A method for integrating place name and address annotation with real-scene three-dimensional model

Abstract

The present invention relates to the field of real-life three-dimensional scene display, and specifically to a method for fusion display of place name and address annotations and real-life three-dimensional models, comprising the following steps: step 1: preprocessing place name and address data; step 2: publishing a spatial search service based on the preprocessed data to obtain a query result; step 3: browsing the client scene camera pose information according to the real-life model to obtain a LOD operator threshold scheduling calculation result; step 4: using the LOD operator threshold scheduling calculation result in step 3 as a query parameter, calling the spatial search service in step 2 for each partition to obtain target data; step 5: fusing the target data searched by S4 with the real-life model for display. Compared with the traditional grid-based online place name and address tile service, this method only requires very few network queries to complete, and the network consumption is reduced by more than 90% on average. Moreover, the client only requires a small amount of calculation to complete the display, and the deployment is simple and the operation is efficient.

Description

Technical Field

[0001] The present invention relates to the field of real-life three-dimensional scene display, and in particular to a method for integrating and displaying place name and address annotations with real-life three-dimensional models. Background Art

[0002] Real-life three-dimensional is a digital space that reflects and expresses human production, life and ecological space within a certain range in a real, three-dimensional and time-series manner. It is a standardized product of new basic surveying and mapping, an important new infrastructure, and provides a unified spatial foundation for economic and social development and the informatization of various departments.

[0003] Real-scene 3D construction mainly includes terrain-level real-scene 3D construction, city-level real-scene 3D construction and component-level real-scene 3D construction. Among them, the city-level real-scene 3D construction requires that the local level will obtain oblique photography images, laser point clouds and other data with a resolution better than 5 cm, and determine the cycle for timely collection and expression based on local actual conditions.

[0004] In the process of expressing real-life 3D data, place name and address annotations are important semantic data. They play a vital role in the promotion and application of real-life 3D in various industries and are important basic data in real-life 3D scenes.

[0005] Place names and addresses include spatial location, data classification and grading, name and address information. According to the requirements of conventional two-dimensional map making, the data display scale is determined according to the data classification and grading information. For example, convenience stores usually start to be displayed at a large scale; the data display icon information is determined according to the data classification information. In the process of labeling place names and addresses, problems such as annotation avoidance and hierarchical display of large amounts of data should be solved. The entire analysis and calculation process is complicated, so it is usually made into sliced ​​data in advance. For example, the image annotation layer released by Tiandi Map contains place name and address information, and map images are generated in advance to provide services in the form of online tile services.

[0006] Real-life 3D data is usually composed of complex spatial 3D grids and a large amount of texture data. The amount of data is huge, and it has high requirements for computer rendering capabilities, network transmission speed and storage management. The core idea of ​​the 3D multi-level of detail (LOD) modeling method is to describe and represent the objects in the same scene with different levels of detail. When the objects are brought closer, they are loaded on demand and the details of the objects are refined layer by layer, so as to reduce network consumption and the total amount of computer rendering.

[0007] Currently, several methods are commonly used to display place name and address information in real three-dimensional scenes.

[0008] First, the real-life 3D model data is superimposed on the traditional 2D electronic map for display. The traditional 2D electronic map provides online services in the form of tile services, which contain place name and address information. The 2D electronic map tiles are usually in the form of pictures and are used as base maps in 3D scenes. They are displayed under the real-life 3D model. The real-life 3D model covers the road network during display, which is an obvious disadvantage of this method.

[0009] The second method is to use image annotations to overlay and fit them onto the real-life 3D model for display. Since the surface of the real-life 3D model is uneven, it is acceptable in flat areas without real-life 3D, but there will be obvious text deformation problems in mountainous areas and areas covered by real-life 3D, and there is also a problem that the text direction cannot change with the scene perspective, resulting in poor results.

