Model generation method and device, computer equipment and storage medium
The real-life model was reconstructed by drone collecting tilted image data, and using coordinate transformation and point cloud fusion technology, the coordinate system inconsistency between satellite models and real-life models was solved, achieving seamless fusion and visual effect improvement of multi-source terrain models.
Patent Information
- Application Number
- CN202410038934.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-01-10
- Publication Date
- 2025-07-11
AI Technical Summary
In the prior art, the coordinate systems and textures of satellite models and real-life models are inconsistent, making it difficult to automatically process multi-source terrain models, especially poorly processed at edges and transition areas.
The real scene model is reconstructed by drone collecting tilted image data, and coordinate conversion, terrain extraction, point cloud fusion, texture map and other operations are adopted to realize the fusion processing of multi-source terrain models at the point cloud level to ensure coordinate consistency and texture matching.
The seamless fusion of transition areas of multi-source topographic models has been achieved, which improves the visual effect and avoids the need for artificial interaction and fusion.
Smart Images
Figure CN120298568A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of geographic information technology, and particularly to a model generation method, device, computer device and storage medium. Background Art
[0002] Satellite elevation images (Geographic Information System, abbreviated as GIS) have the advantages of wide application and large coverage area, and are one of the important ways for users to obtain geographic environment information of a specific area. However, their update efficiency is relatively low and the clarity is poor. Unmanned aerial vehicles have the advantages of small size, low cost, flexible operation, high timeliness, etc., and the images collected have high accuracy, and the resolution can reach the centimeter level.
[0003] The current mainstream method is to complete the fusion of multi-source terrain models through manual interaction and model modification. However, the coordinate systems and textures of satellite models and real-scene models are usually inconsistent, and the terrain is composed of irregular triangular patches with large edge undulations, which is generally difficult to automate, especially in the edge and transition areas where the processing is not good. Summary of the Invention
[0004] In view of this, the present invention provides a model generation method, device, computer device and storage medium to solve the problem of manually interacting to complete the fusion of multi-source terrain models.
[0005] In a first aspect, the present invention provides a model generation method, including: obtaining real-scene images and satellite images of a target area; parsing the surface information of the real-scene images to determine the surface point cloud data corresponding to the real-scene images; parsing the geographic information of the satellite images to determine the geographic point cloud data corresponding to the satellite images; performing data fusion based on the surface point cloud data and the geographic point cloud data to generate target point cloud data; and constructing a three-dimensional model of the target area based on the target point cloud data.
[0006] The model generation method provided by the embodiments of the present invention, after obtaining the real-scene images and satellite images of the target area, parses the surface information of the real-scene images and the geographic information of the satellite images, respectively determines the surface point cloud data corresponding to the real-scene images and the geographic point cloud data corresponding to the satellite images, performs data fusion according to the surface point cloud data and the geographic point cloud data to generate target point cloud data, and constructs a three-dimensional model of the target area according to the generated target point cloud data, thereby realizing the fusion processing of multi-source terrain models at the point cloud level. Without the need for manual interactive fusion, the connection processing of the transition area of the multi-source terrain model has a better visual effect.
[0007] In an alternative embodiment, data fusion is performed based on surface point cloud data and geographic point cloud data to generate target point cloud data, including: performing data processing on the surface point cloud data to obtain target geographic point cloud data matching the surface point cloud data; obtaining a first coordinate system corresponding to the target geographic point cloud data, first coordinate information corresponding to the target geographic point cloud data, and a second coordinate system corresponding to the geographic point cloud data; converting the second coordinate system to the first coordinate system to obtain second coordinate information of the geographic point cloud data in the first coordinate system; and fusing the target geographic point cloud data and the geographic point cloud data based on the first coordinate information and the second coordinate information to generate target point cloud data.
[0008] The model generation method provided by the embodiments of the present invention performs data processing on surface point cloud data to obtain target geographic point cloud data corresponding to a real scene image, obtains a first coordinate system corresponding to the target geographic point cloud data, first coordinate information corresponding to the target geographic point cloud data, and a second coordinate system of the geographic point cloud data corresponding to a satellite image. The second coordinate system is converted to the first coordinate system to obtain second coordinate information of the geographic point cloud data in the first coordinate system, and the target geographic point cloud data and the geographic point cloud data are fused through the obtained first coordinate information and second coordinate information to generate target point cloud data, so as to construct a three-dimensional model of the target area based on the target point cloud data, thereby finding the satellite position area corresponding to the real scene target area from a large number of satellite images, automatically completing coordinate alignment according to the data stream conversion, and automatically fitting without additional manual correction.
[0009] In an alternative embodiment, converting the second coordinate system to the first coordinate system includes: obtaining the origin of the first coordinate system; using the origin as a coordinate system conversion reference point, and converting the second coordinate system to the first coordinate system based on the reference point.
[0010] The model generation method provided by the embodiments of the present invention has different coordinate systems for the target geographic point cloud data corresponding to a real scene image and the geographic point cloud data corresponding to a satellite. Taking the target geographic point cloud data as a reference, using the origin of the first coordinate system as a base point, the alignment of different coordinate systems is completed through mathematical transformation to ensure the coordinate consistency of the two, and automatic fitting can be achieved without any manual correction.
