of the anchoring features, the boundaries are characterized and marked. For example, boundary segments within the distribution range of first-level anchoring features are marked with a specific identifier, and second-level features are marked with another specific identifier. This ensures that the boundary markings accurately reflect the distribution range of the anchoring features, forming a set of shaped spatial characteristic boundaries.

[0161] Step S149: Extract all shaped spatial distribution structures and bound features from the global target area indicator shaping arrangement set, remove features and spatial links that are not directly related to nickel ore formation, retain the core target area indicator features and spatial relationships, and form a simplified target area indicator shaping set.

[0162] In this embodiment, all content in the global target area indicator shaping set is screened. Based on the mineralization alteration sequence rule set, features not directly related to nickel mineralization (such as spectral features of non-mineralization associated alteration) and spatial links (such as links connecting unrelated anchoring features) are identified and eliminated. Core target area indicator features (such as anchoring features at various levels) and effective spatial relationships are retained to form a simplified target area indicator shaping set, improving the accuracy of subsequent target area delineation.

[0163] Step S1410: The simplified target region indicator shaping set is fully fused with all the shaping spatial distribution structures and the bound multi-level target region indicator anchoring feature set, shaping spatial feature association link set, and shaping spatial feature boundary set to form a target region indicator spatial shaping body. The target region indicator spatial shaping body includes the anchoring content and shaping spatial distribution structure of all target region indicator features in the region to be predicted.

[0164] In this embodiment, the shaped spatial distribution structure, multi-level target area indicator anchoring feature set, shaped spatial feature association link set, and shaped spatial feature boundary set of the simplified target area indicator shaping set are fully fused. During the fusion process, the consistency and correlation of various types of data in spatial location are ensured to form a unified target area indicator spatial shaping body. This target area indicator spatial shaping body completely includes all anchoring features and shaped spatial structures related to nickel ore target area indicators within the area to be predicted.

[0165] Step S150: Perform nickel ore target area spatial straightening and delineation processing on the target area indicator space shaping body to form nickel ore target area prediction results.

[0166] In this embodiment, the spatial consolidation and delineation of nickel ore target areas is based on the target area indicator spatial shaping body. Through operations such as boundary processing, spatial range fusion, and feature matching of the shaped spatial distribution structure, the geographic spatial range with nickel ore mineralization potential is finally determined, forming the nickel ore target area prediction result.

[0167] Step S151: Extract each shaped spatial distribution structure in the target area indicator spatial shaping body, extract the characteristic boundary of each shaped spatial distribution structure, obtain the geographic coordinate information corresponding to the characteristic boundary, and form a shaped spatial boundary coordinate extraction set.

[0168] In this embodiment, each shaped spatial distribution structure is extracted from the target area indicator spatial shaping body. By reading the boundary marker information in the shaping space feature boundary set, the geographic coordinate information of the feature boundary of each shaped spatial distribution structure is obtained, including the latitude and longitude coordinates of all points on the boundary, forming a shaping space boundary coordinate extraction set.

[0169] Step S152: Extract the boundary coordinates of the shaping space. Extract the boundary geographic coordinates of the individual shaping spatial distribution structure in the extraction set. Connect the discrete boundary coordinates continuously according to the geographic spatial direction to form a single shaping space geographic range with continuous boundaries, and construct a set of continuous ranges of single shaping space.

[0170] In this embodiment, for a single shaped spatial distribution structure in the shaped space boundary coordinate extraction set, its boundary geographic coordinate points may be discrete. These discrete points are sorted according to geographic spatial direction (e.g., clockwise or counterclockwise), and then linear interpolation or curve fitting methods are used to supplement intermediate points between adjacent coordinate points, making the boundary form a continuous line, thereby determining the geographic extent of a single shaped space. The continuous geographic extents of all single shaped spaces constitute a continuous extent set of single shaped spaces.

[0171] Step S153: Analyze the geographic spatial range of the distribution structure of adjacent shaped spaces in the continuous range of the single shaped space, extract the continuous boundary coordinates of two shaped spaces that are spatially adjacent or spatially overlapping, and perform fusion processing on the boundary coordinates of adjacent or overlapping parts to form a fused continuous geographic boundary.

[0172] In this embodiment, it is achieved through the following sub-steps:

[0173] Step S1531: Extract the geographic spatial range of two adjacent shaped spatial distribution structures from the single-shaped spatial continuous range set, obtain the complete continuous boundary geographic coordinates of the two, determine the boundary coordinate points and spatial orientation of the two spatial ranges, and form a double-shaped spatial boundary coordinate set.

[0174] In this embodiment, within the single-shaping space continuous range set, two geographically adjacent shaping spatial distribution structures are selected, their complete continuous boundary geographic coordinate point sequences are obtained respectively, and their spatial orientation (clockwise or counterclockwise) is recorded to form a double-shaping space boundary coordinate set.

[0175] Step S1532: Compare the geographic coordinate boundaries of the two shaped spaces in the dual-shaped space boundary coordinate set, identify the coordinate segments with adjacent contact or the coordinate regions with spatial overlap, mark the specific geographic coordinates of the adjacent or overlapping parts, and form a dual-shaped space adjacent overlap mark set.

[0176] In this embodiment, the boundary coordinate point sequences of the two shaped spaces are compared, and the shortest distance from each point to the boundary of the other space is calculated. If the distance is less than a set contact threshold, it is determined to be adjacent contact; if the boundary coordinate points of the two spaces intersect or some coordinate points fall within the other space, it is determined to be spatial overlap. All adjacent contact coordinate segments and spatially overlapping coordinate regions are marked, and the specific geographic coordinates of these parts are recorded to form a set of adjacent overlap markers for the dual-shaped spaces.