[0010] Third, when place names and addresses are displayed in a three-dimensional scene in a vector format, for small-scale scenes, the small amount of place names and addresses that need to be displayed can be manually set and superimposed on the real-life three-dimensional scene. However, for large-scale and extremely large amounts of place names and addresses at the city level, they cannot be directly carried, and will result in freezes or jams, and the display will be crowded, with unsatisfactory effects.

[0011] However, currently, several methods are used to display place name and address information. Electronic map annotation services usually publish online grid-based electronic map tile services, such as the Tiandi Map annotation map service. Each display screen contains dozens of tiles, and each tile is usually picture-like content. It has the disadvantages of high network requirements and large amounts of transmitted content, and low efficiency. The tile place name and address information is covered by the real-life three-dimensional model, or is directly displayed above the real-life three-dimensional model with obvious deformation and distortion, which greatly hinders the browsing of the three-dimensional scene and has obvious disadvantages.

[0012] When place names and addresses are directly superimposed and displayed in a real-life three-dimensional scene in a vector format, when the three-dimensional scene is large, display congestion, conflict, and jamming may occur due to the dense information of the place names and addresses and the huge amount of data. This has the disadvantages of being inefficient and incompatible with large scenes and large amounts of data.

[0013] Therefore, the display of place names and addresses in large real-life three-dimensional scenes is poor, and there is a problem of conflict and overlap between place names and addresses and real-life three-dimensional models, which is urgently needed to be solved by technical personnel in this field. Summary of the invention

[0014] In view of the problems existing in the prior art, the purpose of the present invention is to provide a method for efficiently integrating and displaying place name and address information with real-scene three-dimensional models, which achieves high compatibility and ease of use, and at the same time achieves excellent display effects in terms of clarity, aesthetics, stereoscopic effect and adaptability to a large range of real-scene three-dimensional, supplements important semantic data for real-scene three-dimensional, is suitable for integrating and displaying place name and address information with a large range of extremely large data volumes in real-scene three-dimensional, has a good fit between place name and address and real-scene three-dimensional scenes, and effectively solves the problem of conflict between traditional place name and address and real-scene three-dimensional display, thereby effectively improving the application effect of real-scene models.

[0015] To achieve the above object, the technical solution adopted by the present invention is: a method for fusion display of place name address annotation and real scene three-dimensional model, comprising the following steps:

[0016] Step 1: Preprocess the place name and address data, collect the place name and address data, add the LOD operator threshold attribute to each point data through spatial analysis and calculation, and generate the place name and address annotation point data based on the place name and address data;

[0017] Step 2: Publish spatial search services based on preprocessed data, perform spatial search based on any spatial range and LOD operator threshold, return target place name and address annotations, and obtain query results;

[0018] Step 3: Browse the client scene camera pose information according to the real scene model, perform intelligent dynamic partition LOD operator threshold scheduling calculation, calculate the ratio of the real world area and the screen projection area, and obtain the LOD operator threshold scheduling calculation result;

[0019] Step 4: Using the LOD operator threshold scheduling calculation result in step 3 as the query parameter, call the spatial search service in step 2 for each partition, obtain the annotation information required for each partition, and use the partition range as the search space range to obtain the target data;

[0020] Step 5: According to the 3D posture features of the real 3D scene and the spatial structure information of the real 3D data model, the target data searched by S4 is fused and displayed with the real model.

[0021] The above-mentioned method for displaying the fusion of place name and address annotation and real scene three-dimensional model, step 1 comprises:

[0022] Step 1-1: Collect place name and address data and store the place name and address data in a spatial database, wherein the place name and address data includes spatial location data, name, and classification and grading information;

[0023] Step 1-2: Comprehensively consider the classification and grading information, point avoidance, and point density parameters per unit area, and through spatial analysis and calculation, add the LOD operator threshold attribute to each point, and generate place name and address annotation point data based on the place name and address data.

[0024] In the above-mentioned method for fusion display of place name and address annotations and real-life three-dimensional models, the LOD operator threshold is calculated by the ratio of the real-world projection area corresponding to the three-dimensional model to the area projected on the screen. The LOD operator threshold is used as an intelligent LOD operator for dynamic partitioning, and several segmentation thresholds are set. Place name and address data preprocessing and real-time browsing scheduling of three-dimensional scenes are performed according to the segmentation thresholds.