[0011] In an alternative embodiment, in response to a drawing operation on a fusion area, determining a target fusion area of the target geographic point cloud data and the geographic point cloud data; cropping the target geographic point cloud data based on the target fusion area to obtain first geographic point cloud data matching the target fusion area; cropping the geographic point cloud data based on the target fusion area to obtain second geographic point cloud data matching the target fusion area; and fusing the first geographic point cloud data and the second geographic point cloud data to generate geographic point cloud fusion data matching the target fusion area.
[0012] The model generation method provided by an embodiment of the present invention determines target geographic point cloud data and a target fusion area of the geographic point cloud data according to a drawing operation of a fusion area by a user at the front end, trims the target geographic point cloud data and the geographic point cloud data according to the contour of the target fusion area, obtains first geographic point cloud data and second geographic point cloud data that match the target fusion area, and fuses the first geographic point cloud data and the second geographic point cloud data to generate geographic point cloud fusion data that matches the target fusion area. Therefore, by performing point cloud level processing, the first geographic point cloud data and the second geographic point cloud data are fused and spliced into a complete point cloud fusion data, making the connection processing of the transition area of the multi-source terrain model have a better visual effect without the need for manual interactive fusion.
[0013] In an alternative embodiment, fusing the first geographic point cloud data and the second geographic point cloud data to generate geographic point cloud fusion data that matches the target fusion area includes: determining the splicing level of the first geographic point cloud data and the second geographic point cloud data; splicing the first geographic point cloud data and the second geographic point cloud data within the target fusion area according to the splicing level to obtain geographic point cloud fusion data.
[0014] The model generation method provided by an embodiment of the present invention realizes the fusion processing of the multi-source terrain model at the point cloud level by respectively determining the splicing levels of the first geographic point cloud data and the second geographic point cloud data, and splicing the first geographic point cloud data and the second geographic point cloud data within the target fusion area according to the determined splicing levels to obtain geographic point cloud fusion data, making the connection processing of the transition area of the multi-source terrain model have a better visual effect without the need for manual interactive fusion.
[0015] In an alternative embodiment, constructing a three-dimensional model of a target area based on the target point cloud data includes: performing terrain rendering on the target point cloud data to obtain a terrain network model; performing texture mapping on the terrain network model to generate a three-dimensional model of the target area.
[0016] The model generation method provided by an embodiment of the present invention performs terrain rendering on the generated target point cloud data to obtain a terrain network model, and performs texture mapping on the terrain network model to generate a three-dimensional model of the target area, thereby ensuring the rendering continuity of the transition area between the target geographic point cloud data and the geographic point cloud data and completing the seamless fusion processing of the multi-source terrain model.
[0017] In an alternative embodiment, texture mapping is performed on the terrain network model to generate a three-dimensional model of the target area, including: cropping the terrain network model based on the contour of the target fusion area to determine a first terrain network model inside the contour and a second terrain network model outside the contour; determining a first texture corresponding to the first terrain network model and a second texture corresponding to the second terrain network model; and performing texture mapping on the first terrain network model and the second terrain network model based on the first texture and the second texture to generate a three-dimensional model corresponding to the target area.
[0018] The model generation method provided by the embodiments of the present invention crops the terrain network model according to the contour of the target fusion area to determine a first terrain network model inside the contour and a second terrain network model outside the contour. The first texture corresponding to the first terrain network model and the second texture corresponding to the second terrain network model are respectively determined according to the acquired real-scene image and satellite image, and texture mapping is performed on the first terrain network model and the second terrain network model to generate a three-dimensional model corresponding to the target area, thereby ensuring the matching degree and consistency of the textures of the first terrain network model and the second terrain network model to achieve a seamless transition effect and avoiding the problem of poor stitching effect caused by inconsistent accuracies of multi-source models.
[0019] In a second aspect, the present invention provides a model generation device, including: an acquisition module for acquiring a real-scene image and a satellite image of a target area; a first parsing module for parsing the surface information of the real-scene image to determine the surface point cloud data corresponding to the real-scene image; a second parsing module for parsing the geographical information of the satellite image to determine the geographical point cloud data corresponding to the satellite image; a generation module for performing data fusion based on the surface point cloud data and the geographical point cloud data to generate target point cloud data; and a construction module for constructing a three-dimensional model of the target area based on the target point cloud data.
[0020] In a third aspect, the present invention provides a computer device, including: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to execute the model generation method according to the first aspect or any corresponding embodiment thereof.
[0021] In a fourth aspect, the present invention provides a computer-readable storage medium, on which computer instructions are stored. The computer instructions are used to cause a computer to execute the model generation method according to the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in related technologies, the following will briefly introduce the accompanying drawings required for use in the description of the specific embodiments or related technologies. Obviously, the accompanying drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.
[0023] Figure 1 is a schematic flowchart of a model generation method according to an embodiment of the present invention;
[0024] Figure 2 is a schematic flowchart of another model generation method according to an embodiment of the present invention;
[0025] Figure 3 is a schematic flowchart of yet another model generation method according to an embodiment of the present invention;
[0026] Figure 4 is a structural block diagram of a model generation device according to an embodiment of the present invention;
[0027] Figure 5 is a schematic hardware structure diagram of a computer device according to an embodiment of the present invention. Specific Embodiments
[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.
[0029] Satellite geographic information technology (Geographic Information System, abbreviated as GIS) plays an important role in modern society and has wide applications and influences. GIS provides a powerful tool for users by integrating, analyzing, and visualizing geospatial data, promoting the sustainable development and intelligent process of society.
