Natural resource right-confirming intelligent plotting method based on three-dimensional working base map

Abstract

The invention belongs to the technical field of electric digital data processing, and particularly relates to a natural resource right-confirming intelligent plotting method based on a three-dimensional working base map, which comprises the following steps of: 1, carrying out natural resource detection and identification on the three-dimensional working base map, and dividing the three-dimensional working base map into a plurality of different natural resource areas; 2, in each natural resource area, dividing the natural resource area with the terrain elevation conflict again according to a terrain elevation conflict line; and 3, determining a natural resource area upper limit, and for each natural resource area, enabling the area of the combined natural area to be smaller than the determined natural resource area upper limit. According to the invention, high-precision detection and continuous division of natural resource boundaries are realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of electronic digital data processing, and in particular relates to an intelligent mapping method for natural resource rights confirmation based on a three-dimensional working base map. Background Art

[0002] Natural resource rights confirmation and mapping is one of the core tasks in resource management, planning and protection. Its goal is to provide basic data support for scientific resource management and ownership definition by accurately identifying and dividing natural resource areas. With the development of remote sensing technology, geographic information system (GIS) technology and digital earth technology, natural resource rights confirmation has shifted from traditional manual surveying and mapping to automated mapping based on spatial data. However, existing technologies still have many shortcomings in dealing with complex terrain, diverse resource types and high-precision rights confirmation needs, and further improvement and optimization are needed.

[0003] Traditional natural resource mapping mainly relies on two-dimensional remote sensing images. These images capture the spectral reflectance characteristics of the surface through multispectral or hyperspectral sensors, and combine with land object classification algorithms (such as supervised classification, unsupervised classification, etc.) to achieve a preliminary classification of resource types. The advantages of two-dimensional images are wide coverage and convenient data acquisition, especially in flat areas and simple terrain. For example, vegetation cover extraction methods based on the normalized difference vegetation index (NDVI) have been widely used in the classification and monitoring of resources such as forests and grasslands. However, the limitation of two-dimensional images is that they can only reflect the planar distribution characteristics of the surface, and it is difficult to accurately depict the impact of terrain undulations on resource distribution. In areas with complex terrain, such as mountains, hills or river valleys, two-dimensional methods often lead to boundary misjudgment or confusion of resource types due to ignoring elevation information. In addition, two-dimensional images are easily disturbed by factors such as shadows and cloud cover, which limits the accuracy and reliability of resource mapping. To make up for the shortcomings of two-dimensional remote sensing images, digital elevation models (DEMs) are widely used in natural resource mapping. DEM provides elevation data of the earth's surface, which can assist in identifying terrain features (such as ridges, valleys, slopes, etc.) and introduce terrain dimensions in the demarcation of resource boundaries. For example, in the land use classification in mountainous areas, DEM can be used to determine the slope limit of farmland, thereby distinguishing flat areas suitable for farming from unused land with excessive slopes. Although the introduction of DEM has improved mapping accuracy, its application is mostly limited to simple terrain analysis, such as slope calculation and slope aspect determination. Traditional methods find it difficult to deeply integrate elevation information with spectral features, resulting in the failure to fully realize the potential of DEM in complex terrain areas. In addition, the resolution and accuracy limitations of DEM also have an adverse effect on the fine-grained division of natural resources. Summary of the invention

[0004] In view of this, the main purpose of the present invention is to provide an intelligent mapping method for natural resource rights confirmation based on a three-dimensional working base map, which realizes high-precision detection and continuity division of natural resource boundaries. The present invention can adapt well in complex terrain and diversified natural resource environments, effectively solves the problems of insufficient fusion of multi-spectral and elevation information, incoherent boundary detection, and poor adaptability to complex terrain in the prior art, and significantly improves the accuracy, intelligence and efficiency of natural resource mapping.

[0005] The technical solution adopted by the present invention is as follows: A natural resource rights confirmation intelligent mapping method based on a three-dimensional working base map, the method comprising: Step 1: Detect and identify natural resources on a three-dimensional working base map, generate preliminary boundaries, and divide the three-dimensional working base map into a plurality of different natural resource areas based on the preliminary boundaries; the three-dimensional working base map is established by image data captured by remote sensing satellites and geographic data acquired by a GIS system; Step 2: In each natural resource area, terrain elevation conflict detection is performed to obtain the terrain elevation conflict line in each natural resource area, and the natural resource areas with terrain elevation conflicts are re-divided according to the terrain elevation conflict lines; Step 3: Determine an upper limit for the natural resource area. For each natural resource area, calculate the average elevation difference between it and adjacent natural resource areas of different types. If the average elevation difference is within the set threshold range, the two natural resource areas will be merged into one natural resource area. The area of ​​the merged natural area must be smaller than the determined upper limit for the natural resource area. The average elevation difference is defined as the absolute value of the difference between the average altitude of one natural resource area and the average altitude of another natural resource area.

[0006] Furthermore, in step 2, the specific process of performing terrain elevation conflict detection in each natural resource area includes: obtaining a contour distribution map of each natural resource area from the three-dimensional working base map; in the contour distribution map, if the height distance difference between any two adjacent contour lines exceeds a set threshold, it is judged that there is a terrain elevation conflict, and the contour line with the lower height of the two adjacent contour lines is used as the terrain elevation conflict line; the types of natural resource areas include: rivers, wetlands, forests, grasslands, cultivated land and deserts.

[0007] Furthermore, step 1 specifically includes: Step 1.1: Calculate the multispectral features of the three-dimensional working base map; based on the multispectral features, calculate the local feature tensor of each pixel in the three-dimensional working base map; Step 1.2: Based on the multispectral features and the local feature tensor of each pixel, the boundary feature strength of the three-dimensional working base map is calculated; based on the boundary feature strength, a boundary tracking function is constructed; according to the boundary tracking function, a final boundary set is generated as a preliminary boundary; Step 1.3: Based on the preliminary boundaries, divide the 3D working base map into several different natural resource areas.

[0008] Furthermore, the process of calculating the multispectral characteristics of the three-dimensional working base map in step 1.1 specifically includes: obtaining spectral data from the three-dimensional working map, wherein the spectral data specifically includes: near infrared band reflectivity, red light band reflectivity, short infrared band reflectivity, green light band reflectivity and blue light band reflectivity; and then calculating the multispectral characteristics of the three-dimensional working base map by the following formula: ; in, Indicates the coordinate position is The multispectral characteristic value of the pixel at ; Indicates the X-axis coordinate; Indicates the Y-axis coordinate; Indicates the Z-axis coordinate; is the surface temperature; is the surface emissivity; is the reflectivity in the near-infrared band; is the reflectivity of red light band; is the reflectivity in the short infrared band; is the reflectivity of green light band; is the reflectivity of blue light band; The upper limit of the wavelength of remote sensing satellites; is the lower limit of the wavelength of the remote sensing satellite; is the wavelength integration variable; is the spectral reflectance distribution function, indicating that the wavelength is Spectral reflectance at .

[0009] Furthermore, the spectral reflectance distribution function Calculated by the following formula: ; in, is the radiance value of the remote sensing satellite; is the atmospheric radiation path value, and the calculation formula is: ; in, is the Rayleigh scattering reflectivity; is the aerosol scattering reflectivity; To observe the zenith angle; and They are the atmospheric transmittance in the observation direction and the sun direction respectively: is the solar spectrum irradiance, and the calculation formula is: ; in, is the spectral distribution of the solar constant; is the average distance between the sun and the earth; is the actual distance between the sun and the earth; is the atmospheric diffuse irradiance, and the calculation formula is: ; in, is the atmospheric diffuse transmittance; is the average reflectivity of the surface; is the albedo of the large balloon surface.