[0025] The above-mentioned method for displaying the fusion of place name and address annotation and real scene 3D model, step 2 comprises:

[0026] Step 2-1: Store the data in the database. The location information is stored in a spatial location data structure and a spatial index is established for fast search.

[0027] Step 2-2: Publish the spatial search service, support setting the spatial search range and LOD operator threshold parameters, and support descending sorting according to the LOD operator threshold;

[0028] Step 2-3: Deploy the query service on the server. The query results include the location information, name, classification information, and LOD operator threshold information of the annotation point.

[0029] In the above-mentioned method for displaying the fusion of place name and address annotations and real-scene three-dimensional models, in step 3, the LOD operator threshold scheduling calculation includes:

[0030] Step 3-1: Select the 3D scene area visible on the screen;

[0031] Step 3-2: Calculate and determine the three-dimensional scene partitions at different LOD operator threshold levels according to the three-dimensional scene posture, field of view and height information;

[0032] Step 3-3: Calculate the LOD operator value within the partition: calculate the real-world area A corresponding to the small block range, and calculate the screen projection area B corresponding to the spatial range obtained by the small block range, calculate the ratio of the real-world area A to the screen projection area B, and obtain the LOD operator.

[0033] In the above-mentioned method for displaying the fusion of place name and address annotations and real-scene 3D models, in step 3, the camera uses a 3D camera, and the closer the 3D camera is to the target real-scene 3D model, the larger the screen projection area is, and the smaller the LOD operator is;

[0034] On the contrary, the farther the 3D camera is from the target real-scene 3D model, the smaller the screen projection area is and the larger the LOD operator is.

[0035] The above-mentioned method for displaying the fusion of place name and address annotation and real scene three-dimensional model, step 3-3 comprises:

[0036] Step a: Take the four corner points of the displayable range of the 3D scene on the screen, and divide the 3D scene range that can be displayed on the client into several partitions according to the scene posture according to the multiple level changes of the upper and lower LOD operators;

[0037] Step b: After partitioning, calculate the LOD operator for each partition.

[0038] In the above-mentioned method for displaying the fusion of place name and address annotations and real-life three-dimensional models, in step a, when the field of view angle of the camera's visual frustum is α, and the angle between the center line of the visual frustum and the ground is β, the LOD operator ratio function f(α, β) of the apogee C and the perigee B is obtained. , according to the ratio of the ratio function f(α,β), determine to divide the current screen into several areas, the partitions are divided geometrically, the common ratio is 2, and the division number formula Num(α,β) is obtained. , and the range formula for each area relative to Area(n), , where n is the partition number, the perigee is 1, and the apogee direction increases by 1 successively. The value sought is the upper limit of the segmentation ratio range, thus realizing regional classification.

[0039] In the above-mentioned method for displaying the fusion of place name and address annotations and real-life three-dimensional models, in step 4, the query parameter is the LOD operator of each partition calculated by S3, and the search is performed in descending order according to the LOD operator. The maximum number of search results parameter is set to search for a limited number of place name and address information, and the place name and address annotations displayed in the scene are adjusted dynamically in real time to make the total display amount small and the displayed annotations are important annotations.

[0040] The above-mentioned method for integrating and displaying place name and address annotations with real-scene three-dimensional models also includes step 6: when the scene changes, the process of steps 3, 4 and 5 is repeated again, and after the query results are returned, the original annotation information is cleared, and the integrated display operation of place name and address annotations and real-scene three-dimensional is performed again.

[0041] The beneficial effect of the fusion display method of place name and address annotations and real-life three-dimensional models of the present invention is: the present invention adopts an intelligent dynamic partition LOD operator scheduling method, the number of annotations displayed is moderate, there is no display jamming problem, the annotations are dynamically partitioned and adjusted according to the real-time three-dimensional camera posture and viewing angle, the annotation density is moderate, and the real-time analysis process is efficient and fast, and the overall effect is significantly improved.

