An AI-based method and system for building databases of attributeless geological maps

Abstract

This application relates to the field of geological map construction technology, and in particular to an AI-based method and system for building a database of attribute-free geological maps. The method includes: acquiring a remote sensing dataset of the area to be constructed; analyzing the remote sensing dataset to determine a set of geological feature information; analyzing the geological feature information set and performing combined identification analysis on unit geological features to determine a set of geological boundary information; acquiring a sample dataset; based on the geological boundary information set, performing structural morphology analysis on the sample dataset to determine a set of multi-layer structure information; analyzing the geological boundary information set and the multi-layer structure information set to determine a unified spatial mapping coordinate set for each multi-layer structure and its corresponding geological boundary; and constructing and outputting an attribute-free geological map based on the unified spatial mapping coordinate set, the geological boundary information set, and the multi-layer structure information set. This application enables attribute-free geological maps to clearly present the spatial relationships of various geological structures and boundaries, providing important scientific data references for geological research.

Description

Technical Field

[0001] This application relates to the field of geological map construction technology, and in particular to an AI-based method and system for building a database of attribute-free geological maps. Background Technology

[0002] attribute-free geology Figure 1 Generally, a geological map is a type of map that does not contain detailed geological attribute information, such as stratum names, lithological descriptions, and structural features. Such geological maps usually only provide basic geological boundary and morphological information and are often used in situations where it is necessary to show tectonic relationships or spatial distribution patterns without being affected by specific geological details.

[0003] Existing attributeless geological map construction techniques are insufficient for accurately analyzing the boundaries and morphology of complex spatially distributed multi-layered geological structures. This results in the omission of important and minute geological features in the constructed attributeless geological maps, thus limiting geological research based on attributeless geological maps lacking detail. Summary of the Invention

[0004] This application provides an AI-based method and system for building databases of attributeless geological maps to solve the aforementioned technical problems.

[0005] Firstly, this application provides an AI-based method for building a database of attribute-free geological maps, the method comprising:

[0006] A remote sensing dataset of the region to be constructed is acquired, and the dataset is analyzed to determine a set of geological feature information. The geological feature information set is then analyzed, and combined identification analysis is performed on individual geological features to determine a set of geological boundary information. A sample dataset is acquired, and based on the geological boundary information set, structural morphology and orientation analysis is performed on the sample dataset to determine a set of multi-layer structure information. The geological boundary information set and the multi-layer structure information set are analyzed to determine a unified spatial mapping coordinate set for each multi-layer structure and its corresponding geological boundary. Based on the unified spatial mapping coordinate set, an attribute-free geological map is constructed and output according to the geological boundary information set and the multi-layer structure information set.

[0007] By analyzing the geological features in the remote sensing dataset and performing combined analysis of individual geological features, the system achieves dynamic combination and analysis of complex geological structures. This enables the construction of a convenient geological information set, effectively identifying details in geological boundary morphology. Users can capture subtle geological changes and boundaries in attribute-free geological maps, improving their accuracy. Through comprehensive analysis of sample datasets and geological boundary information sets, a multi-layered information set is constructed, allowing attribute-free geological maps to express the relationships and orientations between different geological layers. Furthermore, by establishing a unified spatial mapping coordinate set, the consistency of different geological information in space is ensured. This allows attribute-free geological maps to clearly present the spatial relationships of various geological structures and boundaries, providing important scientific data references for users to conduct research based on attribute-free geological maps.

[0008] Optionally, the remote sensing dataset includes spectral data, basic topographic elevation information, and regional remote sensing image information. Analyzing the remote sensing dataset to determine the geological feature information set includes:

[0009] Based on the preset unit analysis area information, the area to be constructed is divided into several data extraction areas; the spectral data is analyzed to determine the vegetation index, water index, and snow and ice index of each analysis area; the basic topographic elevation information is analyzed to determine the slope information and aspect information of each data extraction area; based on the vegetation index, water index, snow and ice index, slope information, and aspect information corresponding to each data extraction area, several local feature models are determined; based on the several local feature models and the regional remote sensing image information, a geological feature information set is determined.

[0010] Using the above technical solution, the area to be constructed is divided into several data extraction areas. Spectral data is analyzed to determine the vegetation index, water index, and snow index corresponding to the data analysis area. By analyzing topographic elevation information, the slope information corresponding to the data extraction area is determined. Starting from the above multi-dimensional data, a local feature model reflecting the digital geological features of the data extraction area is constructed. Then, based on the local feature model, the remote sensing image information of the area is analyzed to obtain a set of geological feature information reflecting the geological boundary morphology details of the area to be constructed, providing a scientific digital information basis for the subsequent construction of attribute-free geological maps.

[0011] Optionally, the step of determining several local feature models based on the vegetation index, water index, snow and ice index, slope information, and aspect information corresponding to each data extraction region includes:

[0012] According to preset combination rules, several data extraction regions are combined to obtain several feature analysis regions. Based on several vegetation indices, several water indices, several snow and ice indices, several slope information, and several aspect information corresponding to each feature analysis region, the mean and standard deviation of the vegetation index, the mean and standard deviation of the water index, the mean and standard deviation of the snow and ice index, the mean and standard deviation of the slope, and the mean and standard deviation of the aspect are determined for each feature analysis region. The local feature model of each feature analysis region is constructed, specifically as follows:

[0013]

[0014] Where M(i,j) is the local feature model corresponding to the feature analysis region at coordinate (i,j), μ NV Let σ be the mean of the vegetation index. NV Let μ be the standard deviation of the vegetation index. NW σ is the mean of the water index. NW Let μ be the standard deviation of the water index. NS Let σ be the mean of the ice and snow index. NS Let μ be the standard deviation of the ice and snow index. SP Let σ be the average slope. SP Let v be the standard deviation of the slope. AS Let σ be the mean value of the slope aspect. AS Let be the standard deviation of the slope aspect.

[0015] The above technical solution merges several data extraction areas into several feature analysis areas according to preset combination rules, better capturing the changes in geological features within the areas. Based on the mean and standard deviation of plant index, water index, snow and ice index, slope information, and aspect information within the feature analysis areas, a local feature model is constructed through mathematical analysis. By scaling the data, the adaptability of the local feature model under different datasets and complex geological conditions is improved, enabling the local feature model to comprehensively reflect the geological features within the feature analysis areas.

[0016] Optionally, determining the geological feature information set based on several local feature models and the regional remote sensing image information includes:

[0017] Analyze the remote sensing image information of the region to determine the gradient direction pointing value of each feature analysis region; based on the gradient direction pointing value, and according to several local feature models, determine the edge intensity index of each feature analysis region, specifically using the following formula:

[0018]

[0019] Where E(i,j) is the edge intensity index of the feature analysis region at coordinate (i,j), and M k (i,j) represents the k-th feature value in the local feature model corresponding to the current feature analysis region, θ represents the gradient direction, and w k The preset influence weight of the feature type corresponding to the k-th feature value is used; the geological feature information set is constructed based on the edge intensity index corresponding to each feature analysis region.

[0020] Through the above technical solution, using mathematical analysis methods, based on gradient direction pointing value and local feature model, and through designed mathematical formulas, the edge intensity index reflecting whether the feature analysis area is located on the geological boundary is scientifically and accurately quantified, providing scientific quantitative data for subsequent determination of the geological boundary details in the area to be constructed.

