Cross-region multi-source data fusion and modular updating forest land sand table 3D printing method

Abstract

A method for 3D printing a forest sand table using cross-regional multi-source data fusion and modular updates includes the following steps: S1 Data acquisition and preliminary processing → S2 Digital Surface Model (DSM) data inspection and repair → S3 Determination of printing size and scale → S4 Precise trimming of forest area and multi-region mask extraction → S5 Creation of printing boundary transition raster matrix → S6 Spatial range of context base map → S7 Multi-source data fusion calculation → S8 Dynamic determination of printing range based on joint bounding box → S9 Final joint printing raster trimming → S10 Printing raster symbolization and export → S11 Image preprocessing before printing → S12 Generation of 3D printing model → S13 Model slicing and block printing → S14 Establishment of local update mechanism → S15 Model post-processing and decoration. This invention achieves a closed-loop process from multi-source heterogeneous data fusion calculation and standardized model slicing to rapid local replacement and maintenance, generating a new generation of realistic sand tables with natural transition boundaries, macro-geographical correlation, and high-fidelity micro-details.

Description

Technical Field

[0001] This invention relates to a method for printing sand tables, specifically a 3D printing method for forest sand tables that integrates cross-regional multi-source data and modular updates. Background Technology

[0002] Traditional woodland landscape model making relies entirely on manual methods, which has the following inherent defects: the production cycle is long, usually requiring several weeks or even months; the cost is high, heavily dependent on skilled craftsmen; the accuracy is limited, making it difficult to accurately reproduce complex terrain and landforms; the sense of realism is poor, and the expression of surface details (such as vegetation distribution and small land features) is highly subjective.

[0003] In recent years, although methods for creating terrain sand tables based on Digital Elevation Models (DEMs) and 3D printing technology have emerged, improving efficiency to some extent, significant shortcomings remain: 1. Limited data source: Similar to traditional manual methods, existing digital methods mostly rely on low-precision elevation data acquired through satellite remote sensing, making it difficult to capture detailed features such as forest cover, forest gaps, steep slopes, and small structures. The data richness is even less than that of detailed manual surveys. 2. Harsh model boundaries: Whether manually constructed or digitally cut, model boundaries are often treated as unnatural vertical cliffs, severely inconsistent with the smooth transitions of real terrain, resulting in poor aesthetics and practicality. 3. Lack of geographical context: The printed target area exists in isolation, lacking natural connection with the surrounding terrain. Like traditional sand tables, this makes it difficult for observers to establish accurate geospatial relationships. 4. Insufficient information capacity: Printed models are mostly monochrome, unable to reflect surface cover differences like hand-colored models, and lack key map annotations such as place names and scales, resulting in less overall information compared to high-quality handmade sand tables. V. Missing Key Features: For key features such as work areas and management stations, existing digital methods cannot automatically reconstruct their 3D forms, requiring post-production additions similar to manual creation, resulting in low automation. VI. Difficulty in Updating and Maintaining: Traditional sand tables or monolithically formed 3D printed sand tables have a high degree of integration. Once local forest resources or topography change (such as engineering construction or logging), the entire model often needs to be remade, making local dynamic updates impossible and causing a significant waste of resources and costs. Summary of the Invention

[0004] This invention provides a 3D printing method for forest sand table with cross-regional multi-source data fusion and modular updating, the main purpose of which is to overcome the above-mentioned problems existing in the prior art.