[0177] Step S1533: For the coordinate segments of adjacent overlapping markers in the double-shaped space that are in contact with each other, extract the pair of coordinate points that are closest to each other on the adjacent boundaries of the two spatial ranges as connection points, and connect the pair of coordinate points with a straight line segment to form the adjacent boundary connection coordinate segment.

[0178] In this embodiment, for adjacent contact coordinate segments, a point is selected on each of the two spatial boundaries to minimize the distance between them, and these two points are used as connection points. A straight line segment connects these two connection points; this straight line segment is the adjacent boundary connection coordinate segment, used to connect two adjacent spatial boundaries.

[0179] Step S1534: For the coordinate regions of the spatially overlapping adjacent overlapping marker set of the double-shaped space, extract all geographic coordinate points of the two spatial ranges within the overlapping region, calculate the density cluster center of the geographic coordinate points, retain the coordinate points belonging to the main density cluster, and form the core coordinate set of the overlapping region.

[0180] In this embodiment, for spatially overlapping coordinate regions, all geographic coordinate points of the two spatial boundaries within these regions are collected. A density clustering algorithm is used to cluster these points, identifying the cluster with the highest density (the primary density cluster). The center coordinates of this cluster are calculated, and all coordinate points within this cluster are retained to form the core coordinate set of the overlapping region, serving as the basis for merging the overlapping boundaries.

[0181] Step S1535: Integrate the core coordinate set of the overlapping area with the boundary coordinate points of the non-overlapping areas of the two spatial ranges, and connect all coordinate points continuously according to the natural direction of the geographic space to construct a continuous boundary coordinate segment after the overlapping area is merged, forming an overlapping boundary fusion coordinate segment.

[0182] In this embodiment, the core coordinate set of the overlapping area is merged with the boundary coordinate points of the two spatial ranges in the non-overlapping area. All coordinate points are sorted according to the natural direction of geographic space (such as clockwise direction). Then, the above points are continuously connected by curve fitting or linear connection to form a fused boundary coordinate segment covering the overlapping area, that is, the overlapping boundary fused coordinate segment.

[0183] Step S1536: Integrate the adjacent boundary connecting coordinate segment or the overlapping boundary fusion coordinate segment with the boundary coordinates of the non-adjacent and non-overlapping areas of the two spatial ranges to form the initial continuous geographic boundary after the fusion of the two shaped spaces, forming the initial boundary set of the dual-shaped space fusion.

[0184] In this embodiment, for adjacent contact cases, the adjacent boundary connecting coordinate segments replace the original adjacent boundary segments of the two spaces and are integrated with the boundary coordinates of the non-adjacent areas; for spatial overlap cases, the overlapping boundary fusion coordinate segments are integrated with the boundary coordinates of the non-overlapping areas. The integrated boundary forms the initial continuous geographic boundary after the fusion of the two shaped spaces, constituting the initial boundary set of the dual-shaped space fusion.

[0185] Step S1537: Extract all boundary coordinate points in the initial boundary set of the dual-shaping space fusion, sort them in a clockwise or counterclockwise direction, and interpolate or resample the sorted coordinate point sequence to make the interval between adjacent coordinate points less than a set threshold, thus forming a regular boundary set of dual-shaping space fusion.

[0186] In this embodiment, the boundary coordinate points in the initial boundary set of the dual-shaping space fusion are sorted in a clockwise or counterclockwise direction, and then the distance between adjacent coordinate points is checked. If the distance is greater than a set threshold, interpolation is performed between the two points to add new coordinate points; if the distance is too small, resampling is performed to reduce the number of coordinate points. After processing, the distance between adjacent coordinate points is less than the set threshold, forming a regular boundary set of the dual-shaping space fusion.

[0187] Step S1538: Compare the fused geographic boundary corresponding to the dual-shaped spatial fusion regularized boundary set with the geographic spatial range of other surrounding shaped spatial distribution structures again. If there are still adjacent or overlapping boundaries, continue to extract the boundary coordinates and perform fusion processing to form a multi-shaped spatial fusion boundary set.

[0188] In this embodiment, the fused geographic boundary is compared with the geographic spatial range of other surrounding shaped spatial distribution structures. If new adjacent or overlapping relationships are found, the fusion process from step S1531 to step S1537 is repeated to fuse more shaped spatial distribution structures together to form a multi-shaped spatial fusion boundary set.

[0189] Step S1539: Extract all boundary coordinate points in the multi-shaping space fusion boundary set, sort them in a clockwise or counterclockwise direction, calculate the included angle formed by three consecutive points, and delete the middle point if the included angle is within the set threshold range; or calculate the angle change of each vertex of the polygon, retain the vertices whose curvature change is greater than the set threshold, and form the core boundary coordinate set of the fusion space.

[0190] In this embodiment, after sorting the boundary coordinate points in the multi-shape space fusion boundary set, the Douglas-Puk algorithm is used to simplify the boundary. Specifically, the included angle formed by three consecutive points is calculated. If the included angle is close to 180 degrees (within a set threshold range), it indicates that the middle point is a redundant point and can be deleted; or the rate of change of angle of each vertex of the polygon is calculated, and vertices with curvature change (rate of change of angle) greater than the set threshold are retained. These vertices are the feature points of the boundary. The simplified coordinate points form the core boundary coordinate set of the fusion space.

[0191] Step S15310: Connect the coordinate set of the core boundary of the fused space continuously according to the geographic spatial direction to form a continuous geographic boundary after fusion, and determine the complete coordinate direction and coverage of the geographic boundary.

[0192] In this embodiment, the coordinate sets of the fused spatial core boundary are continuously connected in a clockwise or counterclockwise direction to form a smooth, continuous geographic boundary. The geographic spatial range covered by this boundary, including maximum and minimum longitude and latitude values, can be determined by the coordinate orientation of this boundary.

