Fusion optimization method and system for DEM with different precisions

Abstract

The invention provides a fusion optimization method and system for DEMs with different precisions, and the method comprises the steps: generating an elevation correction face through a ground control point, and carrying out the elevation datum conversion of a low-precision LDEM into a GLDEM; establishing a boundary optimization transition region, resampling the GLDEM to a high-precision HDEM based on a transition region range, and outputting a GHDEM; performing terrain mosaic on the HDEM and the GHDEM by keeping high-precision elevation preferentially, and generating a PDEM to be subjected to smooth correction; for an area with a height difference step problem at the boundary of the PDEM, deleting a discontinuous part, and performing filling and Gaussian smoothing optimization processing on the deleted area by using bilinear interpolation; seamless and continuous DEM fusion tiles are output for digital twin system application. According to the method, the problems of inconsistent elevation, abrupt boundary, discontinuous data and the like during different elevation references and low-precision and high-precision DEM data fusion in a drainage basin digital twinning system are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of watershed digital twin applications, and in particular to a fusion optimization method and system for DEMs with different accuracies. Background Art

[0002] Digital Elevation Model (DEM) is an important basic data for terrain modeling and spatial analysis in digital twin systems. It is widely used in watershed hydrological simulation, disaster warning, environmental monitoring and other fields. Digital twin systems require accurate digital modeling of the real world. Especially in watershed digital twin systems, high-precision DEM data is crucial for simulation, analysis and decision-making. However, the acquisition of watershed DEM usually relies on multiple data sources, including satellite remote sensing, aerial photography, ground measurement, etc. These data sources have different spatial resolutions and elevation benchmarks. If the elevation benchmark and spatial resolution are very different, direct fusion will cause obvious stepped abnormal terrain, affecting the visualization and consistency of the fused DEM. Digital twin systems require seamless fusion of DEM data, especially in the refined simulation of watersheds, where boundary transitions, seams and height difference steps have a more significant impact on the results. Especially in microscopic scenes (such as watershed modeling in a small range) and close-up views, seam defects between data are easily exposed, causing significant visual and analytical errors. How to efficiently integrate and optimize these data to meet the high-precision requirements of the watershed digital twin system has become a key issue in the current technology field.

[0003] In order to solve the above problems, some technical methods have been proposed, but there are still some shortcomings. For example, the invention patent CN114612806A proposes to use elevation control points to correct the elevation of DEM data. This method requires the collection of a large number of high-precision elevation control points. If the distribution of the control points is uneven or the number of control points is small, the correction effect may be affected, resulting in a decrease in the fusion effect; the invention patent CN110874613A proposes to assign different weights to DEM data from different sources for weighted averaging. When using the weighted method to correct DEM, the elevation values ​​of the boundary area are often affected by the differences between different data sources. In particular, when the boundary elevation difference is too large, elevation steps or non-smooth transitions may still occur. Especially when fusing low-precision and high-precision data, the transition in the boundary area may produce significant errors; in addition, the invention patent CN112084280A proposes to use terrains of different scales for interpolation and fusion. The interpolation method essentially relies on existing elevation data to infer the elevation values ​​of unknown points, which is often suitable for situations where the elevation difference of the terrain is small. When the difference in elevation datum and resolution is very large, the interpolation effect in the boundary area is often poor, and unnatural transitions or elevation steps will still occur in the boundary area.

[0004] Therefore, a new method needs to be proposed to achieve effective fusion of DEMs with different accuracies. Summary of the invention

[0005] The present invention provides a fusion optimization method and system for DEMs of different precisions, which is used to solve the defect that DEMs of different precisions in the existing technology of watershed digital twin cannot achieve good fusion conversion and optimization, so as to achieve high-precision fusion of DEM data of different resolutions and benchmarks with only a small number of control points, especially for the case of very large elevation differences, the step effect of the boundary area is smoothed, and finally a seamless and continuous high-precision digital elevation model is generated.

[0006] In a first aspect, the present invention provides a fusion optimization method for DEMs of different accuracies, comprising: Acquire preset high-precision DEM data of a first elevation benchmark and preset low-precision DEM data of a second elevation benchmark; Generate an elevation correction face based on the ground control points, convert the elevation benchmark corresponding to the preset low-precision DEM data into an elevation benchmark consistent with the preset high-precision DEM data, and generate a corrected low-precision DEM data; Extract a specified high-precision range from the preset high-precision DEM data, establish a boundary optimization transition zone, resample the modified low-precision DEM data based on the boundary optimization transition zone, and generate modified high-precision DEM data with a resolution consistent with the preset high-precision DEM data; The modified high-precision DEM data and the preset high-precision DEM data are terrain mosaicked by retaining a high-precision priority sampling method to generate DEM data to be smoothed and corrected; Deleting discontinuous areas where there are height difference step problem areas at the boundaries of the DEM data to be smoothed and corrected, filling the deleted areas with a bilinear interpolation algorithm, and removing discontinuities and noise in the terrain with Gaussian smoothing optimization to generate DEM data with smooth boundary transitions; Output seamless and continuous DEM fusion tiles for application in digital twin systems.

