High-efficiency retrieval method for large-scale spatial data of aviation geographic information

Abstract

The invention discloses an efficient retrieval method for large-scale spatial data of aviation geographic information. The efficient retrieval method comprises the following steps: compressing each element in the aviation geographic spatial data by adopting a vector quantization technology; an R tree is adopted as a spatial data index structure, and an index is established through a minimum boundary rectangle MBR surrounding a spatial object; and carrying out parallel retrieval on the geospatial data. According to the method, the data storage space requirement is reduced, the geographic space data retrieval efficiency is effectively improved, and the data retrieval time of a large-scale geographic scene is greatly saved.

Description

Technical Field

[0001] The invention relates to aviation geographic information technology, and more particularly to an efficient retrieval method for large-scale spatial data of aviation geographic information. Background Art

[0002] With the development of aerospace technology, the amount of aeronautical geographic information data is growing, especially large-scale spatial data. However, existing geospatial retrieval technologies are often inefficient and cannot meet the needs of instant retrieval of large-scale geographic scenes. When processing large-scale spatial data, traditional methods usually encounter the following problems:

[0003] 1) Low data processing efficiency: With the development of aerospace technology, the amount of aeronautical geographic information data is growing. Traditional retrieval technology lacks effective compression and indexing strategies, resulting in low efficiency in processing these massive data and cannot meet the needs of instant retrieval of large-scale geographic scenes;

[0004] 2) Large storage space for data to be retrieved: Due to the complexity of geographic information data, it usually requires a large amount of storage space. Traditional retrieval technology often does not take this into account, resulting in a waste of storage space and increasing the difficulty of data transmission and storage;

[0005] 3) Slow query speed: In traditional retrieval technology, due to the lack of effective index structure, the query speed is often slow, which makes users have to wait for a long time when retrieving geospatial data, affecting the user experience. Summary of the invention

[0006] The purpose of the invention is to provide an efficient retrieval method for large-scale spatial data of aeronautical geographic information, which realizes efficient and fast retrieval of aeronautical geographic spatial data by using vector quantization technology to compress aeronautical geographic spatial data, using R-tree as a spatial data index structure, and implementing parallel retrieval and other technical means. The invention not only improves data processing efficiency and reduces data storage space requirements, but also speeds up query speed and solves the difficulties of traditional technologies in processing complex shapes such as polygons.

[0007] The object of the present invention is achieved through the following technical solutions:

[0008] An efficient retrieval method for large-scale spatial data of aviation geographic information includes the following steps:

[0009] Step 1: Use vector quantization technology to compress each element in the aviation geospatial data;

[0010] Step 2: Use R-tree as the spatial data index structure and create an index through the minimum bounding rectangle MBR surrounding the spatial object;

[0011] Step 3: Parallel retrieval of geospatial data.

[0012] Preferably, in step one, for point data compression, quantization technology is used to convert the longitude and latitude coordinates represented by floating-point numbers into integers; for line data compression, incremental encoding technology is used to store the starting point and incremental information of the route; for surface data compression, the boundary is accurately represented by selecting an appropriate "code book" or "vector set".

[0013] Preferably, the specific implementation process of step 2 is as follows:

[0014] 1) Construct the MBR of each element:

[0015] 2) Insert the newly constructed MBR into the R-tree, including:

[0016] 2.1) Traverse the R-tree to find the leaf node with the smallest overlap with the new MBR, and give priority to the leaf node closest to the query range;

[0017] 2.2) Determine whether the selected leaf node is full. If not, directly insert the new MBR into the leaf node; if full, proceed to 2.3);

[0018] 2.3) First select a full leaf node for splitting, and divide the MBR of the node into two parts evenly or split it according to a specific strategy; then create a new leaf node, and if the parent node is full due to the insertion of the new node, recursively split the parent node; then move the split part of the MBR to the new node; finally, after successfully inserting the new MBR or completing the split operation, ensure the correctness and consistency of the R-tree structure.

[0019] Preferably, in constructing the MBR of each element:

[0020] For a point object, first get the coordinates (x, y) of the point, and then build a very small rectangle with the point as the center;

[0021] For a line object, first obtain the coordinates of the two endpoints of the line segment (x1, y1) and (x2, y2); then find the minimum and maximum values ​​of x1 and x2 to determine the boundary of the MBR in the X direction, and find the minimum and maximum values ​​of y1 and y2 to determine the boundary of the MBR in the Y direction; finally, combine the above boundaries in the X and Y directions to construct the MBR of the line segment;

[0022] For a polygon object, first obtain the coordinates of all vertices of the polygon; then traverse the X coordinates of all vertices, find the minimum and maximum values, determine the boundary of the MBR in the X direction, and traverse the Y coordinates of all vertices, find the minimum and maximum values, and determine the boundary of the MBR in the Y direction; finally, combine the above boundaries in the X and Y directions to construct the MBR of the polygon.

