A Modeling Method and System for Complex Terrain

Abstract

The present invention discloses a modeling method and system for complex terrain, relating to the field of modeling technology, including: determining an initial area to be modeled for the complex terrain; obtaining the smallest rectangle containing the initial area as a standard area and mapping to obtain an initial plan view; integrating the initial plan view into a longitude and latitude coordinate system and dividing it into a number of equal rectangular areas; constructing a three-dimensional model for each rectangular area; using a first method to correct the three-dimensional models of adjacent rectangular areas to obtain corrected three-dimensional models; obtaining aerial survey images and determining a number of feature points in each rectangular area, obtaining a first image corresponding to each rectangular area and pasting it onto the corresponding corrected three-dimensional model; using a second method to smooth the first images of adjacent rectangular areas to obtain a first processing result; generating a standard three-dimensional model of the complex terrain based on the correction result and the first processing result. The present invention ensures the accuracy of modeling while improving the modeling efficiency of complex terrain.

Description

Technical Field

[0001] The present invention relates to the technical field of modeling, and specifically, to a method and system for modeling complex terrain. Background Art

[0002] Terrain is a relatively complex scene in nature. After operations such as modeling, perspective projection, and texture mapping on the terrain, a three-dimensional terrain generates a simulation effect, and the generated topographic map can realistically reflect the external real world. Compared with traditional paper topographic maps and computer-generated line-drawn topographic maps, the three-dimensional topographic map after modeling is more realistic and closer to reality. In recent years, three-dimensional topographic maps have been increasingly widely used in fields such as land and resources survey, urban simulation, and virtual reality, and are also of great significance for spatial analysis, planning and decision-making, etc.

[0003] Although currently, for the modeling of conventional three-dimensional terrain, it is already possible to achieve realism and closeness to reality, and the modeling efficiency can also be guaranteed; but for complex three-dimensional terrain, because complex terrain means that a large amount of terrain data needs to be processed, it leads to a large amount of time consumed in modeling, and at the same time, the accuracy of modeling cannot be guaranteed. Therefore, there is an urgent need for a method for modeling complex three-dimensional terrain to improve the accuracy and efficiency of modeling. Summary of the Invention

[0004] To solve the deficiencies of the existing technology, the present invention provides a method for modeling complex terrain, and the method includes the following steps:

[0005] Obtain the basic data of the complex terrain, and based on the basic data, determine the initial area to be modeled of the complex terrain;

[0006] Obtain the smallest rectangle containing the initial area and denote it as the standard area, and map the standard area to the horizontal plane to obtain an initial plan view;

[0007] Construct a coordinate system with latitude as the abscissa and longitude as the ordinate, integrate the initial plan view into the coordinate system, and divide the initial plan view into several equal rectangular areas;

[0008] Obtain the GIS data and elevation data of each rectangular area, and construct an initial three-dimensional model for each rectangular area;

[0009] Use a first method to perform correction processing on the initial three-dimensional models of any two adjacent rectangular areas to obtain a correction result, and based on the correction result, obtain a corrected three-dimensional model for each rectangular area;

[0010] Obtain the aerial survey images of the standard area and process them to obtain a processing result. Determine a preset number of first feature points in each rectangular area. Based on the processing result and all the first feature points in each rectangular area, obtain the first image corresponding to each rectangular area, and map each first image to the corresponding corrected three-dimensional model;

[0011] Use a second method to perform smoothing processing on the first images of any two adjacent rectangular areas to obtain a first processing result;

[0012] Based on the correction result and the first processing result, generate the standard three-dimensional model of the complex terrain.

[0013] The present invention is implemented through the following technical solutions: First, obtain the basic data of the complex terrain (such as drawings or satellite images, etc.). Based on the basic data, determine the initial area that needs to be modeled for the complex terrain. Obtain the smallest rectangle containing this initial area and denote it as the standard area. Map the standard area to the horizontal plane to obtain an initial plan view. Then construct a coordinate system with longitude and latitude as the horizontal and vertical coordinates, integrate the initial plan view into this coordinate system, and divide the initial plan view to obtain several equal rectangular areas. The above steps realize the conversion of the complex terrain into an initial plan view containing multiple rectangular areas and realize searching and positioning through the coordinate system.

[0014] Obtain the GIS data and elevation data of each rectangular area, and construct the initial three-dimensional model of each rectangular area. At this time, if the initial three-dimensional models of all rectangular areas are fused, the initial three-dimensional model of the complex terrain can be obtained. However, when fusing adjacent rectangular areas, due to reasons such as data accuracy, it is impossible to ensure a smooth transition between adjacent three-dimensional models, resulting in an abrupt phenomenon between adjacent three-dimensional models. Therefore, in this solution, a first method is used to perform correction processing on the three-dimensional models of any adjacent rectangular areas to ensure that when all rectangular areas are fused, they can be smoothly transitioned and the abrupt phenomenon between the three-dimensional models of different rectangular areas is reduced.