[0007] According to a fusion optimization method for DEMs of different accuracies provided by the present invention, an elevation correction surface is generated based on ground control points, and the elevation benchmark corresponding to the preset low-precision DEM data is converted into an elevation benchmark consistent with the preset high-precision DEM data to generate corrected low-precision DEM data, including: According to the ground control point containing the first elevation, a control point vector layer is established, and a new field is created to store the first elevation control point; Loading the preset low-precision DEM data that needs to be corrected for elevation, ensuring that the preset low-precision DEM data is consistent with the projection coordinate system of the first elevation control point; Extracting the geodetic elevation value corresponding to the control point on the preset low-precision DEM data using the coordinates of the first elevation control point, and storing it in the control point vector layer field; Performing field operations on the control point vector layer, subtracting the geodetic elevation value corresponding to the preset low-precision DEM data extracted by the control point from the elevation value of the first elevation control point to obtain an elevation difference; Generate an elevation-modified front layer having the same range and resolution as the preset low-precision DEM data by interpolation according to the elevation difference; The corresponding layer of the preset low-precision DEM data and the elevation-corrected front layer are subjected to a raster value overlay operation to generate the corrected low-precision DEM data.

[0008] According to a fusion optimization method for DEMs of different accuracies provided by the present invention, a specified high-precision range is extracted from the preset high-precision DEM data, a boundary optimization transition zone is established, and the corrected low-precision DEM data is resampled based on the boundary optimization transition zone to generate corrected high-precision DEM data with the same resolution as the preset high-precision DEM data, including: Extracting the outer boundary range from the preset high-precision DEM data to generate a first vector range surface; Performing a buffer zone analysis based on the first vector range surface to obtain the boundary optimization transition zone after the first preset distance is expanded, and generating a second vector range surface including a preset high-precision DEM data range and a transition zone range; The modified low-precision DEM data is clipped and resampled based on the second vector range surface to regenerate the modified high-precision DEM data with a resolution consistent with the preset high-precision DEM data.

[0009] According to a fusion optimization method for DEMs of different accuracies provided by the present invention, the modified high-precision DEM data and the preset high-precision DEM data are terrain mosaicked by retaining a high-precision priority sampling method to generate DEM data to be smoothed and corrected, including: Add the preset high-precision DEM data and the revised high-precision DEM data respectively, and superimpose the layer corresponding to the preset high-precision DEM data on the layer corresponding to the revised high-precision DEM data; The preset high-precision DEM data and the corrected high-precision DEM data are subjected to terrain mosaic processing by adopting an upper layer sampling method, and are merged into the DEM data to be smoothed and corrected.

[0010] According to a fusion optimization method for DEMs of different precisions provided by the present invention, the upper layer sampling method includes: The elevation of the overlapping area is taken according to the first terrain layer, and the elevation of the non-overlapping area is taken according to each terrain layer.

[0011] According to a fusion optimization method for DEMs of different accuracies provided by the present invention, a discontinuous area is deleted from the area with height difference step problem at the boundary of the DEM data to be smoothed and corrected, a bilinear interpolation algorithm is used to fill the deleted area, and a Gaussian smoothing optimization is used to remove discontinuities and noise in the terrain to generate DEM data with smooth boundary transition, including: Based on the first vector range plane moving inward by a second preset distance, a first smooth boundary line on the side of a preset high-precision terrain area is generated; Based on the first vector range plane moving outward by a third preset distance, a second smooth boundary line on the side of the preset low-precision terrain area is generated; Within the interval between the first smooth boundary line and the second smooth boundary line, the existing height difference step area is segmented and deleted, wherein the segmentation of the mountainous area is based on the slope change points, and the segmentation of the flat area is based on different elevation planes; The deleted blank areas are filled by bilinear interpolation algorithm based on local terrain features, so that the terrain with step boundaries can be smoothly transitioned to generate seamless and continuous terrain. The interpolated terrain is subjected to Gaussian smoothing optimization processing to remove discontinuities and noises in the terrain, thereby generating the boundary smooth transition DEM data.

