Water and land integrated terrain construction method based on multi-source surveying and mapping data fusion

Abstract

The invention discloses a water-land integrated terrain construction method based on multi-source surveying and mapping data fusion, which comprises the following steps: S1, according to the characteristics of a set surveying and mapping area, data application and precision requirements, synchronously acquiring basic terrain data and local high-precision terrain data of the area by adopting a multi-source surveying and mapping technology; s2, dimension raising is carried out on the discrete river channel section data, and continuous underwater topography is generated through on-way virtual section encryption and differential interpolation algorithm fitting in combination with river channel thalweg and waterline data; s3, fusing the multi-source data by adopting a buffer connection in-band weighted data fusion algorithm, and constructing land and underwater integrated topographic data; and S4, constructing a multi-time-sequence land and water integrated terrain database. On the basis of basic data acquisition and data fusion processing, an underwater continuous terrain rapid fitting construction technology based on a virtual section encryption and differential interpolation method and a weighted multi-source data fusion technology in a buffer connection band are innovatively provided, and regional land and water integrated continuous underwater terrain construction is achieved.

Description

Technical Field

[0001] This invention relates to the field of hydrological feature data mapping, and in particular to a method for constructing integrated land and water terrain based on the fusion of multi-source mapping data. Background Technology

[0002] Underwater topography is crucial hydrological data, forming the basis for hydrodynamic calculations, riverbed sedimentation analysis, and riverbed evolution patterns. For a long time, hydrological surveys have commonly employed the cross-sectional method to obtain extensive cross-sectional data of rivers (reservoirs), involving setting up cross-sections every 1-2 km along the river (reservoir) to acquire elevation data along the cross-section lines. However, due to the discrete layout of cross-sections, the resulting discrete linear cross-sectional data is insufficient to support two-dimensional hydrodynamic calculations (such as two-dimensional hydrodynamic models), and it cannot visually represent the riverbed morphology and sedimentation changes in the river and reservoir areas. With the development of unmanned surface vessels (USVs) and sonar equipment, multibeam echolocation is used on USVs to acquire continuous point cloud data for underwater topographic mapping. However, acquiring mapping data over long distances and large areas is very costly, often only used for key underwater areas with significant localized scour and sedimentation changes. Spatially discontinuous mapping data makes it difficult to digitally map the riverbed and visually represent its underwater morphology and sedimentation changes, failing to provide user-friendly data support for hydraulic engineering analysis. Summary of the Invention

[0003] In view of this, the present invention provides a method for constructing integrated land and water terrain based on the fusion of multi-source surveying and mapping data, so as to realize low-cost, continuous underwater terrain construction of the surveying and mapping area.

[0004] To achieve the above objectives, the present invention adopts the following technical solution: The method for constructing integrated land and water terrain based on multi-source mapping data fusion as described in this invention includes the following steps: S1. Based on the characteristics, data application, and accuracy requirements of the designated surveying area, multi-source surveying technology is used to acquire basic topographic data and local high-precision topographic data of the surrounding land and beach areas and river sections of the area.

[0005] Optionally, the multi-source mapping should be carried out concurrently (e.g., both before or after the flood season, with a time difference not exceeding 2 months), and the multi-source mapping data includes: Satellite remote sensing technology is used to acquire stereo image pairs of the designated survey area during the low water period of the river channel, i.e., high-precision multispectral remote sensing image data. High-precision topographic data of key local land areas such as beach areas, areas with significant river morphology changes (such as sharp bends), and areas where main streams and tributaries converge are acquired by using drones equipped with LiDAR devices. Cross-sectional data of characteristic sections of the river channel are obtained using cross-sectional measurement methods; Multibeam bathymetry was used to collect continuous underwater topographic data in key local underwater areas such as the area near the dam, areas with significant changes in river morphology (such as sharp bends), areas where tributaries and main streams meet, and areas with significant changes in scouring and sedimentation (such as the funnel area in front of the dam and the top slope of the sedimentary delta).

[0006] S2. Up-dimensionalize the discrete river cross-section data, combine the river thalweg line and the waterline data extracted from the multispectral remote sensing image data of the river based on the field survey operation period, and generate continuous underwater topography by densifying the virtual cross-section along the river and fitting the differential interpolation algorithm.