[0042] The present invention adopts an intelligent dynamic LOD scheduling method to achieve efficient scheduling of massive place name and address data in a three-dimensional scene. Compared with the traditional grid-based online place name and address tile service, this method only requires very few network queries to complete, and the network consumption is reduced by more than 90% on average. Moreover, the client only needs a small amount of calculation to complete the display, which is easy to deploy and efficient to run.

[0043] The present invention adopts an intelligent dynamic LOD scheduling method to realize the efficient display of large-scale massive place name and address data in a three-dimensional scene. After the place name and address data are graded, the place name and address annotations of different regions and different importance levels can be dynamically graded and classified according to this method, and the visualization efficiency is improved by about 90% for ultra-large-scale scenes.

[0044] The present invention displays place name and address annotations as three-dimensional text annotations, which are obviously clearer and more beautiful than traditional picture-based place name and address displays. Moreover, during display, the text color, size, font, background and other information can be flexibly adjusted through simple coding. For example, it can be dynamically adjusted according to the three-dimensional application theme color and three-dimensional application scene. Compared with traditional picture-based place name and address displays, it has the advantages of flexibility and beauty. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A simplified flow chart of the display method of the present invention;

[0046] Figure 2 This is an auxiliary graph for LOD operator calculation of the present invention. DETAILED DESCRIPTION

[0047] In order to enable those skilled in the art to better understand the purpose and technical solutions of the present invention, the purpose and technical solutions of the present invention are described below in conjunction with specific implementation methods and drawings.

[0048] Example 1

[0049] like Figure 1 As shown, a method for integrating and displaying place name and address annotations with a real-scene three-dimensional model comprises the following steps:

[0050] Step 1: Preprocess the place name and address data, collect the place name and address data, add the LOD operator threshold attribute to each point data through spatial analysis and calculation, and generate the place name and address annotation point data based on the place name and address data.

[0051] Step 1-1: Collect place name and address data, and store the place name and address data in a spatial database. The place name and address data includes spatial location data, name, and classification and grading information.

[0052] Step 1-2: Comprehensively consider the classification and grading information, point avoidance, and point density parameters per unit area, and through spatial analysis and calculation, add the LOD operator threshold attribute to each point, and generate place name and address annotation point data based on the place name and address data.

[0053] Step 2: Publish spatial search services based on preprocessed data, perform spatial search based on any spatial range and LOD operator threshold, return the target place name and address annotation, and obtain the query results.

[0054] Step 2-1: Store the data in the database. The location information is stored in a spatial location data structure and a spatial index is established for fast search.

[0055] Step 2-2: Publish the spatial search service, support setting the spatial search range and LOD operator threshold parameters, and support descending sorting according to the LOD operator threshold.

[0056] Step 2-3: Deploy the query service on the server. The query results include the location information, name, classification information, and LOD operator threshold information of the annotation point.

[0057] Step 3: Browse the client scene camera pose information according to the real scene model, perform intelligent dynamic partition LOD operator threshold scheduling calculation, calculate the ratio of the real world area and the screen projection area, and obtain the LOD operator threshold scheduling calculation result. LOD operator threshold scheduling calculation includes:

[0058] Step 3-1: Select the 3D scene area visible on the screen.

[0059] Step 3-2: Calculate and determine the three-dimensional scene partitions at different LOD operator threshold levels based on the three-dimensional scene posture, field of view, and height information.

[0060] Step 3-3: Calculate the LOD operator value within the partition: calculate the real-world area A corresponding to the small block range, and calculate the screen projection area B corresponding to the spatial range obtained by the small block range, calculate the ratio of the real-world area A to the screen projection area B, and obtain the LOD operator.

[0061] Step a: Take the four corner points of the displayable range of the 3D scene on the screen, and divide the 3D scene range that can be displayed on the client into several partitions according to the camera tilt angle, based on the multiples of the distance between the upper and lower images.