[0012] According to a fusion optimization method for DEMs of different precisions provided by the present invention, seamless and continuous DEM fusion tiles are output, including: The slicing tool is used to perform fusion slicing, and terrain slicing is performed according to the layer corresponding to the boundary smooth transition DEM data and the layer corresponding to the corrected low-precision DEM data, and the DEM fused tile is output.

[0013] In a second aspect, the present invention further provides a fusion optimization system for DEMs of different accuracies, comprising: An acquisition module, used to acquire preset high-precision DEM data of a first elevation benchmark and preset low-precision DEM data of a second elevation benchmark; A conversion module is used to generate an elevation correction face based on ground control points, convert the elevation benchmark corresponding to the preset low-precision DEM data into an elevation benchmark consistent with the preset high-precision DEM data, and generate a corrected low-precision DEM data; Establishing a module, used for extracting a specified high-precision range from the preset high-precision DEM data, establishing a boundary optimization transition zone, resampling the corrected low-precision DEM data based on the boundary optimization transition zone, and generating corrected high-precision DEM data with a resolution consistent with the preset high-precision DEM data; A mosaic module is used to perform terrain mosaicking on the corrected high-precision DEM data and the preset high-precision DEM data by retaining a sampling method that prioritizes high precision, so as to generate DEM data to be smoothed and corrected; An optimization module is used to delete discontinuous areas in the area with height difference steps at the boundary of the DEM data to be smoothed and corrected, fill the deleted areas with a bilinear interpolation algorithm, and remove discontinuities and noise in the terrain with Gaussian smoothing optimization to generate DEM data with smooth boundary transition; The output module is used to output seamless and continuous DEM fusion tiles for application in the digital twin system.

[0014] In a third aspect, the present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the fusion optimization method for DEMs of different precisions as described in any one of the above is implemented.

[0015] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described fusion optimization methods for DEMs of different precisions.

[0016] The fusion optimization method and system for DEMs of different precisions provided by the present invention can achieve seamless fusion by relying on only a small number of control points, and is particularly suitable for situations where control points are scarce. This is particularly important for DEM fusion in large areas, reducing the demand for ground control points and improving the efficiency of data fusion; the elevation correction is generated by ground control points to perform a benchmark correction on the terrain, achieve preliminary fusion, and solve the problem of large elevation differences; for the remaining discontinuous parts of the terrain boundaries, combined with the design of the boundary optimization transition zone, the discontinuous parts are deleted and filled with a bilinear interpolation algorithm, combined with Gaussian smoothing optimization technology, to achieve seamless fusion. The innovative combination of the two solves the shortcomings of traditional methods in boundary processing, while solving the problems of inconsistent benchmarks and abrupt boundaries, ensuring seamless transition of elevation data, and improving fusion accuracy and continuity of results.

[0017] The present invention can effectively handle the conversion between low-precision and high-precision DEM data, especially for the situation where the elevation difference of different benchmarks is too large, improves the accuracy of data fusion, enables smooth integration of different source data, and eliminates the problems caused by inconsistent elevation. Through the method of the present invention, the problems of inconsistent elevation, abrupt boundaries and step effects caused by different benchmarks and precisions in the traditional DEM fusion process are solved, and the seamless fusion and boundary optimization of multi-source and multi-precision DEM data are realized. Compared with the prior art, the present invention provides an innovative and adaptable DEM data fusion method, which solves the problems of inconsistent elevation, abrupt boundaries and discontinuous data in the prior art by fusing and optimizing DEM data of different benchmarks and precisions, and is particularly suitable for the strict requirements of digital twin systems for high precision and seamless fusion. The method realizes high-precision terrain modeling through a small number of control points, has broad application prospects, and can play an important role in the fields of watershed management, urban modeling, environmental monitoring, etc. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 It is one of the flow charts of the fusion optimization method for DEMs with different precisions provided by the present invention; Figure 2 This is the second flow chart of the fusion optimization method for DEMs with different accuracies provided by the present invention; Figure 3 It is a flow chart of elevation benchmark conversion provided by the present invention; Figure 4 The present invention provides a flow chart for generating a boundary smooth transition C_DEM; Figure 5 It is a study area diagram of the embodiment provided by the present invention; Figure 6 It is a three-dimensional rendering and elevation profile diagram of the high-precision terrain (H_DEM) of the national 85 elevation benchmark and the low-precision terrain (L_DEM) of the reference ellipsoid benchmark before fusion in the embodiment provided by the present invention; Figure 7 It is a three-dimensional rendering and elevation profile of the GL_DEM and H_DEM superimposed after the L_DEM is corrected by the elevation benchmark in the embodiment provided by the present invention; Figure 8 It is the final effect diagram of the DEM fusion and boundary optimization method of different benchmarks and accuracies provided by the present invention; Fig. 9 is a comparison diagram after smoothing and optimizing the terrain boundary in the embodiment provided by the present invention; Fig.10 It is a structural schematic diagram of a fusion optimization system for DEMs of different accuracies provided by the present invention; Fig.11 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] Figure 1 This is one of the flow charts of the fusion optimization method for DEMs with different precisions provided by the embodiment of the present invention. Figure 1 As shown, including: Step 100: obtaining preset high-precision DEM data of a first elevation benchmark and preset low-precision DEM data of a second elevation benchmark; Step 200: Generate an elevation correction face based on ground control points, convert the elevation benchmark corresponding to the preset low-precision DEM data into an elevation benchmark consistent with the preset high-precision DEM data, and generate a corrected low-precision DEM data; Step 300: extracting a specified high-precision range from the preset high-precision DEM data, establishing a boundary optimization transition zone, resampling the revised low-precision DEM data based on the boundary optimization transition zone, and generating revised high-precision DEM data with a resolution consistent with the preset high-precision DEM data; Step 400: Terrain mosaicking is performed on the corrected high-precision DEM data and the preset high-precision DEM data by retaining a sampling method that prioritizes high precision, to generate DEM data to be smoothed and corrected; Step 500: Delete the discontinuous area in the area with height difference step problem at the boundary of the DEM data to be smoothed and corrected, fill the deleted area with bilinear interpolation algorithm, and use Gaussian smoothing optimization to remove discontinuity and noise in the terrain to generate DEM data with smooth boundary transition; Step 600: Output seamless and continuous DEM fusion tiles for application in the digital twin system.