[0007] Optionally, based on multispectral remote sensing image data of the river channel during the field surveying period, waterline data is extracted using band computation or deep learning, and verified against the measured waterline points of the cross-section and the land topographic data obtained in step S1 to extract the extended waterline elevation value; combining the waterline, thalweg (extracted based on the measured cross-section delineation, or longitudinal section data if a longitudinal section is surveyed) and the measured cross-section location, several virtual cross-sections are laid out along the river (along the river channel direction), and values ​​are assigned to the laid-out virtual cross-sections based on the measured cross-section data using a differential interpolation algorithm; combining the measured cross-section data and the virtual cross-section data, underwater topographic data within the main channel is generated with the waterline as the constraint boundary; the generated underwater topographic data is then subjected to certain artificial processing to better reflect the river conditions.

[0008] S3. A weighted data fusion algorithm within the buffer transition zone is used to fuse the above multi-source data to construct integrated land and water terrain data.

[0009] Optionally, based on the regional land surface and beach area obtained in step S1, the underwater topography generated in step S2 is fused to generate basic regional land and water topography data; based on the basic regional land and water topography, UAV LiDAR data and UAV multibeam data of local key core areas are superimposed and fused to improve the data accuracy of local key core areas; when fusing multi-source data, based on high-precision topography data, algorithms such as buffer transition zone weighted fusion algorithm and edge smoothing algorithm are adopted to ensure that the transition at the junction of data with different precision is natural and there are no obvious steps after fusion.

[0010] S4. Establish a multi-time series integrated land and water topographic database to facilitate the management and rapid application of underwater topographic results.

[0011] Optionally, the underwater topographic data generated in step S2 for multiple years can be repeated; based on the integrated land and water topographic data generated in step S3, the underwater topographic data generated in multiple years can be merged to construct a multi-time series integrated land and water topographic database, and a unified data service can be provided to the outside world through a unified interface.

[0012] Starting from basic data acquisition and data fusion processing, this invention comprehensively considers the characteristics and implementation costs of different surveying and mapping methods, and innovatively proposes a rapid underwater continuous terrain fitting and construction technology based on virtual cross-section densification and differential interpolation methods, as well as a weighted multi-source data fusion technology within the buffer zone. By fusing multi-source surveying and mapping data of different acquisition methods, different accuracies, and different types, it realizes the construction of regional integrated and continuous underwater terrain. Attached Figure Description

[0013] Figure 1 This is an overall flowchart of the method of the present invention.

[0014] Figure 2 This is a flowchart of the differential interpolation process according to an embodiment of the present invention.

[0015] Figure 3 This is a flowchart of the edge fusion process according to an embodiment of the present invention.

[0016] Figure 4 This is a schematic diagram of generating division points on a virtual cross-section by dividing the virtual cross-section at equal intervals according to an embodiment of the present invention.

[0017] Figure 5 This is a schematic diagram illustrating the calculation of the elevation value of each segment point on the virtual cross-section using linear differential interpolation, according to an embodiment of the present invention. Detailed Implementation

[0018] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings. These embodiments are implemented based on the technical solution of the present invention, and detailed implementation methods and specific operation processes are given. However, the scope of protection of the present invention is not limited to the following embodiments.

[0019] like Figure 1-4 As shown, the integrated land and water terrain construction method based on multi-source mapping data fusion described in this invention is carried out according to the following steps: Step 1: Construction of basic water and land topography based on remote sensing river morphology and differential interpolation of underwater topography in the reservoir area: Analyze the time of the measured cross-section data to be fitted, find the time and water level of the river in the same time period that are low, collect (or pre-program and order) high-resolution stereo image pairs of the area based on the time, and combine the ground control point data and tie point data of the area to generate the land surface and beach topography of the area through binocular matching fusion and resection algorithm.

[0020] The river cross-section is mapped, and high-precision multispectral remote sensing image data of the same period is collected (or pre-programmed and ordered). The discrete river cross-section data is upgraded in dimensionality. Combined with the thalweg line of the river and the waterline data extracted by the satellite remote sensing technology, a continuous underwater topography is generated by densifying virtual cross-sections along the river and fitting differential interpolation algorithms.