[0010] Furthermore, in step 1.1, the process of calculating the local feature tensor of each pixel in the three-dimensional working base map based on the multispectral feature specifically includes: ; in, Indicates the coordinate position is The local eigenvector of the pixel at ; is the radius of curvature of the earth’s surface; is the reference curvature radius.

[0011] Furthermore, in step 1.2, the boundary feature intensity of the three-dimensional working base map is calculated based on the multispectral features and the local feature tensor of each pixel by the following formula: ; in, Indicates the coordinate position is The boundary feature intensity of the pixel at ; for The gradient of is the absolute value operator.

[0012] Furthermore, in step 1.2, a boundary tracking function is constructed based on the boundary feature strength by the following formula: ; in, is the boundary tracking value; if Within the set threshold range, the coordinates are located at The pixel with coordinates is located at The adjacent pixels are connected to form a boundary segment. When all pixels are connected, all boundary segments are obtained, and each boundary segment is regarded as a boundary vector. is the geodesic curvature; is the normal curvature; is the normalized height value set; Indicates the coordinate position is The boundary feature intensity of the pixel at ; Indicates the coordinate position is The boundary feature intensity of the pixel at ; Indicates the coordinate position is The multispectral characteristic value of the pixel at ; Indicates the coordinate position is The multispectral characteristic value of the pixel at ; The coordinates are located at The pixel with coordinates is located at The distance between adjacent pixels.

[0013] Furthermore, in step 1.2, according to the boundary tracking function, the final boundary set is generated as the preliminary boundary process specifically including: among all boundary vectors, if the starting point coordinates or the ending point coordinates of any boundary vector are the same as the starting point coordinates or the ending point coordinates of any boundary vector, the two boundary vectors are summarized as one boundary, and after traversing all boundary vectors, several boundaries are obtained; if the starting point coordinates or the ending point coordinates of any boundary vector are not the same as the starting point coordinates or the ending point coordinates of any other boundary vector, it is regarded as an isolated boundary and discarded; each retained boundary is taken as an element to generate the final boundary set.

[0014] By adopting the above technical scheme, the present invention produces the following beneficial effects: the multispectral feature analysis method can effectively extract the unique characteristics of different natural resource types in the spectral space, such as the significant difference in reflectance between forests and grasslands in the near-infrared band and the green light band. In addition, by introducing the local feature tensor, the present invention can capture the subtle changes of multispectral features in three-dimensional space, especially at the resource boundary, where the change gradient of the tensor provides a high-precision description of the boundary characteristics. Combined with the boundary tracking function, the present invention further optimizes the consistency and accuracy of boundary extraction, thereby ensuring the high accuracy of the mapping results. For example, in complex terrain areas (such as mountains or hills), traditional methods often cause boundary breaks or misjudgments due to elevation fluctuations, while the present invention successfully overcomes this problem through curvature correction factors and elevation normalization technology. The final boundary set generation process removes isolated boundaries and redundant boundary segments through merging and screening operations, further improving the integrity and accuracy of the mapping results. The present invention adopts a combination of boundary feature strength and boundary tracking function in boundary detection to solve the problems of insufficient boundary detection accuracy and poor continuity in the prior art. The boundary feature intensity can not only accurately capture the significance of resource boundaries, but also reflect the directional characteristics of boundaries through the normalized combination of the determinant of the local feature tensor and the spectral gradient. In complex terrain, this method can accurately locate the resource boundary line, avoiding the boundary ambiguity caused by elevation mutation or insufficient spectral change in traditional methods. In addition, the boundary tracking function further improves the consistency of boundary tracking results by introducing curvature correction and height normalization factors. The final generated boundary set is merged and isolated boundaries are eliminated, so that the boundaries have geometric continuity and consistency, providing a reliable boundary basis for resource confirmation. The existence of isolated boundaries is a difficulty in traditional mapping methods. The present invention can effectively clean up isolated boundaries through a comprehensive traversal and screening mechanism to ensure that the final generated boundary set only contains boundaries that are of practical significance to resource division. This screening mechanism detects whether the starting point and end point of the boundary vector are connected to other boundaries, and eliminates isolated boundaries, thereby improving the overall quality of the boundary set. In addition, by merging adjacent boundary vectors, the present invention further optimizes the compactness of the boundary set, making the mapping results clearer and more concise, and providing efficient data support for the classification management of resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of a method flow of an intelligent mapping method for natural resource rights confirmation based on a three-dimensional working base map provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0016] All features disclosed in this specification, or steps in all methods or processes disclosed, except mutually exclusive features and / or steps, can be combined in any manner.

[0017] Any feature disclosed in this specification (including any additional claims and abstract), unless otherwise stated, may be replaced by other alternative features that are equivalent or have similar purposes. That is, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

[0018] Example 1: Reference Figure 1 , a natural resource rights confirmation intelligent mapping method based on a three-dimensional working base map, the method comprising: Step 1: Detect and identify natural resources on a three-dimensional working base map, generate preliminary boundaries, and divide the three-dimensional working base map into a plurality of different natural resource areas based on the preliminary boundaries; the three-dimensional working base map is established by image data captured by remote sensing satellites and geographic data acquired by a GIS system; The construction of the three-dimensional working base map is based on remote sensing satellite images. The surface feature information, such as vegetation, water bodies, bare land, etc., is extracted through multispectral, hyperspectral and other image data. This information can clearly separate different types of objects through the difference in spectral reflectance. At the same time, the resolution, texture characteristics and multi-phase change information of remote sensing images also provide important basis for dynamic monitoring. In addition, the introduction of the GIS system further enhances the spatial accuracy of the base map. GIS not only integrates geographic coordinates and topological relationships, but also comprehensively models terrain elevation data through digital elevation models (DEM) and digital surface models (DSM), thereby expanding two-dimensional image information into three-dimensional stereo models, so that terrain undulations, slopes and elevation changes can be visualized. This deep fusion of multi-source data gives the three-dimensional working base map an incomparable advantage in dealing with complex terrain, especially in areas with complex terrain, which can significantly improve the ability to distinguish and classify natural resources. Based on the constructed three-dimensional working base map, the present invention applies advanced image processing and machine learning technologies in natural resource detection and identification. These technologies classify and label the types of natural resources in the base map through automated analysis processes. In the specific implementation, the features in the remote sensing image are extracted and classified by deep learning algorithms, such as convolutional neural networks (CNN), and these features include the texture, shape, spectral reflectance characteristics, etc. of the ground objects. The CNN model can accurately distinguish different types of natural resources such as water bodies, forests, and farmlands through multi-layer feature extraction and learning. At the same time, based on the elevation information of the three-dimensional base map, the classification of natural resources can be further refined. For example, for vegetation types, trees and shrubs can be distinguished by elevation information; for water body types, the flow direction and range of the water system can be determined by terrain undulations. This classification method combining two-dimensional images with three-dimensional elevation not only significantly improves the accuracy of detection, but also provides a more scientific basis for the division of ownership of natural resources. In the present invention, another key link in natural resource detection is the automatic generation of boundaries. Through image segmentation technology, the boundary lines of different types of natural resources can be extracted from the three-dimensional working base map. These boundary lines are delineated based on the feature changes of remote sensing images and the spatial distribution in GIS data, ensuring that the position of the boundary line is consistent with the surface texture features and maintains reasonable continuity in the elevation dimension. For example, at the junction of water bodies and vegetation, the system can quickly generate preliminary boundary lines based on spectral differences, and combine elevation data to eliminate misjudgments caused by image shadows or reflections, thereby obtaining more accurate segmentation results. This method of automatically generating boundaries reduces manual intervention and improves mapping efficiency, especially in large-scale regional natural resource rights confirmation tasks. This advantage is particularly obvious.