[0015] Obtain the aerial survey images of the standard area and process them to obtain a processing result. Determine a preset number of first feature points in each rectangular area. Based on the first feature points of each rectangular area and all the obtained aerial survey images, obtain the first image corresponding to each rectangular area. The first feature points of each rectangular area are used to locate the aerial survey images and then crop them to obtain the corresponding first images, and then map each first image to the corresponding corrected three-dimensional model. However, when all the first images are fused, due to reasons such as splicing, an inharmonious phenomenon is likely to occur at the connection of adjacent first images. Therefore, in this solution, a second method is used to perform smoothing processing on the adjacent first images to make the transition between adjacent first images smoother and reduce the inharmonious phenomenon at the connection of adjacent first images.

[0016] Finally, based on the corrected three-dimensional model and all the first images that have been smoothed, a standard three-dimensional model of the complex terrain is generated.

[0017] As an alternative technical solution, the first method includes:

[0018] Transition regions are preset around the perimeter of each rectangular region. The transition region of each rectangular region includes four overlapping regions and four non-overlapping regions;

[0019] Obtain any rectangular region as the second rectangular region, and obtain a non-overlapping region in the second rectangular region as region E and an overlapping region as region F;

[0020] Obtain the adjacent non-overlapping region of region E as region G, and statistically calculate the average elevation value of region E as the first elevation value and the average elevation value of region G as the second elevation value;

[0021] Calculate the first difference between the first elevation value and the second elevation value, and determine whether the first difference is less than the first threshold. If not, do nothing. If so, calculate the average value of the first elevation value and the second elevation value as the third elevation value, and based on the third elevation value, adjust the elevation values of region E and region G to complete the correction of the three-dimensional models corresponding to region E and region G;

[0022] Place region F and all adjacent overlapping regions into a preset region set, and statistically calculate the average elevation value of each region in the preset region set to obtain the calculation result;

[0023] Based on the calculation result, calculate the difference between the average elevation values corresponding to any two regions in the preset region set as the second difference, and determine whether all the second differences are less than the first threshold. If so, calculate the average value of the average elevation values of all regions in the preset region set as the fourth elevation value, and based on the fourth elevation value, adjust the elevation values of all regions in the preset region set to complete the correction of the three-dimensional models corresponding to all regions in the preset region set. If not, do nothing.

[0024] As an alternative technical solution, the second method includes:

[0025] Obtain the first images corresponding to any two adjacent rectangular regions as the second image and the third image respectively;

[0026] Determine whether there is an overlapping area between the second image and the third image. If not, do nothing. If so, obtain the initial pixel value of each pixel point in the overlapping area, and assign a weight value to each pixel point in the overlapping area through the first formula to obtain an assignment result. Based on the assignment result, use the second formula to calculate the initial pixel value of each pixel point in the overlapping area to obtain the corresponding fused pixel value;

[0027] Based on the fused pixel value of each pixel point in the overlapping area, smooth the area where the second image and the third image overlap.

[0028] As an optional technical solution, the first formula is:

[0029]

[0030] δ1 is the weight value of a single pixel point in the overlapping area in the second image, δ2 is the weight value of a single pixel point in the overlapping area in the third image, d1 is the distance between a single pixel point in the overlapping area and the center point of the second image, d2 is the distance between a single pixel point in the overlapping area and the center point of the third image, and B is a constant.

[0031] As an optional technical solution, the second formula is:

[0032] W ′ =δ1×W1 + δ2×W2

[0033] W ′ is the fused pixel value of a single pixel point in the overlapping area, W1 is the initial pixel value of a single pixel point in the overlapping area in the second image, and W2 is the initial pixel value of a single pixel point in the overlapping area in the third image.

[0034] As an optional technical solution, the method further includes optimizing the standard three-dimensional model:

[0035] Obtain the first image of each rectangular area and process it to obtain the corresponding binary image;

[0036] Obtain the first ratio between the shadow area and the area of the corresponding binary image in each binary image to obtain the first acquisition result;

[0037] Based on the first acquisition result, perform grade division on different rectangular areas, and the grade of the rectangular area is proportional to the corresponding first ratio;

[0038] Obtain all aerial survey images of the complex terrain within a preset time period to obtain a second acquisition result. Based on the second acquisition result, process to obtain all first images of each rectangular area and place them in the first data set;

[0039] Based on the level of each rectangular area, select a corresponding number of second feature points in each rectangular area, and the number of the second feature points is proportional to the level of the corresponding rectangular area;

[0040] Obtain the pixel points included in all the second feature points in each rectangular area, and based on the corresponding first data set, process to obtain the change curve of the pixel points corresponding to each second feature point in each rectangular area, to obtain a second processing result;

[0041] Based on the second processing result, within the preset time period, update the pixel points corresponding to each second feature point in each rectangular area to complete the optimization of the standard three-dimensional model.

