CN114972665A - Three-dimensional visual virtual scene modeling method in unmanned aerial vehicle virtual simulation - Google Patents

Abstract

The invention discloses a three-dimensional visual virtual scene modeling method in unmanned aerial vehicle virtual simulation, which belongs to the technical field of visual simulation, and is characterized in that buildings are divided into landmark buildings and non-landmark buildings according to a characteristic attribute similarity normalization principle, the landmark buildings are subjected to refined modeling by utilizing three-dimensional animation production software, and a vector grid of the landmark buildings is extracted from a remote sensing image; and then, carrying out large-scale modeling on the non-landmark building by using three-dimensional visual modeling software, and carrying out integrated multi-element fusion on the landmark building model and the non-landmark building model in the illusion engine software. The invention has the advantages of multi-element fusion modeling, greatly improving the modeling speed while ensuring the modeling quality, realizing the three-dimensional mapping of a real scene through a remote sensing image, and having better immersion and interaction performance when establishing the model.

Description

A 3D visualization virtual scene modeling method in UAV virtual simulation

technical field

The invention belongs to the technical field of visual simulation, and in particular relates to a three-dimensional visualization virtual scene modeling method in the virtual simulation of unmanned aerial vehicles.

Background technique

Visual simulation is an immersive interactive technology based on computer graphics, which supports the visual display of information. Because the flight scene simulation technology can combine the three-dimensional virtual model of the UAV, the simulation process and the simulation data, and simultaneously perform visual output and display, it is favored by scholars at home and abroad. At present, the mainstream flight scene simulation software Creator, X-Plane, and Flight Gear have more or less inefficiencies in 3D virtual scene modeling, and the simulation process is immersive, interactive, and unrealistic.

Most of the early UAV visual simulation systems were based on two-dimensional map data or digital graphics, which made it difficult for researchers to detect the relevant features hidden in the data, and could not intuitively understand the flight status of the UAV. Since the beginning of the 21st century, the development of 3D technology has spawned the emergence of a number of game engines. At the same time, 3D game engines have provided a complete set of solutions for game or visual development, effectively promoting digital twin 3D visualization scene modeling. development of technology. At present, the well-known 3D game engines in the industry mainly include ORGE, Unity 3D, Unreal Engine 4, etc. Shangguan Youbai developed a port visualization simulation demonstration system based on OGRE. The system mainly simulates the operating conditions of port equipment and the impact of different climates on the port system, and has made good progress in the immersion of visualization. Li Xie built a flight scene simulation platform based on the Unity 3D game engine, which is used to reproduce the operation scenes of aircraft carriers and aircraft. The platform has realistic display effects and good immersion effects, but the scene modeling is complicated and the efficiency is low.

SUMMARY OF THE INVENTION

In order to overcome the defects of traditional UAV visual simulation methods such as unsatisfactory immersion and interactive effect, complicated scene construction and low modeling efficiency, the present invention provides a three-dimensional visualization virtual scene modeling in UAV virtual simulation. The method, multi-component fusion modeling, greatly improves the modeling speed while ensuring the modeling quality, realizes the three-dimensional mapping of the real scene through remote sensing images, and establishes a model with good immersion and interaction performance.

The technical solution adopted by the present invention to solve the technical problem is: a three-dimensional visualization virtual scene modeling method in the virtual simulation of unmanned aerial vehicle, comprising: dividing buildings into landmark buildings and For non-landmark buildings, use 3D animation software to perform refined modeling of landmark buildings, and extract the vector grids of landmark buildings from remote sensing images; Model large-scale and integrate iconic and non-iconic building models in Unreal Engine software.

As a further embodiment of the present invention, the use of three-dimensional animation production software to perform refined modeling of landmark buildings includes: importing the CAD data of landmark buildings into 3ds Max, capturing, extruding, chamfering, After the insertion process, use the rectangle tool to outline the edge; according to the height of the top surface of the iconic building and its structure, add different materials and texture elements in different areas of the model.

As a further embodiment of the present invention, in the process of refined modeling of landmark buildings, the invisible inner surfaces in the splicing area of connected buildings are removed, and the number of Boolean operations is minimized for horizontal and vertical structures.

As a further embodiment of the present invention, the use of 3D visualization modeling software to perform large-scale modeling of non-iconic buildings includes: using City Engine to perform large-scale vector data modeling, using the bottom of the building as a standard, using FAME- Net network is trained on the building dataset of aerial remote sensing images to extract the vector data of non-landmark buildings.

As a further embodiment of the present invention, in the large-scale modeling process of non-iconic buildings, the building is divided into multiple structural components, and a large-scale CGA rule is constructed according to the building structure, height and color, and the mapping function is used. Texture mapping the structural components of the building.

