Data distributed storage and retrieval method of live-action three-dimensional model and computer program product

Abstract

The invention discloses a data distributed storage and retrieval method of a live-action three-dimensional model and a computer program product. Comprising the following steps: firstly, reading all tile outer bounding boxes of a live-action three-dimensional model, and generating tile shape information; then, fusing all tile shape information, and generating model shape information; and then, uploading the tile data to a distributed database in parallel. And finally, generating a data space distribution combination graph according to the shape information of the real model. During retrieval, firstly, a to-be-retrieved space retrieval range is obtained; then, the spatial retrieval range is matched with the data spatial distribution combination graphs, and the data spatial distribution combination graphs intersecting with the spatial retrieval range are locked; and then, matching the spatial retrieval range with each piece of tile shape information corresponding to the locked data spatial distribution combination graph, and screening out the tile shape information in the spatial retrieval range. And finally, according to the screened tile shape information, reading tile data from a distributed storage library by adopting a multi-thread parallel downloading mode.

Description

Technical Field

[0001] The present invention relates to the technical field of geographic information database information retrieval, and particularly relates to a method for distributed storage and retrieval of real-scene three-dimensional model data and a computer program product. Background Art

[0002] In the past, the storage, management, and application of real-scene three-dimensional data were separated from each other and not effectively connected. Especially for a large amount of real-scene three-dimensional data, traditional static file storage is difficult to meet the requirements of high-concurrency storage and reading. The multi-terminal data retrieval method is single, inefficient, and inaccurate. The data application transmission efficiency is low, and the data volume is prone to overload in large scenes, making it difficult to cope with the increasing data storage requirements and high-concurrency access requirements. Therefore, for the storage of a large amount of real-scene three-dimensional data, it is urgent to perform multi-node and multi-terminal distributed storage, and combine the spatial geographic information topological relationship of the data itself to connect the whole process of data storage and retrieval. Summary of the Invention

[0003] Aiming at the deficiencies of the existing technology, the present invention proposes a method for distributed storage and retrieval of real-scene three-dimensional model data and a computer program product, which can connect the whole process of data storage and retrieval by combining the spatial geographic information topological relationship of the data itself. The specific technical solutions are as follows:

[0004] In the first aspect, a method for distributed storage of real-scene three-dimensional model data is provided. In the first feasible implementation manner of the first aspect, it includes:

[0005] Obtain all the outer bounding boxes of the tiles of the real-scene three-dimensional model, and generate tile shape information corresponding to each tile in the real-scene three-dimensional model;

[0006] Fuse the tile shape information of all tiles to generate model shape information corresponding to the real-scene three-dimensional model, and establish a topological mapping relationship between the tiles and the real-scene three-dimensional model;

[0007] Based on the number of the tile shape information, use a multi-thread parallel upload method to upload all the tile data corresponding to the real-scene three-dimensional model data to a distributed storage repository;

[0008] Generate a data space distribution combination diagram of the real-scene three-dimensional model according to the model shape information.

[0009] Combined with the first feasible implementation manner of the first aspect, in the second feasible implementation manner of the first aspect, generating tile shape information corresponding to a tile includes:

[0010] Calculate the tile shape information corresponding to each tile respectively according to the extreme coordinates in the corresponding tile outer bounding box, and unify the spatial coordinate systems of all the tile shape information.

[0011] Combined with the first implementation of the first aspect, in the third implementation of the first aspect, uploading the tile data to the distributed repository includes:

[0012] When the number of the tile shape information cannot be evenly divided by the total number of threads, evenly distribute the extra tile data to the threads with earlier sorting.

[0013] In a second aspect, a method for retrieving data of a real scene three-dimensional model is provided. Using the data distributed storage method described in the first or second implementation of the first aspect, in the first implementation of the second aspect, it includes:

[0014] Obtain the spatial retrieval range of the real scene three-dimensional model to be retrieved;

[0015] Topologically intersect the spatial retrieval range with each data spatial distribution combination diagram corresponding to the real scene three-dimensional model respectively, and lock the data spatial distribution combination diagram intersecting with the spatial retrieval range;

[0016] Topologically intersect the spatial retrieval range with all the tile shape information corresponding to the data spatial distribution combination diagram respectively, and screen the tile shape information within the spatial retrieval range;

[0017] Based on the screened tile shape information, use the multi-thread parallel download method to parallel download the corresponding tile data from the distributed repository.