[0042] As an alternative technical solution, the first ratio is calculated by the third formula:

[0043]

[0044] α n is the first ratio, A n is the area of the shadow region in the binary image, S n is the area of the binary image.

[0045] As an alternative technical solution, based on the first acquisition result, the level division of different rectangular areas includes:

[0046] Obtain the historical data of all the first ratios. Based on the historical data, divide the data range of the first ratio, and each data range is assigned a corresponding level value;

[0047] Obtain any rectangular area denoted as the first rectangular area, and obtain the first ratio of the first rectangular area denoted as the second ratio;

[0048] Obtain the data range corresponding to the second ratio denoted as the first data range;

[0049] Obtain the level corresponding to the first data range as the level of the first rectangular area.

[0050] Obtain the first ratio of any rectangular area denoted as the second ratio, and obtain the data range corresponding to the second ratio denoted as the first data range;

[0051] Obtain the level value corresponding to the first data range as the level of the first rectangular area.

[0052] As an alternative technical solution, obtaining and processing the aerial survey images of the standard area includes: based on the aerial survey video stream of the standard area, obtaining multiple frames of aerial survey images of the standard area and performing calibration processing.

[0053] To address the deficiencies of the existing technologies, the present invention also provides a modeling system for complex terrain. The system includes:

[0054] An initial area unit, configured to obtain basic data of the complex terrain and, based on the basic data, determine an initial area of the complex terrain that needs to be modeled;

[0055] A mapping unit, configured to obtain the smallest rectangle containing the initial area, denoted as the standard area, and map the standard area to a horizontal plane to obtain an initial plan view;

[0056] A coordinate system unit, configured to construct a coordinate system with latitude as the abscissa and longitude as the ordinate, integrate the initial plan view into the coordinate system, and divide the initial plan view into a number of equal rectangular areas;

[0057] An initial 3D model unit, configured to obtain GIS data and elevation data of each rectangular area and construct an initial 3D model of each rectangular area;

[0058] A correction unit, configured to perform correction processing on the initial 3D models of any two adjacent rectangular areas using a first method to obtain a correction result, and based on the correction result, obtain a corrected 3D model of each rectangular area;

[0059] A texture mapping unit, configured to obtain and process the aerial survey images of the standard area to obtain a processing result, determine a preset number of first feature points in each rectangular area, and based on the processing result and all the first feature points of each rectangular area, obtain a first image corresponding to each rectangular area, and map each first image to the corresponding corrected 3D model;

[0060] A smoothing processing unit, configured to perform smoothing processing on the first images of any two adjacent rectangular areas using a second method to obtain a first processing result;

[0061] A standard 3D model unit, configured to generate a standard 3D model of the complex terrain based on the first processing result.

[0062] One or more technical solutions provided by the present invention have at least the following technical effects or advantages:

[0063] The present invention discloses a method for modeling complex terrain. By converting the modeling of complex terrain into the processing of multiple rectangular regions, firstly, the difficulty of data processing is reduced, and secondly, the modeling of multiple rectangular regions can be carried out simultaneously, improving the modeling efficiency. However, when separately modeling and texturing multiple rectangular regions, the phenomenon of disharmony during fusion also needs to be considered. Therefore, the present invention also adopts the first method and the second method to process the three-dimensional model fusion of multiple rectangular regions and the first image fusion respectively, so as to ensure the accuracy of complex terrain modeling on the premise of improving the modeling efficiency of complex terrain.

[0064] The present invention also discloses a method for optimizing the standard three-dimensional model of complex terrain. By obtaining and processing the binary images of each rectangular region, grading each rectangular region, judging the vegetation coverage ratio of each rectangular region through the grade, and then based on the grade of each rectangular region, selecting the corresponding number of feature points. The larger the vegetation coverage area, the higher the grade of the rectangular region, and the more the corresponding number of feature points is selected. Then, within a preset time period, obtain the change curve of the pixel points included in the feature points, and based on the corresponding change curve, update the feature points included in each rectangular region, achieving the effect of updating the entire standard three-dimensional model of complex terrain within the preset time period, and further improving the authenticity of the three-dimensional model modeling of complex terrain. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, and constitute a part of the present invention, but do not limit the embodiments of the present invention;

[0066] Figure 1 It is a schematic flowchart of a method for modeling complex terrain in the present invention;

[0067] Figure 2 It is a schematic diagram of the corresponding transition region of the rectangular region in the present invention;

[0068] Figure 3 It is a schematic diagram of the composition of a modeling system for complex terrain in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0069] In order to more clearly understand the above objects, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the drawings and specific embodiments. It should be noted that, without conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other.