[0021] Optionally, the analysis of the basic terrain elevation information to determine the slope and aspect information of each data extraction area includes:

[0022] Based on the basic topographic elevation information, the abscissa, ordinate, and elevation values ​​of the center point of each data extraction area are determined; based on the abscissa, ordinate, and elevation values, the slope direction value of each data extraction area is determined, and the slope direction value is used as the slope information, specifically as follows:

[0023]

[0024] Where SP is the slope direction value of the current data extraction area, x is the horizontal coordinate value, y is the vertical coordinate value, and z is the elevation value; based on the horizontal coordinate value, the vertical coordinate value, and the elevation value, the slope direction value of each data extraction area is determined, and the slope direction value is used as the slope information, specifically as follows:

[0025]

[0026] Where AS is the slope direction value of the current data extraction area, x is the horizontal coordinate value, y is the vertical coordinate value, and z is the elevation value.

[0027] Using the above technical solution and mathematical analysis, based on the x-coordinate, y-coordinate, and elevation values ​​of the center point of the data extraction area, mathematical formulas are used to scientifically quantify the slope direction value (reflecting the magnitude of the slope) and the slope aspect value (reflecting the direction of the slope) of the data extraction area. This clarifies the local slope and aspect of each data extraction area, providing accurate data basis for the construction of local feature models.

[0028] Optionally, the analysis of the geological feature information set, performing combined identification and analysis on unit geological features to determine the geological boundary information set, includes:

[0029] Based on the edge intensity index corresponding to each feature analysis region within the geological feature information set, the mean edge intensity and standard deviation of edge intensity are determined; based on the edge intensity index, the mean edge intensity, and the standard deviation of edge intensity, the boundary pointing value of each feature analysis region is determined, specifically as follows:

[0030]

[0031] Wherein, B(i,j) is the boundary pointing value of the feature analysis region at coordinate (i,j), E(i,j) is the edge strength index of the feature analysis region at coordinate (i,j), mean(E) is the mean edge strength, σ(E) is the standard deviation of the edge strength, and α is a preset adjustment coefficient; based on the distribution position of each feature analysis region in the region to be constructed, according to the boundary pointing value of each feature analysis region, the geological boundary features of adjacent feature analysis regions are combined to determine the geological boundary information set.

[0032] Using the above technical solution and mathematical analysis, based on the edge strength index, the mean edge strength, and the standard deviation of edge strength, the boundary pointing value corresponding to each feature analysis area is automatically quantified through mathematical expressions. This quickly and accurately marks all feature analysis areas located on geological boundaries, providing data basis for clearly presenting the morphological details of geological boundaries in subsequent attribute-free geological maps.

[0033] Optionally, the sample dataset includes several sample extraction coordinates, several sample cross-sectional image data, and several sample interception depths. The step of performing structural morphology orientation analysis on the sample dataset based on the geological boundary information set to determine a multi-layered structural information set includes:

[0034] Analyze the cross-sectional image data of each sample to determine several structural layer boundaries; analyze the geological boundary information set based on the extracted coordinates and the cut-off depth of each sample to determine several structural orientation information; generate several multi-layer structural texture orientation images based on the several structural orientation information and the several structural layer boundaries; construct the multi-layer structural information set based on the extracted coordinates of the several samples and the several multi-layer structural texture orientation images.

[0035] By analyzing sample cross-sectional data, structural stratification boundaries reflecting the boundaries between different geological layers are identified and extracted. Based on the sample extraction coordinates and sample interception depth, the geological boundary information is intelligently analyzed to deduce the structural orientation information reflecting the orientation of different geological layers. By comprehensively analyzing the structural orientation information and structural stratification boundaries, a multi-layer structural texture orientation image is generated. Based on the sample extraction coordinates and the multi-layer structural texture orientation image, a multi-layer structural information set is constructed, enabling the multi-layer structural information set to truly reflect the location and orientation of different geological layers underground in the area to be constructed. This makes the subsequent attribute-free geological map more comprehensive in spatial dimension.

[0036] Optionally, the step of analyzing the geological boundary information set and the multi-layer structure information set to determine a unified spatial mapping coordinate set for each multi-layer structure information and its corresponding geological boundary includes:

[0037] Based on the distribution location of each feature analysis region within the region to be constructed, a unified spatial coordinate system is established for all combined geological boundary features within the geological boundary information set to determine the geological boundary feature coordinate set; based on the geological boundary feature coordinate set, a multi-layer structure coordinate set is determined according to the relative position of each sample extraction coordinate and the corresponding feature analysis region; and the unified spatial mapping coordinate set is determined based on the geological boundary feature coordinate set and the multi-layer structure coordinate set.

[0038] Through the above technical solution, a unified spatial coordinate system is established for the data in the geological boundary information set representing the surface geological boundary of the area to be constructed and the multi-layer structure information set representing the orientation of the underground multi-layer geological structure in the area to be constructed. This results in the geological boundary feature coordinate set and the multi-layer structure coordinate set, effectively avoiding the misalignment problem between data in different information sets. It ensures that the changes in surface geological features and underground geological features are referenced by a unified spatial coordinate system, thereby improving the application value of attribute-free geological maps.

[0039] Optionally, the step of constructing and outputting an attribute-free geological map based on the unified spatial mapping coordinate set, the geological boundary information set, and the multi-layer structure information set includes:

[0040] Based on the unified spatial mapping coordinate set, the geological boundary information set and the multi-layer structure information set are structured to determine boundary vector data and structure vector data. Based on the preset geographic information drawing model, geological boundary morphology maps and several multi-layer geological structure orientation maps are drawn according to the boundary vector data and the structure vector data. Based on the unified spatial mapping coordinate set, the several multi-layer geological structure orientation maps are indexed and mapped on the geological boundary morphology maps to construct and output the attribute-free geological map.

[0041] The above technical solution involves structuring the geological boundary information set and the multi-layer structure information set to obtain boundary vector data and structure vector data, ensuring data format consistency. Then, by using a preset geographic information drawing model, geological boundary morphology maps and several multi-layer geological structure orientation maps are drawn under a unified spatial coordinate system. The positions of the multi-layer geological structure orientation maps are indexed and mapped, enabling users to clearly understand the relative positions of the multi-layer geological structure orientation maps representing the orientation of underground geological structures and the geological boundary morphology maps, accurately conveying geological information, and avoiding mutual interference between geological boundaries and multi-layer geological structures.

[0042] Secondly, this application provides an AI-based system for building databases of attribute-free geological maps, the system comprising:

[0043] The feature analysis module is used to acquire the remote sensing dataset of the area to be constructed, analyze the remote sensing dataset, and determine the set of geological feature information.

[0044] The boundary analysis module is used to analyze the geological feature information set, perform combined identification and analysis of unit geological features, and determine the geological boundary information set.

[0045] The structural analysis module is used to acquire sample datasets, perform structural morphology orientation analysis on the sample datasets based on the geological boundary information set, and determine the multi-layer structure information set.

[0046] The coordinate mapping module is used to analyze the geological boundary information set and the multi-layer structure information set to determine a unified spatial mapping coordinate set for each multi-layer structure information and its corresponding geological boundary.

[0047] An output module is constructed to construct and output an attribute-free geological map based on the unified spatial mapping coordinate set, the geological boundary information set, and the multi-layer structure information set. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a schematic diagram of an application scenario provided in an embodiment of this application;

[0050] Figure 2 A flowchart illustrating an AI-based method for building a database of attributeless geological maps, as provided in an embodiment of this application;

[0051] Figure 3This is a schematic diagram of the structure of an AI-based attributeless geological map database system provided in one embodiment of this application. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0053] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0054] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0055] Existing attributeless geological map construction techniques are insufficient for accurately analyzing the boundaries and morphology of complex spatially distributed multi-layered geological structures. This results in the omission of important and minute geological features in the constructed attributeless geological maps, thus limiting geological research based on attributeless geological maps lacking detail.