[0005] The present invention is implemented using the following technical solution: A method for 3D printing of forest sand tables with cross-regional multi-source data fusion and modular updates includes the following steps: S1. Data Acquisition and Preliminary Processing: Aerial photography of one or more relatively scattered target forest areas is carried out using a drone equipped with an oblique photogrammetry camera to obtain multi-view image data; then the images are processed by oblique photogrammetry 3D modeling software to generate digital surface model (DSM), digital orthophoto image (DOM), and 3D mesh model of key features for each forest area. S2. Digital Surface Model (DSM) Data Inspection and Repair: In GIS software, inspect the initially generated Digital Surface Model (DSM) data to identify voids or outliers caused by occlusion, water surfaces, etc.; then use interpolation or manual editing to repair the abnormal areas and obtain a complete Digital Surface Model (DSM). S3. Determining the printing size and scale: Based on the physical space constraints of the actual placement location of the sand table and the needs of forestry display, determine the overall physical printing size x of the sand table, and calculate the final scale y by combining it with the actual geographical range of the forest land. S4. Precise trimming of forest land area and multi-region mask extraction: For n discrete target forest land plots, obtain their corresponding vector redline data set { L 1, L 2,……, L n}, each vector red line L k (in k =1,2,……, n Transform it into the corresponding binary mask matrix M k Subsequently, each mask matrix is ​​overlaid with the repaired Digital Surface Model (DSM) data, and mask functions are used to extract the core area raster data limited to each forest area one by one. A k ; S5. Creating the Print Boundary Transition Raster Matrix: Combine the above vector red line data set { L 1, L 2,……, L n The edge lines of} are rasterized to generate a set of raster matrices that identify the printing boundaries. B 1, B 2,……, B n Let the scale of the sand table be... y The physical height constant of the boundary subsidence is h 0, then for the first k The boundary grid of each plot B k The formula for assigning values ​​to specific pixels is as follows: ; Among them, Boundary ( Lk ) represents the set of pixels covered by the k-th forest redline vector boundary after equidistant rasterization transformation; S6. Spatial Limitation of Context Base Map: Acquire large-area satellite remote sensing DEM data that can completely cover all target forest plots as a unified terrain transition context base map, denoted as a global raster. C ; S7. Multi-source data fusion calculation: Clearly define the area to be printed. n It consists of several independent woodland plots, with the first... k High-precision raster data for each forest area A k The corresponding print boundary transition grid is B k Before merging, a global merge operation is performed to obtain the complete set of the core area. A And the complete collection of borders B : ; ; Using a raster calculator, the entire set of merged core regions is... A And the complete collection of borders B With global context basemap C Perform pixel-by-pixel fusion and construct a composite function. D ( i , j ); S8. Dynamic determination of printing range based on joint bounding boxes: To ensure that all discrete forest land areas are within the bounding box... A 1, A 2,……, A n The core area and its surrounding geographical context can be included in the final sandbox; the complete set of the core area needs to be calculated. A Let the minimum bounding rectangle be... A The extreme values ​​of the coordinates of all pixels in the projected coordinate system are X min , X max , Y min , Y max Determine the square vector frame E， The side length of the final printable area L Must meet: L =max( X max - X min , Y max -Y min )+2 δ ; in δ Allow sufficient width for the geographic context buffer, and ensure that the vector box is within acceptable limits. E spatial range S E Completely contained in the composite grid D Within the valid domain, i.e. S E ⊆Domain( D ); S9. Final joint printing grid clipping: using the vector frame determined above. E For composite rasters that integrate multi-source data D Spatial clipping is performed to obtain the final printed raster used for modeling. F : F =Crop( D , E ); F It is a continuous raster matrix containing all target woodland plots, subsidence boundary transition zones, and satellite remote sensing base maps, with a resolution equal to that of the entire core area. A Maintain consistency; S10. Printed Raster Symbolization and Export: To convert geospatial elevation values ​​into grayscale intensities recognizable by the 3D model, the raster is symbolized and exported. F Perform global normalization processing, F The true elevation values ​​in the image are linearly mapped to the interval [0, 255] to generate a GeoTIFF format image. P ; S11. Image preprocessing before printing: In image processing software, process the image... P Converted into a final displacement intensity map with physical constraints and information annotations. H ; S12. 3D Printing Model Generation: In 3D graphics software, associate grids. H The intensity map is used for mesh replacement to generate a 3D printable model. S13, Model Slicing and Block Printing: Import the 3D printing model file into the slicing software, scale the model according to the scale y, and use a standard grid to regularly divide the overall sand table model according to the size of the 3D printer's molding chamber or transportation requirements to form multiple independent square printing modules. Then, print the key ground feature models in step S1 separately, and then generate the corresponding G-code instructions to complete the block printing of each module. S14. Establish a local update mechanism: When a local area needs to be updated, there is no need to recreate the global sandbox. Only steps S1 to S12 need to be re-executed for the specific changed area to obtain the latest spatial fusion data and generate the corresponding 3D printing model. Then, based on the coordinate index of the original standard grid, the affected local modules are printed and replaced separately. S15. Model post-processing and decoration: Color the main model blocks according to the digital orthophoto DOM, and then assemble the independently printed key feature models into the corresponding positions; finally, insert marker flags to mark special business areas to complete the final assembly.