[0070] Many specific details are set forth in the following description in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described within the scope here. Therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0071] Example 1

[0072] Please refer to Figure 1 , Figure 1 which is a schematic flowchart of a modeling method for complex terrain in the present invention. The method includes the following steps:

[0073] Obtain the basic data of the complex terrain, and based on the basic data, determine the initial area that needs to be modeled for the complex terrain;

[0074] Obtain the smallest rectangle containing the initial area and denote it as the standard area, and map the standard area to the horizontal plane to obtain the initial plan view;

[0075] Construct a coordinate system with latitude as the abscissa and longitude as the ordinate, integrate the initial plan view into the coordinate system, and divide the initial plan view into several equal rectangular areas;

[0076] Obtain the GIS data and elevation data of each rectangular area, and construct the initial three-dimensional model of each rectangular area;

[0077] Adopt the first method to perform correction processing on the initial three-dimensional models of any two adjacent rectangular areas to obtain a correction result, and based on the correction result, obtain the corrected three-dimensional model of each rectangular area;

[0078] Obtain the aerial survey image of the standard area and perform processing to obtain a processing result. Determine a preset number of first feature points in each rectangular area. Based on the processing result and all the first feature points of each rectangular area, obtain the first image corresponding to each rectangular area, and map each first image to the corresponding corrected three-dimensional model;

[0079] Adopt the second method to perform smoothing processing on the first images of any two adjacent rectangular areas to obtain a first processing result;

[0080] Based on the correction result and the first processing result, generate the standard three-dimensional model of the complex terrain.

[0081] The specific embodiments of the present invention are as follows:

[0082] Modeling preprocessing:

[0083] Obtain the basic data of the complex terrain, which includes but is not limited to drawings, surveying and mapping data, satellite images, etc. Determine the initial area that needs to be modeled for the complex terrain through the basic data;

[0084] Obtain the smallest rectangle containing the initial area and denote it as the standard area, and map the standard area to the horizontal plane to obtain the initial plan view;

[0085] Construct a coordinate system with latitude as the abscissa and longitude as the ordinate, and integrate the initial plane map into the coordinate system. For example, make a vertex of the standard area in the initial plane map coincide with the origin of the coordinate system, and two sides of the standard area coincide with the horizontal and vertical axes of the coordinate system. Then divide the initial plane map into several equal rectangular areas.

[0086] Among them, the above steps are converted into the modeling of multiple rectangular areas in the plane map through the modeling of complex terrain, and the plane map is integrated into the coordinate system with latitude and longitude as the horizontal and vertical coordinates, which is more conducive to searching and data matching.

[0087] Modeling processing:

[0088] Obtain the GIS data and elevation data of each rectangular area, and construct the initial three-dimensional model of each rectangular area. By converting the modeling of complex terrain into the processing of multiple rectangular areas, on the one hand, the difficulty of data processing is reduced, and on the other hand, the modeling of multiple rectangular areas can be carried out simultaneously to improve the modeling efficiency. However, when modeling multiple rectangular areas separately, the phenomenon of incoordination in subsequent fusion needs to be considered. Therefore, this embodiment also provides a first method to correct the three-dimensional models of adjacent rectangular areas. The first method includes:

[0089] Preset transition areas around each rectangular area. The transition area of each rectangular area includes four overlapping areas and four non-overlapping areas;

[0090] Obtain any rectangular area denoted as the second rectangular area, and obtain a non-overlapping area in the second rectangular area denoted as area E and an overlapping area denoted as area F;

[0091] Obtain the adjacent non-overlapping area of area E denoted as area G, and statistically calculate the average elevation value of area E as the first elevation value, and the average elevation value of area G as the second elevation value;

[0092] Calculate the first difference between the first elevation value and the second elevation value, and determine whether the first difference is less than the first threshold. If not, do nothing. If so, calculate the average value of the first elevation value and the second elevation value denoted as the third elevation value, and based on the third elevation value, adjust the elevation values of area E and area G to complete the correction of the three-dimensional models corresponding to area E and area G;

[0093] Put area F and all adjacent overlapping areas into the preset area set, and statistically calculate the average elevation value of each area in the preset area set to obtain the calculation result;

[0094] Based on the calculated results, calculate the difference between the average elevation values corresponding to any two regions in the preset region set, which is denoted as the second difference. Determine whether all the second differences are less than the first threshold. If so, calculate the average value of the average elevation values of all regions in the preset region set, which is denoted as the fourth elevation value. Based on the fourth elevation value, adjust the elevation values of all regions in the preset region set to complete the correction of the three-dimensional models corresponding to all regions in the preset region set. If not, then do nothing.