[0019] Step 2: In each natural resource area, terrain elevation conflict detection is performed to obtain the terrain elevation conflict line in each natural resource area, and the natural resource areas with terrain elevation conflicts are re-divided according to the terrain elevation conflict lines; Step 2: By detecting and subdividing the terrain elevation conflicts within the natural resource area, the consistency of the natural resource area in three-dimensional space is ensured, providing accurate and scientific division results for subsequent rights confirmation and mapping. This process is based on the strong data support of the three-dimensional working base map. By deeply analyzing the elevation characteristics of the natural resource area, it systematically handles the possible sudden changes or inconsistencies in terrain elevation within the area. Its core principle is to detect the abnormal elevation changes within the area through terrain elevation conflict detection, and generate elevation conflict lines based on this to refine the natural resource area. This process not only improves the accuracy of natural resource area division, but also solves the division error problem caused by the lack of elevation information in the traditional two-dimensional mapping method. The three-dimensional working base map plays a key role in this step. The elevation data it contains not only reflects the terrain characteristics of the surface, but also provides a comprehensive description of the elevation changes within the area. After the preliminary division of the natural resource area is completed, each area may cover complex terrain features, such as hills, mountains or plains. In traditional two-dimensional mapping, the elevation changes of these terrains are often ignored, resulting in the inability of regional division to accurately reflect the actual geographical situation. However, the present invention can accurately capture the elevation mutation within the region by performing terrain elevation conflict detection on each natural resource area. This elevation mutation is usually manifested as a significant difference in elevation between different locations within the region, such as an area containing both ridges and valleys, or an area with a large elevation span that exceeds the reasonable range of natural resource classification. By detecting these conflict points, the present invention can effectively identify elevation anomalies within the region and provide a scientific basis for subsequent refinement. The detection of terrain elevation conflicts depends on the in-depth analysis of elevation data. In the three-dimensional working base map, the elevation data of each natural resource area is extracted and statistically analyzed, including parameters such as the minimum, maximum, average, and elevation change rate of the elevation. These parameters comprehensively reflect the elevation characteristics within the region and provide a basis for judging elevation conflicts. For example, when the elevation change rate of a natural resource area is significantly higher than the average, this indicates that there may be a large elevation mutation in the area. At the same time, by analyzing the continuity of the elevation distribution within the region, the specific location and range of the elevation mutation can be further identified. This elevation analysis method utilizes the spatial stereo information of the three-dimensional working base map, and can detect elevation conflicts with higher accuracy and reliability, thus providing support for subsequent detailed divisions.

[0020] Once it is detected that there is an elevation conflict inside the natural resource area, the present invention will generate elevation conflict lines based on the elevation mutation points, and use these conflict lines to divide the original area again. The generation of elevation conflict lines is based on the analysis of the gradient of elevation data changes. Specifically, the system will search for points where the elevation change rate exceeds the set threshold along the elevation change trajectory inside the area, and connect these points into lines to form complete elevation conflict lines. Such conflict lines usually correspond to significant boundaries in actual terrain, such as ridge lines, valley lines, or steep slope boundaries. These terrain boundaries are not only important features of natural landforms, but also potential boundaries for different types of natural resources. For example, a ridge line may separate a forest area from a grassland area, while a valley line may mark the watershed boundary of different water bodies. By generating conflict lines based on elevation data, the present invention can handle complex terrain areas in a scientific way and provide a basis for the reasonable division of natural resources. In the division process, the system not only pays attention to the consistency of the elevation inside the area, but also takes into account the integrity and rationality of the divided area. For example, a natural resource area may be divided into multiple sub-areas due to elevation conflicts, but these sub-areas still need to meet the requirements of relative consistency in elevation, and also maintain clear boundaries with other adjacent areas. This division method effectively avoids the problem of blurred boundaries or division errors caused by complex terrain in traditional methods. In addition, through the dynamic adjustment of elevation conflict lines, the present invention can adapt to the specific needs of different natural resource types. For example, for forest areas, the division may pay more attention to changes in slope and altitude, while for agricultural land, more attention may be paid to terrain flatness and drainage conditions. This dynamic adaptability makes the division results more in line with actual natural resource characteristics and management needs.

[0021] Step 3: Determine an upper limit for the natural resource area. For each natural resource area, calculate the average elevation difference between it and adjacent natural resource areas of different types. If the average elevation difference is within the set threshold range, the two natural resource areas will be merged into one natural resource area. The area of ​​the merged natural area must be smaller than the determined upper limit for the natural resource area. The average elevation difference is defined as the absolute value of the difference between the average altitude of one natural resource area and the average altitude of another natural resource area.

[0022] The average elevation difference is an important indicator to measure the consistency of two adjacent natural resource areas in the vertical direction. The core of this method is to use the elevation data in the three-dimensional working base map to calculate the average elevation value of each area, and then compare the average elevation difference between the two adjacent areas. The elevation data in the three-dimensional working base map has extremely high accuracy and can fully reflect the terrain undulations and vertical change characteristics of the natural resource area. Therefore, the average elevation difference calculation based on the elevation data is not only mathematically reliable, but also highly related to the actual distribution characteristics of natural resources. Through this indicator, the present invention can scientifically evaluate whether two natural resource areas have the potential to merge in terms of terrain characteristics. For example, if two adjacent natural resource areas appear to be independent on the plane, but their average elevation difference falls within the set threshold range, it can be inferred that they may belong to the same type or fused natural resources. This judgment method makes up for the deficiency of the traditional method of dividing the area only based on the plane distribution and texture characteristics, and introduces the vertical information into the rules of regional merging, so as to carry out the mapping of natural resources more comprehensively and scientifically. In the process of regional merging, the setting of the threshold plays a key role. The threshold reflects the allowable average elevation difference range, which is essentially the acceptable consistency of the natural resource area in the vertical direction. The setting of the threshold needs to be dynamically adjusted according to the specific natural resource type and geographical environment. For example, in a flat agricultural area, the threshold can be small to ensure that only areas with very similar elevations can be merged; while in mountainous or hilly areas, the threshold can be appropriately relaxed to adapt to the undulating characteristics of the terrain itself. By dynamically adjusting the threshold, the present invention can adapt to the mapping needs of different natural resource types, making regional merging both scientific and reasonable and highly flexible. This merging principle based on the elevation difference threshold, combined with the precise elevation data of the three-dimensional working base map, makes the regional division of natural resources more in line with the actual geographical situation, and can significantly improve the accuracy and reliability of right confirmation. In addition to the average elevation difference, the present invention also introduces the concept of an upper limit on the area of ​​natural resource areas to constrain the results of regional merging. The setting of the area upper limit is based on the actual needs of resource management and right confirmation accuracy, because an overly large natural resource area may not only lead to a decrease in the accuracy of mapping, but also bring difficulties to subsequent management and right confirmation work. By setting a reasonable area upper limit, the present invention can ensure that even if multiple natural resource areas meet the merging conditions, the area of ​​the merged area will not exceed the scope of management requirements. For example, in some nature reserves, the area of ​​a single area may need to be controlled within a certain range to better conduct ecological monitoring and resource management. In land rights confirmation, the setting of an area cap can effectively avoid the emergence of large plots that do not meet the standards, thereby improving the operability and scientificity of rights confirmation. The area cap and the elevation difference threshold work together to provide a scientific constraint framework for regional mergers, so that the merged areas not only meet the requirements of terrain consistency, but also meet the actual management needs.The process of regional merging is an intelligent optimization process based on three-dimensional spatial information, and its goal is to optimize the division of natural resource regions by comprehensively considering the consistency of elevation and rationality of area of ​​adjacent regions. In actual operation, the present invention accurately locates the boundaries and elevation data of adjacent natural resource regions through a three-dimensional working base map, and then uses an algorithm to calculate the average elevation difference and the merged area. When the average elevation difference between two natural resource areas is within the set threshold range and the area of ​​the merged area is less than the upper limit of the area, the system will automatically complete the merger of the areas. This intelligent merging method not only reduces the need for manual intervention, but also significantly improves the efficiency of mapping and the accuracy of the results. Through such an optimization process, the present invention can complete the intelligent division of large-scale natural resources in a short period of time, providing reliable technical support for the implementation of rights confirmation work.