[0056] Based on this, this application provides an AI-based method and system for building a database of attribute-free geological maps. Through comprehensive analysis of sample datasets and geological boundary information sets, a multi-layered information set is constructed, enabling the attribute-free geological map to express the relationships and orientations between different geological layers. Furthermore, by establishing a unified spatial mapping coordinate set, the spatial consistency of different geological information is ensured, allowing the attribute-free geological map to clearly present the spatial relationships of various geological structures and boundaries. This provides important scientific data references for users conducting research based on attribute-free geological maps. Specific implementation methods can be found in the following embodiments.

[0057] Figure 1 This application provides an illustration of an application scenario. In the process of constructing a non-attribute geological map, the method provided in this application is used to accurately present the surface geological boundary morphology and underground geological structure orientation within the area to be constructed, thereby helping to improve the efficiency and effectiveness of geological research based on non-attribute maps.

[0058] Specifically, the method provided in this application is applied to any server that communicates with remote sensing data collection equipment. Through this server, it acquires remote sensing datasets provided by the equipment and sample datasets provided by geological exploration personnel. It analyzes the geological feature information in the remote sensing datasets, identifies and analyzes combinations of unit geological features, and achieves dynamic combination and analysis of complex geological structures to construct a convenient geological information set. This effectively identifies details in geological boundary morphology, enabling users to capture subtle geological changes and boundaries in attribute-free geological maps, thus improving the accuracy of attribute-free geological maps. Through comprehensive analysis of sample datasets and geological boundary information sets, a multi-layered information set is constructed, allowing attribute-free geological maps to express the relationships and orientations between different geological layers. By establishing a unified spatial mapping coordinate set, it ensures the spatial consistency of different geological information, enabling attribute-free geological maps to clearly present the spatial relationships of various geological structures and boundaries. This provides important scientific data references for users to conduct research based on attribute-free geological maps. Specific implementation methods can be found in the following embodiments.

[0059] Figure 2 This is a flowchart illustrating an AI-based method for building a database of attribute-free geological maps, provided as an embodiment of this application. The method of this embodiment can be applied to servers in the above scenarios. Figure 2 As shown, the method includes:

[0060] S201. Obtain the remote sensing dataset of the area to be constructed, analyze the remote sensing dataset, and determine the geological feature information set.

[0061] The area to be constructed can be the corresponding area for which a geological map needs to be constructed.

[0062] Remote sensing datasets can be data obtained by analyzing the area to be constructed using remote sensing data collection devices, such as satellites, and using remote sensing analysis techniques. These datasets reflect the geological conditions within the area to be constructed.

[0063] A geological feature information set can be a collection of information that reflects the geological features of the area to be constructed.

[0064] Specifically, remote sensing technology has the advantages of wide coverage and rapid acquisition. It can provide a large amount of basic data for the creation of attribute-free geological maps. Compared with simple ground exploration, remote sensing technology is more cost-effective. By using remote sensing data collection equipment to analyze the geology of the area to be constructed, a remote sensing dataset is collected, and a preliminary extensive survey of the geology of the area to be constructed is achieved. Then, mathematical analysis methods are used to extract information that reflects the geological characteristics of the area to be constructed, so as to construct a geological feature information set. This provides a scientific and important data basis for further analysis of the geological boundary morphology details of the area to be constructed.

[0065] S202. Analyze the geological feature information set, perform combined identification and analysis of unit geological features, and determine the geological boundary information set.

[0066] Unit geological features can be characteristic information that reflects the local geological boundary morphology, obtained through analysis.

[0067] Combination identification analysis can be an analytical process that combines and identifies different local geological boundary morphologies.

[0068] A geological boundary information set can be a collection of information reflecting the geological boundary morphology within the area to be constructed.

[0069] Specifically, since the geological feature information set contains all the geological feature information of the area to be constructed, directly analyzing the geological feature information set as a whole would be computationally burdensome and could easily lead to the loss of geological feature details in the analysis results. By performing mathematical analysis on the geological feature information corresponding to different local areas within the geological feature information set in the area to be constructed, the geological boundary morphological features of different local areas can be accurately extracted. The geological boundary morphological features of different local areas can then be combined and identified to construct a geological boundary information set. This provides the geological boundary feature details of the area to be constructed for the construction of attribute-free geological maps, thereby improving the precision and accuracy of attribute-free geological maps.

[0070] S203. Obtain the sample dataset. Based on the geological boundary information set, perform structural morphology analysis on the sample dataset to determine the multi-layer structure information set.

[0071] The sample dataset can be core sample data that reflects the geological structural characteristics of the area to be constructed.

[0072] Structural orientation analysis can be a process of deducing and analyzing the orientation of geological structures reflected in core samples based on geological boundary information sets.

[0073] A multi-layered structure information set can be a collection of information reflecting the orientation of the underground multi-layered geological structure within the area to be constructed.

[0074] Specifically, the geological boundary morphology can reflect the apparent changes in geological features within the area to be constructed. In order to make the attributeless geological map more comprehensively reflect the geological features within the area to be constructed, based on the geological boundary, it is necessary to analyze the local subsurface multi-layer geological structure changes reflected in the core samples within the area to be constructed. Combined with the trend of surface geological feature changes reflected by the geological boundary, the trend of subsurface multi-layer geological structure changes within the area to be constructed can be deduced. This enables scientific analysis of the geological features of the area to be constructed from different spatial dimensions, making the attributeless geological map obtained later more comprehensive and helping relevant geological researchers to conduct more scientific geological research.

[0075] S204. Analyze the geological boundary information set and the multi-layer structure information set to determine the unified spatial mapping coordinate set of each multi-layer structure information and the corresponding geological boundary.

[0076] A unified spatial mapping coordinate set can be the set of coordinates corresponding to each piece of information within a unified spatial coordinate system, including geological boundary information set and multi-layer structure information set.

[0077] Specifically, after obtaining the geological boundary information set reflecting changes in surface geological features within the area to be constructed and the multi-layered structure information set reflecting changes in subsurface geological features, to avoid misalignment between data in different information sets, a unified spatial coordinate system is established and mapped for the information in the geological boundary information set and the multi-layered structure information set. This results in a unified spatial mapping coordinate system, ensuring that the changes in surface and subsurface geological features are referenced to a unified spatial coordinate system. This allows professionals conducting research using attribute-free geological maps to accurately combine surface and subsurface geological features within the area to be constructed, enabling scientific research on the overall geological features of the area. Ensuring the spatial consistency and comparability of surface and subsurface geological features is crucial for arriving at correct geological research conclusions.

[0078] S205. Based on a unified spatial mapping coordinate set, construct and output an attribute-free geological map according to the geological boundary information set and the multi-layer structure information set.

[0079] A geological map without attributes may be a geological map that does not contain detailed geological attribute information and focuses on displaying the geological structure and morphology.

[0080] Specifically, based on a unified spatial mapping coordinate set, geological analysis techniques, such as GIS (Geographic Information System), are used to visualize geological boundary information sets and multi-layer structure information sets based on spatial coordinates. Different geological features and boundaries are drawn using different line types, colors, or symbols to construct attribute-free geological maps. These attribute-free geological maps are then provided to relevant users through human-computer interaction devices, such as high-definition displays, for their research and reference.