[0095] Among them, please refer to Figure 2 , Figure 2 which is a schematic diagram of the corresponding transition region for a rectangular region in the present invention. It should be noted that Figure 2 the legends for the overlapping regions and non-overlapping regions in the transition region only serve to explain and illustrate. For example, if the Figure 2 rectangular region in is denoted as the second rectangular region, after presetting the transition region around the second rectangular region, the entire transition region will be divided into four overlapping regions and four non-overlapping regions. Denote one of the non-overlapping regions as region E and one of the overlapping regions as region F; obtain the adjacent non-overlapping region of region E, which is denoted as region G, calculate the difference between the average elevation values of region E and region G, and determine whether the difference is less than the first threshold. If so, it indicates that the possibility of the same mountain type (including flat land, hilly land, and high mountains, etc.) between region E and region G is relatively high. At this time, calculate the average elevation value of the region composed of region E and region G, which is denoted as the third elevation value. Based on this third elevation value, adjust the elevation values of region E and region G. If there is an elevation value in region E or region G that exceeds (or is lower than) the third elevation value by a certain margin, adjust the exceeded elevation value to the third elevation value. If not, it indicates that the possibility of different mountain types between region E and region G is relatively high, and at this time, do nothing. Similarly, the above method is also used to process the overlapping region F. The difference is that the adjacent regions of region F may be 1 or 3. Therefore, in this embodiment, region F and the adjacent overlapping regions are all placed in the preset region set, and then calculate the second difference between the average elevation values of any two regions in this preset region set, determine whether all the second differences are less than the first threshold. If so, calculate the average value of the average elevation values of all regions in the preset region set, which is denoted as the fourth elevation value. Based on the fourth elevation value, adjust the elevation values of all regions in the preset region set. The adjustment method has been described in the above embodiment. If not, then do nothing.

[0096] Texture processing:

[0097] Based on the aerial survey video stream of the standard area, multiple frames of aerial survey images of the standard area are obtained and corrected; a preset number of first feature points are determined in each rectangular area, and then the first feature points are used to match and search for the relative aerial survey image, and then the aerial survey images are integrated and cropped to obtain the first image of each rectangular area, and all the first images are mapped to the corresponding corrected three-dimensional model. However, when mapping all the first rectangular areas, there is also a phenomenon of incoordination in the subsequent fusion. Therefore, this embodiment also provides a second method for smoothing the first images of any two adjacent rectangular areas. The second method includes:

[0098] Obtaining first images corresponding to any two adjacent rectangular areas and recording them as second images and third images respectively;

[0099] Determine whether there is an overlapping area between the second image and the third image, if not, do not take any action, if yes, obtain the initial pixel value of each pixel in the overlapping area, assign a weight value to each pixel in the overlapping area using the first formula to obtain an assignment result, and based on the assignment result, use the second formula to calculate the initial pixel value of each pixel in the overlapping area to obtain a corresponding fused pixel value;

[0100] Based on the fused pixel value of each pixel point in the overlapping area, a smoothing process is performed on the overlapping area of the second image and the third image.

[0101] The first formula is:

[0102]

[0103] δ1 is the weight value of a single pixel point in the overlapping area in the second image, δ2 is the weight value of a single pixel point in the overlapping area in the third image, d1 is the distance between a single pixel point in the overlapping area and the center point of the second image, d2 is the distance between a single pixel point in the overlapping area and the center point of the third image, and B is a constant.

[0104] The second formula is:

[0105] W ′ =δ1×W1+δ2×W2

[0106] W ′ is the fused pixel value of a single pixel in the overlapping area, W1 is the initial pixel value of a single pixel in the overlapping area in the second image, and W2 is the initial pixel value of a single pixel in the overlapping area in the third image.