[0006] Furthermore, in step S4, A k The mathematical expression is: A k ( i , j =DSM( i , j )× M k ( i , j ); in,( i , j ) represents the planar coordinates of the raster pixel; when ( i , j ) located in the k When within the forest land vector red line range M k ( i , j )=1, otherwise M k ( i , j If the value is 0 or null, this step yields a set of high-precision raster data for each forest area. A 1, A 2,……, A n Furthermore, the resolution of all models remains consistent with the original digital surface model, and the same projection coordinate system is used.

[0007] Furthermore, in step S6, to ensure a natural transition, the global grid... C spatial coverage S C Must meet: ; in, The sum of the spatial extent of all n target forest plots. To ensure the outward expansion of the buffer distanceS C The area shall not be less than 1 times the total target area.

[0008] Furthermore, in step S7, to reflect the logic of concurrent fusion of multiple land parcels, the following composite function is constructed. D ( i , j ): ; Among them, Domain( A k (Refers to the first) k A localized core area of ​​forest land grid A k The effective spatial range (pixel coordinates of effective elevation values ​​(non-null)) i , j The set of ), similarly Domain( B k (Refers to local boundary grid) B k The effective spatial range.

[0009] Furthermore, in step S10, the mapping formula is: ; in, F global_max and F global_min They are grids F The maximum and minimum elevation values ​​within the range are determined. This step ensures that the vertical scale of multiple discrete plots is absolutely consistent within the same sand table, avoiding elevation misalignment caused by segmentation.

[0010] Furthermore, in step S11, a preset information annotation area is defined. R info The set of pixel coordinates is S info Then paste the image. H The pixel assignment logic is as follows: ; Among them, Edge ( P ,1) is P The surrounding physical outer boundary is 1 pixel wide.

[0011] Furthermore, in step S12, to ensure that the digital model and the final physical model are completely consistent in physical space, the calculation formulas for the relevant forming parameters are as follows: Replacement strength parameters n : n=2( F global_max - F global_min ) y / x； Base plate thickness parameters z : To ensure the overall strength of the sand table, a cube is added as a base plate, with the following thickness parameters: z With print size x Inversely proportional mapping relationship: z = h base / x ; Finally, export the completed model with text and scale engraving as STL / obj / gly format files.

[0012] As can be seen from the above description of the present invention, compared with the prior art, the present invention has the following advantages: 1. Strong overall and macro-level display capabilities: Through global GIS calculations, multiple discontinuous low-altitude high-precision UAV data are seamlessly integrated with large-scale, low-cost satellite remote sensing base maps; overcoming the display challenges of cross-regional management for forestry management units, showcasing macro-geographical relationships and topographical trends in the same sand table while preserving high-precision details of the core management area, significantly enhancing the value of business guidance.

[0013] 2. Supports rapid local updates and low-cost maintenance: The use of standardized grid segmentation and modular printing technology not only facilitates the disassembly and transportation of the sand table, but also allows for the collection of data on the changed areas and the printing of individual modules to replace local modules when dealing with "hot spots" of terrain changes. This avoids the need to rebuild the entire sand table and achieves efficient synchronization between data and entities.

[0014] 3. Natural transition boundary: The vertical boundary is transformed into a natural descending slope through grid calculation logic, which avoids the "island feeling" when multiple plots are integrated and enhances the realism and aesthetics of the sand table.

[0015] 4. Absolutely consistent physical scale: Through rigorous mathematical bounding box definition and global elevation mapping algorithm, it is ensured that different discretely distributed plots enjoy an absolutely consistent horizontal and vertical scale on the same physical sand table, which is convenient for actual engineering measurement.

[0016] 5. High degree of automation: From data fusion to physical border generation, it is supported by rigorous mathematical expressions, supports batch processing and scripted operation, and improves the efficiency and repeatability of large-scale complex forest mapping. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the process of the present invention.

[0018] Figure 2 This is a schematic diagram illustrating the principle of multi-source data composite calculation to generate the result in step S7 of the present invention.

[0019] Figure 3 This is a schematic diagram of the preprocessing effect of step S11 of the present invention, where box A shows the outer border of the printout and box B shows the reserved space for information labeling. Detailed Implementation

[0020] Reference Figures 1 to 3 A method for 3D printing of forest sand table with cross-regional multi-source data fusion and modular updates includes the following steps: S1. Data Acquisition and Preliminary Processing: Aerial photography of one or more relatively dispersed target forest areas is carried out using a drone equipped with an oblique photogrammetry camera to obtain multi-view image data; then the images are processed by oblique photogrammetry 3D modeling software to generate digital surface model (DSM), digital orthophoto image (DOM), and 3D mesh model (OBJ format) of key features (such as houses, towers, etc. for key forestry land) for each forest area.