[0022] Specifically, Figure 2As shown, the specific steps of the overall implementation technical process of the embodiment of the present invention include: (1) In this embodiment, the DEM data of the study area come from high-precision H_DEM and low-precision L_DEM respectively. The H_DEM data adopts the national 85 elevation benchmark with high accuracy, while the L_DEM data adopts the reference ellipsoid benchmark with low accuracy. On this basis, the ground control points (image control points collected by GPS RTK) are used to establish a modified front through a difference algorithm. The modified front represents the correction value of the low-precision L_DEM under the target elevation benchmark, which is used to convert the benchmark of the L_DEM data to a benchmark consistent with the H_DEM, and generate a low-precision GL_DEM; (2) Extract the designated high-precision areas from the H_DEM data and establish the boundary optimization transition zone based on these areas. The design of the transition zone takes into account the smooth transition of the boundary height difference to ensure that the data gradually transitions to high-precision data in the transition zone. On this basis, resample the GL_DEM to generate a GH_DEM with the same resolution as the H_DEM. The resampling method uses a bilinear interpolation algorithm to ensure the accuracy and continuity of the data; (3) The resampled GH_DEM and H_DEM are mosaicked. A high-precision elevation priority sampling strategy is adopted to retain the high-precision H_DEM data during the mosaicking process to ensure the accuracy of the high-precision area. The final generated P_DEM contains the fusion results from two different precision data, but still needs to be further optimized to eliminate the boundary height difference; (4) At the boundary of the P_DEM, there is a height difference step problem. The discontinuous parts are deleted, and the bilinear interpolation algorithm is used to fill the discontinuous areas. The Gaussian smoothing algorithm is used to further optimize the data to generate a C_DEM with a smooth boundary transition. This C_DEM can be used in the watershed digital twin system to accurately display infrastructure such as rivers and roads; (5) Finally, seamless and continuous DEM fusion tiles are output, which meet the requirements of the digital twin system for high accuracy and detailed display. Rivers can be paved normally, roads are connected smoothly, and slopes are in line with actual conditions, ensuring the efficient operation of the watershed digital twin system in terrain modeling and visualization applications.

[0023] Alternatively, if Figure 3 As shown in step (1), the process of elevation benchmark conversion includes the following steps: (1-1) Import the ground control points containing the national 85 elevation, create a control point vector layer, and create a new field to store the control point national 85 elevation N_Height; (1-2) Load the low-precision L_DEM data of the geodetic high datum that needs to be corrected for elevation, ensuring that it is consistent with the projection coordinate system of the control points.

[0024] (1-3) Use the coordinates of the control points to extract the geodetic elevation values ​​corresponding to the control points on the L_DEM data to be corrected and store them in the control point vector layer field G_Height; (1-4) Perform field operations on the control point vector layer to obtain the elevation difference Dh. The calculation method is: Dh=N_Height-G_Height In the above formula, N_Height is the national 85 elevation value of the control point, and G_Height is the geodetic elevation value corresponding to the low-precision L_DEM extracted from the control point; (1-5) Generate the elevation modified front layer C_DEM with the same range and resolution as L_DEM by interpolating the elevation difference Dh; (1-6) The L_DEM layer to be corrected is subjected to raster value overlay operation with the elevation correction front layer C_DEM to generate the terrain data GL_DEM of the national 85 elevation benchmark.