[0193] Step S154: Extract the entire geographic space range covered by the fused continuous geographic boundary, take it as the whole mineralization potential space range, extract the multi-level target area indicator anchoring feature set and shaping spatial feature association link set bound within the mineralization potential space range, and form a fused mineralization potential space feature set.

[0194] In this embodiment, the entire geographic space covered by the fused continuous geographic boundary is determined based on its coordinate range. Within the target area indicator spatial shaping body, all bound multi-level target area indicator anchoring feature sets and shaping space feature association link sets within this range are extracted. These features are then associated with the mineralization potential spatial range to form a fused mineralization potential spatial feature set.

[0195] Step S155: Extract the geographic spatial range of independent shaped spatial distribution structures that do not form spatial adjacencies or overlaps in the single shaped spatial continuous range set, retain their original continuous geographic boundaries and the bound target area indicator features, and form an independent mineralization potential spatial feature set.

[0196] In this embodiment, within the continuous range of a single shaped space, independent structures that do not form adjacent or overlapping relationships with other shaped spatial distribution structures are selected. The geographic spatial range, original continuous geographic boundaries, and bound multi-level target area indicator anchoring feature sets and shaped space feature association link sets of these independent structures are extracted to form an independent mineralization potential spatial feature set.

[0197] Step S156: Integrate the fused mineralization potential spatial feature set with the independent mineralization potential spatial feature set across the entire region, and arrange all mineralization potential spatial ranges according to the geographic spatial partitions of the region to be predicted to form a preliminary set of mineralization potential spatial features across the entire region.

[0198] In this embodiment, the fused mineralization potential spatial feature set and the independent mineralization potential spatial feature set are merged, and all mineralization potential spatial ranges are classified and arranged according to the geographic spatial partitions of the region to be predicted (such as sub-regions divided by latitude and longitude grids), ensuring that the mineralization potential spatial ranges in each partition are correctly placed, forming a preliminary set of global mineralization potential space.

[0199] Step S157: Extract the complete geographic boundary coordinate sequence of all mineral potential spatial ranges in the preliminary set of global mineral potential space, use the Douglas-Puk algorithm to thin the boundary coordinate sequence, retain the feature coordinate points within the set tolerance range, and form a normalized mineral potential space boundary set.

[0200] In this embodiment, for each mineralization potential spatial range in the initial set of global mineralization potential space, its complete geographic boundary coordinate sequence is extracted. The Douglas-Pock algorithm is applied, with a tolerance threshold set. It iteratively removes points that have little impact on the boundary shape, retaining key feature coordinate points, thereby thinning the boundary coordinate sequence, reducing the amount of data, and maintaining the basic shape of the boundary. The thinned boundary coordinate sequence forms a regularized mineralization potential spatial boundary set.

[0201] Step S158: Accurately match the geographical boundary of each mineralization potential spatial range in the regularized mineralization potential spatial boundary set with the corresponding multi-level target area indicator anchoring feature set and mineralization alteration combination characterization content to form a one-to-one correspondence between spatial range and feature characterization, and construct a precise matching set of spatial features.

[0202] In this embodiment, for each mineralization potential spatial range within the regularized mineralization potential spatial boundary set, its geographical boundary coordinates are associated with the multi-level target area indicator anchoring feature set and mineralization alteration combination characterization content (such as alteration type, sequential evolution information, etc.) in the target area indicator spatial shaping body. This ensures that each spatial range corresponds to a unique feature characterization content, forming a precise spatial feature matching set.

[0203] Step S159: Extract the geographic coordinate range, core anchoring features, multi-level anchoring features, and mineralization alteration combination evolution relationship of all mineralization potential spatial ranges in the spatial feature precise matching set, and record all content completely region by region to form a global mineralization potential spatial feature record set.

[0204] In this embodiment, the geographic coordinate range (maximum and minimum latitude and longitude), core anchoring features (type, coordinates), multi-level anchoring features (type and distribution of features at each level), and the evolutionary relationship of mineralization alteration combinations (alteration type sequence, evolution direction, etc.) of each mineralization potential spatial range are extracted from the precise matching set of spatial features. The above information is organized and recorded by region to form a global mineralization potential spatial feature record set, which describes in detail the characteristics of each mineralization potential region.

[0205] Step S1510: Summarize the spatial feature record set of the whole region mineralization potential to form the nickel ore target area prediction result. The nickel ore target area prediction result includes all geographic spatial ranges with nickel ore mineralization potential in the area to be predicted and the complete mineralization feature characterization content bound to each geographic spatial range.

[0206] In this embodiment, all information from the spatial feature record set of mineralization potential across the entire region is summarized and sorted and organized according to the size of the mineralization potential or geographical region. The final nickel ore target area prediction result includes the boundary coordinates of all geographic spatial ranges with nickel ore mineralization potential within the area to be predicted, as well as complete mineralization feature representations such as the core anchoring features, multi-level anchoring features, and mineralization alteration combination evolution relationships bound to each range.

[0207] Based on the same inventive concept, please refer to Figure 2 The diagram shows a schematic block diagram of a nickel ore target area prediction system 100 based on hyperspectral remote sensing data analysis provided in this application embodiment. The nickel ore target area prediction system 100 based on hyperspectral remote sensing data analysis may include a communication unit 110, a machine-readable storage medium 120, and a processor 130.

[0208] In this embodiment, the machine-readable storage medium 120 can also be integrated into the processor 130 and can communicate and interact with external systems through the communication unit 110. The machine-readable storage medium 120 stores machine-executable instructions for executing the scheme of this application, and the processor 130 executes the machine-executable instructions stored in the machine-readable storage medium 120 to implement the nickel ore target area prediction method based on hyperspectral remote sensing data analysis provided in the aforementioned method embodiments.