[0021] S2. Digital Surface Model (DSM) Data Inspection and Repair: In GIS software, inspect the initially generated Digital Surface Model (DSM) data to identify voids or outliers caused by occlusion, water surfaces, etc.; then use interpolation or manual editing to repair the abnormal areas and obtain a complete Digital Surface Model (DSM).

[0022] S3. Determining the printing size and scale: Based on the physical space constraints of the actual placement location of the sand table and the needs of forestry display, determine the overall physical printing size x of the sand table, and calculate the final scale y by combining it with the actual geographical range of the forest land.

[0023] S4. Precise trimming of forest land area and multi-region mask extraction: For n discrete target forest land plots (such as different forest compartments, management areas, or hotspot areas), obtain their corresponding vector redline data sets { L 1, L 2,……, L n}, each vector red line L k (in k =1,2,……, n Transform it into the corresponding binary mask matrix M k Subsequently, each mask matrix is ​​overlaid with the repaired Digital Surface Model (DSM) data, and mask functions are used to extract the core area raster data limited to each forest area one by one. A k , A k The mathematical expression is: A k ( i , j =DSM( i , j )× M k ( i , j ); in,( i , j ) represents the planar coordinates of the raster pixel; when ( i , j ) located in the k When within the forest land vector red line range M k ( i , j )=1, otherwise M k ( i , j If the value is 0 or null, this step yields a set of high-precision raster data for each forest area. A 1, A 2,……, A n Furthermore, the resolution of all models remains consistent with the original digital surface model, and the same projection coordinate system is used.

[0024] S5. Creating the Boundary Transition Raster Matrix: To achieve a natural sinking and smooth transition of multiple discrete forest plots on the sand table, a boundary raster needs to be generated separately for each plot. This involves combining the aforementioned vector redline data set { L 1, L 2,……, L n The edge lines of} are rasterized to generate a set of raster matrices that identify the printing boundaries. B 1, B 2,……, B n Let the scale of the sand table be... y The physical height constant of the boundary subsidence is h 0, then for the first k The boundary grid of each plot B k The formula for assigning values ​​to specific pixels is as follows: ; Among them, Boundary ( L k ) represents the set of pixels covered by the k-th forest redline vector boundary after equidistant rasterization transformation.

[0025] S6. Spatial Limitation of Context Base Map: Acquire large-area satellite remote sensing DEM data that can completely cover all target forest plots as a unified terrain transition context base map, denoted as a global raster. C To ensure a smooth transition, the global grid... C spatial coverage S C Must meet: ; in, The sum of the spatial extent of all n target forest plots. To ensure the outward expansion of the buffer distance S C The area shall not be less than 1 times the total target area.

[0026] S7. Multi-source data fusion calculation: Clearly define the area to be printed. n It consists of several independent woodland plots, with the first... k High-precision raster data for each forest area A k The corresponding print boundary transition grid is B k Before merging, a global merge operation is performed to obtain the complete set of the core area. A And the complete collection of borders B : ; ; Using a raster calculator, the entire set of merged core regions is... A And the complete collection of borders B With global context basemap C Perform pixel-by-pixel fusion and construct a composite function. D ( i , j ); To reflect the logic of concurrent integration of multiple land parcels, the following composite function is constructed. D ( i , j ): ; Among them, Domain( A k (Refers to the first) k A localized core area of ​​forest land grid A k The effective spatial range (pixel coordinates of effective elevation values ​​(non-null)) i , j The set of ), similarly Domain( Bk (Refers to local boundary grid) B k The effective spatial range.