[0025] Optionally, in step (2), the process of generating a GH_DEM with the same resolution as the H_DEM includes the following steps: (2-1), extract the outer boundary range from the high-precision H_DEM data and generate the vector range surface H_POLYGON; (2-2), perform buffer analysis based on the vector range surface H_POLYGON, obtain the boundary optimization transition zone after a certain distance of expansion, and generate the vector range surface B_POLYGON including the H_DEM range and the transition zone range; (2-3), based on the range surface B_POLYGON, the low-precision GL_DEM of the national 85 elevation datum is clipped and resampled to regenerate a GH_DEM with the same resolution as the H_DEM.

[0026] Optionally, in step (3), the process of generating the P_DEM to be smoothed and corrected includes the following steps: (3-1), add high-precision H_DEM and resampled GH_DEM respectively, and adjust the layer order so that H_DEM is superimposed on GH_DEM to ensure that the high-precision terrain covers the low-precision terrain; (3-2), the H_DEM and GH_DEM are processed by terrain mosaic using the sampling method of the upper layer and merged into a P_DEM to be smoothed and corrected; Among them, the sampling method of the upper layer is that the elevation of the overlapping area is based on the first terrain layer, and the elevation of the non-overlapping area is based on the respective terrain layers.

[0027] As a preferred embodiment of the present invention, Figure 4As shown, in step (4), the process of generating the boundary smooth transition C_DEM includes the following steps: (4-1), from the boundary of H_POLYGON to a certain distance inward (about 1-2 grid distances of high-precision terrain), retain as much high-precision terrain as possible, and generate a smooth boundary line S_POLYGON on the side of the high-precision terrain area; (4-2), generate a smooth boundary line E_POLYGON on the low-precision terrain area side from the boundary of H_POLYGON to a certain distance outward (approximately one grid distance of low-precision terrain); (4-3), within the range of S_POLYGON and E_POLYGON, the existing height difference step areas are segmented and deleted. The segmentation of the mountainous area is based on the slope change points, and the segmentation of the flat area is based on different elevation planes; (4-4), the deleted blank areas are reasonably filled by bilinear interpolation based on local terrain features, so that the terrain with step boundaries transitions smoothly and generates seamless and continuous terrain; (4-5), Gaussian smoothing optimization processing is performed on the interpolated terrain to remove discontinuities and noise in the terrain, making the DEM data smoother and more natural, and finally generating a terrain C_DEM with smooth boundary transitions.

[0028] Optionally, in step (5), the process of outputting seamless and continuous DEM fused tiles includes: The terrain is sliced ​​together according to the layer order of C_DEM and GL_DEM by using the slicing tool to fuse the slices, and the DEM fused tiles are output for application in the digital twin system.

[0029] In one embodiment, Figure 5 As shown in the figure, the study area is located in the upstream reservoir and downstream flooding area of ​​a certain basin, and contains many important geographical elements, such as rivers, roads and man-made facilities. The terrain in this area is complex, with hills and plains alternating, and rivers meandering along the valley. The DEM data of this area comes from two types of terrain data with different precision and benchmarks: one is a high-precision DEM (H_DEM), which uses a drone equipped with a lidar device to collect three-dimensional point cloud data. After filtering, ground point classification, registration, simplification and other processes, a high-precision DEM result is generated. The result uses the national 85 elevation benchmark, which has high precision and is suitable for accurate terrain modeling and digital twin applications; the other is a low-precision DEM (L_DEM), which is obtained by remote sensing satellites. The data uses a reference ellipsoid benchmark. Compared with high-precision DEM, it has lower precision, simpler acquisition, and lower price. In addition, due to the use of a reference ellipsoid benchmark, there is an elevation difference between it and the high-precision DEM.

[0030] In this area, due to the difference in elevation benchmarks and accuracy, there will be very obvious faults at the boundary after the two DEMs are superimposed (such as Figure 6 The difference in elevation is 10 to 40 meters, which brings certain difficulties to the application of watershed digital twins. In addition, the study area contains a large number of infrastructure such as rivers and roads, which have high requirements for the continuity of terrain. The discontinuity of data and inconsistent elevation will make it impossible to accurately display these important geographical elements.