[0209] It should be noted that, in order to simplify the description of the present invention and thus help to understand one or more embodiments of the invention, multiple features may sometimes be grouped into one embodiment, drawing or description thereof in the foregoing description of the embodiments of the present invention.

Claims

1. A method for predicting nickel ore target areas based on hyperspectral remote sensing data analysis, characterized in that, The method includes: The hyperspectral remote sensing data volume of the region to be predicted is obtained. The hyperspectral remote sensing data volume includes the full-band spectral response characteristics and the geospatial distribution characteristics of each geographic detection point in the region to be predicted. The full-band spectral response characteristics are the continuous response characteristics of spectral reflectance and spectral absorption formed by the geographic detection points within the hyperspectral remote sensing detection band. The geospatial distribution characteristics are the geographic coordinates of each geographic detection point and the spatial adjacency and spatial extension relationships between the points. The hyperspectral remote sensing data volume is subjected to alteration feature spatial clustering construction processing to form an alteration feature spatial cluster body. The alteration feature spatial cluster body is a continuous cluster distribution structure formed by various geological alteration types based on geographic spatial distribution characteristics, and the continuous cluster distribution structure is bound to the full-band spectral response characteristics of the corresponding geological alteration type. The alteration feature spatial cluster is subjected to sequential spatial evolution processing of mineralization alteration combination to form mineralization alteration sequential evolution body. The mineralization alteration sequential evolution body is a spatial sequential extension structure formed by geological alteration combination adapted to nickel ore mineralization characteristics in the order of mineralization alteration occurrence. The spatial sequential extension structure is bound to the full-band spectral response feature evolution content of the corresponding alteration combination. The target area indicator feature anchoring and shaping construction process is performed on the mineralization alteration sequence evolution body to form a target area indicator spatial shaping body. The target area indicator spatial shaping body is a shaped spatial distribution structure constructed with the core alteration feature of nickel ore mineralization as the anchoring center, and the shaped spatial distribution structure is bound to multi-level target area indicator alteration features. The target area indicator space shaping body is subjected to nickel ore target area spatial straightening and delineation processing to form nickel ore target area prediction results.

2. The method for predicting nickel ore target areas based on hyperspectral remote sensing data analysis according to claim 1, characterized in that, The step of performing alteration feature space clustering construction processing on the hyperspectral remote sensing data volume to form an alteration feature space cluster volume includes: Extract the full-band spectral response features from the hyperspectral remote sensing data volume, and bind the full-band spectral response features of each geographic detection point to the geographic coordinates of the corresponding geographic detection point point one by one to form a spectral feature geographic point binding set. Each geographic coordinate in the spectral feature geographic point binding set uniquely corresponds to a set of full-band spectral response features. The standard spectral response library of geological alteration types related to nickel ore formation is retrieved, and the standard full-band spectral response features corresponding to various geological alteration types in the standard spectral response library are extracted. All standard full-band spectral response features are then normalized according to the band division rules of hyperspectral remote sensing to form a normalized standard spectral feature set. The single-class standard full-band spectral response features in the normalized standard spectral feature set are compared with the single-group detection full-band spectral response features in the spectral feature geographic location binding set. Spectral fragments that match the standard full-band spectral response features are selected from the detection full-band spectral response features to form a single-location spectral feature matching set. For the spectral feature geographic point binding set, the full-band spectral response features of all geographic detection points are sequentially compared and spectral fragments are screened. All geographic detection points that form a single-point spectral feature matching set are collected. The corresponding geographic detection points are classified according to the matching geological alteration type to form an alteration type geographic point classification set. Extract the geospatial distribution characteristics of all geographic detection points corresponding to a single geological alteration type in the geographic point classification set of the alteration type, analyze the spatial adjacency relationship between each geographic detection point, identify the geographic detection point group with continuous spatial adjacency relationship, and form a single-type alteration continuous point group set. Extract the geographic coordinates of all geographic detection points within a single geographic detection point group in the single-type alteration continuous point group set, and connect the geographic coordinates of each point continuously according to the spatial extension direction of the geographic detection points to form a spatial continuous distribution line of single-type geological alteration, and construct a single-type alteration spatial line set. Extract the full-band spectral response features of all geographic detection points on a single spatially continuous distribution line in the single-type alteration spatial line set, arrange the full-band spectral response features of each point continuously according to the spatial extension direction to form the continuous arrangement content of the spectral features of a single line, and bind it to the corresponding spatially continuous distribution line to form an alteration line spectral binding set. The spectral binding of the alteration lines integrates all spatially continuous distribution lines and their bound spectral features under the same geological alteration type. Combining the spatial positional relationship between the lines, an overall cluster distribution structure of a single type of geological alteration is constructed, forming a single-type alteration cluster structure set. The cluster distribution structures of all geological alteration types in the single-type alteration cluster structure set are arranged according to the actual geographic spatial distribution relationship in the area to be predicted, and the actual spatial location relationship between various types of cluster distribution structures is preserved to form a global alteration cluster arrangement set. The entire distribution structure of all clusters in the global alteration cluster arrangement is fused with the full-band spectral response characteristics and geographic spatial distribution characteristics of the corresponding geological alteration types to form an alteration feature spatial cluster body. The alteration feature spatial cluster body includes the cluster distribution structure of all nickel ore mineralization-related geological alteration types in the area to be predicted and the corresponding bound feature content.