[0021] Specifically, such as Figure 2 As shown, based on high-precision multispectral remote sensing image data of the river channel during the field surveying period, waterline data is extracted using band computation or deep learning. This data is then compared with the measured waterline points and acquired land topographic data to extract the extended waterline elevation values. Combining the waterline, thalweg (extracted based on the measured cross-section, or longitudinal profile data if available), and the measured cross-section location, several virtual cross-sections are established along the river's course. Based on the measured cross-section data, differential interpolation algorithms are used to assign values ​​to these virtual cross-sections. By integrating the measured cross-section data and the virtual cross-section data, underwater topographic data within the main channel is generated, using the waterline as the constraint boundary. The generated underwater topographic data undergoes some manual processing to better reflect the river conditions.

[0022] Based on underwater continuous terrain fitting using virtual cross-section densification and differential interpolation, the following principles are established for virtual cross-section layout: —Virtual cross-section perpendicular to the thalweg line; — Cross sections are laid out at locations such as sharp bends in the river channel and significant changes in gradient; —The upstream and downstream layout sections of river-blocking and cross-river structures with obvious water-blocking effects.

[0023] The river channel is divided using the thalweg, which serves as the waterline of the divided channel. Using virtual cross-section measurement data, line elements for each virtual cross-section are generated based on its name, constructing a virtual cross-section line vector layer. Intersection analysis is performed between the virtual cross-section line vector data and the waterlines of the left and right banks to generate left and right bank waterline point layers. The elevation values ​​at these waterline points are calculated using a polynomial function fitted from the original cross-section measurement data. Finally, the left and right bank waterlines are divided using these waterline points to obtain the waterlines between each pair of virtual cross-sections.

[0024] First, the waterline is traversed, and each segment is divided at equal intervals to obtain dividing points on the waterline. The elevation value of each dividing point is calculated using linear differential interpolation. Then, the virtual cross-section lines are traversed, and each virtual cross-section line is divided at equal intervals to obtain dividing points on the cross-section. The elevation value of each dividing point is calculated using the aforementioned polynomial function interpolation. Finally, the dividing points on the left and right bank waterlines are traversed, and the corresponding dividing points form a virtual cross-section. The virtual cross-section is then divided at equal intervals to generate dividing points on the virtual cross-section, such as... Figure 4 As shown; then, based on the segmentation points on the corresponding virtual cross-section, the elevation value of each segmentation point is calculated using linear differential interpolation, such as... Figure 5 As shown.

[0025] Step 2, improving data accuracy in key local areas: For key land areas of concern, such as areas with significant river morphology changes (e.g., sharp bends) and areas where main streams and tributaries converge, high-precision topographic data is collected using drones equipped with LiDAR devices.

[0026] For key underwater areas such as the area near the dam, areas with significant changes in river morphology (such as sharp bends), areas where tributaries and main streams converge, and areas with significant changes in scouring and sedimentation (such as the funnel area in front of the dam and the top slope of the sedimentary delta), high-precision continuous underwater topographic data are collected by using an unmanned surface vessel equipped with a multibeam echo sounder.

[0027] Step 3: Multi-source terrain data fusion and weighted fusion within the multi-source mapping data buffer zone: like Figure 3 As shown, based on the basic land and water topography, a frequency domain weighted optimal fusion algorithm within the buffer zone and a topographic line smoothing algorithm at the junction are used to superimpose and fuse high-precision topographic data of key local land areas and high-precision underwater topographic data of local underwater areas, thereby constructing a land and water topography that meets the accuracy requirements for hydraulic analysis.

[0028] Generally, resampling large-area low-precision terrain data into high-precision terrain data results in a large amount of data and slow computation time, causing difficulties for subsequent digital twin work. Therefore, it is generally not possible to directly resample large-area low-precision terrain. In order to solve the problem of elevation abrupt changes at the junction, this invention mainly processes the terrain of the junction area.

[0029] First, the overlapping area of ​​high-precision and low-precision terrain is obtained. This overlapping area is then identified through the boundaries of the high-precision terrain, and buffered inwards based on the boundary lines. An excessively large buffer range will significantly impact the original high-precision data and increase the amount of data processing required; an excessively small buffer range will result in an unsatisfactory fusion effect and will not effectively mitigate abrupt changes. Multi-level buffers are constructed based on the contour lines of the high-precision terrain data, and low-precision terrain within these buffer areas is obtained through masking.

[0030] Secondly, low-precision terrain is resampled in multi-level buffer areas. When merging raster data of different grid sizes, in order to preserve the original high-precision data to the greatest extent, the terrain data of large grids is resampled to ensure that the grid size of low-precision raster data is consistent with that of high-precision raster data.