[0031] like Figure 6 As shown, Figure 6 The middle left picture shows the three-dimensional effect of the superposition of the high-precision terrain (H_DEM) of the national 85 elevation benchmark and the low-precision terrain (L_DEM) of the reference ellipsoid benchmark. Due to the large gap between the benchmarks, an obvious fault phenomenon appears at the boundary between the two terrains. Figure 6 The middle right picture is the elevation profile at the boundary of two terrains. The height difference reaches more than 20 meters, resulting in abnormal terrain elevation.

[0032] like Figure 7 As shown, Figure 7 The middle left picture shows the front view of the low-precision terrain L_DEM generated by the control points (Table 1). The GL_DEM generated after the elevation benchmark correction is superimposed with the H_DEM. It can be seen from the figure that the obvious fault phenomenon of the terrain has basically disappeared, but there are still terrain discontinuities and seams. Figure 7 The middle right picture is the elevation profile at the boundary of two terrains after the elevation benchmark correction. The large elevation difference at the boundary is corrected to a smaller elevation difference range of about 2 meters, which provides a basis for the subsequent smooth and seamless fusion. As shown in Table 1: Table 1 Coordinates of elevation control points

[0033] like Figure 8 As shown, Figure 8 The middle left picture shows the removal of discontinuous terrain parts, and the use of bilinear interpolation algorithm to fill the discontinuous areas to generate a three-dimensional effect image after terrain fusion. As can be seen from the picture, a smooth transition and seamless terrain is achieved at the boundary. Figure 8 The middle right picture is an elevation profile after the fusion of two terrains with different benchmarks and different accuracies. It can be seen from the figure that there is no elevation anomaly at the terrain boundary, the slope is normal, the boundary has a smooth transition and is natural, and the fusion effect is relatively good, which verifies the practicability and effectiveness of the present invention.

[0034] like Fig. 9 As shown, Fig. 9 The middle left picture shows the effect before Gaussian smoothing optimization. Fig. 9The middle right picture shows the effect after Gaussian smoothing optimization. It can be seen from the picture that if there are some terrain noise or slight discontinuities after terrain fusion, Gaussian window can be used for further optimization.

[0035] The fusion optimization system for DEMs with different precisions provided by the present invention is described below. The fusion optimization system for DEMs with different precisions described below and the fusion optimization method for DEMs with different precisions described above can be referred to each other.

[0036] Fig.10 is a schematic diagram of the structure of a fusion optimization system for DEMs of different accuracies provided by an embodiment of the present invention, such as Fig.10 As shown, it includes: an acquisition module 1001, a conversion module 1002, a creation module 1003, a mosaic module 1004, an optimization module 1005 and an output module 1006, wherein: The acquisition module 1001 is used to acquire the preset high-precision DEM data of the first elevation benchmark and the preset low-precision DEM data of the second elevation benchmark; the conversion module 1002 is used to generate an elevation correction face based on the ground control points, convert the elevation benchmark corresponding to the preset low-precision DEM data into an elevation benchmark consistent with the preset high-precision DEM data, and generate corrected low-precision DEM data; the establishment module 1003 is used to extract a specified high-precision range from the preset high-precision DEM data, establish a boundary optimization transition zone, resample the corrected low-precision DEM data based on the boundary optimization transition zone, and generate a resolution that is consistent with the preset high-precision DEM data. The modified high-precision DEM data is consistent; the mosaic module 1004 is used to mosaic the modified high-precision DEM data with the preset high-precision DEM data by retaining the high-precision priority sampling method to generate the DEM data to be smoothed and corrected; the optimization module 1005 is used to delete the discontinuous area where there is a height difference step problem area at the boundary of the DEM data to be smoothed and corrected, fill the deleted area with a bilinear interpolation algorithm, and use Gaussian smoothing optimization to remove discontinuities and noise in the terrain to generate DEM data with smooth boundary transition; the output module 1006 is used to output seamless and continuous DEM fusion tiles for application in the digital twin system.