[0081] This embodiment analyzes geological feature information in remote sensing datasets, identifies and analyzes combinations of unit geological features, and achieves dynamic combination and analysis of complex geological structures to construct a convenient geological information set. This effectively identifies details in geological boundary morphology, enabling users to capture subtle geological changes and boundaries in attribute-free geological maps, thus improving the accuracy of attribute-free geological maps. Through comprehensive analysis of sample datasets and geological boundary information sets, a multi-layered information set is constructed, allowing attribute-free geological maps to express the relationships and orientations between different geological layers. By establishing a unified spatial mapping coordinate set, the spatial consistency of different geological information is ensured, enabling attribute-free geological maps to clearly present the spatial relationships of various geological structures and boundaries. This provides important scientific data references for users to conduct research based on attribute-free geological maps.

[0082] In some embodiments, the region to be constructed is divided into several data extraction regions based on the preset unit analysis region information; spectral data is analyzed to determine the vegetation index, water index, and snow and ice index of each analysis region; basic topographic elevation information is analyzed to determine the slope information and aspect information of each data extraction region; several local feature models are determined based on the vegetation index, water index, snow and ice index, slope information, and aspect information corresponding to each data extraction region; and a set of geological feature information is determined based on several local feature models and regional remote sensing image information.

[0083] The remote sensing dataset includes spectral data, basic topographic elevation information, and regional remote sensing image information.

[0084] Spectral data can be obtained by remote sensing data collection equipment, such as artificial satellites, which transmit electromagnetic waves to the ground of the area to be constructed.

[0085] Basic topographic elevation information can be information reflecting the topographic elevation of the area to be constructed, which can be collected through remote sensing technology.

[0086] Regional remote sensing image information can be surface image information of the area to be constructed, collected through remote sensing technology.

[0087] The preset unit analysis area information can be preset information that includes the size and number of unit areas to be divided into regions to be constructed. The preset unit analysis area information can be set according to the actual situation of the region to be constructed.

[0088] The data extraction area can be the smallest unit area for geological data analysis and extraction, obtained by dividing the area to be constructed based on the preset unit analysis area information.

[0089] The vegetation index can be a quantitative data reflecting the vegetation cover in the data extraction area. The vegetation index can be calculated using the following formula (1):

[0090]

[0091] Wherein, NV is the vegetation index, N is the near-infrared index in the spectral data, and R is the red band index in the spectral data.

[0092] The water index can be a quantitative data reflecting the distribution of water bodies within the data extraction area. The water index can be calculated using the following formula (2):

[0093]

[0094] Wherein, NW is the water index, G is the green band index in the spectral data, and N is the near-infrared band index in the spectral data.

[0095] The snow and ice index can be used as quantitative data to reflect the snow and ice coverage in the data extraction area. Since snow and ice reflect strongly in the green light band, the snow and ice index can be calculated by the following formula (3):

[0096]

[0097] Wherein, ND is the snow and ice index, G is the green band index in the spectral data, and S is the shortwave infrared band index in the spectral data.

[0098] Slope information can reflect the changes in terrain slope within the data extraction area.

[0099] Slope aspect information can be information that reflects the changes in topographic slope aspect within the data extraction area.

[0100] A local feature model can be a data model that includes local geological features within a unit data extraction area.

[0101] Specifically, based on the preset unit analysis area information, the area to be constructed is divided into several data extraction areas, which helps to dynamically adjust the analysis granularity and achieve a more detailed analysis of geological features within the acceptable computational pressure. During the analysis of geological features, large areas of surface cover will cover the actual geological morphology, which will directly affect the accuracy of the geological feature analysis. However, the characteristics of surface cover can also reflect changes in geological morphology. For example, the vegetation cover density in fault areas will change significantly, and the continuity of water cover will also change significantly. Therefore, by comprehensively considering the vegetation cover, water distribution, and vegetation cover in the data extraction area, the surface geological features can be fully reflected. The vegetation cover, water distribution, and vegetation cover can be quantified by mathematical analysis based on spectral data, that is, the vegetation index, water index, and snow index are calculated by formulas 1-3 respectively.

[0102] Furthermore, the influence of topography cannot be ignored in the analysis of surface geological morphology. The shaping of geological features is affected by topography. For example, steep slopes with large gradients are prone to erosion, while steep slopes with small gradients are prone to deposition. By analyzing and extracting the slope and aspect information reflected by the topography in the data extraction area, the influence of slope and aspect information on geological morphology can be reflected. Then, by integrating vegetation index, water index, snow and ice index, slope information, and aspect information, a local feature model corresponding to each data extraction area can be constructed through mathematical analysis to express the geological boundary characteristics of the data extraction area. Based on the local feature model, mathematical analysis is further performed on the regional remote sensing images to achieve accurate positioning of geological boundary morphology and construct a set of geological feature information that can reflect the details of geological boundary morphology in the area to be constructed.

[0103] The method provided in this embodiment divides the area to be constructed into several data extraction areas, analyzes spectral data, and determines the vegetation index, water index, and snow index corresponding to the data analysis area. By analyzing topographic elevation information, the slope information corresponding to the data extraction area is determined. Based on the above multi-dimensional data, a local feature model reflecting the digital geological features of the data extraction area is constructed. Then, based on the local feature model, the remote sensing image information of the area is analyzed to obtain a set of geological feature information reflecting the geological boundary morphology details of the area to be constructed, providing a scientific digital information basis for the subsequent construction of attribute-free geological maps.

[0104] In some embodiments, several data extraction regions are combined according to preset combination rules to obtain several feature analysis regions; based on several vegetation indices, several water indices, several snow and ice indices, several slope information and several aspect information corresponding to each feature analysis region, the mean and standard deviation of vegetation indices, the mean and standard deviation of water indices, the mean and standard deviation of snow and ice indices, the mean and standard deviation of slope, and the mean and standard deviation of aspect are determined respectively for each feature analysis region; a local feature model for each feature analysis region is constructed, specifically as the following expression (4):

[0105]

[0106] Where M(i,j) is the local feature model of the feature analysis region corresponding to coordinate (i,j), and μ NV σ is the mean of the vegetation index. NV μ represents the standard deviation of the vegetation index. NW σ is the mean of the water index. NW The standard deviation of the water index is μ. NS The average value of the snow and ice index is σ. NS Let μ be the standard deviation of the snow and ice index.SP σ is the average slope. SP μ represents the standard deviation of the slope. AS The slope mean, σ AS This represents the standard deviation of slope aspect.

[0107] The preset combination rule can be a pre-defined rule for combining several data extraction regions. The preset combination rule can be set to 3×3, or it can be set as needed.

[0108] The feature analysis region can be an analysis region that is easy to extract geological boundary features by combining several data extraction regions.

[0109] The mean and standard deviation mentioned above can both be understood as the mathematically defined average and standard deviation of corresponding parameters within several data extraction regions contained in the feature analysis region.

[0110] Specifically, in the data extraction stage, the area to be constructed is divided into several smaller data extraction areas, which effectively ensures the accuracy of the data. When constructing the local feature model, in order to better capture the changes of geological features in the area, the smaller data extraction areas need to be merged into several larger feature analysis areas according to the preset combination rules. Since local details are very important in complex terrain or geological structure, the local mean and standard deviation can capture the small variations and anomalies of geological features and help identify the dynamic changes of potential geological features. At the same time, normalization of different features helps to reflect the performance of different features at similar scales. Therefore, based on the mean and standard deviation of plant index, water index, snow index, slope information and aspect information in the feature analysis area, the corresponding local feature model is constructed through expression (4) to fully reflect the geological features in the feature analysis area.