[0030] Due to the characteristics of wide application and large coverage area of satellite geographic information technology, it has become one of the important ways for users to obtain geographical environment information of specific regions. However, the satellite geographic information technology has a low update efficiency and low resolution and accuracy of geographical images, providing a poor user experience for users.
[0031] Drones have the advantages of small size, low cost, flexible operation, high timeliness, etc., and the images collected have high accuracy, and the resolution can reach the centimeter level. Therefore, it is very meaningful to seamlessly fuse the real scene model reconstructed from the drone oblique images with the satellite model to generate a panoramic model with a broader view, which is beneficial to the analysis and display of the target area, and thus better serves the business scenario.
[0032] The current mainstream method is to complete it through manual interaction and model modification. However, the coordinate systems of the satellite model and the real scene model are usually inconsistent, and the textures are also inconsistent. Moreover, the terrain is composed of irregular triangular patches with large edge undulations, which are generally difficult to automate, especially in the edge and transition areas where the processing is not good.
[0033] In view of this, the technical solution of the present invention can reconstruct the real scene model by collecting oblique image data of the drone, and realizes the fusion processing of multi-source terrain models at the point cloud level through a series of operations such as coordinate transformation, terrain extraction, point cloud fusion, terrain restoration, and texture mapping, making the connection processing of the transition area between the real scene model and the satellite model smoother and with better visual effects.
[0034] According to an embodiment of the present invention, an embodiment of a model generation method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.
[0035] In this embodiment, a model generation method is provided, which can be used in computer devices such as laptops and tablets. Figure 1 It is a flowchart of the model generation method according to an embodiment of the present invention, as Figure 1 shown, and this process includes the following steps.
[0036] Step S101, obtain the real scene image and satellite image of the target area.
[0037] The real scene image is the actual terrain image of the target area, that is, it may include building information, vegetation information, vehicle information, etc. located in the target area. Specifically, through the drone oblique photogrammetry technology, the multi-angle cameras carried by the drone are used to synchronously obtain the high-resolution aerial photography real scene images of the ground objects at various angles in the target area, and the aerial photography real scene images are input into the computer device.
[0038] Satellite images are obtained by remote sensing satellites in space detecting the reflection of electromagnetic waves by surface objects in the target area and the electromagnetic waves emitted by them, so as to extract object information and complete long-distance object recognition. These electromagnetic waves are converted and recognized to obtain the satellite images of the target area. Specifically, the satellite images of the target area can be determined through the satellite map program carried by the computer device, or after the computer device is connected to the network, by searching for the satellite map program containing the satellite images to determine the satellite images of the target area.
[0039] Step S102: Analyze the surface information of the real-scene image and determine the surface point cloud data corresponding to the real-scene image.
[0040] The parsing algorithm carried by the computer device can automatically parse the real-scene images collected by the above-mentioned UAV. The real-scene images contain the surface information of the target area. The surface information can include, for example, information such as buildings, big trees, and the ground. Specifically, the surface point cloud data corresponding to the real-scene image can be determined according to the stereo vision method and the deep learning method carried by the computer device. Among them, the stereo vision method is to take pictures of the target area from different angles by two or more cameras carried by the UAV, and then calculate the differences between the images of different target areas to infer the depth information. Combining the length information and width information included in the real-scene image, a three-dimensional reconstruction of the real-scene image is performed to generate a real-scene model corresponding to the real-scene image, and then the real-scene model is converted into a three-dimensional point cloud to determine the surface point cloud data corresponding to the real-scene image. The deep learning method is to use a deep neural network to directly predict the depth information from the real-scene image pictures, and then combine the length information and width information included in the real-scene image to determine the surface point cloud data corresponding to the real-scene image.
[0041] Step S103: Analyze the geographical information of the satellite image and determine the geographical point cloud data corresponding to the satellite image.
[0042] Specifically, when the computer device obtains the satellite image corresponding to the target area, the satellite image is parsed according to the above-mentioned parsing algorithm carried by the computer device to obtain the elevation information of the satellite image, and a digital elevation model (Digital Elevation Model, DEM) corresponding to the satellite image is generated according to the elevation information of the satellite image. Further, according to the three-dimensional point cloud conversion algorithm carried by the computer device, the digital elevation model corresponding to the satellite image is converted into a three-dimensional point cloud to determine the geographical point cloud data corresponding to the satellite image.
[0043] Step S104: Perform data fusion based on the surface point cloud data and the geographical point cloud data to generate target point cloud data.
[0044] After processing the ground point cloud data corresponding to the real scene image, the target geographic point cloud data corresponding to the real scene image is obtained. The coordinate system of the satellite image geographic point cloud data is converted to the coordinate system corresponding to the real scene image target geographic point cloud data, so that the coordinate systems of the geographic point cloud data and the target geographic point cloud data are aligned. Then, the geographic point cloud data and the target geographic point cloud data are fused according to the preset point cloud level stitching method to obtain the target point cloud data corresponding to the target area.
[0045] Step S105, construct a three-dimensional model of the target area based on the target point cloud data.
[0046] After determining the target point cloud data corresponding to the target area, the computer device performs terrain rendering on the target point cloud data according to the carried terrain rendering tool to obtain the terrain network model corresponding to the target area. Texture mapping is performed on the terrain network model according to the texture images corresponding to the real scene image and the satellite image to generate the three-dimensional model of the target area.