As a further embodiment of the present invention, the integration of the iconic building model and the non-iconic building model in the Unreal Engine software includes: importing DEM digital elevation data into the virtual engine UE4 for terrain data design , take the GDEMV2 elevation dataset as the original data source, perform interpolation and noise reduction processing on the original data, and then import the terrain data into Global Mapper for 3D expansion to obtain the terrain elevation map file in hfz format; according to the width and height of the elevation map , set the resolution and data range in World Machine to obtain a UE4-compatible RAW16 format elevation map file; import the RAW16 format height map in UE4, select the material corresponding to the remote sensing image to create a 3D terrain, and convert the constructed The iconic building model and the non-iconic building model are imported into UE4 in the same proportion and placed on the built 3D terrain.

The beneficial effects of the present invention include:

1. Design a multivariate fusion modeling method, which greatly improves the modeling speed while ensuring the modeling quality;

2. Realize 3D mapping of real scenes through remote sensing images, and establish models with good immersion and interaction performance.

Description of drawings

Fig. 1 is the flow chart of the modeling method of the present invention;

2 is a schematic diagram of refined modeling in Embodiment 1 of the present invention;

3 is a schematic diagram of large-scale modeling vector grid data in Embodiment 1 of the present invention;

Fig. 4 is the CGA code diagram of the embodiment 1 of the present invention;

FIG. 5 is a schematic diagram of a three-dimensional virtual scene modeling of a campus in Embodiment 1 of the present invention;

6 is a three-dimensional scene diagram of a campus in Embodiment 1 of the present invention;

FIG. 7 is a test diagram of virtual scene immersion and interaction in Embodiment 2 of the present invention.

Detailed ways

The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

In the description of the present invention, it should be noted that the orientations or positional relationships indicated by the terms "vertical", "horizontal", "inside", "outside", etc. are based on the orientations or positional relationships shown in the drawings, only In order to facilitate the description of the present invention and simplify the description, it is not indicated or implied that the device or component referred to must have a particular orientation, be constructed and operate in a particular orientation, and therefore should not be construed as a limitation of the present invention. Furthermore, the terms "first", "second", and "third" are only used to distinguish components and should not be construed to indicate or imply relative importance.

In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

Example 1

Establishing a virtual world that can map the physical world is the cornerstone of UAV visual simulation. For this reason, this embodiment builds a twin virtual model of a real flight scene. Buildings are the most critical elements in the construction of virtual scenes. Using 3dsMax to build simulation objects, although more realistic effects can be obtained, the modeling speed is too slow. To this end, this embodiment proposes a multi-integrated 3D visualization virtual scene modeling method, which is divided into landmark buildings (such as libraries, stadiums) and non-symbolic buildings ( Such as: dormitory building, teaching building). Use 3ds Max to model the iconic buildings in detail, extract the vector meshes of the buildings from the remote sensing images, and then use the City Engine software to model non-signature buildings in a large scale, and combine the 3ds Max and City Engine The model is integrated and multi-integrated in UE4, and the modeling process is shown in Figure 1.

1. Refinement modeling

First, carry out refined modeling of important buildings in the scene, take N campus library as an example, import its CAD data into 3ds Max, after capturing, extruding, chamfering, inserting, and then use the rectangle tool to outline wall edge. Secondly, according to the height and structure of the top surface of the building, different materials and texture elements are added in different areas of the model to increase the texture and fidelity of the main body of the building, as shown in Figure 2.

In order to reduce the redundancy and complexity of the model and improve the running speed of the model, this embodiment proposes a method for optimizing the number of surfaces. In the modeling process, the invisible inner surfaces in the splicing area of the connected buildings are removed, the generation of invalid surfaces is reduced, and the redundancy caused by a large number of copied structured models is avoided. For horizontal and vertical structures, the number of Boolean operations is minimized and the model complexity is reduced. In order to minimize the number of models, the fine threshold in the model modifier is optimized to improve the running speed of the model on the basis of ensuring the authenticity of the building.

2. Large-scale modeling

The large-scale modeling mainly solves the modeling of non-signature buildings, trees and roads. In order to improve the modeling speed, based on the rule method, this embodiment uses the City Engine to perform large-scale vector data modeling. The traditional oblique photography technology uses the building roof as the standard vector data extraction method, and the existing problems of building inclination, displacement and missing will lead to modeling deviation. As shown in Figure 3, in this embodiment, the building bottom is used as the standard, and the FAME-Net network is used to train the building data set of aerial remote sensing images, so as to avoid the extraction deviation existing in the traditional method, and extract the vector data of the building.

The CGA (Computer Generated Architecture) rule is the core of building a large-scale modeling method, which mainly focuses on the modeling speed and efficiency, and can ignore some details of the building. To this end, modeling rules are written according to the structure type, floor height, and roof color of the building, and corresponding types of buildings are quickly generated in large batches. The building details are related to the binding force of the rules. The more rules, the more perfect the model details will be.

In this embodiment, taking the N campus scene as an example, in order to establish a CGA rule, it is necessary to find out the relationship between the number of floors and the height of buildings, and the height and number of floors of some buildings in the study area are measured, and Table 2 is drawn. .

Table 2 Building height and number of floors

According to Table 2, the relationship between the building height and the number of floors is obtained by fitting, as shown in the following formula:

H=3.46N+0.69 (6)

Specifically, the building is divided into small structural components, and large-scale rules are constructed according to the building structure, floor height and color, and texture maps are used to map the doors, windows, roofs, and exterior walls of the building, and some CGAs are used. The code is shown in Figure 4.