[0037] Fig.11 An example of a physical structure diagram of an electronic device is shown in FIG. Fig.11As shown, the electronic device may include: a processor 1110, a communication interface 1120, a memory 1130 and a communication bus 1140, wherein the processor 1110, the communication interface 1120 and the memory 1130 communicate with each other through the communication bus 1140. The processor 1110 may call the logic instructions in the memory 1130 to execute a fusion optimization method for DEMs of different accuracies, the method comprising: obtaining a preset high-precision DEM data of a first elevation benchmark and a preset low-precision DEM data of a second elevation benchmark; generating an elevation correction face based on ground control points, converting the elevation benchmark corresponding to the preset low-precision DEM data into an elevation benchmark consistent with the preset high-precision DEM data, and generating a corrected low-precision DEM data; extracting a specified high-precision range from the preset high-precision DEM data, establishing a boundary optimization transition zone, and correcting the low-precision DEM data based on the boundary optimization transition zone. The data is resampled to generate revised high-precision DEM data with the same resolution as the preset high-precision DEM data; the revised high-precision DEM data and the preset high-precision DEM data are terrain mosaicked by retaining the high-precision priority sampling method to generate the DEM data to be smoothed and corrected; the discontinuous areas where there are height difference step problem areas at the boundaries of the DEM data to be smoothed and corrected are deleted, the deleted areas are filled by the bilinear interpolation algorithm, and the discontinuity and noise in the terrain are removed by Gaussian smoothing optimization to generate DEM data with smooth boundary transition; seamless and continuous DEM fusion tiles are output for application in the digital twin system.

[0038] In addition, the logic instructions in the above-mentioned memory 1130 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0039] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the fusion optimization method for DEMs of different accuracies provided by the above-mentioned methods, the method comprising: obtaining preset high-precision DEM data of a first elevation benchmark and preset low-precision DEM data of a second elevation benchmark; generating an elevation correction surface based on ground control points, converting the elevation benchmark corresponding to the preset low-precision DEM data into an elevation benchmark consistent with the preset high-precision DEM data, and generating corrected low-precision DEM data; extracting a specified high-precision range from the preset high-precision DEM data, establishing a boundary optimization transition zone, and based on the boundary The modified low-precision DEM data is resampled in the boundary optimized transition zone to generate modified high-precision DEM data with the same resolution as the preset high-precision DEM data; the modified high-precision DEM data and the preset high-precision DEM data are terrain mosaicked by retaining the high-precision priority sampling method to generate the DEM data to be smoothed and corrected; the discontinuous areas in the height difference step problem area at the boundary of the DEM data to be smoothed and corrected are deleted, the deleted areas are filled by the bilinear interpolation algorithm, and the discontinuity and noise in the terrain are removed by Gaussian smoothing optimization to generate boundary smooth transition DEM data; seamless and continuous DEM fusion tiles are output for application in the digital twin system.

[0040] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0041] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0042] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fusion optimization method for DEMs of different precisions, characterized in that: include: Acquire preset high-precision DEM data of a first elevation benchmark and preset low-precision DEM data of a second elevation benchmark; Generate an elevation correction face based on the ground control points, convert the elevation benchmark corresponding to the preset low-precision DEM data into an elevation benchmark consistent with the preset high-precision DEM data, and generate a corrected low-precision DEM data; Extract a specified high-precision range from the preset high-precision DEM data, establish a boundary optimization transition zone, resample the modified low-precision DEM data based on the boundary optimization transition zone, and generate modified high-precision DEM data with a resolution consistent with the preset high-precision DEM data; The modified high-precision DEM data and the preset high-precision DEM data are terrain mosaicked by retaining a high-precision priority sampling method to generate DEM data to be smoothed and corrected; Deleting discontinuous areas where there are height difference step problem areas at the boundaries of the DEM data to be smoothed and corrected, filling the deleted areas with a bilinear interpolation algorithm, and removing discontinuities and noise in the terrain with Gaussian smoothing optimization to generate DEM data with smooth boundary transitions; Output seamless and continuous DEM fusion tiles for application in digital twin systems.

2. The fusion optimization method for DEMs of different precisions according to claim 1 is characterized in that: Generate an elevation correction face based on the ground control points, convert the elevation benchmark corresponding to the preset low-precision DEM data into an elevation benchmark consistent with the preset high-precision DEM data, and generate a corrected low-precision DEM data, including: According to the ground control point containing the first elevation, a control point vector layer is established, and a new field is created to store the first elevation control point; Loading the preset low-precision DEM data that needs to be corrected for elevation, ensuring that the preset low-precision DEM data is consistent with the projection coordinate system of the first elevation control point; Extracting the geodetic elevation value corresponding to the control point on the preset low-precision DEM data using the coordinates of the first elevation control point, and storing it in the control point vector layer field; Performing field operations on the control point vector layer, subtracting the geodetic elevation value corresponding to the preset low-precision DEM data extracted by the control point from the elevation value of the first elevation control point to obtain an elevation difference; Generate an elevation-to-front layer with the same range and resolution as the preset low-precision DEM data by interpolation based on the elevation difference; The corresponding layer of the preset low-precision DEM data and the elevation-corrected front layer are subjected to a raster value overlay operation to generate the corrected low-precision DEM data.