[0047] The model generation method provided by the embodiments of the present invention, after obtaining the real scene image and the satellite image of the target area, analyzes the ground information of the real scene image and the geographic information of the satellite image, respectively determines the ground point cloud data corresponding to the real scene image and the geographic point cloud data corresponding to the satellite image, performs data fusion according to the ground point cloud data and the geographic point cloud data to generate the target point cloud data, and constructs a three-dimensional model of the target area according to the generated target point cloud data, thereby realizing the fusion processing of multi-source terrain models at the point cloud level. Without the need for manual interactive fusion, the connection processing of the transition area of the multi-source terrain model has a better visual effect.
[0048] In this embodiment, a model generation method is provided, which can be used in the above computer devices, such as laptops, tablets, etc. Figure 2 is a flowchart of the model generation method according to the embodiments of the present invention, as Figure 2 shown, and this process includes the following steps.
[0049] Step S201, obtain the real scene image and the satellite image of the target area. For details, please refer to Figure 1 the steps of embodiment shown in S101, which will not be elaborated here.
[0050] Step S202, analyze the ground information of the real scene image and determine the ground point cloud data corresponding to the real scene image. For details, please refer to Figure 1 the steps of embodiment shown in S102, which will not be elaborated here.
[0051] Step S203, analyze the geographic information of the satellite image and determine the geographic point cloud data corresponding to the satellite image. For details, please refer to Figure 1 the steps of embodiment shown in S103, which will not be elaborated here.
[0052] Step S204: Perform data fusion based on the surface point cloud data and the geographic point cloud data to generate target point cloud data.
[0053] Specifically, the above step S204 includes:
[0054] Step S2041: Process the surface point cloud data to obtain target geographic point cloud data that matches the surface point cloud data.
[0055] For the surface point cloud data corresponding to the real-scene image, according to the AI algorithm and the terrain algorithm improved by normalization, such as the improved CSF cloth simulation algorithm that does not require parameter adjustment, filter the surface point cloud data corresponding to the target area to remove the point cloud data of the obstacles in the surface point cloud data. The obstacles can be objects such as buildings and big trees, and the ground point cloud data is separated.
[0056] Perform interpolation processing on the ground point cloud data to generate target geographic point cloud data that matches the surface point cloud data. Specifically, the target geographic point cloud data that matches the surface point cloud data can be generated by the bilinear interpolation method.
[0057] Step S2042: Obtain the first coordinate system corresponding to the target geographic point cloud data, the first coordinate information corresponding to the target geographic point cloud data, and the second coordinate system corresponding to the geographic point cloud data.
[0058] The 3D reconstruction tool carried by the computing device completes the 3D reconstruction of the real-scene image according to the length information, width information, and depth information corresponding to the real-scene image, and outputs the reconstructed real-scene model. Determine the central origin of the reconstructed real-scene model according to the center point of the real-scene image picture, and establish the reconstructed real-scene model according to the central origin, so as to determine the first coordinate system corresponding to the target geographic point cloud data and the first coordinate information corresponding to the target geographic point cloud data.
[0059] It can be understood that the length information, width information, and depth information correspond to the coordinates (x, y, z), that is, the length information is x, the width information is y, and the depth information is z. Determine the coordinate information of each target geographic point cloud data, and determine the coordinate set information of each target geographic point cloud data as the first coordinate information.
[0060] The coordinate system used by the satellite map program is usually the geodetic coordinate system, that is, the second coordinate system corresponding to the geographic point cloud data determined according to the satellite map program is also the geodetic coordinate system.
[0061] Step S2043: Convert the second coordinate system to the first coordinate system to obtain the second coordinate information of the geographic point cloud data in the first coordinate system.
[0062] Based on the reconstructed real - scene model, with the central origin of the reconstructed real - scene model as the reference point, the geodetic coordinates of the geographic point cloud data are converted to the first coordinate system where the reconstructed real - scene model is located, aligning its coordinates with the coordinate system of the reconstructed real - scene model to obtain the second coordinate information of the geographic point cloud data in the first coordinate system.
[0063] In some alternative embodiments, the above - mentioned step S2043 includes:
[0064] Step a1, obtain the origin of the first coordinate system.
[0065] The first coordinate system of the reconstructed real - scene model is the local - level - east - north - up (ENU) coordinate system. The computer device can determine the center point of the real - scene image according to the length information and width information of the real - scene image, determine the central origin of the reconstructed real - scene model with the center point of the real - scene image, and establish the reconstructed real - scene model based on the central origin, that is, the central origin of the reconstructed real - scene model is the origin of the first coordinate system.
[0066] Step a2, taking the origin as the reference point for coordinate system conversion, and converting the second coordinate system to the first coordinate system based on the reference point.
[0067] Determine the central origin of the reconstructed real - scene model as the origin of the geodetic coordinates, earth - centered earth - fixed coordinates, and local - level - east - north - up coordinate system. Directly determine the geodetic coordinates (B, L, H) corresponding to each geographic point cloud data through the satellite map program, convert the geodetic coordinates corresponding to each geographic point cloud data to the earth - centered earth - fixed coordinates (X, Y, Z) according to the translation matrix and rotation matrix, and then convert the earth - centered earth - fixed coordinates (X, Y, Z) to the local - level - east - north - up coordinates (E, N, U) through the coordinate system conversion formula.