[0107] Among them, based on the aerial survey images and the first feature points in each rectangular area, the first image of each rectangular area is obtained through processing. When pasting all the first images onto the corresponding corrected 3D model, however, it is inevitable that there will be overlapping areas between adjacent rectangular areas. Therefore, in this embodiment, it is determined whether there is an overlapping area between the second image and the third image. If not, no action is taken. If so, corresponding weight values are assigned to each pixel point in the overlapping area through the first formula, and the initial pixel value of each pixel point in the overlapping area is obtained. Then, the initial pixel value of each pixel point in the overlapping area is calculated through the second formula to obtain the fused pixel value of a single pixel point in the overlapping area, thereby realizing smooth transition between adjacent images.

[0108] Finally, after preprocessing for modeling, modeling processing, and pasting processing, a standard 3D model of complex terrain is generated, ensuring the accuracy of the 3D model while improving the modeling efficiency.

[0109] Furthermore, this embodiment also provides an optimization method for the 3D model of complex terrain, including:

[0110] Obtain the first image of each rectangular area and process it to obtain the corresponding binary image;

[0111] Obtain the first ratio between the area of the shadow area and the area of the corresponding binary image in each binary image to obtain the first acquisition result;

[0112] Based on the first acquisition result, different rectangular areas are classified. The level of the rectangular area is proportional to the corresponding first ratio;

[0113] Obtain all the aerial survey images of the complex terrain within a preset time period to obtain the second acquisition result. Based on the second acquisition result, process all the first images of each rectangular area and place them in the first data set;

[0114] Based on the level of each rectangular area, select the corresponding number of second feature points in each rectangular area. The number of the second feature points is proportional to the level of the corresponding rectangular area;

[0115] Obtain the pixel points included in all the second feature points in each rectangular area, and based on the corresponding first data set, process to obtain the change curve of the pixel points corresponding to each second feature point in each rectangular area to obtain the second processing result;

[0116] Based on the second processing result, within the preset time period, update the pixel points corresponding to each second feature point in each rectangular area to complete the optimization of the standard 3D model.

[0117] The first ratio is calculated through the third formula:

[0118]

[0119] α n is the first ratio, A n is the area of the shaded region in the binarized image, S n is the area of the binarized image.

[0120] Based on the first acquisition result, the hierarchical division of different rectangular regions includes:

[0121] Obtain the historical data of all the first ratios. Based on the historical data, divide the data range of the first ratio, and each data range is assigned a corresponding level value;

[0122] Obtain any rectangular region denoted as the first rectangular region, and obtain the first ratio of the first rectangular region denoted as the second ratio;

[0123] Obtain the data range corresponding to the second ratio denoted as the first data range;

[0124] Obtain the level corresponding to the first data range as the level of the first rectangular region.

[0125] Obtain the first ratio of any rectangular region denoted as the second ratio, and obtain the data range corresponding to the second ratio denoted as the first data range;

[0126] Obtain the level value corresponding to the first data range as the level of the first rectangular region.

[0127] Among them, in this embodiment, by obtaining the binarized image of each rectangular region and processing it, by calculating the ratio between the area of the shaded region in each binarized image and the area of the entire image, the hierarchical division of each rectangular region is carried out. The vegetation coverage ratio of each rectangular region is judged by the level. Then, based on the level of each rectangular region, the corresponding number of feature points is selected. The larger the vegetation coverage area, the higher the level of the rectangular region, and the more the corresponding number of feature points is selected. Then, the change curve of the pixel points included in the feature points within a preset time period is obtained, and based on the corresponding change curve, the feature points included in each rectangular region are updated; for complex terrains, if the entire three-dimensional model needs to be updated in real time, the amount of data to be processed will be huge. In this example, by selecting different numbers of feature points in different rectangular regions and updating the selected feature points, the effect of updating the standard three-dimensional model of the entire complex terrain within a preset time period is achieved, thereby further improving the authenticity of the three-dimensional model modeling of complex terrains.

[0128] Embodiment 2

[0129] Please refer to Figure 3 , Figure 3It is a schematic diagram of the composition of a modeling system for complex terrain in the present invention. The system includes:

[0130] An initial area unit, which is used to obtain the basic data of the complex terrain and determine the initial area to be modeled for the complex terrain based on the basic data;

[0131] A mapping unit, which is used to obtain the smallest rectangle containing the initial area as the standard area, and map the standard area to the horizontal plane to obtain an initial plan view;

[0132] A coordinate system unit, which is used to construct a coordinate system with latitude as the abscissa and longitude as the ordinate, integrate the initial plan view into the coordinate system, and divide the initial plan view into several equal rectangular areas;

[0133] An initial 3D model unit, which is used to obtain the GIS data and elevation data of each rectangular area and construct the initial 3D model of each rectangular area;

[0134] A correction unit, which is used to correct the initial 3D models of any two adjacent rectangular areas by using a first method to obtain a correction result, and obtain the corrected 3D model of each rectangular area based on the correction result;

[0135] A texture mapping unit, which is used to obtain the aerial survey image of the standard area and perform processing to obtain a processing result, determine a preset number of first feature points in each rectangular area, obtain the first image corresponding to each rectangular area based on the processing result and all the first feature points of each rectangular area, and map each first image to the corresponding corrected 3D model;

[0136] A smoothing processing unit, which is used to smooth the first images of any two adjacent rectangular areas by using a second method to obtain a first processing result;

[0137] A standard 3D model unit, which is used to generate the standard 3D model of the complex terrain based on the first processing result.