[0111] The method provided in this embodiment merges several data extraction areas into several feature analysis areas according to preset combination rules, thereby better capturing the changes in geological features within the areas. Based on the mean and standard deviation of plant index, water index, snow and ice index, slope information, and aspect information within the feature analysis areas, a local feature model is constructed through mathematical analysis. By scaling the data, the adaptability of the local feature model under different datasets and complex geological conditions is improved, enabling the local feature model to comprehensively reflect the geological features within the feature analysis areas.

[0112] In some embodiments, remote sensing image information of the region is analyzed to determine the gradient direction pointing value of each feature analysis region; based on the gradient direction pointing value, the edge intensity index of each feature analysis region is determined according to several local feature models, specifically as follows (5):

[0113]

[0114] Where E(i,j) is the edge intensity index of the feature analysis region at coordinate (i,j), and M k (i,j) represents the k-th feature value in the local feature model corresponding to the current feature analysis region, θ is the gradient direction, and w k The preset influence weights for the feature type corresponding to the k-th feature value are used; a geological feature information set is constructed based on the edge intensity index corresponding to each feature analysis area.

[0115] The gradient direction can be the direction of gradient change within the image region occupied by each feature analysis area in a remote sensing image. The gradient direction value is an angle between 0 and 360 degrees. It can be obtained by performing convolution operations, such as using the Sobel operator to calculate the horizontal and vertical gradients within the feature analysis region, and then calculating the gradient using the arctangent function.

[0116] The edge intensity index can be a quantitative index that reflects the probability that the current feature analysis area is on a geological boundary. The larger the edge intensity index, the greater the probability that the current feature analysis area is on a geological boundary.

[0117] The preset influence weight can be a weight value that expresses the magnitude of the influence of different features on the edge intensity index. The preset influence weight can be set by expert experience or obtained by fitting analysis of authoritative geological research data.

[0118] Specifically, through formula (5) The degree of change of each feature value in the local feature model along the corresponding image gradient direction is evaluated, and each feature is evaluated using a preset influence weight w. k By adjusting the contribution of different features to the edge pre-degree, the overall edge intensity index corresponding to the current feature analysis region is finally calculated. This edge intensity index represents the comprehensive rate of change of the feature values ​​of multiple local feature models in a given feature analysis region along the image gradient direction. In this way, different local features are combined to identify and quantify the abrupt change properties of the local image, i.e., the edge. In reality, this is the geological boundary in the remote sensing image information, thus reflecting the probability of each feature analysis region being located on the geological boundary.

[0119] The method provided in this embodiment utilizes mathematical analysis techniques, based on gradient direction values ​​and local feature models, to scientifically and accurately quantify the edge intensity index that reflects whether the feature analysis area is located on a geological boundary using a designed mathematical formula. This provides scientific quantitative data for subsequently determining the geological boundary details within the area to be constructed.

[0120] In some embodiments, based on the basic topographic elevation information, the abscissa, ordinate, and elevation values ​​of the center point of each data extraction area are determined; based on the abscissa, ordinate, and elevation values, the slope direction value of each data extraction area is determined, and the slope direction value is used as slope information, specifically as shown in the following formula (6):

[0121]

[0122] Where SP is the slope direction value of the current data extraction area, x is the horizontal coordinate value, y is the vertical coordinate value, and z is the elevation value; based on the horizontal coordinate value, vertical coordinate value and elevation value, the slope direction value of each data extraction area is determined, and the slope direction value is used as the slope information, specifically as the following formula (7);

[0123]

[0124] Where AS is the slope direction value of the current data extraction area, x is the horizontal coordinate value, y is the vertical coordinate value, and z is the elevation value.

[0125] The elevation value can be the numerical elevation value corresponding to the center point of the data extraction area in the basic terrain elevation information.

[0126] The slope indicator value can be a value that reflects the slope of the data extraction area in the area to be constructed. The larger the slope indicator value, the greater the terrain slope of the data extraction area.

[0127] The slope direction value can be a direction value that reflects the slope direction of the data extraction area in the area to be constructed. The slope direction value is an angle value used to indicate the slope angle.

[0128] Specifically, through formula (6) and The slope is described by describing the rate of change of the terrain in the horizontal and vertical directions respectively. Based on the slope calculation formula, the slope intensity is represented by the sum of the squares of the two. At the same time, the ratio of the rate of change of the terrain in the horizontal and vertical directions is processed by the arctangent function through formula (7) to obtain the slope direction angle, i.e., the slope aspect. Through the above two formulas, the slope information and slope aspect information of the data extraction area are scientifically quantified.

[0129] The method provided in this embodiment utilizes mathematical analysis to scientifically quantify the slope direction value (reflecting the magnitude of the slope) and the slope aspect value (reflecting the direction of the slope) based on the x-coordinate, y-coordinate, and elevation values ​​of the center point of the data extraction area using mathematical formulas. This clarifies the local slope and aspect corresponding to each data extraction area, providing accurate data basis for the construction of local feature models.

[0130] In some embodiments, the mean edge strength and standard deviation of edge strength are determined based on the corresponding edge strength index of each feature analysis region within the geological feature information set; the boundary pointing value of each feature analysis region is determined based on the edge strength index, mean edge strength, and standard deviation of edge strength, specifically as shown in the following expression (8):

[0131]

[0132] Where B(i,j) is the boundary pointing value of the feature analysis region at coordinate (i,j), E(i,j) is the edge strength index of the feature analysis region at coordinate (i,j), mean(E) is the mean edge strength, σ(E) is the standard deviation of edge strength, and α is a preset adjustment coefficient; based on the distribution location of each feature analysis region in the region to be constructed, the geological boundary features of adjacent feature analysis regions are combined according to the boundary pointing value of each feature analysis region to determine the geological boundary information set.

[0133] The mean edge strength can be the average edge strength index calculated by combining the edge strength indices of all feature analysis regions within the region to be constructed.

[0134] The standard deviation of edge strength can be the standard deviation of the edge strength index calculated using the standard deviation quantification formula based on the mean edge strength.

[0135] The preset adjustment coefficient can be a value used to adjust the influence of the standard deviation of edge strength on the determination of boundary orientation. The preset adjustment coefficient can be obtained through fitting analysis of authoritative geological data.

[0136] Boundary pointing value can be a quantitative value used to determine whether the feature analysis area is on a geological boundary. The boundary pointing value is a binary value. If the boundary pointing value is 1, it means that the current feature analysis area is on a geological boundary. If the boundary pointing value is 0, it means that the current feature analysis area is not on a geological boundary.

[0137] Specifically, when the edge intensity index within the feature analysis area satisfies E(i,j)>mean(E)+α·σ(E) in expression (8), the boundary pointing value of the current feature analysis area is set to 1. The specific process is as follows: when the edge intensity index within the feature analysis area exceeds mean(E)+α·σ(E) (the influence of standard deviation is adjusted by a preset adjustment coefficient), it is considered that the edge intensity index exceeds the typical intensity range. At this time, the boundary pointing value of the feature analysis area is set to 1. The above process uses the mean and standard deviation to determine the geological boundary reflected by the "abrupt change". If the edge intensity index does not meet the above conditions, the boundary pointing value of the feature analysis area is set to 0.