[0023] Preferably, the implementation process of step three is as follows:

[0024] 3.1) Divide the large spatial data set into multiple small data subsets and construct a corresponding R-tree index for each data subset;

[0025] 3.2) Assign a data subset and a corresponding R-tree index to each processing node;

[0026] 3.3) All processing nodes retrieve their respective data subsets;

[0027] 3.4) After all processing nodes complete the search, the search results of each node are gathered to the central node to merge and remove duplicate results.

[0028] Preferably, when allocating the data subset to each processing node, a dynamic scheduling strategy is used to dynamically allocate the data subset according to the real-time load condition of the node.

[0029] The beneficial effects of the present invention are:

[0030] The present invention provides a method for efficiently retrieving large-scale spatial data of geographic information in the field of aviation data retrieval. The data retrieval function realized based on this method not only reduces the demand for data storage space, but also effectively improves the efficiency of geographic spatial data retrieval, greatly saving the data retrieval time of large-scale geographic scenes. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 Schematic diagram of an efficient retrieval method for large-scale spatial data of aerial geographic information. DETAILED DESCRIPTION

[0032] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0033] See also Figure 1 As shown, the present embodiment shows an efficient retrieval method for large-scale spatial data of aviation geographic information, comprising the following steps:

[0034] Step 1: Spatial data compression: Use vector quantization technology to compress each element in the aeronautical geographic spatial data.

[0035] Data compression is an important part of large-scale spatial data processing of aeronautical geographic information. Different compression strategies are used according to different data types (points, lines, and surfaces).

[0036] 1. Point data compression

[0037] Use quantization technology to convert the longitude and latitude coordinates represented by floating point numbers into integers. The longitude and latitude coordinates can be quantized into integers with a certain number of digits while ensuring that the accuracy of the coordinates is not affected. In addition, this compression method can significantly reduce the storage space requirements and accelerate the efficiency of data transmission and storage. The specific implementation process is as follows:

[0038] 1) Read in point coordinate data;

[0039] 2) Convert the longitude and latitude into integers according to the preset quantization interval;

[0040] 3) Establish a coordinate-integer mapping table;

[0041] 4) Output the compressed integer coordinates.

[0042] 2. Line data compression

[0043] Incremental encoding technology is used to store only the starting point and incremental information (such as direction and length) of the route. This can greatly reduce the storage space requirements of line data while ensuring the calculation and rendering efficiency of routes and flight paths. The specific implementation process is as follows:

[0044] 1) Read line coordinate data;

[0045] 2) Record the starting point;

[0046] 3) Calculate the increments segment by segment and output the encoded line segment description.

[0047] 3. Surface data compression

[0048] Use vector quantization technology to compress the boundary information of complex areas. By selecting an appropriate "codebook" or "vector set" to accurately represent the boundary, ensure the optimization of storage requirements and rendering efficiency. The specific implementation process is as follows:

[0049] 1) Read the coordinates of the polygon vertices;

[0050] 2) Select a vector set as the "codebook";

[0051] 3) Use a vector set to encode the vertex sequence to describe the polygon.

[0052] Step 2: Construct a spatial index R-tree: Use the R-tree as the spatial data index structure and establish an index through the minimum bounding rectangle (MBR) around the spatial object.

[0053] Spatial data indexing is the key to improving retrieval efficiency. We choose R-tree as the index structure. We build an index around the minimum bounding rectangle (MBR) of the spatial object. We ensure the balance and query efficiency of the R-tree through operations such as insertion and splitting. The specific implementation process is as follows:

[0054] 1. Construct the MBR of each element;

[0055] 1.1 Object MBR Construction

[0056] For point objects, each point can be regarded as a tiny rectangular area. Therefore, the MBR calculation of point objects is relatively simple. The process is as follows:

[0057] 1) Determine the location of the point: first obtain the coordinates (x, y) of the point;

[0058] 2) Construct a tiny rectangle: Construct a very small rectangle with the point as the center. This rectangle is essentially the point itself.

[0059] 1.2 Line Object MBR Construction

[0060] The MBR calculation of the line object needs to consider the start and end points of the line segment. The process is as follows:

[0061] 1) Get the endpoints of the line segment: First, get the coordinates of the two endpoints of the line segment (x1, y1) and (x2, y2);

[0062] 2) Calculate the range in the X direction: find the minimum and maximum values ​​of x1 and x2 to determine the boundary of MBR in the X direction;

[0063] 3) Calculate the range in the Y direction: Similarly, find the minimum and maximum values ​​in y1 and y2 to determine the boundary of the MBR in the Y direction;

[0064] 4) Construct MBR: Combine the above boundaries in the X and Y directions to construct the MBR of the line segment.

[0065] 1.3 MBR construction of polygonal objects

[0066] The MBR construction of polygon objects is similar to that of line objects, and multiple vertices must be considered. The calculation process is as follows:

[0067] 1) Get all the vertices of the polygon: First get the coordinates of all the vertices of the polygon;

[0068] 2) Calculate the range in the X direction: traverse the X coordinates of all vertices, find the minimum and maximum values, and determine the boundary of the MBR in the X direction;

[0069] 3) Calculate the range in the Y direction: Similarly, traverse the Y coordinates of all vertices, find the minimum and maximum values, and determine the boundary of the MBR in the Y direction;

[0070] 4) Construct MBR: Combine the above boundaries in the X and Y directions to construct the MBR of the polygon.