[0027] S8. Dynamic determination of printing range based on joint bounding boxes: To ensure that all discrete forest land areas are within the bounding box... A 1, A 2,……, A n The core area and its surrounding geographical context can be included in the final sandbox; the complete set of the core area needs to be calculated. A Let the minimum bounding rectangle be... A The extreme values ​​of the coordinates of all pixels in the projected coordinate system are X min , X max , Y min , Y max Determine the square vector frame E， The side length of the final printable area L Must meet: L =max( X max - X min , Y max - Y min )+2 δ ; in δ Allow sufficient width for the geographic context buffer, and ensure that the vector box is within acceptable limits. E spatial range S E Completely contained in the composite grid D Within the valid domain, i.e. S E ⊆Domain( D ); S9. Final joint printing grid clipping: using the vector frame determined above. E For composite rasters that integrate multi-source data D Spatial clipping is performed to obtain the final printed raster used for modeling. F : F =Crop( D , E ); F It is a continuous raster matrix containing all target woodland plots, subsidence boundary transition zones, and satellite remote sensing base maps, with a resolution equal to that of the entire core area. A Maintain consistency; S10. Printed Raster Symbolization and Export: To convert geospatial elevation values ​​into grayscale intensities recognizable by the 3D model, the raster is symbolized and exported. F Perform global normalization processing, F The true elevation values ​​in the image are linearly mapped to the interval [0, 255] to generate a GeoTIFF format image. P Its mapping formula is: ; in, F global_max and F global_min They are grids F The maximum and minimum elevation values ​​within the range are determined. This step ensures that the vertical scale of multiple discrete plots is absolutely consistent within the same sand table, avoiding elevation misalignment caused by segmentation.

[0028] S11. Image preprocessing before printing: In image processing software, process the image... P Converted into a final displacement intensity map with physical constraints and information annotations. H Preset information labeling area R info The set of pixel coordinates is S info Then paste the image. H The pixel assignment logic is as follows: ; Among them, Edge ( P ,1) is P The surrounding physical outer boundary is 1 pixel wide.

[0029] S12. 3D Printing Model Generation: In 3D graphics software, associate grids. H The intensity map is used for mesh replacement to generate a 3D printing model. To ensure that the digital model and the final physical model are completely consistent in physical space, the calculation formulas for the relevant molding parameters are as follows: Replacement strength parameters n : n =2( F global_max - F global_min ) y / x； Base plate thickness parameters z : To ensure the overall strength of the sand table, a cube is added as a base plate, with the following thickness parameters: z With print size x Inversely proportional mapping relationship:z = h base / x ; Finally, export the completed model with text and scale engraving as STL / obj / gly format files.

[0030] S13, Model Slicing and Block Printing: Import the 3D printing model file into the slicing software, scale the model according to the scale bar y, and use a standard grid to regularly divide the overall sand table model according to the size of the 3D printer's molding chamber or transportation requirements to form multiple independent square printing modules. Then, print the key feature models in step S1 separately, and then generate the corresponding G-code instructions to complete the block printing of each module.

[0031] S14. Establish a local update mechanism: For "hotspot areas" in the forest area where the topography and landform have changed due to logging, engineering construction or natural disasters, a dynamic update process is established. When a local area needs to be updated, there is no need to recreate the global sand table. Only steps S1 to S12 need to be repeated for the specific changed area to obtain the latest spatial fusion data and generate the corresponding 3D printing model. Then, based on the coordinate index of the original standard grid, the affected local modules are printed and replaced separately, thereby achieving low-cost and high-efficiency synchronous updates between the sand table entity and the actual forest stand status.

[0032] S15. Model post-processing and decoration: Color the main model blocks according to the digital orthophoto DOM, and then assemble the independently printed key feature models into the corresponding positions; finally, insert marker flags to mark special business areas to complete the final assembly.

[0033] Through the above-described methods and steps, the present invention can achieve the following technical effects: 1. Strong overall and macro-level display capabilities: Through global GIS calculations, multiple discontinuous low-altitude high-precision UAV data are seamlessly integrated with large-scale, low-cost satellite remote sensing base maps; overcoming the display challenges of cross-regional management for forestry management units, showcasing macro-geographical relationships and topographical trends in the same sand table while preserving high-precision details of the core management area, significantly enhancing the value of business guidance.

[0034] 2. Supports rapid local updates and low-cost maintenance: The use of standardized grid segmentation and modular printing technology not only facilitates the disassembly and transportation of the sand table, but also allows for the collection of data on the changed areas and the printing of individual modules to replace local modules when dealing with "hot spots" of terrain changes. This avoids the need to rebuild the entire sand table and achieves efficient synchronization between data and entities.

[0035] 3. Natural transition boundary: The vertical boundary is transformed into a natural descending slope through grid calculation logic, which avoids the "island feeling" when multiple plots are integrated and enhances the realism and aesthetics of the sand table.

[0036] 4. Absolutely consistent physical scale: Through rigorous mathematical bounding box definition and global elevation mapping algorithm, it is ensured that different discretely distributed plots enjoy an absolutely consistent horizontal and vertical scale on the same physical sand table, which is convenient for actual engineering measurement.