[0138] The method provided in this embodiment utilizes mathematical analysis techniques to automatically quantify the boundary pointing value corresponding to each feature analysis region based on the edge strength index, the mean edge strength, and the standard deviation of edge strength. This allows for the rapid and accurate marking of all feature analysis regions located on geological boundaries, providing data support for clearly presenting the morphological details of geological boundaries in subsequent attribute-free geological maps.

[0139] In some embodiments, the cross-sectional image data of each sample is analyzed to determine several structural layer boundaries; based on the extracted coordinates and the interception depth of each sample, the geological boundary information set is analyzed to determine several structural orientation information; based on the several structural orientation information and several structural layer boundaries, several multi-layer structural texture orientation images are generated; based on the extracted coordinates of several samples and several multi-layer structural texture orientation images, a multi-layer structural information set is constructed.

[0140] The sample dataset includes several sample extraction coordinates, several sample cross-sectional image data, and several sample cut-out depths.

[0141] The sample extraction coordinates can be the latitude and longitude coordinates corresponding to the current core sample collection point. The sample extraction coordinates can be obtained from the records of geological exploration personnel when collecting core samples.

[0142] Sample cross-sectional image data can be the image data corresponding to the longitudinal section of the core sample. Sample cross-sectional image data can be obtained by geological exploration personnel using professional image acquisition equipment to analyze the longitudinal section of the core sample.

[0143] The sample interception depth can be the depth at which the core sample is collected, and this depth can be obtained from the records of geological exploration personnel when collecting core samples.

[0144] The structural stratification boundary can be the dividing line between different geological layers reflected in the cross section of a core sample.

[0145] Structural orientation information can be derived by combining geological boundary information sets to obtain the orientation information of different geological layers.

[0146] Multi-layered texture orientation images can be generated image information used to express the orientation of underground multi-layered geological structures.

[0147] Specifically, due to sedimentation, there are boundaries between different geological layers. By analyzing the evolution interface image data using edge detection algorithms in image analysis technology, several structural layer boundaries are determined. Then, using AI technology, such as a convolutional neural network trained on a large amount of geological data, the orientation of different underground geological layers is deduced based on the geological boundary morphology changes near the sample extraction coordinates and the sample extraction depth. Combined with several structural layer boundaries, a two-dimensional image describing the orientation of different geological layers is generated, namely a multi-layer structural texture orientation image. Then, using the sample extraction coordinates as an index and the corresponding multi-layer structural texture orientation image as content, a multi-layer structural information set is constructed.

[0148] The method provided in this embodiment analyzes sample cross-sectional data, identifies and extracts structural stratification boundaries that reflect the boundaries between different geological layers, and intelligently analyzes geological boundary information based on sample extraction coordinates and sample interception depth to deduce structural orientation information reflecting the orientation of different geological layers. By comprehensively analyzing the structural orientation information and structural stratification boundaries, a multi-layer structural texture orientation image is generated. Based on the sample extraction coordinates and the multi-layer structural texture orientation image, a multi-layer structural information set is constructed, enabling the multi-layer structural information set to truly reflect the location and orientation of different geological layers underground in the area to be constructed. This makes the subsequently obtained attribute-free geological map more comprehensive in spatial dimension.

[0149] In some embodiments, based on the distribution location of each feature analysis region within the region to be constructed, a unified spatial coordinate system is established for all combined geological boundary features within the geological boundary information set to determine the geological boundary feature coordinate set; based on the geological boundary feature coordinate set, a multi-layer structure coordinate set is determined according to the relative position of the extracted coordinates of each sample with the corresponding feature analysis region; and a unified spatial mapping coordinate set is determined based on the geological boundary feature coordinate set and the multi-layer structure coordinate set.

[0150] A geological boundary feature coordinate set can be a collection containing the coordinate information of all geological boundaries in a unified spatial coordinate system.

[0151] A multi-layer structure coordinate set can be a collection containing the coordinate information of all multi-layer structure texture images in a unified spatial coordinate system.

[0152] Specifically, the unification of the coordinate system ensures that each geological feature has a precise location on the map, facilitating the generation of accurate and reliable attribute-free geological maps. This has direct application value for subsequent geological research based on attribute-free geological maps. By integrating the latitude and longitude positions corresponding to each feature analysis area in reality, GIS technology is used to establish a unified spatial coordinate system for the geological boundary features of the geological boundary information set, thereby constructing a geological boundary feature coordinate set. On this basis, according to the relative position of the coordinates extracted from each sample within the corresponding feature analysis area, the coordinate information of each multi-layer structure texture direction image in the unified spatial coordinate system is generated, thereby constructing a multi-layer structure coordinate set. By combining the geological boundary feature coordinate set and the multi-layer structure coordinate set, a unified spatial mapping coordinate set is constructed.

[0153] The method provided in this embodiment establishes unified spatial coordinates for the data in the geological boundary information set representing the surface geological boundary of the area to be constructed and the multi-layer structure information set representing the orientation of the underground multi-layer geological structure within the area to be constructed. This results in the geological boundary feature coordinate set and the multi-layer structure coordinate set, effectively avoiding the misalignment problem between data in different information sets. It ensures that changes in surface geological features and underground geological features are referenced to a unified spatial coordinate system, thereby improving the application value of attribute-free geological maps.

[0154] In some embodiments, based on a unified spatial mapping coordinate set, the geological boundary information set and the multi-layer structure information set are structured to determine the boundary vector data and structure vector data. Based on a preset geographic information drawing model, a geological boundary morphology map and several multi-layer geological structure orientation maps are drawn according to the boundary vector data and structure vector data. Based on the unified spatial mapping coordinate set, the several multi-layer geological structure orientation maps are indexed and mapped on the geological boundary morphology map to construct and output an attribute-free geological map.

[0155] Boundary vector data can be vector data formatted data corresponding to information within a geological boundary information set.

[0156] Structural vector data can be vector data formatted to correspond to information within a multi-layered structural information set.

[0157] The preset geographic information mapping model can be a technical model used to generate geological maps based on massive parameters, such as GIS.

[0158] A geological boundary morphology map can be a non-attribute two-dimensional geological image used to describe the different surface geological boundary morphologies within the area to be constructed.

[0159] A multi-layered geological structure trend map can be a non-attribute two-dimensional geological image that describes the trend of different underground geological structures in the area to be constructed.

[0160] Specifically, using vector data format conversion tools, such as GDAL (Geospatial Data Abstraction Library), the geological boundary information set and multi-layer structure information set are structured and converted into corresponding vector data formats, such as GeoJSON format boundary vector data and structure vector data. Then, using a preset geographic information drawing model, such as GIS, geological boundary morphology maps and several multi-layer geological structure orientation maps are drawn using different lines and colors under a unified spatial coordinate system based on the boundary vector data and structure vector data. Based on the relative positions of the surface geological boundaries and underground multi-layer structures within a unified spatial mapping coordinate set, the multi-layer geological structure orientation maps are indexed and mapped, thereby realizing the construction and output of attribute-free geological maps.

[0161] The method provided in this embodiment performs structured processing on the geological boundary information set and the multi-layer structure information set to obtain boundary vector data and structure vector data, ensuring data format consistency. Then, through a preset geographic information drawing model, a geological boundary morphology map and several multi-layer geological structure orientation maps are drawn under a unified spatial coordinate system. The positions of the multi-layer geological structure orientation maps are indexed and mapped, enabling users to clearly know the relative positions of the multi-layer geological structure orientation maps representing the orientation of underground geological structures and the geological boundary morphology maps, accurately conveying geological information, while avoiding mutual interference between geological boundaries and multi-layer geological structures.