3. The method for predicting nickel ore target areas based on hyperspectral remote sensing data analysis according to claim 1, characterized in that, The step of performing a sequential spatial evolution process on the alteration feature spatial clusters to form a sequential evolution body of mineralized alteration includes: Extract the cluster distribution structure of various geological alterations and their corresponding geographic spatial distribution range in the alteration feature spatial cluster body, and arrange all cluster distribution structures as a whole according to the geographic spatial orientation of the area to be predicted to form an alteration cluster spatial orientation arrangement set; The spatial orientation of the alteration clusters is analyzed to determine the positional connection between different geological alteration cluster distribution structures in geographic space. Combinations of geological alteration cluster distribution structures that are adjacent or directly connected in spatial distribution are identified, forming a set of alteration cluster spatial connection combinations. The sequence rules of geological alteration during nickel ore mineralization are retrieved, and the order of occurrence of various geological alterations and the evolutionary relationship between alterations are extracted from the sequence rules. The core geological alteration types and combination forms corresponding to each stage of nickel ore mineralization are extracted to form a set of mineralization alteration sequence rules. The alteration types and their spatial arrangement order of a single alteration combination in the alteration cluster spatial connection combination set are compared with the main sequence of alteration occurrence defined in the mineralization alteration sequence rule set. If the type and arrangement order of the alteration combination is a continuous subsequence of the main sequence, it is determined to be a match, and all matching alteration combinations are grouped into the initial selection set of mineralization-related alteration combinations. Extract the cluster-shaped distribution structure of various geological alterations within a single mineralization-related alteration combination in the preliminary selection set of the mineralization-related alteration combination, and arrange the cluster-shaped distribution structure sequentially according to the order of alteration occurrence in nickel mineralization to form a single combination alteration cluster-shaped sequential arrangement structure. Extract the geographic spatial coordinates of the cluster distribution structure of geological alteration at each stage in the single-combination alteration cluster sequential arrangement structure, and continuously connect the cluster distribution structures of the preceding alteration and the subsequent alteration according to the spatial adjacency relationship to form a spatial sequential extension structure of the mineralization alteration combination, and construct a single-combination spatial sequential extension set. Extract the full-band spectral response features of geological alteration at each stage in the single-combination spatial sequential extension concentrated mineralization alteration combination spatial sequential extension structure, arrange the full-band spectral response features of each stage continuously according to the alteration sequential evolution direction, form the spectral feature sequential evolution arrangement content, and bind it to the corresponding spatial sequential extension structure. Extract the alteration cluster distribution structure and the bound spectral feature sequential evolution arrangement content corresponding to the core stage of nickel mineralization from the single combination spatial sequential extension set, focus on extracting the feature content of the core stage, form the core mineralization stage alteration feature set, and bind it to the spatial sequential extension structure of the corresponding mineralization alteration combination. The spatial sequential extension structure of all mineralization-related alteration combinations in the preliminary selection set of mineralization-related alteration combinations, the sequential evolution arrangement of the bound spectral features, and the core mineralization stage alteration feature set are arranged as a whole according to the actual geographic spatial distribution of the area to be predicted, forming a global mineralization alteration sequential arrangement set. The spatial sequential extension structure of all mineralization alteration combinations and the various associated features in the global mineralization alteration sequential arrangement are fully integrated to form a mineralization alteration sequential evolution body. The mineralization alteration sequential evolution body contains the dual characteristics of the sequential evolution and spatial distribution of all mineralization-related alteration combinations in the area to be predicted.

4. The method for predicting nickel ore target areas based on hyperspectral remote sensing data analysis according to claim 1, characterized in that, The process of anchoring and shaping the target area indicator features of the sequential evolution body of mineralization alteration to form a target area indicator spatial shaping body includes: Extract the spatial sequential extension structure of each mineralization alteration combination in the mineralization alteration sequential evolution body, identify the alteration cluster distribution structure corresponding to the core stage of nickel mineralization in each mineralization alteration combination, determine the geospatial center coordinates of the alteration cluster distribution structure, and use it as the core anchor point for target area indication to form a core anchor point set. Extract the spatial sequential extension structure of a single mineralization alteration combination in the mineralization alteration sequential evolution body, identify all preceding alteration cluster distribution structures before the core alteration cluster distribution structure corresponding to the core anchoring point, and all subsequent alteration cluster distribution structures after it, extract the geographic spatial range covered by the outer polygon of the alteration cluster distribution structure, and form the initial alteration combination spatial association range. Centered on the core anchoring point, and based on the outer polygon of the initial alteration combination spatial association range, a single combination shaped spatial distribution structure is generated by fitting a preset geometric template corresponding to the typical nickel ore mineralization spatial morphology. Extract the geological alteration cluster distribution structure and full-band spectral response features corresponding to each geometric part in the single-combination shaping spatial distribution structure, and bind various feature contents to specific geometric parts of the shaping space point by point to form a binding set of feature parts of the shaping space; The shaping space feature parts are bound together, and the alteration feature corresponding to the core anchoring point is marked as the first-level anchoring feature. With the core anchoring point as the center, multiple concentric rings are divided according to the preset distance radius. The alteration features located in different rings are marked as multi-level anchoring features in sequence to form a multi-level target area indicator anchoring feature set. Extract the geospatial coordinates of each level of anchoring features in the multi-level target area indicator anchoring feature set, construct spatial association links between the first-level anchoring features and each level of anchoring features, and bind all spatial association links to the corresponding shaped spatial distribution structure to form a set of shaped spatial feature association links; The shaping spatial distribution structure corresponding to all mineralization alteration combinations in the sequential evolution of mineralization alteration, as well as the binding set of shaping spatial feature parts, the multi-level target area indicator anchoring feature set, and the shaping spatial feature association link set, are arranged as a whole according to the actual geographic spatial location of the area to be predicted, forming a global target area indicator shaping arrangement set. Extract the boundary features of each shaped spatial distribution structure in the global target region indicator shaping arrangement set, and combine them with the corresponding multi-level target region indicator anchoring feature set to characterize and mark the boundary of the shaped space so that the boundary marking content is consistent with the distribution range of the anchoring features, thus forming a shaped space characteristic boundary set. Extract all the shaped spatial distribution structures and bound features from the global target area indicator shaping arrangement set, remove features and spatial links that are not directly related to nickel ore formation, retain the core target area indicator features and spatial relationships, and form a simplified target area indicator shaping set. The simplified target region indicator shaping set is fully fused with all the shaping spatial distribution structures and the bound multi-level target region indicator anchoring feature set, shaping spatial feature association link set, and shaping spatial feature boundary set to form a target region indicator spatial shaping body. The target region indicator spatial shaping body includes the anchoring content and shaping spatial distribution structure of all target region indicator features in the region to be predicted.