[0018] Combined with the first implementation of the second aspect, in the second implementation of the second aspect, topologically intersecting the spatial retrieval range with the data spatial distribution combination diagram includes:

[0019] Analyze the shape information of the spatial retrieval range, and unify the coordinate system of the shape information with the coordinate system of the tile shape information;

[0020] Topologically intersect the shape information after unifying the coordinate systems with the data spatial distribution combination diagram.

[0021] Combined with the first implementation of the second aspect, in the third implementation of the second aspect, topologically intersecting the spatial retrieval range with the tile shape information includes:

[0022] According to the topological mapping relationship corresponding to the data spatial distribution combination diagram, determine all the tile shape information corresponding to the data spatial distribution combination diagram.

[0023] Combined with the first implementation manner of the second aspect, in the fourth implementation manner of the second aspect, screening the tile shape information within the spatial retrieval range includes:

[0024] Using the tile shape information as child nodes and the corresponding model shape information as parent nodes, record the screened tile shape information in a tree structure.

[0025] Combined with the fourth implementation manner of the second aspect, in the fifth implementation manner of the second aspect, parallel downloading of the corresponding tile data from the distributed repository includes:

[0026] Traverse all the parent nodes in the tree structure, and count the number of tile shape information under all the parent nodes;

[0027] Divide the download tasks of each thread evenly according to the number of the tile shape information, and parallel download the corresponding tile data from the distributed repository according to the download tasks of each thread.

[0028] Combined with the fourth implementation manner of the second aspect, in the sixth implementation manner of the second aspect, parallel downloading of the corresponding tile data from the distributed repository includes:

[0029] When the number of the tile shape information cannot be evenly divided by the total number of threads, distribute the extra download tasks to the threads with the previous sorting.

[0030] In a third aspect, a computer program product is provided, including a computer program / instructions, characterized in that when the computer program / instructions are executed by a processor, the steps of the data retrieval method according to any one of the first to sixth implementation manners of the second aspect are implemented.

[0031] Beneficial effects: By using the data distributed storage, retrieval method and computer program product of the real scene three-dimensional model of the present invention, by fusing the tile shape information of all tiles of the real scene three-dimensional model, the model shape information corresponding to the real scene three-dimensional model can be generated, a topological mapping relationship between the tiles and the real scene three-dimensional model is established, and combined with the storage locations of the respective tile data of the real scene three-dimensional model in the distributed repository, a data space distribution combination diagram of the real scene three-dimensional model is generated, and the storage and retrieval of data are connected in series by the topological relationship of spatial geographic information, so as to improve the storage and retrieval efficiency of the real scene three-dimensional model. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the specific implementation manners of the present invention, the drawings required for the specific implementation manners will be briefly introduced below. In all the drawings, the components or parts do not necessarily draw according to the actual ratio.

[0033] Figure 1 Flow chart of the data distributed storage method for the real scene three-dimensional model provided by an embodiment of the present invention;

[0034] Figure 2 Flow chart of the retrieval method for the real scene three-dimensional model provided by an embodiment of the present invention;

[0035] Figure 3 Schematic diagram of the mapping between the outer bounding box of the tile and the tile shape information;

[0036] Figure 4 Schematic diagram of the mapping between the model shape information and the tile shape information;

[0037] Figure 5 Schematic diagram of the locked model shape information;

[0038] Figure 6 Schematic diagram of the intersection between the spatial retrieval range and the tile shape information;

[0039] Figure 7 Schematic diagram of the filtered tile shape information. Detailed implementation manners

[0040] Hereinafter, embodiments of the technical solution of the present invention will be described in detail with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and thus are only examples and cannot be used to limit the protection scope of the present invention.