[0162] Figure 3 This application provides a schematic diagram of the structure of an AI-based attributeless geological map database system, as shown in one embodiment. Figure 3 As shown, the AI-based attributeless geological map database system 300 of this embodiment includes: a feature analysis module 301, a boundary analysis module 302, a structure analysis module 303, a coordinate mapping module 304, and a construction output module 304.

[0163] Feature analysis module 301 is used to acquire the remote sensing dataset of the area to be constructed, analyze the remote sensing dataset, and determine the set of geological feature information.

[0164] Boundary analysis module 302 is used to analyze the geological feature information set, perform combined identification and analysis of unit geological features, and determine the geological boundary information set;

[0165] The structural analysis module 303 is used to acquire a sample dataset, perform structural morphology orientation analysis on the sample dataset based on the geological boundary information set, and determine the multi-layer structure information set.

[0166] The coordinate mapping module 304 is used to analyze the geological boundary information set and the multi-layer structure information set, and determine the unified spatial mapping coordinate set of each multi-layer structure information and the corresponding geological boundary.

[0167] The output module 305 is used to construct and output an attribute-free geological map based on the unified spatial mapping coordinate set, the geological boundary information set, and the multi-layer structure information set.

[0168] Optionally, the feature analysis module 301 is specifically used for: dividing the region to be constructed into several data extraction regions according to the preset unit analysis region information; analyzing the spectral data to determine the vegetation index, water index, and snow index of each analysis region; analyzing the basic topographic elevation information to determine the slope information and aspect information of each data extraction region; determining several local feature models according to the vegetation index, water index, snow index, slope information, and aspect information corresponding to each data extraction region; and determining a geological feature information set according to the several local feature models and the regional remote sensing image information.

[0169] Optionally, when the feature analysis module 301 determines several local feature models based on the vegetation index, water index, snow and ice index, slope information, and aspect information corresponding to each data extraction region, it is specifically used to: combine several data extraction regions according to preset combination rules to obtain several feature analysis regions; determine the mean and standard deviation of the vegetation index, the mean and standard deviation of the water index, the mean and standard deviation of the snow and ice index, the mean and standard deviation of the slope, and the mean and standard deviation of the aspect for each feature analysis region based on the several vegetation indices, water indices, snow and ice indices, slope information, and aspect information corresponding to each feature analysis region; and construct the local feature model for each feature analysis region, specifically as follows:

[0170]

[0171] Where M(i,j) is the local feature model corresponding to the feature analysis region at coordinate (i,j), μ NV Let σ be the mean of the vegetation index. NV Let μ be the standard deviation of the vegetation index. NW σ is the mean of the water index. NW Let μ be the standard deviation of the water index. NS Let σ be the mean of the ice and snow index. NS Let μ be the standard deviation of the ice and snow index. SP Let σ be the average slope. SP Let μ be the standard deviation of the slope. AS Let σ be the mean value of the slope aspect. AS Let be the standard deviation of the slope aspect.

[0172] Optionally, when determining the geological feature information set based on the plurality of local feature models and the regional remote sensing image information, the feature analysis module 301 is specifically used to: analyze the regional remote sensing image information to determine the gradient direction pointing value of each feature analysis region; and based on the gradient direction pointing value, determine the edge intensity index of each feature analysis region according to the plurality of local feature models, specifically using the following formula:

[0173]

[0174] Where E(i,j) is the edge intensity index of the feature analysis region at coordinate (i,j), and M k (i,j) represents the k-th feature value in the local feature model corresponding to the current feature analysis region, θ represents the gradient direction, and w k The preset influence weight of the feature type corresponding to the k-th feature value is used; the geological feature information set is constructed based on the edge intensity index corresponding to each feature analysis region.

[0175] Optionally, when analyzing the basic terrain elevation information and determining the slope and aspect information of each data extraction area, the feature analysis module 301 is specifically used to: determine the abscissa, ordinate, and elevation values ​​of the center point of each data extraction area based on the basic terrain elevation information; determine the slope direction value of each data extraction area based on the abscissa, ordinate, and elevation values, and use the slope direction value as the slope information, specifically using the following formula:

[0176]

[0177] Where SP is the slope direction value of the current data extraction area, x is the horizontal coordinate value, y is the vertical coordinate value, and z is the elevation value; based on the horizontal coordinate value, the vertical coordinate value, and the elevation value, the slope direction value of each data extraction area is determined, and the slope direction value is used as the slope information, specifically as follows:

[0178]

[0179] Where AS is the slope direction value of the current data extraction area, x is the horizontal coordinate value, y is the vertical coordinate value, and z is the elevation value.

[0180] Optionally, the boundary analysis module 302 is specifically used to: determine the mean edge strength and standard deviation of edge strength based on the corresponding edge strength index of each feature analysis region within the geological feature information set; and determine the boundary pointing value of each feature analysis region based on the edge strength index, the mean edge strength, and the standard deviation of edge strength, specifically as follows:

[0181]

[0182] Wherein, B(i,j) is the boundary pointing value of the feature analysis region at coordinate (i,j), E(i,j) is the edge strength index of the feature analysis region at coordinate (i,j), mean(E) is the mean edge strength, σ(E) is the standard deviation of the edge strength, and α is a preset adjustment coefficient; based on the distribution position of each feature analysis region in the region to be constructed, according to the boundary pointing value of each feature analysis region, the geological boundary features of adjacent feature analysis regions are combined to determine the geological boundary information set.

[0183] Optionally, the structural analysis module 303 is specifically used for: analyzing the cross-sectional image data of each sample to determine several structural layer boundaries; analyzing the geological boundary information set based on the extracted coordinates and the cut-off depth of each sample to determine several structural orientation information; generating several multi-layer structural texture orientation images based on the several structural orientation information and the several structural layer boundaries; and constructing the multi-layer structural information set based on the extracted coordinates of the several samples and the several multi-layer structural texture orientation images.

[0184] Optionally, the coordinate mapping module 304 is specifically used for: establishing a unified spatial coordinate system for all combined geological boundary features within the geological boundary information set according to the distribution position of each feature analysis region within the region to be constructed, and determining a geological boundary feature coordinate set; determining a multi-layer structure coordinate set based on the geological boundary feature coordinate set and the relative position of each sample extraction coordinate with the corresponding feature analysis region; and determining the unified spatial mapping coordinate set based on the geological boundary feature coordinate set and the multi-layer structure coordinate set.

[0185] Optionally, the output construction module 305 is specifically used for: performing structured processing on the geological boundary information set and the multi-layer structure information set based on the unified spatial mapping coordinate set to determine boundary vector data and structure vector data; drawing a geological boundary morphology map and several multi-layer geological structure orientation maps based on the boundary vector data and the structure vector data according to a preset geographic information drawing model; and indexing and mapping several multi-layer geological structure orientation maps on the geological boundary morphology map according to the unified spatial mapping coordinate set to construct and output the attribute-free geological map.

[0186] The system in this embodiment can be used to execute the methods of any of the above embodiments, and its implementation principle and technical effect are similar, so they will not be described again here.