[0138] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they know the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications falling within the scope of the present invention.

[0139] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A modeling method for complex terrain, characterized in that The method includes the following steps: Obtain the basic data of the complex terrain, and based on the basic data, determine the initial area to be modeled for the complex terrain; Obtain the smallest rectangle containing the initial area and denote it as the standard area, and map the standard area to the horizontal plane to obtain the initial plan view; Construct a coordinate system with latitude as the abscissa and longitude as the ordinate, integrate the initial plan view into the coordinate system, and divide the initial plan view into several equal rectangular areas; Obtain the GIS data and elevation data of each rectangular area, and construct the initial 3D model of each rectangular area; Use the first method to perform correction processing on the initial 3D models of any two adjacent rectangular areas to obtain a correction result, and based on the correction result, obtain the corrected 3D model of each rectangular area; Obtain the aerial survey image of the standard area and perform processing to obtain a processing result, determine a preset number of first feature points in each rectangular area, and based on the processing result and all the first feature points of each rectangular area, obtain the first image corresponding to each rectangular area, and map each first image to the corresponding corrected 3D model; Use the second method to perform smoothing processing on the first images of any two adjacent rectangular areas to obtain a first processing result; Based on the correction result and the first processing result, generate the standard 3D model of the complex terrain; The optimization of the standard 3D model includes: Obtain the first image of each rectangular area and perform processing to obtain the corresponding binary image; Obtain the first ratio between the area of the shadow region and the area of the corresponding binary image in each binary image to obtain a first acquisition result; Based on the first acquisition result, perform grade division on different rectangular areas, and the grade of the rectangular area is proportional to the corresponding first ratio; Obtain all the aerial survey images of the complex terrain within a preset time period to obtain a second acquisition result, and based on the second acquisition result, process and obtain all the first images of each rectangular area and place them in the first data set; Based on the grade of each rectangular area, select the corresponding number of second feature points in each rectangular area, and the number of the second feature points is proportional to the grade of the corresponding rectangular area; Obtain the pixel points included in all the second feature points in each rectangular area, and based on the corresponding first data set, process and obtain the change curve of the pixel points corresponding to each second feature point in each rectangular area to obtain a second processing result; Based on the second processing result, update the pixel points corresponding to each second feature point in each rectangular area within the preset time period to complete the optimization of the standard 3D model.

2. The modeling method for a complex terrain according to claim 1, wherein The first method includes: Preset transition areas around each rectangular area, and each transition area of each rectangular area includes four overlapping areas and four non-overlapping areas; Obtain any rectangular area and denote it as the second rectangular area, and obtain an un-overlapping area in the second rectangular area and denote it as area E and an overlapping area and denote it as area F; Obtain the adjacent un-overlapping area of area E and denote it as area G, and statistically calculate and obtain the average elevation value of area E as the first elevation value, and the average elevation value of area G as the second elevation value; Calculate the first difference between the first elevation value and the second elevation value, and determine whether the first difference is less than a first threshold. If not, do nothing. If so, calculate the average value of the first elevation value and the second elevation value and denote it as the third elevation value. Based on the third elevation value, adjust the elevation values of region E and region G to complete the correction of the 3D models corresponding to region E and region G. Place region F and all adjacent overlapping regions into a preset region set, and statistically calculate the average elevation value of each region in the preset region set to obtain a calculation result. Based on the calculation result, calculate the difference between the average elevation values of any two regions in the preset region set and denote it as the second difference. Determine whether all second differences are less than the first threshold. If so, calculate the average value of the average elevation values of all regions in the preset region set and denote it as the fourth elevation value. Based on the fourth elevation value, adjust the elevation values of all regions in the preset region set to complete the correction of the 3D models corresponding to all regions in the preset region set. If not, do nothing.