[0031] Finally, the elevation value of each grid within the multi-level buffer area is calculated by weighted summation; the terrain data of the multi-level buffer is processed step by step, and the weights corresponding to the high-precision terrain grid and the low-precision terrain grid are set according to the terrain changes in the buffer, so that the new value obtained by weighted summation changes slowly in a step-like manner, thereby solving the problem of abrupt terrain changes.

[0032] Step 4: Establishment of a multi-temporal integrated land and water terrain database: Based on the aforementioned regional land and water topography, underwater topographic data fitted from cross-sectional data of different years are overlaid and integrated to construct a multi-temporal integrated land and water topography database. The multi-temporal integrated land and water topography database is visualized through topographic rendering. The underwater topographic data of adjacent periods generated by overlay are analyzed to show changes in scour and sedimentation, and visualized through topographic rendering. Corresponding metadata is established and a unified data interface service is provided for access.

Claims

1. A method for constructing integrated land and water terrain based on multi-source mapping data fusion, characterized in that: Includes the following steps: S1. Based on the characteristics, data application and accuracy requirements of the designated surveying area, multi-source surveying technology is used to simultaneously acquire basic topographic data and local high-precision topographic data of the surrounding land and beach areas and river cross-sections of the area. S2. Up-dimensionalize the discrete river cross-section data, combine the river thalweg line and the waterline data extracted by satellite remote sensing technology, and generate continuous underwater topography by densifying virtual cross-sections along the river and fitting differential interpolation algorithms. S3. The above multi-source data are fused using a weighted data fusion algorithm within the buffer zone to construct integrated land and water topographic data. S4. Construct a multi-time series integrated land and water topography database to facilitate the management and application of the underwater topography data.

2. The method for constructing integrated land and water terrain based on multi-source mapping data fusion according to claim 1, characterized in that: In step S1, the multi-source mapping data mainly includes: Using satellite remote sensing technology, i.e., multispectral remote sensing image data, stereo image pairs of the designated survey area are obtained during the low water period of the river channel. High-precision topographic data of key local land areas, such as beach areas, areas with significant river morphology changes, and areas where main streams and tributaries converge, were acquired by using drones equipped with LiDAR devices. Cross-sectional data of characteristic sections of the river channel are obtained using cross-sectional measurement methods; Multibeam unmanned depth sounding technology was used to collect continuous underwater topographic data in key local underwater areas, such as the area near the dam, the area with large changes in river morphology, the area where the main stream and tributaries meet, and the area with obvious changes in scouring and deposition. High-precision multispectral remote sensing image data of the river channel during the field surveying period, obtained using satellite remote sensing technology.

3. The method for constructing integrated land and water terrain based on multi-source mapping data fusion according to claim 1, characterized in that: In step S2, based on the collected multispectral remote sensing image data of the river channel during the field surveying period, waterline data is extracted using band operations or deep learning. This data is then compared with the measured waterline points of the river cross-section and the land topographic data obtained in step S1 to verify and extract the extended waterline elevation values. Combining the waterline, the thalweg line, and the measured cross-section location, several virtual cross-sections are established along the river. Based on the measured cross-section data, differential interpolation algorithms are used to assign values ​​to these virtual cross-sections. By integrating the measured cross-section data and the virtual cross-section data, underwater topographic data within the main channel is generated, with the waterline as the constraint boundary. The generated underwater topographic data undergoes certain manual processing to better reflect the river conditions.

4. The method for constructing integrated land and water terrain based on multi-source mapping data fusion according to claim 3, characterized in that: In step S3, based on the regional land surface and beach area obtained in step S1, the underwater topography generated in step S2 is fused to generate basic regional land and water topography data. Based on the basic regional land and water topography, UAV LiDAR data and UAV multibeam data of local key core areas are superimposed and fused to improve the data accuracy of local key core areas. When fusing multi-source data, based on high-precision topography data, algorithms such as buffer transition zone weighted fusion algorithm and edge smoothing algorithm are adopted to ensure that the transition at the junction of data with different precision is natural and there are no obvious steps after fusion.

5. The method for constructing integrated land and water terrain based on multi-source mapping data fusion according to claim 1, characterized in that: In step S4, the underwater topographic data generated in step S2 for multiple years are repeated; based on the integrated land and water topographic data generated in step S3, the underwater topographic data generated in multiple years are merged to construct a multi-time series integrated land and water topographic database, and a unified data service is provided to the outside world through a unified interface.

Images