[0062] When the field of view angle of the camera's frustum is α, and the angle between the center line of the frustum and the ground is β, the LOD operator ratio function f(α, β) of the apogee C and the perigee B is obtained. .

[0063] According to the ratio of the ratio function f(α, β), the current screen is divided into several areas, and the partitions are divided geometrically with a common ratio of 2. The division number formula Num(α, β) is obtained. , and the range formula for each area relative to Area(n), , where n is the partition number, the perigee is 1, and the apogee direction increases by 1 successively. The value sought is the upper limit of the segmentation ratio range, thus realizing regional classification.

[0064] Step b: After partitioning, calculate the threshold of the LOD operator for each partition.

[0065] Step 4: Using the LOD operator threshold scheduling calculation result in step 3 as the query parameter, call the spatial search service in step 2 for each partition to obtain the annotation information required for each partition. The partition boundaries are the spatial range of the search to obtain the target data. The query parameter is the LOD operator of each partition calculated by S3. Set the LOD operator parameters and the maximum number of search results parameters. Search in descending order according to the LOD operator. Search for a limited number of place name and address information with higher weights. Dynamically adjust the place name and address annotations displayed in the scene in real time to make the total display small and the displayed annotations important and highly concerned.

[0066] Step 5: Construct a real-life model based on the 3D posture features of the real-life 3D scene and the spatial structure information of the real-life 3D data model, and fuse the target data searched by S4 with the real-life model for display.

[0067] Example 2

[0068] The present invention discloses a method for fusion display of place name and address annotations and real-life 3D models. The method first collects place name and address data, and performs preprocessing according to information such as data classification and grading, and adds the best LOD operator attribute value to all data, that is, it is most suitable for display when the LOD operator is used, and the greater the display priority, the greater the LOD operator value; the second step is to publish a spatial search service based on the preprocessed data, and perform spatial search according to any spatial range and LOD operator threshold, and return the target place name and address annotation; then, according to the real-life model, the camera posture of the client scene is browsed, and intelligent dynamic partition LOD scheduling calculation is performed, and the above-mentioned spatial search service is called in real time according to the calculation result to perform search query, and the target place name and address annotation data is dynamically obtained from the place name and address database; finally, according to the 3D posture characteristics of the real-life 3D scene and the spatial structure information of the real-life 3D data model, the target place name and address are fused and displayed with the real-life model, and a brief flow chart is shown as follows: Figure 1 The height information of the place name and address annotation can be calculated in advance according to the three-dimensional model and stored in the annotation point attributes.

[0069] S1: Preprocessing of place name and address data.

[0070] Collect place name and address data, and pre-process the place name and address data; the place name and address data includes spatial location data, name, classification and grading information; store the place name and address data in a spatial database; the place name and address spatial data is the point data type used by the geographic information system;

[0071] Generate place name and address annotation point data based on place name and address data; in the process of point data generation, comprehensively consider classification and grading information, point avoidance, and point density parameters per unit area, and add LOD operator attributes to each point through spatial analysis and calculation. The point data in each table are guaranteed not to overlap each other when displayed in the corresponding LOD level. When the density of similar data is too high, high-level data is displayed first, such as giving priority to tertiary hospitals, high-level schools, large shopping malls, large parks and other landmarks or important place names with high attention. The LOD operator threshold attribute directly affects the priority of data display. The larger the value, the greater the display priority; this step uses conventional database processing software. Example LOD operator threshold classification is shown in Table 1 below.

[0072] Table 1 Example of LOD operator threshold classification

[0073] .

[0074] LOD operator threshold: By calculating the ratio of the real-world projection area corresponding to the three-dimensional model to the area projected on the screen, an intelligent LOD operator for dynamic partitioning is used to set several segmentation thresholds. Place name and address data preprocessing and three-dimensional scene real-time browsing scheduling are performed based on the thresholds, thereby realizing a fusion display method for a wide range of place names and addresses and real-life three-dimensional models.