Claims

1. A method for constructing a database of attribute-free geological maps based on AI, characterized in that, include: Obtain the remote sensing dataset of the area to be constructed, analyze the remote sensing dataset, and determine the set of geological feature information; Analyze the geological feature information set, perform combined identification and analysis on the unit geological features, and determine the geological boundary information set; Obtain a sample dataset, and based on the geological boundary information set, perform structural morphology orientation analysis on the sample dataset to determine the multi-layer structure information set; Analyze the geological boundary information set and the multi-layer structure information set to determine the unified spatial mapping coordinate set of each multi-layer structure information and the corresponding geological boundary; Based on the unified spatial mapping coordinate set, and according to the geological boundary information set and the multi-layer structure information set, an attribute-free geological map is constructed and output. The sample dataset includes several sample extraction coordinates, several sample cross-sectional image data, and several sample cut-off depths. Based on the geological boundary information set, the structural morphology orientation analysis of the sample dataset is performed to determine a multi-layer structural information set, including: Analyze the cross-sectional image data of each sample to determine several structural layer boundaries; Based on the extracted coordinates and the cut-off depth of each sample, the geological boundary information set is analyzed to determine several structural orientation information. Based on the structural orientation information and the structural layer boundaries, generate several multi-layer structural texture orientation images; Based on the coordinates extracted from the samples and the multi-layer structure texture direction images, the multi-layer structure information set is constructed. The remote sensing dataset includes spectral data, basic topographic elevation information, and regional remote sensing image information. Analyzing the remote sensing dataset to determine the geological feature information set includes: Based on the preset unit analysis area information, the area to be constructed is divided into several data extraction areas; Analyze the spectral data to determine the vegetation index, water index, and snow and ice index for each analysis area; Analyze the basic topographic elevation information to determine the slope and aspect information of each data extraction area; Based on the vegetation index, water index, snow and ice index, slope information, and aspect information corresponding to each data extraction area, several local feature models are determined; Based on several local feature models and remote sensing image information of the region, a set of geological feature information is determined.

2. The method according to claim 1, characterized in that, The method involves determining several local feature models based on the vegetation index, water index, snow and ice index, slope information, and aspect information corresponding to each data extraction region, including: According to preset combination rules, several data extraction regions are combined to obtain several feature analysis regions. Based on the vegetation indices, water indices, snow and ice indices, slope information, and aspect information corresponding to each feature analysis area, the mean and standard deviation of the vegetation indices, the mean and standard deviation of the water indices, the mean and standard deviation of the snow and ice indices, the mean and standard deviation of the slope, and the mean and standard deviation of the aspect are determined for each feature analysis area. The local feature model for each feature analysis region is constructed as follows: ； in, coordinates The local feature model corresponding to the feature analysis region. The mean of the vegetation index is... The standard deviation of the vegetation index is given. The average value of the water index. The standard deviation of the water index is given. The average value of the ice and snow index. The standard deviation of the ice and snow index is given. The average slope is... Let the slope standard deviation be... The average value of the slope aspect. Let be the standard deviation of the slope aspect.

3. The method according to claim 2, characterized in that, The step of determining a set of geological feature information based on several local feature models and regional remote sensing image information includes: Analyze the remote sensing image information of the region to determine the gradient direction pointing value of each feature analysis region; Based on the gradient direction value, and according to several local feature models, the edge strength index of each feature analysis region is determined, specifically by the following formula: ； in, coordinates The edge strength index of the feature analysis region at the location. The current feature analysis region corresponds to the first feature in the local feature model. 1 eigenvalue, The gradient direction is the value. For the first Each feature value corresponds to a preset influence weight for the feature type; The geological feature information set is constructed based on the edge intensity index corresponding to each feature analysis region.

4. The method according to claim 1, characterized in that, The analysis of the basic terrain elevation information, determining the slope and aspect information of each data extraction area, includes: Based on the aforementioned basic terrain elevation information, determine the x-coordinate, y-coordinate, and elevation values ​​of the center point of each data extraction area; Based on the horizontal coordinate value, the vertical coordinate value, and the elevation value, the slope direction value of each data extraction area is determined, and the slope direction value is used as the slope information, specifically as follows: ； in, The slope direction value for the currently extracted data area. The x-coordinate value is... The ordinate value is... The elevation value; Based on the horizontal coordinate value, the vertical coordinate value, and the elevation value, the slope direction value of each data extraction area is determined, and the slope direction value is used as the slope information, specifically as follows: ； in, The slope direction value for the currently extracted data area. The x-coordinate value is... The ordinate value is... The elevation value is given.

5. The method according to claim 3, characterized in that, The analysis of the geological feature information set, the combined identification and analysis of unit geological features, and the determination of the geological boundary information set include: Based on the edge strength index corresponding to each feature analysis region within the geological feature information set, determine the mean edge strength and the standard deviation of edge strength; Based on the edge strength index, the mean edge strength, and the standard deviation of the edge strength, the boundary pointing value of each feature analysis region is determined, specifically as follows: ； in, coordinates The boundary pointing value of the feature analysis region at that location. coordinates The edge strength index of the feature analysis region at the location. The mean edge strength, The standard deviation of the edge strength is... The preset adjustment coefficient; Based on the distribution location of each of the feature analysis regions within the region to be constructed, and according to the boundary pointing value of each of the feature analysis regions, the geological boundary features of adjacent feature analysis regions are combined to determine the geological boundary information set.

6. The method according to claim 5, characterized in that, The analysis of the geological boundary information set and the multi-layer structure information set, determining the unified spatial mapping coordinate set of each multi-layer structure information and the corresponding geological boundary, includes: Based on the distribution location of each feature analysis region within the region to be constructed, a unified spatial coordinate system is established for all combined geological boundary features within the geological boundary information set to determine the geological boundary feature coordinate set. Based on the geological boundary feature coordinate set, the multi-layer structure coordinate set is determined according to the relative position of the extracted coordinates of each sample and the corresponding feature analysis area; The unified spatial mapping coordinate set is determined based on the geological boundary feature coordinate set and the multi-layer structure coordinate set.

7. The method according to claim 6, characterized in that, The process of constructing and outputting an attribute-free geological map based on the unified spatial mapping coordinate set, the geological boundary information set, and the multi-layer structure information set includes: Based on the unified spatial mapping coordinate set, the geological boundary information set and the multi-layer structure information set are subjected to structured processing to determine the boundary vector data and structure vector data; Based on the preset geographic information drawing model, a geological boundary morphology map and several multi-layer geological structure orientation maps are drawn according to the boundary vector data and the structure vector data. Based on the unified spatial mapping coordinate set, the several multi-layer geological structure orientation maps are indexed and mapped on the geological boundary morphology map to construct and output the attribute-free geological map.

8. An AI-based system for building a database of attribute-free geological maps, characterized in that, Applied to performing the method as described in any one of claims 1-7, comprising: The feature analysis module is used to acquire the remote sensing dataset of the area to be constructed, analyze the remote sensing dataset, and determine the set of geological feature information. The boundary analysis module is used to analyze the geological feature information set, perform combined identification and analysis of unit geological features, and determine the geological boundary information set. The structural analysis module is used to acquire sample datasets, perform structural morphology orientation analysis on the sample datasets based on the geological boundary information set, and determine the multi-layer structure information set. The coordinate mapping module is used to analyze the geological boundary information set and the multi-layer structure information set to determine a unified spatial mapping coordinate set for each multi-layer structure information and its corresponding geological boundary. An output module is constructed to construct and output an attribute-free geological map based on the unified spatial mapping coordinate set, the geological boundary information set, and the multi-layer structure information set.

Images