[0023] In the prior art, remote sensing satellite data, as the basic source of three-dimensional working base maps, obtains images of the earth's surface through multispectral and hyperspectral sensors carried on satellites. These image data cover information such as the spectral reflectance characteristics of the surface, terrain texture, and distribution of objects. Through different orbit designs and resolution settings of remote sensing satellites, high-precision image data of different times and different geographical areas can be obtained. The core advantage of remote sensing data lies in its macroscopic and real-time nature. It can cover a large area of ​​the surface in a short time interval, providing rich data support for dynamic monitoring and terrain modeling. However, the original image data is usually affected by factors such as the curvature of the earth, sensor distortion, and atmospheric scattering. Therefore, it needs to go through a series of preprocessing steps, including radiation correction, atmospheric correction, geometric correction, and orthorectification, to ensure that the data truly reflects the actual characteristics of the geographic space.

[0024] At the same time, the GIS system provides key information such as geographic coordinates, terrain elevation and topological relationships for the three-dimensional working base map by acquiring geospatial data. Terrain elevation data usually comes from digital elevation models (DEMs) or digital surface models (DSMs), which can describe the undulating characteristics of the surface. DEMs are usually obtained through laser radar (LiDAR) technology or photogrammetry technology, which extracts elevation information based on the principle of stereo geometry by analyzing multiple overlapping images. DEMs store the elevation values ​​of each geographical location in the form of regular grids or triangulated networks, providing the core elevation support for base map modeling. The GIS system not only stores these terrain elevation data, but also integrates them with vector data (such as roads, water systems, building boundaries, etc.) and attribute data (such as land use types and geological features) to provide a multi-dimensional information foundation for three-dimensional modeling. In the data integration stage, remote sensing images and GIS data are fused through spatial superposition and coordinate unification. Since remote sensing data and GIS data usually come from different sources and may use different coordinate reference systems, coordinate system conversion and registration processing are required first. The goal of registration is to ensure that each pixel in the remote sensing image completely corresponds to the geographic spatial point in the GIS data. After completing the spatial matching, the fusion stage combines the texture and spectral information of the remote sensing image with the elevation and attribute information of the GIS data to generate a basic three-dimensional model.

[0025] In order to achieve three-dimensional visualization, the fused data needs to be further processed. The texture information of remote sensing images is usually mapped to a terrain elevation model (such as DEM or DSM), a process called texture mapping. Through texture mapping, the spectral characteristics of the surface and the elevation information together constitute a visual expression of the three-dimensional terrain. For complex land features (such as buildings, vegetation, etc.), vector data and three-dimensional modeling technology can be combined for refinement, such as by building a three-dimensional building model (CityGML) or a vegetation coverage model to make the base map more realistic and practical. In addition, in order to improve the applicability of the base map, the existing technology often uses multi-resolution expressions for three-dimensional models to support application scenarios with different accuracy requirements. For example, in large-scale terrain analysis, low-resolution models can be used to reduce the computational burden, while when fine-area modeling is performed, high-resolution mode is switched to obtain more accurate geographic information. This multi-resolution expression is achieved through technologies such as mesh simplification algorithms, quadtree segmentation, or viewpoint-dependent rendering.

[0026] Embodiment 2: In step 2, the specific process of performing terrain elevation conflict detection in each natural resource area includes: obtaining a contour distribution map of each natural resource area from a three-dimensional working base map; in the contour distribution map, if the height distance difference between any two adjacent contour lines exceeds a set threshold, it is judged that there is a terrain elevation conflict, and the contour line with the lower height of the two adjacent contour lines is used as the terrain elevation conflict line; the types of natural resource areas include: rivers, wetlands, forests, grasslands, cultivated land and deserts.

[0027] Specifically, the three-dimensional working base map is a comprehensive expression of geographic information, and the digital elevation model (DEM) it contains provides the data basis for generating contour distribution maps. In specific operations, the system generates contour distribution maps of natural resource areas at set contour intervals by extracting elevation data from the DEM. These contour lines are horizontal slices of the surface at specific elevation intervals and are an intuitive representation of terrain undulations. For example, areas with dense contour lines usually represent steep slopes or landforms with large elevation differences, while areas with sparse contour lines reflect flat landforms or landforms with small elevation changes. These characteristics make contour lines a key tool for analyzing terrain elevation conflicts. After the contour distribution map is generated, the system will analyze the height difference between each pair of adjacent contour lines in turn. Specifically, the system calculates the difference in elevation values ​​corresponding to two adjacent contour lines and compares this difference with a preset threshold. The threshold reflects the range of elevation changes allowed within the natural resource area. For example, in flat areas such as farmland or wetlands, the threshold may be small to ensure the precision of the division results; while in areas with large terrain fluctuations such as forests or deserts, the threshold can be appropriately relaxed to adapt to actual terrain conditions. If the height difference between two adjacent contour lines exceeds this threshold, the system determines that there is a terrain elevation conflict in the area.

[0028] In order to further clarify the location and scope of the conflict, the system marks a lower contour line as a terrain elevation conflict line. This choice is based on the characteristics of natural resources and topographic principles, because areas with lower altitudes are usually more able to reflect the demarcation points of the terrain, such as the bottom of a river or valley is often the point where the conflict appears. By clearly marking the terrain elevation conflict line, the present invention can accurately locate the conflict area, thereby providing a basis for the subsequent refinement of the natural resource area. The type of natural resources also has an important impact on the detection rules of terrain elevation conflicts. In different types of natural resource areas, the elevation characteristics and conflicts are manifested in different forms. For example, in river areas, contour lines may appear parallel and relatively dense, and conflict points often appear near river valleys; in wetland areas, contour lines are usually sparse and evenly distributed, and conflict points may be related to local terrain uplift; while in forest and grassland areas, the range of elevation changes may be larger, and conflict points are mostly concentrated on ridge lines or places where slopes change suddenly. Cultivated land is usually distributed on relatively flat terrain, so elevation conflicts are mostly related to artificially reclaimed slopes or terrain transformation; elevation conflicts in desert areas may be related to wind erosion landforms or sand dune distribution. This detection rule for different resource types enables the method of the present invention to not only adapt to diverse natural resource environments, but also show high detection accuracy and practicality in each environment.