[0041] As shown in the Figure 1 flow chart of the data distributed storage method for the real scene three-dimensional model, the storage method includes:

[0042] Step 1, obtain all the outer bounding boxes of the tiles of the real scene three-dimensional model, and generate the tile shape information corresponding to each tile in the real scene three-dimensional model;

[0043] Step 2, fuse the tile shape information of all the tiles to generate the model shape information corresponding to the real scene three-dimensional model, and establish a topological mapping relationship between the tiles and the real scene three-dimensional model;

[0044] Step 3, based on the quantity of the tile shape information, upload all the tile data corresponding to the real scene three-dimensional model data to the distributed storage library by using a multi-thread parallel upload method;

[0045] Step 4, generate a data space distribution combination diagram of the real scene three-dimensional model according to the model shape information.

[0046] Specifically, first, the outer bounding box of all tile data corresponding to the real-scene 3D model can be read. It should be understood that the outer bounding box of the tile includes the spatial range covered by the tile, and the tile shape information corresponding to each tile in the real-scene 3D model is generated according to the outer bounding box of the tile. The tile shape information includes data parameter information and graphic data of the spatial range covered by the tile. Then, the tile shape information of all tiles included in the real-scene 3D model can be fused to generate the model shape information of the real-scene 3D model. As Figure 4 shown, a topological mapping relationship between the real-scene 3D model and the tiles is established.

[0047] After that, according to the number of tile shape information included in the model shape information and the number of threads for uploading data, the upload task of each thread can be allocated, and according to the allocated upload task, the corresponding tile data can be uploaded to the distributed database in parallel. Finally, a data space distribution combination diagram corresponding to the real-scene 3D model can be generated according to the model shape information of the real-scene 3D model. Repeating this way, a large amount of real-scene 3D models can be stored in the distributed repository in sequence, improving the data storage efficiency of the real-scene 3D model.

[0048] In this embodiment, optionally, in step 1, generating the tile shape information corresponding to the tile includes:

[0049] Calculating the tile shape information corresponding to each tile respectively according to the extreme coordinates in the corresponding outer bounding box of the tile, and unifying the spatial coordinate systems of all tile shape information.

[0050] Specifically, as Figure 3 shown, parsing the outer bounding box data of the tile to obtain the maximum three-dimensional coordinates and the minimum three-dimensional coordinates of the range covered by the bounding box. Then, calculating the tile shape information of the tile according to the maximum three-dimensional coordinates and the minimum three-dimensional coordinates, and at the same time, unifying the tile shape information of all tiles into the same spatial coordinate system.

[0051] In this embodiment, optionally, uploading the tile data to the distributed repository includes:

[0052] When the number of the tile shape information cannot be evenly divided by the total number of threads, the extra tile data is evenly distributed to the threads sorted in the front.

[0053] Specifically, the data volume of each tile in the real-scene three-dimensional model is not much different. The upload tasks of each thread can be allocated according to the number of tile shape information included in the model shape information corresponding to the real-scene three-dimensional model and the number of threads used for uploading data, ensuring that the number of tile data uploaded by each thread is balanced, avoiding excessive data volume uploaded by a single thread, and reducing the data storage efficiency. When the number of tile shape information cannot be evenly divided by the total number of threads, the remaining tile data can be evenly distributed to the first few threads in the sorting. All threads store the respective tile data in the distributed repository in parallel according to the allocated upload tasks, improving the storage efficiency.

[0054] Such as Figure 2 The flowchart of the data retrieval method for the real-scene three-dimensional model shown, adopting the above data distributed storage method, the retrieval method includes:

[0055] Step S1, obtain the spatial retrieval range of the real-scene three-dimensional model to be retrieved;

[0056] Step S2, respectively perform topological intersection of the spatial retrieval range with the data space distribution combination diagrams corresponding to each real-scene three-dimensional model, and lock the data space distribution combination diagrams intersecting with the spatial retrieval range;

[0057] Step S3, respectively perform topological intersection of the spatial retrieval range with all the tile shape information corresponding to the data space distribution combination diagram, and screen the tile shape information within the spatial retrieval range;

[0058] Step S4, based on the screened tile shape information, adopt a multi-thread parallel download method to parallel download the corresponding tile data from the distributed repository.