[0068] In the above - mentioned embodiment, the target geographic point cloud data corresponding to the real - scene image and the geographic point cloud data corresponding to the satellite have different coordinate systems. Taking the target geographic point cloud data as the benchmark, with the origin of the first coordinate system as the base point, the alignment of different coordinate systems is completed through mathematical transformation to ensure the coordinate consistency of the two, and they can be automatically matched without any manual correction.
[0069] Step S2044, fuse the target geographic point cloud data and the geographic point cloud data based on the first coordinate information and the second coordinate information to generate the target point cloud data.
[0070] Based on the reconstructed real - scene model as the benchmark, with the central origin of the reconstructed real - scene model as the reference point, convert the geodetic coordinates of the geographic point cloud data to the local - level - east - north - up coordinate system where the reconstructed real - scene model is located. After aligning its coordinates with the coordinate system of the reconstructed real - scene model, fuse the target geographic point cloud data and the geographic point cloud data to generate the target point cloud data.
[0071] Step S2045, in response to a drawing operation on the fusion region, determine the target geographic point cloud data and the target fusion region of the geographic point cloud data.
[0072] When the second coordinate system is converted to the first coordinate system, the target geographic point cloud data and the geographic point cloud data are virtually superimposed and displayed on the display device corresponding to the computer device. The user can draw an area of interest in the virtual superimposed area, and the drawn area is determined as the target fusion region of the target geographic point cloud data and the geographic point cloud data.
[0073] Step S2046, crop the target geographic point cloud data based on the target fusion region to obtain the first geographic point cloud data that matches the target fusion region.
[0074] According to the user's drawing operation, determine the contour of the target fusion region, and perform a point cloud cropping operation on the target geographic point cloud data according to the contour of the target fusion region, retaining the first geographic point cloud data within the contour of the target fusion region.
[0075] Step S2047, crop the geographic point cloud data based on the target fusion region to obtain the second geographic point cloud data that matches the target fusion region.
[0076] According to the user's drawing operation, determine the contour of the target fusion region, and perform a point cloud cropping operation on the geographic point cloud data according to the contour of the target fusion region, retaining the second geographic point cloud data outside the contour of the target fusion region.
[0077] Step S2048, fuse the first geographic point cloud data and the second geographic point cloud data to generate geographic point cloud fusion data that matches the target fusion region.
[0078] Perform point cloud stitching and fusion on the first geographic point cloud data and the second geographic point cloud data. Specifically, point cloud stitching and fusion of the first geographic point cloud data and the second geographic point cloud data can be performed according to point cloud registration and point cloud interpolation to generate geographic point cloud fusion data that matches the target fusion region.
[0079] In some alternative embodiments, the above step S2048 includes:
[0080] Step b1, determine the stitching level of the first geographic point cloud data and the second geographic point cloud data.
[0081] The splicing level is the splicing position level of the geographic point cloud data. Specifically, since the target geographic point cloud data within the contour of the target fusion area is determined as the first geographic point cloud data and the geographic point cloud data outside the contour of the target fusion area is determined as the second geographic point cloud data, the computer device can determine that the splicing level of the first geographic point cloud data is the internal splicing level and the splicing level of the second geographic point cloud data is the external splicing level according to the position information of the first geographic point cloud data and the second geographic point cloud data.
[0082] Step b2: Splice the first geographic point cloud data and the second geographic point cloud data within the target fusion area according to the splicing level to obtain the geographic point cloud fusion data.
[0083] The computer device splices the first geographic point cloud data and the second geographic point cloud data within the target fusion area according to the determined internal splicing level corresponding to the first geographic point cloud data and the external splicing level corresponding to the second geographic point cloud data, that is, the geographic point cloud data within the contour of the target fusion area is the first geographic point cloud data, and the geographic point cloud data outside the contour of the target fusion area is the second geographic point cloud data, to obtain the geographic point cloud fusion data corresponding to the target area.
[0084] In the above embodiment, by separately determining the splicing levels of the first geographic point cloud data and the second geographic point cloud data, and splicing the first geographic point cloud data and the second geographic point cloud data within the target fusion area according to the determined splicing levels, the geographic point cloud fusion data is obtained, thereby realizing the fusion processing of multi-source terrain models at the point cloud level. Without the need for manual interactive fusion, the connection processing of the transition area of the multi-source terrain model has a better visual effect.
[0085] Step S205: Construct a three-dimensional model of the target area based on the target point cloud data. For details, please refer to Figure 1 Step S105 of the illustrated embodiment, which will not be elaborated here.
[0086] The model generation method provided by the embodiments of the present invention processes the surface point cloud data to obtain the target geographic point cloud data corresponding to the real scene image, and obtains the first coordinate system corresponding to the target geographic point cloud data, the first coordinate information corresponding to the target geographic point cloud data, and the second coordinate system of the geographic point cloud data corresponding to the satellite image. The second coordinate system is converted to the first coordinate system to obtain the second coordinate information of the geographic point cloud data in the first coordinate system. The target geographic point cloud data and the geographic point cloud data are fused through the obtained first coordinate information and second coordinate information to generate the target point cloud data, so as to construct a three-dimensional model of the target area according to the target point cloud data, thereby finding the satellite position area corresponding to the real scene target area from the huge satellite images, automatically completing coordinate alignment according to the data stream conversion, and automatically matching without additional manual correction. According to the drawing operation of the fusion area by the user at the front end, the target fusion area of the target geographic point cloud data and the geographic point cloud data is determined. According to the contour of the target fusion area, the target geographic point cloud data and the geographic point cloud data are cropped to obtain the first geographic point cloud data and the second geographic point cloud data matching the target fusion area, and the first geographic point cloud data and the second geographic point cloud data are fused to generate the geographic point cloud fusion data matching the target fusion area. Therefore, by adopting point cloud level processing, the first geographic point cloud data and the second geographic point cloud data are fused and spliced into a complete point cloud fusion data, making the connection processing of the transition area of the multi-source terrain model have a better visual effect without the need for manual interactive fusion.