[0029] Embodiment 3: Step 1 specifically comprises: Step 1.1: Calculate the multispectral features of the three-dimensional working base map; based on the multispectral features, calculate the local feature tensor of each pixel in the three-dimensional working base map; Step 1.2: Based on the multispectral features and the local feature tensor of each pixel, the boundary feature strength of the three-dimensional working base map is calculated; based on the boundary feature strength, a boundary tracking function is constructed; according to the boundary tracking function, a final boundary set is generated as a preliminary boundary; Step 1.3: Based on the preliminary boundaries, divide the 3D working base map into several different natural resource areas.

[0030] Specifically, in step 1.1, the multispectral features of the three-dimensional working base map are calculated by multispectral or hyperspectral data in remote sensing satellite images. Multispectral data can provide reflectance information of the surface in different bands. Different natural resource types (such as forests, cultivated land, grasslands, etc.) have significant differences in multispectral features. For example, vegetation usually has a higher reflectance in the near-infrared band, while water bodies have a lower reflectance in the visible light band. These multispectral features are converted into spatial reflections through specific algorithms. For example, spectral indices (such as the normalized difference vegetation index NDVI) can intuitively display the vegetation coverage. In order to further improve the spatial resolution and feature expression capabilities, this step also calculates the local feature tensor of each pixel in the three-dimensional working base map. The local feature tensor is a mathematical expression that describes the spatial relationship around the pixel. It combines spectral features and spatial texture features. By calculating the difference between each pixel and the surrounding pixels in the multispectral space, it can capture the subtle changes in the local surface. This tensor representation method can effectively enhance the clarity of the boundary, especially at the junction of different natural resource types. Next, in step 1.2, the boundary feature intensity of the three-dimensional working base map is further calculated based on the multispectral features and local feature tensors calculated in the previous step. Boundary feature intensity is an indicator used to measure the degree of feature difference between pixels, and its purpose is to locate the location where the natural resource type changes significantly. The calculation method usually includes gradient analysis or edge detection algorithm. For example, by calculating the gradient amplitude of the multispectral features, an intensity map reflecting the boundary position can be obtained. The higher the boundary feature intensity, the more likely the location is to be the boundary line of different natural resource types. After obtaining the boundary feature intensity, a boundary tracking function is constructed based on its value. The function of the boundary tracking function is to connect possible boundary points in the intensity map into continuous boundary lines. This function combines global optimization and local tracking strategies, and can accurately extract boundaries in areas with drastic intensity changes, while avoiding boundary breaks or offsets caused by noise or local anomalies. Finally, by solving the boundary tracking function, a preliminary boundary set is generated to identify the boundary lines of different natural resource types. Finally, in step 1.3, the three-dimensional working base map is divided into multiple natural resource areas based on the generated preliminary boundary set. The process of regional division is based on the spectral consistency and spatial connectivity of natural resource types, ensuring that pixels in the same area have high similarity in spectral characteristics and form a complete connected area in space. For example, a divided forest area may cover a plot of densely covered vegetation, while its boundary is clearly distinguished from the adjacent grassland or cultivated land. Through this boundary feature-based division method, the present invention can realize the automatic identification and preliminary division of natural resource areas, laying an important foundation for subsequent terrain elevation conflict detection and regional merging.

[0031] Embodiment 4: The process of calculating the multispectral characteristics of the three-dimensional working base map in step 1.1 specifically includes: obtaining spectral data from the three-dimensional working map, the spectral data specifically including: near infrared band reflectivity, red light band reflectivity, short infrared band reflectivity, green light band reflectivity and blue light band reflectivity; and then calculating the multispectral characteristics of the three-dimensional working base map by the following formula: ; in, Indicates the coordinate position is The multispectral characteristic value of the pixel at ; Indicates the X-axis coordinate; Indicates the Y-axis coordinate; Indicates the Z-axis coordinate; is the surface temperature; is the surface emissivity; is the reflectivity in the near-infrared band; is the reflectivity of red light band; is the reflectivity in the short infrared band; is the reflectivity of green light band; is the reflectivity of blue light band; The upper limit of the wavelength of remote sensing satellites; is the lower limit of the wavelength of the remote sensing satellite; is the wavelength integration variable; is the spectral reflectance distribution function, indicating that the wavelength is Spectral reflectance at .

[0032] Specifically, the formula (near infrared reflectivity) and (Red light band reflectivity) combination The basic principle of vegetation index is reflected. The near-infrared band is highly sensitive to vegetation. Areas with dense vegetation have higher reflectivity in this band, while the reflectivity of the red band reflects more the health and density of vegetation. By calculating the difference between the near-infrared and red reflectivity and standardizing them, it is possible to effectively distinguish high-density vegetation from other surface types, such as water bodies or bare land. Short-infrared band reflectivity The introduction of further enhances the sensitivity to soil moisture and vegetation moisture content, making the combination not only able to identify vegetation coverage, but also able to distinguish different types of vegetation and soil conditions. Utilizes the green light band With blue band The spectral characteristics of surface vegetation are captured by the reflectivity ratio of the green light band. The green light band is usually closely related to the greenness and health of vegetation, while the blue light band is greatly affected by atmospheric scattering and water reflection. Through this ratio calculation, the influence of atmospheric interference and water reflection on vegetation detection can be effectively suppressed, thereby improving the accuracy of vegetation boundary detection. This exponential function not only enhances the nonlinear expression ability of features, but also further strengthens the distinction between different types of natural resources through ratio operations, so that the formula can still maintain high-precision recognition capabilities in complex environments.

[0033] Integral Item Represents the integral of the spectral reflectance distribution function within the wavelength range of the remote sensing satellite. This part provides the overall response characteristics of the surface material to spectra of different wavelengths by accumulating the reflectance within the entire wavelength range. This integral term not only integrates multi-band information, but also further enhances the ability to identify complex surface types through an overall analysis of the spectral reflectance. For example, in diverse natural resource areas such as forests, grasslands, and cultivated land, the overall distribution patterns of spectral reflectance are different. These differentiated features can be effectively extracted through integral operations, thereby achieving more accurate resource classification and boundary demarcation. Finally, the surface temperature and surface emissivity Combination of Thermal infrared information is introduced to reflect the thermal characteristics of the surface. The combination of surface temperature and emissivity not only reveals the thermal inertia and heat dissipation characteristics of surface materials, but also helps to distinguish different types of natural resources. For example, water bodies usually show lower surface temperatures due to their high specific heat capacity and low emissivity, while urban buildings and bare land may show higher surface temperatures. By incorporating these thermal characteristics into the calculation of multispectral features, the formula not only expands the dimension of multispectral features, but also enhances the ability to distinguish different types of natural resources, thereby improving the accuracy and reliability of natural resource boundary detection.

[0034] Example 5: Spectral reflectance distribution function Calculated by the following formula: ; in, is the radiance value of the remote sensing satellite; is the atmospheric radiation path value, and the calculation formula is: ; in, is the Rayleigh scattering reflectivity; is the aerosol scattering reflectivity; To observe the zenith angle; and They are the atmospheric transmittance in the observation direction and the sun direction respectively: is the solar spectrum irradiance, and the calculation formula is: ; in, is the spectral distribution of the solar constant; is the average distance between the sun and the earth; is the actual distance between the sun and the earth; is the atmospheric diffuse irradiance, and the calculation formula is: ; in, is the atmospheric diffuse transmittance; is the average reflectivity of the surface; is the albedo of the large balloon surface.