[0037] 5. High degree of automation: From data fusion to physical border generation, it is supported by rigorous mathematical expressions, supports batch processing and scripted operation, and improves the efficiency and repeatability of large-scale complex forest mapping.

[0038] The above are merely specific embodiments of the present invention, but the design concept of the present invention is not limited thereto. Any non-substantial modifications made to the present invention using this concept shall be considered as infringing upon the protection scope of the present invention.

Claims

1. A method for 3D printing of forest land sand tables with cross-regional multi-source data fusion and modular updating, characterized by: Includes the following steps: S1. Data Acquisition and Preliminary Processing: Using a drone equipped with an oblique photography camera, aerial photography is conducted on one or more relatively scattered target woodlands to acquire multi-view image data. Then, the images were processed using oblique photogrammetry 3D modeling software to generate digital surface models (DSM), digital orthophoto images (DOM), and 3D mesh models of key features for each forest plot. S2. Digital Surface Model (DSM) Data Inspection and Repair: In GIS software, inspect the initially generated Digital Surface Model (DSM) data to identify voids or outliers caused by occlusion, water surfaces, etc.; then use interpolation or manual editing to repair the abnormal areas and obtain a complete Digital Surface Model (DSM). S3. Determining the printing size and scale: Based on the physical space constraints of the actual placement location of the sand table and the needs of forestry display, determine the overall physical printing size x of the sand table, and calculate the final scale y by combining it with the actual geographical range of the forest land. S4. Precise trimming of forest land area and multi-region mask extraction: For n discrete target forest land plots, obtain their corresponding vector redline data set { L 1, L 2,……, L n }, each vector red line L k (in k =1,2,……, n Transform it into the corresponding binary mask matrix M k Subsequently, each mask matrix is ​​overlaid with the repaired Digital Surface Model (DSM) data, and mask functions are used to extract the core area raster data limited to each forest area. A k ; S5. Creating the Print Boundary Transition Raster Matrix: Combine the above vector red line data set { L 1, L 2,……, L n The edge lines of} are rasterized to generate a set of raster matrices that identify the printing boundaries. B 1, B 2,……, B n }, Let the scale of the sand table be... y The physical height constant of the boundary subsidence is h 0, then for the first k The boundary grid of each plot B k The formula for assigning values ​​to specific pixels is as follows: ； Among them, Boundary ( L k ) represents the set of pixels covered by the k-th forest redline vector boundary after equidistant rasterization transformation; S6. Spatial Limitation of Context Base Map: Acquire large-area satellite remote sensing DEM data that can completely cover all target forest plots as a unified terrain transition context base map, denoted as a global raster. C ; S7. Multi-source data fusion calculation: Clearly define the area to be printed. n It consists of several independent woodland plots, with the first... k High-precision raster data for each forest area A k The corresponding print boundary transition grid is B k Before merging, a global merge operation is performed to obtain the complete set of the core area. A And the complete collection of borders B : ； ； Using a raster calculator, the entire set of the merged core regions is... A And the complete collection of borders B With global context basemap C Perform pixel-by-pixel fusion and construct a composite function. D ( i , j ); S8. Dynamic determination of printing range based on joint bounding box: To ensure that all discrete forest land areas are within the bounding box... A 1, A 2,……, A n The core area and its surrounding geographical context can be included in the final sandbox; the complete set of the core area needs to be calculated. A Let the minimum bounding rectangle be... A The extreme values ​​of the coordinates of all pixels in the projected coordinate system are X min , X max , Y min , Y max Determine the square vector frame E， The side length of the final printable area L Must meet: L =max( X max - X min , Y max - Y min )+2 δ ; in δ Allow sufficient width for the geographic context buffer, and ensure that the vector box is within acceptable limits. E spatial range S E Completely contained in the composite grid D Within the valid domain, i.e. S E ⊆Domain( D );in δ Allow sufficient width for the geographic context buffer, and ensure that the vector box is within acceptable limits. E spatial range S E Completely contained in the composite grid D Within the valid domain, i.e. S E ⊆Domain( D ); S9. Final joint printing grid clipping: using the vector frame determined above. E For composite rasters that integrate multi-source data D Spatial clipping is performed to obtain the final printed raster used for modeling. F : F =Crop( D , E ); F It is a continuous raster matrix containing all target woodland plots, subsidence boundary transition zones, and satellite remote sensing base maps, with a resolution equal to that of the entire core area. A Maintain consistency; S10. Printed Raster Symbolization and Export: To convert geospatial elevation values ​​into grayscale intensities recognizable by the 3D model, the raster is symbolized and exported. F Perform global normalization processing, F The true elevation values ​​in the image are linearly mapped to the interval [0, 255] to generate a GeoTIFF format image. P ; S11. Image preprocessing before printing: In image processing software, process the image... P Converted into a final displacement intensity map with physical constraints and information annotations. H ; S12. 3D Printing Model Generation: In 3D graphics software, associate grids. H The intensity map is used for mesh replacement to generate a 3D printable model. S13, Model Slicing and Block Printing: Import the 3D printing model file into the slicing software, scale the model according to the scale y, and use a standard grid to regularly divide the overall sand table model according to the size of the 3D printer's molding chamber or transportation requirements to form multiple independent square printing modules. Then, print the key ground feature models in step S1 separately, and then generate the corresponding G-code instructions to complete the block printing of each module. S14. Establish a local update mechanism: When a local area needs to be updated, there is no need to recreate the global sandbox. Only steps S1 to S12 need to be re-executed for the specific changed area to obtain the latest spatial fusion data and generate the corresponding 3D printing model. Then, based on the coordinate index of the original standard grid, the affected local modules are printed and replaced separately. S15. Model Post-processing and Decoration: Based on the digital orthophoto DOM, the main model is divided into blocks and colored. Then, the independently printed key feature models are assembled into their corresponding positions. Finally, signage flags are inserted for special business areas. Label and complete the final assembly.