[0087] In this embodiment, a model generation method is provided, which can be used in the above computer devices, such as laptops, tablets, etc. Figure 3 It is a flowchart of the model generation method according to the embodiments of the present invention, as Figure 3 shown, and this process includes the following steps.
[0088] Step S301, obtain the real scene image and satellite image of the target area. For details, please refer to Figure 1 Step S101 of the shown embodiment, which will not be elaborated here.
[0089] Step S302, analyze the surface information of the real scene image to determine the surface point cloud data corresponding to the real scene image. For details, please refer to Figure 1 Step S102 of the shown embodiment, which will not be elaborated here.
[0090] Step S303, analyze the geographic information of the satellite image to determine the geographic point cloud data corresponding to the satellite image. For details, please refer to Figure 1 Step S103 of the shown embodiment, which will not be elaborated here.
[0091] Step S304: Perform data fusion based on the surface point cloud data and the geographic point cloud data to generate target point cloud data. For details, please refer to Figure 1 Step S104 of the embodiment shown in
[0092] Step S305: Construct a 3D model of the target area based on the target point cloud data.
[0093] Specifically, the above-mentioned step S305 includes:
[0094] Step S3051: Perform terrain rendering on the target point cloud data to obtain a terrain network model.
[0095] Perform surface reconstruction on the target point cloud data. By topologically connecting each point cloud included in the target point cloud data, a terrain network model corresponding to the target point cloud data is obtained to restore the surface shapes of the reconstructed real scene model and the satellite model. Specifically, the Poisson surface reconstruction algorithm and the Delaunay triangulation algorithm can be used to perform terrain rendering on the target point cloud data to construct a Mesh terrain grid.
[0096] Step S3052: Perform texture mapping on the terrain network model to generate a 3D model of the target area.
[0097] Determine the texture image of the terrain network model, and perform texture mapping on the terrain network model through the texture image to generate a 3D model of the target area. Among them, the texture mapping is the actual image corresponding to the real scene image and the satellite image.
[0098] In some optional embodiments, the above-mentioned step S3052 includes:
[0099] Step c1: Based on the contour of the target fusion area, crop the terrain network model to determine the first terrain network model inside the contour and the second terrain network model outside the contour.
[0100] Crop the terrain network model according to the contour of the target fusion area by the cropping algorithm carried by the computer device to obtain the real scene terrain Mesh inside the cropping contour and the satellite Mesh outside the cropping contour. For example, the cropping algorithm can be the GGP cropping algorithm, but it is not limited here.
[0101] Step c2: Determine the first texture corresponding to the first terrain network model and the second texture corresponding to the second terrain network model.
[0102] When the computer device receives the real - scene images collected by the drone, it can directly obtain the Digital Orthophoto Map (DOM) corresponding to the real - scene images. The Digital Orthophoto Map is the image data generated by radiometric correction, differential rectification, and mosaicking for each pixel of the digitized aerial real - scene image after scanning processing, and then cropping according to the specified map sheet range. It is a planimetric map with kilometer grids, map border (inner and outer) decoration, and annotations. The data information included in the Digital Orthophoto Map is determined as the first texture corresponding to the first terrain network model.
[0103] When the computer device obtains satellite images, it can directly obtain the True Ortho / True Digital Ortho Map (TDOM) corresponding to the satellite images. The True Ortho Map is the image data generated by radiometric correction, differential rectification, and mosaicking for each pixel of the remotely sensed image after scanning processing, and then cropping according to the specified map sheet range. It is a planimetric map with kilometer grids, map border (inner and outer) decoration, and annotations. The data information included in the orthophoto map is determined as the second texture corresponding to the second terrain network model.
[0104] Step c3: Based on the first texture and the second texture, perform texture mapping on the first terrain network model and the second terrain network model to generate the three - dimensional model corresponding to the target area.
[0105] After determining the first texture corresponding to the first terrain network model and the second texture corresponding to the second terrain network model, the computer device uses the first texture to perform texture mapping on the first terrain network model and the second texture to perform texture mapping on the second terrain network model according to the carried texture mapping tool, and then generates the three - dimensional model corresponding to the target area.
[0106] The model generation method provided by the embodiments of the present invention crops the terrain network model according to the contour of the target fusion area, and determines the first terrain network model inside the contour and the second terrain network model outside the contour. The first texture corresponding to the first terrain network model and the second texture corresponding to the second terrain network model are respectively determined according to the obtained real - scene images and satellite images, and texture mapping is performed on the first terrain network model and the second terrain network model to generate the three - dimensional model corresponding to the target area, thereby ensuring the matching degree and consistency of the textures of the first terrain network model and the second terrain network model, achieving a seamless transition effect, and avoiding the problem of poor stitching effect caused by inconsistent accuracies of multi - source models.