[0035] Specifically, the formula The core principle is to use the radiance observed by remote sensing satellites at the top of the atmosphere , by deducting the atmospheric path radiation The influence of the sun's irradiance and the atmospheric transmittance is combined to calculate the reflection characteristics of the surface to light of a specific wavelength. Atmospheric path radiation is caused by Rayleigh scattering and aerosol scattering, which are interference factors that cannot be ignored in remote sensing observations. The calculation formula shows that atmospheric path radiation is related to The relationship between is controlled by the Rayleigh scattering reflectivity and the aerosol scattering reflectivity, which are respectively expressed in the formula and The combination of the two reflects how particles and air molecules in the atmosphere affect the transmission and scattering of solar radiation. By correcting the atmospheric path radiation, the surface contribution can be extracted from the total radiation to ensure the accuracy of the spectral reflectance calculation. Next, the calculation of solar irradiance further reflects the dynamic radiation relationship between the earth and the sun. Solar irradiance Spectral distribution by solar constant This correction is made to take into account the elliptical shape of the Earth's orbit, which causes the Earth-Sun distance to change over time. , the difference in solar radiation intensity caused by distance changes can be adjusted to obtain the actual solar irradiance. This processing ensures that the surface radiation data observed by remote sensing satellites at different times are comparable and avoids the introduction of errors due to natural changes in solar radiation intensity.

[0036] In addition, the atmospheric diffuse irradiance The calculation formula describes the diffuse behavior of solar radiation when it passes through the atmosphere. Diffuse radiation refers to the radiation part of sunlight that is scattered in all directions by atmospheric molecules. The atmospheric diffuse transmittance is introduced into the formula , average surface reflectivity and the large balloon surface albedo , which together describe the complex radiation exchange relationship between the atmosphere and the surface. In particular represents the average reflectance characteristic of the surface, and It describes the reflectivity of the atmosphere itself. These parameters make the calculation of diffuse radiation more consistent with the actual atmospheric and surface conditions, ensuring the accuracy of the final reflectivity calculation. Finally, the transmittance parameter in the formula and The transmittance in the observation direction and the direction of the sun is respectively. Transmittance is a key factor in remote sensing data correction. It reflects the degree of attenuation of radiation due to molecular absorption, scattering, etc. when passing through the atmosphere. Transmittance in the observation direction Combined with the observation zenith angle to describe the attenuation of radiation in the transmission path, and the transmittance in the direction of the sun It shows the loss of solar radiation after entering the earth's atmosphere and before reaching the earth's surface. These transmittance parameters directly affect the denominator in the spectral reflectance calculation, ensuring that the corrected data can accurately reflect the actual spectral characteristics of the earth's surface.

[0037] Embodiment 6: The process of calculating the local feature tensor of each pixel in the three-dimensional working base map based on the multispectral feature in step 1.1 specifically includes: ; in, Indicates the coordinate position is The local eigenvector of the pixel at ; is the radius of curvature of the earth’s surface; is the reference curvature radius.

[0038] Specifically, the basis of the formula is the characteristic tensor matrix , whose structure is composed of multi-spectral eigenvalues These partial derivatives describe the pixel in different spatial directions ( ) and their correlation. Each item in the feature tensor matrix, such as , indicating that the pixel is The second-order change in direction, i.e. the acceleration characteristic of the local change. Similarly, the cross terms (e.g. ) reflects the joint changes between the two directions. This tensor representation can capture subtle changes in multispectral features in three-dimensional space, especially at the junction of natural resource types, where the gradient of feature changes tends to increase significantly. By calculating these second-order derivatives, the feature tensor matrix describes the local characteristics of the area where the pixel is located, including smooth areas, boundary areas, and mutation points. For example, in areas with flat terrain, the diagonal terms of the tensor matrix (i.e., the values ​​of the second-order partial derivatives in a single direction) are close to zero, while in areas with steep boundaries or significant changes in natural resource types, these values ​​will grow rapidly. In addition, the cross terms can further capture the synergy of multi-directional changes, providing a more comprehensive description for boundary detection of complex terrain or natural resource types.

[0039] Curvature Modulation Factor The influence of surface geometry information is introduced into the formula, where is the radius of curvature of the earth's surface, is the reference radius of curvature. The radius of curvature reflects the curvature of the surface at a specific point. A large radius of curvature corresponds to a flatter terrain, while a small radius of curvature indicates a steep area or an area with a dramatic height change. Through this exponential function, the value of the local feature tensor matrix will be nonlinearly adjusted according to the magnitude of the curvature. In a flat area ( When the modulator is larger, the modulation factor is close to 1, and the effect on the feature tensor is small; while in areas with large curvature (such as ridges or valleys), the modulation factor decreases rapidly, thereby amplifying the value of the feature tensor matrix. This nonlinear correction makes the formula more sensitive to complex terrain and can more accurately describe local features. The final result of the feature tensor matrix is ​​a comprehensive local descriptor, which not only captures the subtle changes of pixels in multispectral space, but also combines the surface curvature characteristics, enabling it to effectively distinguish different types of natural resources and complex terrain features. For example, at the junction of forests and grasslands, the gradient of the feature tensor changes significantly, while in the transition zone between wetlands and cultivated land, its changes are relatively gentle. This detailed description capability provides important data support for the intelligent mapping of natural resource rights confirmation.

[0040] Embodiment 7: In step 1.2, the boundary feature intensity of the three-dimensional working base map is calculated based on the multispectral features and the local feature tensor of each pixel by the following formula: ; in, Indicates the coordinate position is The boundary feature intensity of the pixel at ; for The gradient of is the absolute value operator.

[0041] Specifically, the determinant of the local feature tensor is It is the core part of the formula. Local feature tensor It is a matrix composed of the second-order partial derivatives of multispectral features, each component of which describes the change characteristics of the pixel in a certain direction in three-dimensional space. For example, Indicates that the pixel is The local curvature characteristics of the direction, and the cross terms such as It describes and The determinant calculation compresses the overall information of the tensor matrix into a scalar value, whose size reflects the comprehensive intensity of local feature changes. When the pixel is at the boundary of natural resources, the value of the feature tensor will increase significantly due to the drastic change of multispectral features at the boundary, resulting in a larger determinant value. By taking the cubic root of the determinant , the formula nonlinearly scales the determinant value, making the distribution of feature intensity smoother. This processing can not only suppress abnormal spikes caused by noise, but also enhance the ability to respond to large-scale boundary changes. For example, at the junction of forest and grassland, due to the significant difference in the characteristics of the two in multispectral space, the determinant of the tensor matrix will increase significantly, and the cubic root operation can adjust this range of change to a range suitable for subsequent boundary detection.

[0042] Second, the gradient part of the formula It is the gradient of the multispectral feature, indicating the multispectral feature value of the pixel The direction and strength of the change in space. The magnitude of the gradient It reflects the change amplitude of the pixel in space, while the gradient direction points to the direction of the fastest change. In the formula, through the normalization operation , the directional information of the gradient is extracted, so that the feature intensity can highlight the position and direction of the boundary. The role of gradient normalization is to enhance the expression of directionality, so that the boundary feature can not only describe the intensity, but also reflect the location attribute of the boundary. For example, when the gradient direction is perpendicular to the boundary, its normalized result will significantly enhance the boundary feature intensity, while when the gradient direction is parallel to the boundary, the feature intensity is relatively weak. This feature enables the formula to effectively distinguish boundary features in different directions, thereby achieving more accurate boundary detection in scenes with complex terrain or diverse natural resource types. Combining these two parts, the formula as a whole constructs a comprehensive boundary feature intensity descriptor. The determinant of the local feature tensor provides a global description of the boundary significance, while the normalization of the gradient introduces local directional information, so that the feature intensity can reflect both the change intensity of the boundary and the directionality of the boundary. For example, at the junction of wetlands and cultivated land, the multispectral feature values ​​show significant changes in the gradient direction, and the value of the local feature tensor also increases due to the severity of this change, resulting in the boundary feature intensity. The formula also has strong noise resistance. Since the determinant operation can smooth out small local outliers, and the gradient normalization operation can further reduce the impact of noise on directionality, the formula can still maintain high robustness when processing remote sensing data containing more noise or slight disturbances. This feature is particularly important when dealing with complex terrain and diverse natural resource types, because natural resource boundaries are often accompanied by a variety of interference factors, such as irregular changes in surface reflectivity or the influence of atmospheric conditions.