5. The method for predicting nickel ore target areas based on hyperspectral remote sensing data analysis according to claim 1, characterized in that, The process of performing nickel ore target area spatial realignment and delineation on the target area indicator space shaping body to form nickel ore target area prediction results includes: Extract each shaped spatial distribution structure in the target area indicator spatial shaping body, extract the characteristic boundary of each shaped spatial distribution structure, obtain the geographic coordinate information corresponding to the characteristic boundary, and form a shaped spatial boundary coordinate extraction set; Extract the boundary coordinates of the shaping space boundary. Extract the boundary geographic coordinates of a single shaping spatial distribution structure in the extraction set. Connect the discrete boundary coordinates continuously according to the geographic spatial direction to form a single shaping space geographic range with continuous boundaries. Construct a set of continuous ranges of single shaping space. The geographic spatial range of the distribution structure of adjacent shaped spaces in the continuous range of the single shaped space is analyzed. The continuous boundary coordinates of two shaped spaces that are spatially adjacent or spatially overlapping are extracted. The boundary coordinates of adjacent or overlapping parts are fused to form a fused continuous geographic boundary. Extract the entire geographic space range covered by the fused continuous geographic boundary and take it as the whole mineral potential space range. Extract the multi-level target area indicator anchoring feature set and shaping spatial feature association link set bound within the mineral potential space range to form a fused mineral potential space feature set. Extract the geographic spatial range of independent shaped spatial distribution structures that do not form spatial adjacency or overlap in the single shaped spatial continuous range set, retain their original continuous geographic boundaries and the bound target area indicator features, and form an independent mineralization potential spatial feature set. The fused mineralization potential spatial feature set and the independent mineralization potential spatial feature set are integrated across the entire region. All mineralization potential spatial ranges are arranged according to the geographic spatial partitions of the region to be predicted, forming a preliminary set of mineralization potential spatial features across the entire region. Extract the complete geographic boundary coordinate sequence of all mineral potential spatial ranges in the preliminary set of the global mineral potential space, use the Douglas-Puk algorithm to thin the boundary coordinate sequence, retain the feature coordinate points within the set tolerance range, and form a regularized mineral potential space boundary set. The geographic boundary of each mineral potential spatial range in the regularized mineralization potential spatial boundary set is precisely matched with the corresponding multi-level target area indicator anchoring feature set and mineralization alteration combination characterization content to form a one-to-one correspondence between spatial range and feature characterization, and to construct a precise spatial feature matching set. Extract the geographic coordinate range, core anchoring features, multi-level anchoring features, and mineralization alteration combination evolution relationship of all mineralization potential spatial ranges in the precise matching set of the spatial features; record all content completely region by region to form a global mineralization potential spatial feature record set. The entire set of spatial characteristics of mineralization potential is summarized to form the prediction results of nickel ore target areas. The prediction results of nickel ore target areas include all geographic spatial ranges with nickel mineralization potential in the area to be predicted, as well as the complete mineralization characteristic representation content bound to each geographic spatial range.

6. The method for predicting nickel ore target areas based on hyperspectral remote sensing data analysis according to claim 2, characterized in that, The step involves performing a band-by-band feature comparison between the single-class standard full-band spectral response features in the normalized standard spectral feature set and the single-group detection full-band spectral response features in the spectral feature geographic location binding set, and selecting spectral fragments from the detection full-band spectral response features that have feature matching with the standard full-band spectral response features to form a single-location spectral feature matching set, including: Extract the standard full-band spectral response features of a single type of geological alteration from the regularized standard spectral feature set, and divide it into multiple continuous band units according to the band division rules of hyperspectral remote sensing. Each band unit retains complete spectral reflectance and spectral absorption response features, forming a set of single-type standard spectral band units. Extract the full-band spectral response features of a single geographic detection point in the geographic location binding set of the spectral features, and split it into multiple band units corresponding to the standard spectrum according to the same band division rules to form a single-point detection spectral band unit set, so as to realize the synchronous splitting of band units of the detection spectrum and the standard spectrum. Extract a single standard band unit from the single-class standard spectral band unit set and a single detection band unit corresponding to the single-point detection spectral band unit set; analyze the changing trends of spectral reflectance and spectral absorption within the two band units; compare the characteristic morphological matching of the two to form a single band unit feature comparison result. For the feature comparison results of single-band units with consistent spectral variation trends, the spectral angle of the detected band unit and the corresponding standard band unit in the same band range is calculated. If the spectral angle is less than the set threshold, it is determined to be a feature detail match, and the matching band unit detail confirmation content is formed. The standard band units in the single-class standard spectral band unit set are compared with the corresponding detection band units in the single-point detection spectral band unit set one by one, and all band units that form feature morphology matching and detail matching are collected to form a single-point matching band unit set. Extract the band numbers of all matching band units in the single-point matching band unit set, and connect each matching band unit continuously according to the order of the band numbers to form a continuous spectral segment corresponding to a single geographic detection point and a single type of standard spectrum, thus forming a single-point single-type spectral matching segment. The number of band units contained in the single-point single-class spectral matching segment is counted, and the continuous distribution of the band units is used to confirm that the spectral matching segment is a continuous band distribution pattern, thus forming the validity confirmation content of the single-point single-class spectral matching segment. For the full-band spectral response characteristics of a single geographic detection point, the above-mentioned band-by-band comparison and spectral matching fragment extraction are completed with the standard full-band spectral response characteristics of all geological alteration types in the normalized standard spectral feature set, forming a single-point multi-class spectral matching fragment set; Extract all spectral matching segments with continuous band distribution from the single-point multi-class spectral matching segment set, and bind each spectral matching segment with the corresponding geological alteration type to form a single-point spectral feature matching initial set; The initial set of single-point spectral feature matching is organized to determine the geological alteration type and corresponding continuous spectral matching fragments corresponding to a single geographic detection point. Non-continuous spectral fragments are removed to form a single-point spectral feature matching set.