3. A modeling method for complex terrain according to claim 1, characterized in that The second method includes: Obtain the first images corresponding to any two adjacent rectangular regions and denote them as the second image and the third image respectively. Determine whether there is an overlapping region between the second image and the third image. If not, do nothing. If so, obtain the initial pixel value of each pixel point in the overlapping region, and assign a weight value to each pixel point in the overlapping region through the first formula to obtain an assignment result. Based on the assignment result, use the second formula to calculate the initial pixel value of each pixel point in the overlapping region to obtain the corresponding fused pixel value. Based on the fused pixel value of each pixel point in the overlapping region, perform smoothing processing on the overlapping region of the second image and the third image.

4. A modeling method for complex terrain according to claim 3, characterized in that, The first formula is: is the weight value of a single pixel point in the overlapping area in the second image, is the weight value of a single pixel point in the overlapping area in the third image, is the distance between a single pixel point in the overlapping area and the center point of the second image, is the distance between a single pixel point in the overlapping area and the center point of the third image, where B is a constant.

5. A modeling method for complex terrain according to claim 4, characterized in that, The second formula is: is the fused pixel value of a single pixel in the overlapping area, is the initial pixel value of a single pixel in the overlapping area in the second image, is the initial pixel value of a single pixel in the overlapping area in the third image.

6. A modeling method for complex terrain according to claim 1, characterized in that The first ratio is calculated through the third formula: is the first ratio, is the area of the shadow region in the binarized image, is the area of the binarized image.

7. A modeling method for complex terrain according to claim 1, characterized in that Based on the first acquisition result, the hierarchical division of different rectangular regions includes: Obtain the historical data of all first ratios. Based on the historical data, divide the data interval of the first ratio, and assign a corresponding grade value to each data interval. Obtain any rectangular region and denote it as the first rectangular region, and obtain the first ratio of the first rectangular region and denote it as the second ratio. Obtain the data interval corresponding to the second ratio and denote it as the first data interval. Obtain the grade corresponding to the first data interval as the grade of the first rectangular region. Obtain the first ratio of any rectangular region and denote it as the second ratio, and obtain the data interval corresponding to the second ratio and denote it as the first data interval. Obtain the grade value corresponding to the first data interval as the grade of the first rectangular region.

8. A modeling method for complex terrain according to claim 1, characterized in that Obtain the aerial survey image of the standard region and perform processing, including: based on the aerial survey video stream of the standard region, obtain multiple frames of aerial survey images of the standard region and perform calibration processing.

9. A modeling system for complex terrains, characterized in that, The system includes: An initial region unit for obtaining the basic data of the complex terrain and determining the initial region that needs to be modeled based on the basic data. A mapping unit, configured to obtain the smallest rectangle containing the initial area as the standard area, and map the standard area to a horizontal plane to obtain an initial plan view; A coordinate system unit, configured to construct a coordinate system with latitude as the abscissa and longitude as the ordinate, integrate the initial plan view into the coordinate system, and divide the initial plan view into a plurality of equal rectangular areas; An initial 3D model unit, configured to obtain GIS data and elevation data of each rectangular area, and construct an initial 3D model of each rectangular area; A correction unit, configured to perform correction processing on the initial 3D models of any two adjacent rectangular areas by using a first method to obtain a correction result, and based on the correction result, obtain a corrected 3D model of each rectangular area; A texture mapping unit, configured to obtain an aerial image of the standard area and perform processing to obtain a processing result, determine a preset number of first feature points in each rectangular area, and based on the processing result and all the first feature points of each rectangular area, obtain a first image corresponding to each rectangular area, and map each first image to the corresponding corrected 3D model; A smoothing processing unit, configured to perform smoothing processing on the first images of any two adjacent rectangular areas by using a second method to obtain a first processing result; A standard 3D model unit, configured to generate a standard 3D model of the complex terrain based on the first processing result, and the optimization of the standard 3D model includes: Obtaining the first image of each rectangular area and performing processing to obtain a corresponding binary image; Obtaining a first ratio between the area of the shadow area and the area of the corresponding binary image in each binary image to obtain a first obtaining result; Based on the first obtaining result, performing grade division on different rectangular areas, and the grade of the rectangular area is proportional to the corresponding first ratio; Obtaining all the aerial images of the complex terrain within a preset time period to obtain a second obtaining result, and based on the second obtaining result, processing to obtain all the first images of each rectangular area and storing them in a first data set; Based on the grade of each rectangular area, selecting a corresponding number of second feature points in each rectangular area, and the number of the second feature points is proportional to the grade of the corresponding rectangular area; Obtaining the pixel points included in all the second feature points in each rectangular area, and based on the corresponding first data set, processing to obtain a change curve of the pixel points corresponding to each second feature point in each rectangular area to obtain a second processing result; Based on the second processing result, updating the pixel points corresponding to each second feature point in each rectangular area within the preset time period to complete the optimization of the standard 3D model.

Images