[0059] Specifically, first, the spatial retrieval range to be retrieved can be obtained through a human-computer interaction terminal. The human-computer interaction terminal can display a visualization interface, which includes basic remote sensing images, administrative divisions, and the data space distribution combination diagrams of all real-scene three-dimensional models. The user can manually draw the spatial retrieval range on the interface. Then, the data space distribution combination diagrams of each real-scene three-dimensional model can be traversed in turn, and the data space distribution combination diagram can be matched with the spatial retrieval range, so as to lock the data space distribution combination diagrams intersecting with the spatial retrieval range.

[0060] After that, the respective tile shape information in the locked data space distribution combination diagrams can be matched with the spatial retrieval range, and the tile shape information located within the spatial retrieval range can be screened out. Finally, according to the screened tile shape information, a multi-thread parallel download method can be adopted to parallel download the corresponding tile data from the distributed repository, realizing the efficient retrieval of the real-scene three-dimensional model data.

[0061] In this embodiment, optionally, performing topological intersection on the spatial search range and the combined graph of data spatial distribution includes:

[0062] Analyzing the shape information of the spatial search range and unifying the coordinate system of the shape information with the coordinate system of the tile shape information;

[0063] Performing topological intersection on the shape information after unifying the coordinate system and the combined graph of data spatial distribution.

[0064] Specifically, first, the drawn spatial search range can be analyzed to obtain the shape information corresponding to the spatial search range, and the shape information is unified into the spatial coordinate system of the tile shape information. Then, traverse all the combined graphs of data spatial distribution stored in the distributed repository, and compare the model shape information in the combined graph of data spatial distribution with the shape information of the spatial search range, so as to determine the combined graphs of data spatial distribution that intersect with the spatial search range. As Figure 5 shown, A, B, and C in the figure are the combined graphs of data spatial distribution that intersect with the spatial search range.

[0065] In this embodiment, optionally, performing topological intersection on the spatial search range and the tile shape information includes:

[0066] Determining all the tile shape information corresponding to the combined graph of data spatial distribution according to the topological mapping relationship corresponding to the combined graph of data spatial distribution.

[0067] Specifically, as Figure 6 shown, according to the model shape information in the combined graph of data spatial distribution and in combination with the topological mapping relationship between the tile and the real scene 3D model, all the tile shape information corresponding to the combined graph of data spatial distribution can be quickly determined, thereby improving the retrieval efficiency of the real scene 3D model.

[0068] In this embodiment, optionally, screening the tile shape information within the spatial search range includes:

[0069] Taking the tile shape information as the child node and the corresponding model shape information as the parent node, and recording the screened tile shape information in a tree structure.

[0070] Specifically, as Figure 7As shown, after screening out the tile shape information within the spatial retrieval range, to facilitate subsequent distribution of download tasks and improve retrieval efficiency, the screened tile shape information can be recorded in a tree structure. The parent node of the tree structure is the locked model shape information, and the child nodes are the tile shape information corresponding to the model shape information.

[0071] In this embodiment, optionally, parallel downloading of corresponding tile data from the distributed repository includes:

[0072] Traverse all the parent nodes in the tree structure and count the number of tile shape information under all the parent nodes;

[0073] Divide the download tasks of each thread evenly according to the number of tile shape information, and parallel download the corresponding tile data from the distributed repository according to the download tasks of each thread.

[0074] Specifically, when distributing download tasks, the tile data of each real-scene three-dimensional model can be downloaded in sequence according to the sorting of the parent nodes in the tree structure. When downloading the tile data of the real-scene three-dimensional model, first, the number of child nodes included in the parent node can be counted. Then, according to the number of child nodes and the total number of threads used for downloading data, the number of tiles to be downloaded by each thread is calculated. The download tasks of each thread are allocated according to the number of tiles to be downloaded, and each thread parallel downloads the corresponding tile data from the distributed repository, thereby improving the retrieval efficiency of the real-scene three-dimensional model.

[0075] In this embodiment, optionally, parallel downloading of corresponding tile data from the distributed repository includes: when the number of tile shape information cannot be evenly divided by the total number of threads, distribute the extra download tasks evenly to the threads with earlier sorting.