7. The method for predicting nickel ore target areas based on hyperspectral remote sensing data analysis according to claim 3, characterized in that, The process involves matching individual alteration combinations from the spatially connected alteration clusters with the set of mineralization alteration sequence rules to select alteration combinations that match the mineralization alteration sequence of nickel ore, forming a preliminary selection set of mineralization-related alteration combinations, including: Extract all geological alteration types contained in a single alteration combination within the alteration cluster spatial connection combination set. Based on the geometric center point coordinates of various alteration cluster distribution structures, sort the alteration types along a preset spatial reference direction to form a sequential structure of single-combination alteration spatial distribution. Extract the main sequence of geological alteration corresponding to the core stage of nickel mineralization from the mineralization alteration sequence rule set, determine the order of occurrence and type combination requirements of various geological alterations in the main sequence, and form the main sequence set of mineralization core alteration. The sequential structure of the spatial distribution of single-combination alteration is matched with the main sequential set of the mineralization core alteration. The overall matching of the combination and arrangement order of geological alteration types in the two is analyzed to form the overall sequential matching result of single combination. For the overall sequential matching results of the single combination that form the overall matching, the spatial extension direction of the cluster distribution structure of each geological alteration type in the alteration spatial distribution sequence of the single combination is extracted, and the matching of the spatial extension direction with the alteration evolution direction of the main sequence set of the mineralization core alteration is analyzed to form the spatial evolution direction matching result of the single combination. Extract the full-band spectral response characteristics of various geological alteration types in the sequential structure of the spatial distribution of single-combination alteration, analyze the evolutionary correlation between each characteristic, compare the matching of the evolutionary correlation with the spectral evolution characteristics of the concentrated alteration of the sequential rule of mineralization alteration, and form the matching result of single-combination spectral evolution characteristics. For single-combination overall sequential matching results that do not form an overall match, extract the geological alteration type fragments in the alteration spatial distribution sequence of the single combination that have a local match with the main sequence set of alteration of the mineralization core, analyze whether the geological alteration type fragment is an alteration combination of the core stage of nickel mineralization, and form a single-combination local sequential matching result. For the local sequential matching results of a single combination that forms a local match, extract the other geological alteration types in the alteration combination, analyze whether the corresponding alteration type is an associated alteration type of nickel mineralization alteration, and whether it has a spatial distribution and evolution relationship with the core alteration type, and form a single combination associated alteration matching result; Alteration assemblages that form overall matching and those that form local matching and have associated alteration matching are grouped together to form a candidate set of mineralization-related alteration assemblages. Extract the cluster-shaped distribution structure corresponding to each alteration combination in the candidate set of mineralization-related alteration combinations, calculate the ratio of the total length of the continuous distribution lines in each cluster-shaped distribution structure to the area enclosed by the lines. If the ratio is lower than a set threshold, the alteration cluster-shaped distribution structure is determined to be a combination with poor spatial extensibility and is removed to form a mineralization-related alteration combination screening set. The mineralization-related alteration combination screening set is used as the initial selection set of mineralization-related alteration combinations to determine all geological alteration combinations in the area to be predicted that match the mineralization characteristics of nickel ore.