2. The 3D printing method for forest land sand table with cross-regional multi-source data fusion and modular updating as described in claim 1, characterized in that: In step S4 A k The mathematical expression is: A k ( i , j )=DSM( i , j )× M k ( i , j ); in,( i , j ) represents the planar coordinates of the raster pixel; when ( i , j ) located in the k When within the forest land vector red line range M k ( i , j )=1, otherwise M k ( i , j If the value is 0 or null, this step yields a set of high-precision raster data for each forest area. A 1, A 2,……, A n Furthermore, the resolution of all models remains consistent with the original digital surface model, and the same projection coordinate system is used.

3. The 3D printing method for forest land sand table with cross-regional multi-source data fusion and modular updating as described in claim 1, characterized in that: In step S6, to ensure a natural transition, the global grid... C spatial coverage S C Must meet: ； in, The sum of the spatial extent of all n target forest plots. To ensure the outward expansion of the buffer distance S C The area shall not be less than 1 times the total target area.

4. The 3D printing method for forest land sand table with cross-regional multi-source data fusion and modular updating as described in claim 1, characterized in that: In step S7, to reflect the logic of concurrent fusion of multiple land parcels, the following composite function is constructed. D ( i , j ): ; Among them, Domain( A k (Refers to the first) k A localized core area of ​​forest land grid A k The effective spatial range, similarly, Domain( B k (Refers to local boundary grid) B k The effective spatial range.

5. The 3D printing method for forest land sand table with cross-regional multi-source data fusion and modular updating as described in claim 1, characterized in that: In step S10, the mapping formula is: ； in, F global_max and F global_min They are grids F The maximum and minimum elevation values ​​within the range are determined. This step ensures that the vertical scale of multiple discrete plots is absolutely consistent within the same sand table, avoiding elevation misalignment caused by segmentation.

6. The 3D printing method for forest land sand table with cross-regional multi-source data fusion and modular updating as described in claim 1, characterized in that: In step S11, a preset information labeling area is defined. R info The set of pixel coordinates is S info Then paste the image. H The pixel assignment logic is as follows: ； Among them, Edge ( P ,1) is P The surrounding physical outer boundary is 1 pixel wide.

7. The 3D printing method for forest land sand table with cross-regional multi-source data fusion and modular updating as described in claim 5, characterized in that: In step S12, to ensure that the digital model and the final physical model are completely consistent in physical space, the calculation formulas for the relevant forming parameters are as follows: Replacement strength parameters n : n =2( F global_max - F global_min ) y / x； Base plate thickness parameters z : To ensure the overall strength of the sand table, a cube is added as a base plate, with the following thickness parameters: z With print size x Inversely proportional mapping relationship: z = h base / x ; Finally, export the completed model with text and scale engraving as STL / obj / gly format files.

Images