[0076] Specifically, when the number of child nodes cannot be evenly divided by the total number of threads, the remaining child nodes can be evenly distributed to the first few threads in the sorting. All threads parallel read the corresponding tile data from the distributed repository according to the allocated download tasks, thereby improving the storage efficiency.

[0077] A computer program product includes a computer program / instructions. When the computer program / instructions are executed by a processor, the steps of the above data retrieval method are implemented.

[0078] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered by the scope of the claims and the specification of the present invention.

Claims

1. A data distributed storage method for a real - scene three - dimensional model, characterized in that, Including: Obtain the outer bounding boxes of all tiles of the real-scene 3D model, and generate tile shape information corresponding to each tile in the real-scene 3D model; Fuse the tile shape information of all tiles to generate the model shape information corresponding to the real-scene 3D model, and establish a topological mapping relationship between the tiles and the real-scene 3D model; Based on the quantity of the tile shape information, use a multi-thread parallel upload method to upload all tile data corresponding to the real-scene 3D model data to the distributed repository; Generate a data space distribution combination diagram of the real-scene 3D model according to the model shape information.

2. The data distributed storage method for the real scene three-dimensional model according to claim 1, wherein, Generating the tile shape information corresponding to the tile includes: Calculate the tile shape information corresponding to each tile respectively according to the extreme value coordinates in the corresponding tile outer bounding box, and unify the spatial coordinate systems of all tile shape information.

3. The data distributed storage method for the real scene three-dimensional model according to claim 1, characterized in that Uploading the tile data to the distributed repository includes: When the quantity of the tile shape information cannot be evenly divided by the total number of threads, evenly distribute the extra tile data to the threads sorted in the front.

4. A data retrieval method for a real-scene three-dimensional model, adopting the data distributed storage method as described in claim 1 or 2, characterized in that, Including: Obtain the spatial retrieval range of the real-scene 3D model to be retrieved; Topologically intersect the spatial retrieval range with the data space distribution combination diagrams corresponding to each real-scene 3D model respectively, and lock the data space distribution combination diagrams intersecting with the spatial retrieval range; Topologically intersect the spatial retrieval range with all tile shape information corresponding to the data space distribution combination diagram respectively, and screen the tile shape information within the spatial retrieval range; Based on the screened tile shape information, use a multi-thread parallel download method to parallelly download the corresponding tile data from the distributed repository.

5. The data retrieval method for the true three-dimensional model according to claim 4, wherein Topologically intersect the spatial retrieval range with the data space distribution combination diagram, including: Analyze the shape information of the spatial retrieval range, and unify the coordinate system of the shape information with the coordinate system of the tile shape information; Topologically intersect the shape information after unifying the coordinate systems with the data space distribution combination diagram.

6. The data retrieval method for the true three-dimensional model according to claim 4, wherein Topologically intersect the spatial retrieval range with the tile shape information, including: Determine all tile shape information corresponding to the data space distribution combination diagram according to the topological mapping relationship corresponding to the data space distribution combination diagram.

7. The data retrieval method for the real scene three-dimensional model according to claim 4, characterized in that, Screening the tile shape information within the spatial retrieval range includes: Use a tree structure to record the screened tile shape information with the tile shape information as the child node and the corresponding model shape information as the parent node.

8. The data retrieval method for the real scene three-dimensional model according to claim 7, characterized in that Parallelly download the corresponding tile data from the distributed repository, including: Traverse all parent nodes in the tree structure, and count the quantity of tile shape information under all parent nodes; Evenly divide the download tasks of each thread according to the quantity of the tile shape information, and parallelly download the corresponding tile data from the distributed repository according to the download tasks of each thread.

9. The method for retrieving data of the true three-dimensional model according to claim 7, wherein Parallelly download the corresponding tile data from the distributed repository, including: When the quantity of the tile shape information cannot be evenly divided by the total number of threads, evenly distribute the extra download tasks to the threads sorted in the front.

10. A computer program product, comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by a processor, the steps of the data retrieval method according to any one of claims 4-9 are implemented.

Images