[0075] S2: Based on the preprocessed data, a spatial search service is published. A spatial search can be performed according to any spatial range and LOD operator threshold, and the target place name address annotation is returned. The data is stored in the database, and the location information is stored in a spatial location data structure and a spatial index is established for fast search. For example, the open source PostgreSQL database can be used with the PostGIS plug-in for database construction and spatial data management. Then the RESTFul spatial search service is published, which supports setting the spatial search range and LOD operator threshold parameters, and supports descending sorting according to the LOD operator threshold. The query results include the annotation point location information, name, classification information, and LOD operator threshold information; the query service is deployed in the server.

[0076] S3: Perform intelligent dynamic partition LOD scheduling calculation based on the camera pose information of the scene browsed by the real-life model. The camera uses a three-dimensional camera, and the LOD operator calculation method of the real-life three-dimensional browser client mainly includes three steps: the first step is to select the three-dimensional scene area visible on the screen; the second step is to calculate and determine the three-dimensional scene partitions of different LOD levels based on the three-dimensional scene pose, field of view, height and other information; the third step is to calculate the LOD operator value in the partition: calculate the area A of the real world corresponding to the small block range, and calculate the screen projection area B corresponding to the spatial range obtained by the small block range, and calculate the ratio of the real world area A to the screen projection area B, that is, the LOD operator. It is easy to infer that the closer the three-dimensional camera is to the target real-life three-dimensional model, the larger the screen projection area and the smaller the LOD operator.

[0077] On the contrary, the farther the 3D camera is from the target real-scene 3D model, the smaller the screen projection area is and the larger the LOD operator is.

[0078] First, select the four corner points of the displayable range of the three-dimensional scene on the screen, and then divide the three-dimensional scene range that can be displayed by the client into several partitions according to the camera tilt angle. It is usually divided according to 2 times the distance between the upper and lower maps. This method does not fix the level and threshold, and other level multiples are also available. It can be set according to the actual scene size and the classification and grading information of place name and address data. The larger the multiple, the fewer the spatial data classification levels and the greater the jump. According to the characteristics of the three-dimensional scene camera, the LOD operators of the two boundary corners above the screen are very different, and the LOD operators of the lower boundary corners are very different, so the four boundaries can be divided into horizontal strip partitions. LOD query space division, this method uses the horizontal strip partitioning method, and other shapes of segmentation methods can also be used. The purpose is to query place name and address annotation data according to the LOD level.

[0079] If the difference between the LOD operator thresholds of the upper corner point and the lower corner point is small, no partition is required. The specific difference threshold can be set according to the needs. After partitioning, the LOD operator threshold of each partition is calculated. Example calculation method, such as Figure 2 As shown, when the field of view angle of the camera's viewing frustum is α, and the angle between the center line of the viewing frustum and the ground is β, the LOD operator ratio function f(α, β) of the apogee C and the perigee B can be approximately derived. It can be approximately calculated using the following formula or obtained using other algorithms.

[0080] .

[0081] According to the ratio calculated by the above formula, the current screen is divided into several areas. The partitions can be divided in equal proportions with a common ratio of 2. The number of divisions can refer to the following Num(α,β) formula. The relative range of each area can refer to the Area(n) formula. , n is the partition number, the perigee is 1, and the apogee direction increases by 1 successively. The value sought is the upper limit of the segmentation ratio range. Other common ratio values ​​or other reasonable schemes can also be used to achieve regional classification.

[0082] .

[0083] S4: According to the LOD scheduling calculation results, call the S2 spatial query service for each partition to obtain the annotation information required for each partition. The partition boundaries are the spatial range of the query and the LOD operator of each partition calculated by S3 as query parameters. Search in descending order according to the LOD operator. The maximum number of search results parameter can be set. This setting can search for a limited number of place name and address information with higher weights, and can dynamically adjust the place name and address annotations displayed in the scene in real time, so that the total display amount is small and the displayed annotations are important and highly concerned.