In addition, in addition to the building model, the flowers, trees, street lights and road parts are built using existing rules to generate a 3D virtual scene of the N campus as shown in Figure 5.

3. Model multiple fusion

In order to improve the immersion and interactivity of the simulation, the DEM (Digital Elevation Map) digital elevation data was imported into UE4 for uneven terrain design. This embodiment uses the GDEMV2 elevation data set as the original data source, but the massive DEM data will affect the running speed of subsequent virtual scenes, and the amount of data describing flat terrain and complex areas varies, so the original DEM data needs to be processed. To this end, the present embodiment performs interpolation and noise reduction processing on the original data to improve data utilization. Then, import terrain data into Global Mapper for 3D expansion to obtain terrain files in hfz format. After that, according to the width and height of the elevation map, set the resolution and data range in World Machine to obtain a UE4-compatible RAW16 format height map file. In order to perform multiple fusion of the scene in UE4, import the height map in RAW16 format, select the material corresponding to the remote sensing image to create a real 3D terrain with different bumps, and import the buildings constructed by 3ds Max and CityEngine in the same proportion. In UE4, it is placed on top of the built 3D terrain. In order to solve the dynamic interaction between models, collision settings are added between different objects to realize the conversion of the scene from 2D static to 3D dynamic, as shown in Figure 6.

Example 2

Virtual scene immersion and interaction performance test:

In order to conduct the immersive and interactive test of the virtual scene, the digital model of the UAV is operated in the main entrance scene of the N campus constructed in Figure 6 for autonomous flight, as shown in Figure 7, since the 3D virtual scene construction is a real scene twin mapping, Mountains and terrain are consistent with the real environment, and elements such as trees and red flags in the scene will also move with the wind, and the whole scene has good immersion. When the drone is in the circle position in Figure 7 in the virtual scene, the small frame in the lower left corner is the real-time perception of the surrounding environment information by the fisheye camera. At this time, the drone perceives the obstacle flag. Due to the collision setting in the scene, At that time, the action of avoiding obstacles will be performed, which has good interaction with the environment, and can provide technical support for performance tests such as 3D survey and obstacle avoidance.

Obviously, the above-mentioned embodiments are only examples for clear description, and are not intended to limit the implementation manner. For those of ordinary skill in the art, changes or modifications in other different forms can also be made on the basis of the above description. There is no need and cannot be exhaustive of all implementations here. And the obvious changes or changes derived from this are still within the protection scope of the present invention.

Claims (6)

1. A three-dimensional visual virtual scene modeling method in unmanned aerial vehicle virtual simulation is characterized by comprising the following steps: dividing the buildings into landmark buildings and non-landmark buildings according to a characteristic attribute similarity normalization principle, carrying out fine modeling on the landmark buildings by using three-dimensional animation software, and extracting a vector grid of the landmark buildings from a remote sensing image; and then, carrying out large-scale modeling on the non-landmark building by using three-dimensional visual modeling software, and carrying out integrated multi-element fusion on the landmark building model and the non-landmark building model in the illusion engine software.

2. The method of claim 1, wherein the three-dimensional animation software is used for fine modeling of landmark buildings, and comprises: importing CAD data of a landmark building into 3ds Max, and after the CAD data are subjected to capturing, extruding, chamfering and inserting, using a rectangular tool to outline edges; according to the height and the structure of the top surface of the landmark building, different materials and texture elements are added in different areas of the model.

3. The method of claim 2, wherein invisible inner surfaces in the splicing area of connected buildings are removed during the fine modeling of landmark buildings, and Boolean operation times are minimized for horizontal and vertical structures.

4. The method according to claim 1, wherein the large-scale modeling of the non-landmark buildings by using the three-dimensional visualization modeling software comprises: and (3) performing large-range vector data modeling by using the City Engine, training an aviation remote sensing image building data set by using a FAME-Net network by taking the bottom of the building as a standard, and extracting the vector data of the non-landmark building.

5. The method of claim 4, wherein in the process of modeling the non-landmark building in a large range, the building is split into a plurality of structural components, the building is subjected to large-range CGA rule construction according to the structure, height and color of the building, and the structural components of the building are subjected to texture mapping by using a mapping function.

6. The method of claim 1, wherein the integrating multivariate fusion of landmark building models and non-landmark building models in the illusion engine software comprises: importing DEM digital elevation data into a virtual engine UE4 for topographic data design, taking a GDEMV2 elevation data set as an original data source, interpolating and denoising the original data, importing topographic data into a GlobalMapper for three-dimensional expansion to obtain a topographic elevation map file in a hfz format; setting resolution and data range in WorldMachine according to the width and height of the elevation map to obtain an elevation map file compatible with UE4 in RAW16 format; importing a height map in a RAW16 format into the UE4, selecting a material corresponding to the remote sensing image to create a three-dimensional terrain, importing the constructed landmark building model and the non-landmark building model into the UE4 according to the same proportion, and placing the model on the created three-dimensional terrain.

Images