[0107] In this embodiment, a model generation device is further provided. This device is used to implement the above-mentioned embodiments and preferred implementation manners, and those that have been described will not be repeated. As used hereinafter, the term "module" can be a combination of software and / or hardware that realizes a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.
[0108] This embodiment provides a model generation device, as Figure 4 shown, including:
[0109] An acquisition module 401, configured to acquire real-scene images and satellite images of a target area.
[0110] A first analysis module 402, configured to analyze the surface information of the real-scene image and determine the surface point cloud data corresponding to the real-scene image.
[0111] A second analysis module 403, configured to analyze the geographical information of the satellite image and determine the geographical point cloud data corresponding to the satellite image.
[0112] A generation module 404, configured to perform data fusion based on the surface point cloud data and the geographical point cloud data to generate target point cloud data.
[0113] A construction module 405, configured to construct a three-dimensional model of the target area based on the target point cloud data.
[0114] In some alternative embodiment manners, the above-mentioned generation module 404 may include:
[0115] A processing sub-module, configured to perform data processing on the surface point cloud data to obtain target geographical point cloud data that matches the surface point cloud data.
[0116] An acquisition sub-module, configured to acquire the first coordinate system corresponding to the target geographical point cloud data, the first coordinate information corresponding to the target geographical point cloud data, and the second coordinate system corresponding to the geographical point cloud data.
[0117] A conversion sub-module, configured to convert the second coordinate system to the first coordinate system to obtain the second coordinate information of the geographical point cloud data in the first coordinate system.
[0118] A generation sub-module, configured to perform fusion on the target geographical point cloud data and the geographical point cloud data based on the first coordinate information and the second coordinate information to generate target point cloud data.
[0119] In some alternative embodiment manners, the above-mentioned conversion sub-module may include:
[0120] An acquisition unit, configured to acquire the origin of the first coordinate system.
[0121] A conversion unit for using the origin as a reference point for coordinate system conversion and converting the second coordinate system to the first coordinate system based on the reference point.
[0122] In some alternative embodiment modes, the above-mentioned generation module 404 may further include:
[0123] A determination sub-module for determining the target geographic point cloud data and the target fusion area of the geographic point cloud data in response to a drawing operation on the fusion area.
[0124] A first clipping sub-module for clipping the target geographic point cloud data based on the target fusion area to obtain first geographic point cloud data matching the target fusion area.
[0125] A second clipping sub-module for clipping the geographic point cloud data based on the target fusion area to obtain second geographic point cloud data matching the target fusion area.
[0126] A fusion sub-module for fusing the first geographic point cloud data and the second geographic point cloud data to generate geographic point cloud fusion data matching the target fusion area.
[0127] In some alternative embodiment modes, the above-mentioned fusion sub-module may include:
[0128] A first determination unit for determining the splicing level of the first geographic point cloud data and the second geographic point cloud data.
[0129] A splicing unit for splicing the first geographic point cloud data and the second geographic point cloud data within the target fusion area according to the splicing level to obtain geographic point cloud fusion data.
[0130] In some alternative embodiment modes, the above-mentioned construction module 405 may include:
[0131] A rendering sub-module for performing terrain rendering on the target point cloud data to obtain a terrain network model.
[0132] A texturing sub-module for performing texture mapping on the terrain network model to generate a three-dimensional model of the target area.
[0133] In some alternative embodiment modes, the above-mentioned texturing sub-module may include:
[0134] A second determination unit for clipping the terrain network model based on the contour of the target fusion area to determine a first terrain network model inside the contour and a second terrain network model outside the contour.
[0135] A third determination unit for determining a first texture corresponding to the first terrain network model and a second texture corresponding to the second terrain network model.
[0136] A generation unit, configured to perform texture mapping on a first terrain network model and a second terrain network model based on a first texture and a second texture, and generate a three-dimensional model corresponding to a target area.
[0137] The further function descriptions of the above-mentioned various modules and units are the same as those in the corresponding embodiments above, and will not be repeated here.
[0138] The model generation device in this embodiment is presented in the form of functional units. Here, the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.
[0139] The model generation device provided by the embodiment of the present invention, after obtaining the real scene image and satellite image of the target area, analyzes the surface information of the real scene image and the geographical information of the satellite image, respectively determines the surface point cloud data corresponding to the real scene image and the geographical point cloud data corresponding to the satellite image, performs data fusion according to the surface point cloud data and the geographical point cloud data to generate target point cloud data, and constructs a three-dimensional model of the target area according to the generated target point cloud data, thereby realizing the fusion processing of multi-source terrain models at the point cloud level. Without the need for manual interactive fusion, the connection processing of the transition area of the multi-source terrain model has a better visual effect.
[0140] The embodiment of the present invention further provides a computer device having the above Figure 5 shown model generation device.
[0141] Please refer to Figure 5 , Figure 5 is a schematic structural diagram of a computer device provided by an optional embodiment of the present invention. As Figure 5 shown, the computer device includes: one or more processors 10, a memory 20, and an interface for connecting each component, including a high-speed interface and a low-speed interface. Each component communicates with each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed within the computer device, including instructions stored in the memory or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (such as an array of servers, a set of blade servers, or a multi-processor system). Figure 5Take a processor 10 as an example.
[0142] The processor 10 can be a central processing unit, a network processor, or a combination thereof. Among them, the processor 10 can further include a hardware chip. The above-mentioned hardware chip can be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The above-mentioned programmable logic device can be a complex programmable logic device, a field-programmable gate array, a generic array logic, or any combination thereof.