[0084] S5: According to the three-dimensional posture characteristics of the real-life three-dimensional scene and the spatial structure information of the real-life three-dimensional data model, the target place name and address searched by S4 are integrated and displayed with the real-life model. The result of the annotation of the queried place name and address point includes the name of the annotation and the classification and grading information. According to the classification and grading information, the text color, text size, and auxiliary icon of the text annotation are set to improve the aesthetics and readability of the expression of the place name and address information. This configuration can refer to the map style of Tiandi Map or be determined according to the project interface style. The height of the annotation display is calculated in real time by the three-dimensional scene, which is specifically obtained by adding a lift value to the height information value of the intersection point of the spherical normal of the placement point and the real-life three-dimensional model. The lift value is to solve the conflict between the annotation display position and the three-dimensional scene display, especially the annotation of the building-intensive area, which may cause more annotation occlusion. After the elevation, the annotation is not easy to be occluded.

[0085] Whenever the scene changes, the S3\S4\S5 process is repeated. After the query result is returned, the original annotation information is cleared, and the place name and address annotation is re-displayed in fusion with the real scene three-dimensional.

[0086] Example 3

[0087] The real-life three-dimensional place name and address annotation display effect constructed by the method used in the present invention initially selected a camera viewing angle height of 30,000 meters, and further selected a camera viewing angle height of 4 kilometers. The real-life three-dimensional scene with a resolution of 15 cm and 104,532 place name and address data in the entire city were used as test objects. The client graphics card was NVIDIA GeForce GTX 1060 6GB. After testing, the overall operation was efficient and beautiful, and it performed well at oblique viewing angles, top viewing angles, high viewing angles, and low viewing angles, proving that the method can be applied to large-scale real-life three-dimensional scenes and large amounts of place name and address data, and it runs efficiently and beautifully. The efficiency test records are shown in Table 2 below.

[0088] Table 2 Efficiency test record table of the method used in the present invention

[0089] .

[0090] By calculating the ratio of the real-world projection area corresponding to the three-dimensional model to the area projected on the screen, a number of segmentation thresholds are set as the dynamic partitioning intelligent LOD operator, and place name and address data preprocessing and three-dimensional scene real-time browsing scheduling are performed according to the thresholds to achieve the fusion display of a wide range of place names and addresses and real-life three-dimensional models.

[0091] During scene browsing, the 3D scene client queries the place name and address interface service in real time based on scene posture, camera height and other information. During the interface query process, the LOD operator threshold is used as a query parameter, and the LOD operator threshold calculated in advance is sorted in descending order. The maximum return number parameter is set to search for a limited number of place name and address information with higher weights, effectively avoiding the problem of too much annotation data.

[0092] The query results are integrated and displayed in the real-life 3D scene. The text height is calculated in real time, and is obtained by adding an elevation value to the height information value of the intersection of the spherical normal of the placement point and the real-life 3D model.

[0093] The above embodiments are only for illustrating the inventive concept and features of the present invention, and their purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the protection scope of the present invention. Any equivalent changes or modifications made based on the essence of the content of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for displaying the integration of place name and address annotations and real-scene three-dimensional models, characterized in that: The following steps are involved: Step 1: Preprocess the place name and address data, collect the place name and address data, add LOD operator threshold attributes to each point data through spatial analysis and calculation, and generate place name and address annotation point data according to the place name and address data. The LOD operator threshold is calculated by the ratio of the real-world projection area corresponding to the three-dimensional model to the area projected on the screen. The LOD operator threshold is used as an intelligent LOD operator for dynamic partitioning, and several segmentation thresholds are set. The place name and address data are preprocessed and the three-dimensional scene is browsed and scheduled in real time according to the segmentation threshold, including: Step 1-1: Collect place name and address data and store the place name and address data in a spatial database, wherein the place name and address data includes spatial location data, name, and classification and grading information; Step 1-2: Comprehensively consider the classification and grading information, point avoidance, and point density parameters per unit area, and add LOD operator threshold attributes to each point through spatial analysis and calculation, and generate place name and address annotation point data based on place name and address data; Step 2: Publish spatial search services based on preprocessed data, perform spatial search based on any spatial range and LOD operator threshold, return target place name and address annotations, and obtain query results; Step 3: Browse the client scene camera pose information according to the real scene model, perform intelligent dynamic partition LOD operator threshold scheduling calculation, calculate the ratio of the real world area and the screen projection area, and obtain the LOD operator threshold scheduling calculation result. The LOD operator threshold scheduling calculation includes: Step 3-1: Select the 3D scene area visible on the screen; Step 3-2: Calculate and determine the three-dimensional scene partitions at different LOD operator threshold levels according to the three-dimensional scene posture, field of view and height information; Step 3-3: Calculate the LOD operator value in the partition: calculate the real-world area A corresponding to the small block range, and calculate the screen projection area B corresponding to the spatial range obtained by the small block range, calculate the ratio of the real-world area A to the screen projection area B, and obtain the LOD operator; Step 4: Using the LOD operator threshold scheduling calculation result in step 3 as the query parameter, call the spatial search service in step 2 for each partition, obtain the annotation information required for each partition, and use the partition range as the search space range to obtain the target data; Step 5: According to the 3D posture features of the real 3D scene and the spatial structure information of the real 3D data model, the target data searched by S4 is fused and displayed with the real model.