8. The method for predicting nickel ore target areas based on hyperspectral remote sensing data analysis according to claim 4, characterized in that, The process involves extracting the spatial sequential extension structure of a single mineralized alteration assemblage within the mineralized alteration sequential evolution body, identifying all preceding alteration cluster distribution structures before the core alteration cluster distribution structure corresponding to the core anchoring point, and all subsequent alteration cluster distribution structures. The geographic spatial range covered by the circumscribed polygon of the alteration cluster distribution structure is then extracted to form the initial spatial association range of the alteration assemblages, including: Extract the spatial sequential extension structure of a single mineralization alteration combination in the mineralization alteration sequential evolution body, extract the core alteration cluster distribution structure corresponding to the core anchoring point, obtain the complete geospatial distribution coordinates of the cluster distribution structure, and form a core alteration cluster spatial feature set. Based on the alteration sequence defined in the spatial sequential extension structure, a geological alteration cluster distribution structure that precedes the core alteration cluster distribution structure is extracted from the sequence as a first-level preceding alteration cluster distribution structure. Its geographic spatial distribution coordinates are obtained to form a first-level preceding alteration spatial feature set. Based on the alteration sequence defined in the spatial sequential extension structure, a geological alteration cluster distribution structure located after the core alteration cluster distribution structure is extracted from the sequence as a first-level subsequent alteration cluster distribution structure. Its geographic spatial distribution coordinates are obtained to form a first-level subsequent alteration spatial feature set. Continue to extract the multi-level pre-sequence geological alteration cluster distribution structure corresponding to the first-level pre-sequence alteration spatial feature set along the pre-sequence evolution direction until the alteration type that is not directly related to nickel mineralization is extracted. Record the geographic spatial distribution coordinates of each level of pre-sequence alteration and the spatial distance from the core alteration to form a multi-level pre-sequence alteration spatial feature set. Continue to extract the multi-level geological alteration cluster distribution structure corresponding to the first-level alteration spatial feature set along the subsequent evolution direction until the alteration type that is not directly related to nickel mineralization is extracted. Record the geographic spatial distribution coordinates of each level of alteration and the spatial distance from the core alteration to form a multi-level alteration spatial feature set. By integrating the geographic spatial distribution coordinates of all alteration cluster distribution structures in the core alteration cluster spatial feature set, the multi-level preceding alteration spatial feature set, and the multi-level subsequent alteration spatial feature set, the overall spatial distribution outline of all alteration cluster distribution structures is drawn to form a global spatial outline map of alteration combination. Taking the geographic spatial center of the core alteration cluster spatial feature set as the origin, and combining the spatial distance between the pre- and post-alteration alteration cluster distribution structures at each level and the origin, the minimum geographic spatial range that can include all mineralization-related alteration cluster distribution structures is delineated, forming an initial outline of the alteration combination spatial association range. Extract the geographic coordinate boundary of the initial outline of the spatial association range of the alteration combination, and expand the boundary of the initial outline according to the spatial extension edge of all mineralization-related alteration cluster distribution structures to ensure that the whole of all mineralization-related alteration cluster distribution structures is within the expanded range, thus forming the extended outline of the spatial association range of the alteration combination. Extract all alteration cluster distribution structures within the initial outline of the alteration combination spatial association range. If an unrelated alteration cluster distribution structure that does not conform to the definition type of the mineralization alteration sequence rule set is identified, remove the vertex closest to the geometric center of the unrelated alteration cluster distribution structure from the vertices of the circumscribed polygon, and regenerate the convex polygon boundary to form the alteration combination spatial association range adjustment outline. Extract the geographic spatial distribution range and corresponding geographic coordinate boundary of the adjusted contour of the alteration combination spatial association range to form the initial alteration combination spatial association range, and determine all mineralization-related alteration cluster distribution structures included in the geographic spatial distribution range.

9. The method for predicting nickel ore target areas based on hyperspectral remote sensing data analysis according to claim 5, characterized in that, The analysis of the geographic spatial range of the continuous range of the single-shaped space focuses on the distribution structure of adjacent shaped spaces, extracts the continuous boundary coordinates of two spatially adjacent or overlapping shaped spaces, and performs fusion processing on the boundary coordinates of adjacent or overlapping parts to form a fused continuous geographic boundary, including: Extract the geographic spatial range of two adjacent shaped spatial distribution structures from the single-shaped spatial continuous range set, obtain the complete continuous boundary geographic coordinates of the two, determine the boundary coordinate points and spatial orientation of the two spatial ranges, and form a double-shaped spatial boundary coordinate set; By comparing the geographic coordinate boundaries of the two shaped spaces in the dual-shaped space boundary coordinate set, we can identify coordinate segments with adjacent contact or coordinate regions with spatial overlap, mark the specific geographic coordinates of adjacent or overlapping parts, and form a dual-shaped space adjacent overlap mark set. For the coordinate segments of adjacent overlapping markers in the double-shaping space that are in contact with each other at the boundary, extract the pair of coordinate points that are closest to each other on the adjacent boundaries of the two spatial ranges as connection points, and connect the pair of coordinate points with a straight line segment to form the adjacent boundary connecting coordinate segment; For the coordinate regions of the spatially overlapping marker set of the dual-shaping space, extract all geographic coordinate points within the overlapping region of the two spatial ranges, calculate the density cluster center of the geographic coordinate points, retain the coordinate points belonging to the main density cluster, and form the core coordinate set of the overlapping region. The core coordinate set of the overlapping area is integrated with the boundary coordinate points of the non-overlapping areas of the two spatial ranges. All coordinate points are continuously connected according to the natural direction of the geographic space to construct a continuous boundary coordinate segment after the overlapping area is merged, forming an overlapping boundary fusion coordinate segment. The adjacent boundary connecting coordinate segment or the overlapping boundary fusion coordinate segment are integrated with the boundary coordinates of the non-adjacent and non-overlapping areas of the two spatial ranges to form the initial continuous geographic boundary after the fusion of the two shaped spaces, thus forming the initial boundary set of the dual-shaped space fusion. Extract all boundary coordinate points from the initial boundary set of the dual-shaping space fusion, sort them in a clockwise or counterclockwise direction, and interpolate or resample the sorted coordinate point sequence to make the interval between adjacent coordinate points less than a set threshold, thus forming a regular boundary set of dual-shaping space fusion. The fused geographic boundary corresponding to the dual-shaped spatial fusion regularized boundary set is compared again with the geographic spatial range of other surrounding shaped spatial distribution structures. If there are still adjacent or overlapping boundaries, the boundary coordinates are extracted and fused to form a multi-shaped spatial fusion boundary set. Extract all boundary coordinate points from the multi-shape space fusion boundary set, sort them in a clockwise or counterclockwise direction, calculate the included angle formed by three consecutive points, and delete the middle point if the included angle is within a set threshold range; or calculate the angle change of each vertex of the polygon, retain vertices with curvature changes greater than the set threshold, and form a fusion space core boundary coordinate set. Connect the fusion space core boundary coordinate set continuously according to the geographic spatial direction to form a fusion continuous geographic boundary, and determine the complete coordinate direction and coverage of the geographic boundary.

10. A nickel ore target area prediction system based on hyperspectral remote sensing data analysis, characterized in that, include: processor; A machine-readable storage medium for storing machine-executable instructions of the processor; The processor is configured to execute the nickel ore target area prediction method based on hyperspectral remote sensing data analysis as described in any one of claims 1 to 9 by executing the machine-executable instructions.

Images