[0143] Among them, the memory 20 stores instructions that can be executed by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiments.
[0144] The memory 20 can include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the computer device, etc. In addition, the memory 20 can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some alternative embodiments, the memory 20 can optionally include a memory remotely set relative to the processor 10, and these remote memories can be connected to the computer device through a network. Examples of the above-mentioned network include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and a combination thereof.
[0145] The memory 20 can include a volatile memory, such as a random access memory; the memory can also include a non-volatile memory, such as a flash memory, a hard disk, or a solid-state drive; the memory 20 can also include a combination of the above types of memories.
[0146] The computer device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30, and the output device 40 can be connected through a bus or other means, Figure 5 Take the connection through the bus as an example.
[0147] The input device 30 can receive input digital or character information, and generate key signal inputs related to the user settings and function controls of the computer device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 can include a display device, an auxiliary lighting device (such as an LED), and a tactile feedback device (such as a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light-emitting diode, a display, and a plasma display. In some alternative embodiments, the display device can be a touch screen.
[0148] Embodiments of the present invention also provide a computer-readable storage medium. The methods according to the embodiments of the present invention can be implemented in hardware, firmware, or be implemented as computer code that can be recorded on a storage medium, or be implemented as computer code that is originally stored in a remote storage medium or a non-transitory machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the methods described herein can be stored as such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code, and when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the methods shown in the above embodiments are implemented.
[0149] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations all fall within the scope defined by the appended claims.
Claims
1. A model generation method, characterized in that, The method includes: Obtaining the real-scene image and satellite image of the target area; Analyzing the surface information of the real-scene image to determine the surface point cloud data corresponding to the real-scene image; Analyzing the geographical information of the satellite image to determine the geographical point cloud data corresponding to the satellite image; Performing data fusion based on the surface point cloud data and the geographical point cloud data to generate target point cloud data; Constructing a three-dimensional model of the target area based on the target point cloud data.
2. The method according to claim 1, wherein The performing data fusion based on the surface point cloud data and the geographical point cloud data to generate target point cloud data includes: Performing data processing on the surface point cloud data to obtain target geographical point cloud data matching the surface point cloud data; Obtaining the first coordinate system corresponding to the target geographical point cloud data, the first coordinate information corresponding to the target geographical point cloud data, and the second coordinate system corresponding to the geographical point cloud data; Converting the second coordinate system to the first coordinate system to obtain the second coordinate information of the geographical point cloud data in the first coordinate system; Fusing the target geographical point cloud data and the geographical point cloud data based on the first coordinate information and the second coordinate information to generate the target point cloud data.
3. The method according to claim 2, characterized in that, The converting the second coordinate system to the first coordinate system includes: Obtaining the origin of the first coordinate system; Using the origin as the coordinate system conversion reference point and converting the second coordinate system to the first coordinate system based on the reference point.
4. The method according to claim 2, wherein It further includes: In response to a drawing operation on the fusion area, determining the target fusion area of the target geographical point cloud data and the geographical point cloud data; Cropping the target geographical point cloud data based on the target fusion area to obtain the first geographical point cloud data matching the target fusion area; Cropping the geographical point cloud data based on the target fusion area to obtain the second geographical point cloud data matching the target fusion area; Fusing the first geographical point cloud data and the second geographical point cloud data to generate geographical point cloud fusion data matching the target fusion area.
5. The method according to claim 4, wherein The fusing the first geographical point cloud data and the second geographical point cloud data to generate geographical point cloud fusion data matching the target fusion area includes: Determining the splicing level of the first geographical point cloud data and the second geographical point cloud data; Splicing the first geographical point cloud data and the second geographical point cloud data within the target fusion area according to the splicing level to obtain the geographical point cloud fusion data.
6. The method according to claim 1, characterized in that, The constructing a three-dimensional model of the target area based on the target point cloud data includes: Performing terrain rendering on the target point cloud data to obtain a terrain network model; Performing texture mapping on the terrain network model to generate a three-dimensional model of the target area.
7. The method according to claim 6, wherein The performing texture mapping on the terrain network model to generate a three-dimensional model of the target area includes: Based on the contour of the target fusion area, cropping the terrain network model to determine a first terrain network model inside the contour and a second terrain network model outside the contour; Determine the first texture corresponding to the first terrain network model and the second texture corresponding to the second terrain network model; Based on the first texture and the second texture, perform texture mapping on the first terrain network model and the second terrain network model to generate a three-dimensional model corresponding to the target area.
8. A model generation device, characterized in that, The device includes: An acquisition module, configured to acquire real-scene images and satellite images of a target area; A first parsing module, configured to parse the surface information of the real-scene images to determine the surface point cloud data corresponding to the real-scene images; A second parsing module, configured to parse the geographical information of the satellite images to determine the geographical point cloud data corresponding to the satellite images; A generation module, configured to perform data fusion based on the surface point cloud data and the geographical point cloud data to generate target point cloud data; A construction module, configured to construct a three-dimensional model of the target area based on the target point cloud data.
9. A computer device, characterized in that, Includes: A memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to execute the model generation method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, Computer instructions are stored on the computer-readable storage medium, and the computer instructions are used to cause a computer to execute the model generation method according to any one of claims 1 to 7.
Citation Information
Cited By
Fusion method and system of multi-source three-dimensional geographic information data
CN121639998A