2. The method for displaying the fusion of place name and address annotation and real scene three-dimensional model according to claim 1 is characterized in that: The step 2 comprises: Step 2-1: Store the data in the database. The location information is stored in a spatial location data structure and a spatial index is established for fast search. Step 2-2: Publish the spatial search service, support setting the spatial search range and LOD operator threshold parameters, and support descending sorting according to the LOD operator threshold; Step 2-3: Deploy the query service in the server. The query results include the location information, name, classification information, and LOD operator threshold information of the annotation point.

3. The method for displaying the fusion of place name and address annotation and real scene three-dimensional model according to claim 1, characterized in that: In step 3, the camera uses a three-dimensional camera. The closer the three-dimensional camera is to the target real-scene three-dimensional model, the larger the screen projection area is and the smaller the LOD operator is. On the contrary, the farther the 3D camera is from the target real-scene 3D model, the smaller the screen projection area is and the larger the LOD operator is.

4. The method for displaying the fusion of place name and address annotation and real scene three-dimensional model according to claim 1, characterized in that: The step 3-3 comprises: Step a: Take the four corner points of the displayable range of the 3D scene on the screen, and divide the 3D scene range that can be displayed on the client into several partitions according to the scene posture according to the multiple level changes of the upper and lower LOD operators; Step b: After partitioning, calculate the LOD operator for each partition.

5. The method for displaying the fusion of place name and address annotation and real scene three-dimensional model according to claim 4 is characterized in that: In step a, when the field of view angle of the camera frustum is α, and the angle between the center line of the frustum and the ground is β, the LOD operator ratio function f(α, β) of the apogee C and the perigee B is obtained. , according to the ratio of the ratio function f(α,β), determine to divide the current screen into several areas, the partitions are divided geometrically, the common ratio is 2, and the division number formula Num(α,β) is obtained. , and the range formula for each area relative to Area(n), , where n is the partition number, the perigee is 1, and the apogee direction increases by 1 successively. The value sought is the upper limit of the segmentation ratio range, thus realizing regional classification.

6. The method for displaying the fusion of place name and address annotation and real scene three-dimensional model according to claim 1, characterized in that: In step 4, the query parameter is the LOD operator of each partition calculated by S3. The search is performed in descending order according to the LOD operator. The maximum number of search results parameter is set to search for a limited number of place name and address information. The place name and address annotations displayed in the scene are adjusted dynamically in real time to make the total display amount small and the displayed annotations are important annotations.

7. The method for displaying the fusion of place name and address annotation and real scene three-dimensional model according to claim 1, characterized in that: It also includes step 6: when the scene changes, the process of steps 3, 4 and 5 is repeated again, and after the query result is returned, the original annotation information is cleared, and the fusion display operation of the place name and address annotation and the real scene three-dimensional is performed again.

Images