3. The fusion optimization method for DEMs of different precisions according to claim 1 is characterized in that: Extracting a specified high-precision range from the preset high-precision DEM data, establishing a boundary optimization transition zone, resampling the revised low-precision DEM data based on the boundary optimization transition zone, and generating revised high-precision DEM data with a resolution consistent with the preset high-precision DEM data, including: Extracting the outer boundary range from the preset high-precision DEM data to generate a first vector range surface; Performing a buffer zone analysis based on the first vector range surface to obtain the boundary optimization transition zone after the first preset distance is expanded, and generating a second vector range surface including a preset high-precision DEM data range and a transition zone range; The modified low-precision DEM data is clipped and resampled based on the second vector range surface to regenerate the modified high-precision DEM data with a resolution consistent with the preset high-precision DEM data.

4. The fusion optimization method for DEMs of different precisions according to claim 1 is characterized in that: The modified high-precision DEM data and the preset high-precision DEM data are terrain mosaicked by retaining a high-precision priority sampling method to generate DEM data to be smoothed and corrected, including: Add the preset high-precision DEM data and the revised high-precision DEM data respectively, and superimpose the layer corresponding to the preset high-precision DEM data on the layer corresponding to the revised high-precision DEM data; The preset high-precision DEM data and the corrected high-precision DEM data are subjected to terrain mosaic processing by adopting an upper layer sampling method, and are merged into the DEM data to be smoothed and corrected.

5. The fusion optimization method for DEMs of different precisions according to claim 4 is characterized in that: The upper layer sampling method includes: The elevation of the overlapping area is taken according to the first terrain layer, and the elevation of the non-overlapping area is taken according to each terrain layer.

6. The fusion optimization method for DEMs of different precisions according to claim 3 is characterized in that: The discontinuous area in the area with height difference step problem at the boundary of the DEM data to be smoothed and corrected is deleted, the deleted area is filled by using bilinear interpolation algorithm, and the discontinuity and noise in the terrain are removed by Gaussian smoothing optimization to generate DEM data with smooth boundary transition, including: Based on the first vector range plane moving inward by a second preset distance, a first smooth boundary line on the side of a preset high-precision terrain area is generated; Based on the first vector range, the first vector range is moved outward by a third preset distance to generate a second smooth boundary line on the side of the preset low-precision terrain area; Within the interval between the first smooth boundary line and the second smooth boundary line, the existing height difference step area is segmented and deleted, wherein the segmentation of the mountainous area is based on the slope change points, and the segmentation of the flat area is based on different elevation planes; The deleted blank areas are filled by bilinear interpolation algorithm based on local terrain features, so that the terrain with step boundaries can be smoothly transitioned to generate seamless and continuous terrain. The interpolated terrain is subjected to Gaussian smoothing optimization processing to remove discontinuities and noises in the terrain, thereby generating the boundary smooth transition DEM data.

7. The fusion optimization method for DEMs of different precisions according to claim 1 is characterized in that: Output seamless and continuous DEM fused tiles, including: The slicing tool is used to perform fusion slicing, and terrain slicing is performed according to the layer corresponding to the boundary smooth transition DEM data and the layer corresponding to the corrected low-precision DEM data, and the DEM fused tile is output.

8. A fusion optimization system for DEMs of different precisions, characterized in that: include: An acquisition module, used to acquire preset high-precision DEM data of a first elevation benchmark and preset low-precision DEM data of a second elevation benchmark; A conversion module is used to generate an elevation correction face based on ground control points, convert the elevation benchmark corresponding to the preset low-precision DEM data into an elevation benchmark consistent with the preset high-precision DEM data, and generate a corrected low-precision DEM data; Establishing a module, used for extracting a specified high-precision range from the preset high-precision DEM data, establishing a boundary optimization transition zone, resampling the corrected low-precision DEM data based on the boundary optimization transition zone, and generating corrected high-precision DEM data with a resolution consistent with the preset high-precision DEM data; A mosaic module is used to perform terrain mosaicking on the corrected high-precision DEM data and the preset high-precision DEM data by retaining a sampling method that prioritizes high precision, so as to generate DEM data to be smoothed and corrected; An optimization module is used to delete discontinuous areas in the area with height difference steps at the boundary of the DEM data to be smoothed and corrected, fill the deleted areas with a bilinear interpolation algorithm, and remove discontinuities and noise in the terrain with Gaussian smoothing optimization to generate DEM data with smooth boundary transition; The output module is used to output seamless and continuous DEM fusion tiles for application in the digital twin system.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the fusion optimization method for DEMs of different precisions is implemented as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the fusion optimization method for DEMs of different precisions is implemented as described in any one of claims 1 to 7.

Images