[0043] Example 8: In step 1.2, a boundary tracking function is constructed based on the boundary feature strength using the following formula: ; in, is the boundary tracking value; if Within the set threshold range, the coordinates are located at The pixel with coordinates is located at The adjacent pixels are connected to form a boundary segment. When all pixels are connected, all boundary segments are obtained, and each boundary segment is regarded as a boundary vector. is the geodesic curvature; is the normal curvature; is the normalized height value set; Indicates the coordinate position is The boundary feature intensity of the pixel at ; Indicates the coordinate position is The boundary feature intensity of the pixel at ; Indicates the coordinate position is The multispectral characteristic value of the pixel at ; Indicates the coordinate position is The multispectral characteristic value of the pixel at ; The coordinates are located at The pixel with coordinates is located at The distance between adjacent pixels.

[0044] Specifically, the beginning of the formula is a curvature combination term, which is used to measure the local characteristics of the surface geometry of the pixel. represents the geodesic curvature, which reflects the curvature of the surface along the tangent direction around the pixel, and represents the normal curvature, which describes the curvature of the surface along the normal direction. The sum of the squares of the two reflects the comprehensive complexity of the geometric morphology of the area. The introduction of curvature in the boundary tracking function enables the system to sensitively capture areas where the surface curvature changes significantly, such as steep slopes, ridges or valleys, thereby enhancing the terrain adaptability of boundary detection. Next, is the normalized height difference correction term, which is used to weaken the adverse effect of height difference on the correlation of boundary features. represents the vertical height difference between two pixels, and is a preset normalized scale that controls the sensitivity of height differences. The exponential function ensures that small height differences have little effect on boundary tracking, while the effect of large height differences is suppressed by the exponential. This is especially important when the boundary crosses different elevation areas, such as on a hillside or stepped terrain. By weakening the effect of height differences, the continuity of the boundary can be tracked more stably.

[0045] Then, the boundary feature strength term Describes the difference in boundary intensity between two pixels. This difference reflects the degree of significant change in boundary characteristics in the local area. The greater the intensity difference, the more significant the difference in boundary characteristics between the two pixels, and the larger the tracking function value, the more likely it is to identify them as part of the boundary. The combination of multi-spectral feature values ​​and gradients is introduced to further enhance the characterization of spectral feature similarity. Here, and is the multispectral gradient of two pixels, indicating the rate of change of spectral characteristics. The product of the absolute values ​​of the gradients in the denominator reflects the comprehensive amplitude of multispectral changes of the two pixels, while the product of the spectral eigenvalues ​​in the numerator describes the spectral correlation of the pixels. Through this normalization process, the formula can dynamically adjust the sensitivity of the boundary tracking function to the spectral eigenvalues ​​and gradients, thereby more accurately capturing the boundaries of different natural resource types. The integral operation in the formula globally quantifies the entire tracking path, and the path length It indicates that arrive The integration ensures that boundary tracking can comprehensively consider all local characteristic changes on the path, not just the characteristics of the starting and ending points. Overall, the boundary tracking function comprehensively quantifies the boundary correlation between pixels by integrating geometric curvature, multispectral features, gradient information, and height normalization correction. This multidimensional method that combines geometric and spectral features can achieve high-precision boundary detection and tracking in environments with complex terrain and diverse natural resource types. For example, at the junction of wetlands and cultivated land, due to the large spectral characteristic gradients of the two and the significant changes in boundary intensity, the value of the tracking function will be higher, thereby accurately identifying the continuity of the boundary. When the calculation is completed, the system filters the boundary tracking value by setting a threshold. If If the threshold value is within the range, two pixels are connected to form a boundary segment. After traversing all pixels, these boundary segments eventually form a complete boundary vector set, which provides a basis for the precise division and confirmation of natural resource areas.

[0046] Example 9: In step 1.2, according to the boundary tracking function, the final boundary set is generated as the preliminary boundary process specifically including: among all boundary vectors, if the starting point coordinates or the ending point coordinates of any boundary vector are the same as the starting point or the ending point coordinates of any boundary vector, the two boundary vectors are summarized as one boundary, and after traversing all boundary vectors, several boundaries are obtained; if the starting point coordinates or the ending point coordinates of any boundary vector are not the same as the starting point or the ending point coordinates of any other boundary vector, it is regarded as an isolated boundary and discarded; each retained boundary is taken as an element to generate the final boundary set.

[0047] Specifically, by traversing all boundary vectors, check one by one whether the starting point or end point coordinates of each boundary vector match the starting point or end point coordinates of other boundary vectors. If it is found that the starting point and end point of two boundary vectors are the same, or the end point of one boundary vector coincides with the starting point of another boundary vector, it can be considered that the two boundary segments are continuous in space and should be merged into a longer boundary segment. The core of the merge operation is to merge these adjacent boundary vectors according to the spatial connection relationship, so as to generate a larger and more complete boundary. This process is similar to the splicing of spatial paths. By checking the consistency of coordinate points, the boundary segments are ensured to be geometrically coherent and possible discontinuities are excluded. However, when traversing all boundary vectors, it is possible to encounter some boundary vectors whose starting point and end point cannot match any other boundary vector. These vectors are regarded as isolated boundaries, which are usually generated due to detection noise or local errors and have no practical significance. The existence of isolated boundaries will interfere with the clarity and practicality of the boundary set, so they need to be removed from the final boundary set. The rule for removing isolated boundaries is very clear, that is, check whether the starting point and end point of each boundary vector have a matching relationship with other vectors. If there is no matching relationship, it will be directly discarded. Through this screening step, meaningless isolated boundaries can be effectively removed, making the final generated boundary set more concise and of practical application value. After completing the merging and removal of isolated boundaries, the retained boundary vectors constitute the final boundary set. Each boundary segment has a clear spatial position and continuity, which can fully describe the boundary between different natural resource areas. These boundary sets provide a basic framework for the preliminary division of natural resources, forming a set of boundary networks in three-dimensional space that can reflect the distribution law of resources. Through this process, the boundary set is not only geometrically coherent, but also can truly reflect the differences in multispectral characteristics and terrain elevation of natural resources. For example, at the junction of wetlands and grasslands, the generation of boundary sets can accurately capture the transition zone between different vegetation types; and at the junction of mountains and plains, the boundary set can reflect the changes in resource distribution caused by terrain elevation differences. This generation process is further optimized with the support of a three-dimensional working base map. The three-dimensional working base map provides accurate terrain elevation and multispectral data, enabling the boundary tracking function to capture more detailed boundary characteristics, while the merging and filtering operations ensure the consistency and integrity of the generated boundary set in three-dimensional space by combining three-dimensional spatial coordinates. The final boundary set not only has clear boundaries on the two-dimensional plane, but also reflects the resource distribution law in three-dimensional space, providing a scientific basis for the division of natural resources under complex terrain. Through this process, the final boundary set provides high-quality data support for the intelligent mapping of natural resource rights. These boundary sets can be used to define the preliminary boundaries of different resource areas, providing a basis for subsequent regional mergers, optimization and rights confirmation, while also ensuring the scientificity and practicality of the mapping results.In a complex natural resource environment, this method of generating boundary sets through traversal, merging and screening can effectively cope with diverse resource types and complex terrain conditions, and improve the accuracy and efficiency of natural resource division.

[0048] Although the specific embodiments of the present invention are described above, it should be understood by those skilled in the art that these specific embodiments are only illustrative, and those skilled in the art may omit, replace, and change the details of the above methods and systems in various ways without departing from the principles and essence of the present invention. For example, merging the above method steps so as to perform substantially the same functions in substantially the same manner to achieve substantially the same results is within the scope of the present invention. Therefore, the scope of the present invention is limited only by the appended claims.

Claims

1. An intelligent mapping method for natural resource rights confirmation based on a three-dimensional working base map, characterized in that: The method comprises: Step 1: Detect and identify natural resources on a three-dimensional working base map, generate preliminary boundaries, and divide the three-dimensional working base map into a plurality of different natural resource areas based on the preliminary boundaries; the three-dimensional working base map is established by image data captured by remote sensing satellites and geographic data acquired by a GIS system; Step 2: In each natural resource area, terrain elevation conflict detection is performed to obtain the terrain elevation conflict line in each natural resource area, and the natural resource areas with terrain elevation conflicts are re-divided according to the terrain elevation conflict lines; Step 3: Determine an upper limit for the natural resource area. For each natural resource area, calculate the average elevation difference between it and adjacent natural resource areas of different types. If the average elevation difference is within the set threshold range, the two natural resource areas will be merged into one natural resource area. The area of ​​the merged natural area must be smaller than the determined upper limit for the natural resource area. The average elevation difference is defined as the absolute value of the difference between the average altitude of one natural resource area and the average altitude of another natural resource area.

2. The method for intelligent mapping of natural resource rights confirmation based on a three-dimensional working base map according to claim 1, characterized in that: In step 2, the specific process of performing terrain elevation conflict detection in each natural resource area includes: obtaining a contour distribution map of each natural resource area from the three-dimensional working base map; in the contour distribution map, if the height distance difference between any two adjacent contour lines exceeds a set threshold, it is judged that there is a terrain elevation conflict, and the contour line with the lower height of the two adjacent contour lines is used as the terrain elevation conflict line; the types of natural resource areas include: rivers, wetlands, forests, grasslands, cultivated land and deserts.

3. The method for intelligent mapping of natural resource rights confirmation based on a three-dimensional working base map as claimed in claim 2, characterized in that: Step 1 specifically includes: Step 1.1: Calculate the multispectral features of the three-dimensional working base map; based on the multispectral features, calculate the local feature tensor of each pixel in the three-dimensional working base map; Step 1.2: Based on the multispectral features and the local feature tensor of each pixel, the boundary feature strength of the three-dimensional working base map is calculated; based on the boundary feature strength, a boundary tracking function is constructed; according to the boundary tracking function, a final boundary set is generated as a preliminary boundary; Step 1.3: Based on the preliminary boundaries, divide the 3D working base map into several different natural resource areas.

4. The method for intelligent mapping of natural resource rights confirmation based on a three-dimensional working base map as claimed in claim 3 is characterized in that: The process of calculating the multispectral characteristics of the three-dimensional working base map in step 1.1 specifically includes: obtaining spectral data from the three-dimensional working map, and the spectral data specifically includes: near infrared band reflectivity, red light band reflectivity, short infrared band reflectivity, green light band reflectivity and blue light band reflectivity; and then calculating the multispectral characteristics of the three-dimensional working base map by the following formula: ； in, Indicates the coordinate position is The multispectral characteristic value of the pixel at ; Indicates the X-axis coordinate; Indicates the Y-axis coordinate; Indicates the Z-axis coordinate; is the surface temperature; is the surface emissivity; is the reflectivity in the near-infrared band; is the reflectivity of red light band; is the reflectivity in the short infrared band; is the reflectivity of green light band; is the reflectivity of blue light band; The upper limit of the wavelength of remote sensing satellites; is the lower limit of the wavelength of the remote sensing satellite; is the wavelength integration variable; is the spectral reflectance distribution function, indicating that the wavelength is Spectral reflectance at .

5. The method for intelligent mapping of natural resource rights confirmation based on a three-dimensional working base map as claimed in claim 4, characterized in that: Spectral reflectance distribution function Calculated by the following formula: ； in, is the radiance value of the remote sensing satellite; is the atmospheric radiation path value, and the calculation formula is: ； in, is the Rayleigh scattering reflectivity; is the aerosol scattering reflectivity; To observe the zenith angle; and They are the atmospheric transmittance in the observation direction and the sun direction respectively: is the solar spectrum irradiance, and the calculation formula is: ； in, is the spectral distribution of the solar constant; is the average distance between the sun and the earth; is the actual distance between the sun and the earth; is the atmospheric diffuse irradiance, and the calculation formula is: ； in, is the atmospheric diffuse transmittance; is the average surface reflectivity; is the albedo of the large balloon surface.

6. The method for intelligent mapping of natural resource rights confirmation based on a three-dimensional working base map according to claim 5, characterized in that: The process of calculating the local feature tensor of each pixel in the three-dimensional working base map based on multispectral features in step 1.1 specifically includes: ； in, Indicates the coordinate position is The local eigenvector of the pixel at ; is the radius of curvature of the earth’s surface; is the reference curvature radius.

7. The method for intelligent mapping of natural resource rights confirmation based on a three-dimensional working base map according to claim 6, characterized in that: In step 1.2, the boundary feature intensity of the three-dimensional working base map is calculated based on the multispectral features and the local feature tensor of each pixel using the following formula: ； in, Indicates the coordinate position is The boundary feature intensity of the pixel at ; for The gradient of is the absolute value operator.

8. The method for intelligent mapping of natural resource rights confirmation based on a three-dimensional working base map according to claim 7, characterized in that: In step 1.2, the boundary tracking function is constructed based on the boundary feature strength using the following formula: ； in, is the boundary tracking value; if Within the set threshold range, the coordinates are located at The pixel with coordinates is located at The adjacent pixels are connected to form a boundary segment; when all the pixels are connected, all the boundary segments are obtained, and each boundary segment is regarded as a boundary vector; is the geodesic curvature; is the normal curvature; is the normalized height value set; Indicates the coordinate position is The boundary feature intensity of the pixel at ; Indicates the coordinate position is The boundary feature intensity of the pixel at ; Indicates the coordinate position is The multispectral characteristic value of the pixel at ; Indicates the coordinate position is The multispectral characteristic value of the pixel at ; The coordinates are located at The pixel with coordinates is located at The distance between adjacent pixels.

9. The method for intelligent mapping of natural resource rights confirmation based on a three-dimensional working base map according to claim 8, characterized in that: In step 1.2, the final boundary set is generated according to the boundary tracking function. The process of the preliminary boundary specifically includes: among all boundary vectors, if the starting point coordinates or the ending point coordinates of any boundary vector are the same as the starting point coordinates or the ending point coordinates of any boundary vector, the two boundary vectors are summarized as one boundary. After traversing all boundary vectors, several boundaries are obtained; if the starting point coordinates or the ending point coordinates of any boundary vector are different from the starting point coordinates or the ending point coordinates of any other boundary vector, it is regarded as an isolated boundary and discarded; each retained boundary is taken as an element to generate the final boundary set.

Images