[0071] 2 Insert the newly constructed MBR into the R-tree

[0072] 2.1 Select leaf nodes

[0073] 1) Traverse the R-tree and find the leaf node with the smallest overlap with the new MBR;

[0074] 2) Prioritize the leaf nodes closest to the query range to optimize the efficiency of subsequent queries.

[0075] 2.2 Check leaf node capacity

[0076] 1) Determine whether the selected leaf node is full (based on the pre-set maximum capacity);

[0077] 2) If it is not full, insert the new MBR directly into the leaf node;

[0078] 3) If it is full, perform subsequent split operations.

[0079] 2.3 Node Splitting and Updating

[0080] 1) Select a full leaf node (usually the fullest one) to split;

[0081] 2) Divide the MBR of the node into two parts equally or according to a specific strategy;

[0082] 3) Create a new leaf node and move the split part of the MBR to the new node;

[0083] 4) If the parent node is full due to inserting a new node, recursively split the parent node;

[0084] 5) After successfully inserting a new MBR or completing a split operation, ensure the correctness and consistency of the R-tree structure.

[0085] Step 3: Parallel retrieval of geospatial data.

[0086] Parallel retrieval can further improve the processing efficiency of large-scale spatial data: First, the large spatial data set is divided into multiple small data subsets, each subset has a corresponding R-tree index. Then, multiple processing nodes process their respective data subsets in parallel, and finally collect and merge the results. The implementation process is as follows:

[0087] 1) Spatial data partitioning: Divide a large spatial data set into multiple small data subsets to ensure that there is no overlap between the data in each subset. At the same time, build a corresponding R-tree index for each data subset;

[0088] 2) Parallel processing preparation: Allocate a data subset and the corresponding R-tree index to each processing node. Ensure that each processing node has all the necessary resources and indexes required for retrieval;

[0089] 3) Parallel search execution: All processing nodes start searching their respective data subsets at the same time. Each node uses the same or similar search algorithm, but processes different data. This ensures that the search operation can be performed on multiple data subsets simultaneously;

[0090] 4) Result merging: After all processing nodes have completed the search, the search results of each node are gathered to the central node for merging and deduplication. This step ensures that the results retrieved from all data subsets are merged into a unified result set;

[0091] 5) Optimization and load balancing: To optimize the performance of parallel retrieval, a load balancing strategy can be used to ensure that the size of the data subset matches the computing power of the processing node. A dynamic scheduling strategy can be used to dynamically allocate data subsets based on the real-time load of the node.

[0092] In summary, by combining technical means such as data compression, spatial data indexing and parallel retrieval, large-scale spatial data of aeronautical geographic information can be efficiently processed to achieve efficient and fast aeronautical geographic spatial data retrieval to meet the needs of the aviation field.

[0093] It is understandable that those skilled in the art can make equivalent substitutions or changes based on the technical solution and inventive concept of the present invention, and all these changes or substitutions should fall within the protection scope of the claims attached to the present invention.

Claims

1. An efficient retrieval method for large-scale spatial data of aviation geographic information, characterized in that The following steps are involved: Step 1: Use vector quantization technology to compress each element in the aviation geospatial data; Step 2: Use R-tree as the spatial data index structure and create an index through the minimum bounding rectangle MBR surrounding the spatial object; Step 3: Parallel retrieval of geospatial data.

2. The method for efficiently retrieving large-scale spatial data of aeronautical geographic information according to claim 1, characterized in that In step 1, for point data compression, quantization technology is used to convert the longitude and latitude coordinates represented by floating point numbers into integers; For line data compression, incremental encoding technology is used to store the starting point and incremental information of the route; For surface data compression, the boundaries are accurately represented by choosing an appropriate "codebook" or "vector set".

3. The method for efficiently retrieving large-scale spatial data of aeronautical geographic information according to claim 1, characterized in that The specific implementation process of step 2 is as follows: 1) Construct the MBR of each element: 2) Insert the newly constructed MBR into the R-tree, including: 2.1) Traverse the R-tree to find the leaf node with the smallest overlap with the new MBR, and give priority to the leaf node closest to the query range; 2.2) Determine whether the selected leaf node is full. If not, directly insert the new MBR into the leaf node; if full, proceed to 2.3); 2.3) First select a full leaf node for splitting, and divide the MBR of the node into two parts evenly or split it according to a specific strategy; then create a new leaf node, and if the parent node is full due to the insertion of the new node, recursively split the parent node; then move the split part of the MBR to the new node; finally, after successfully inserting the new MBR or completing the split operation, ensure the correctness and consistency of the R-tree structure.

4. The method for efficiently retrieving large-scale spatial data of aeronautical geographic information according to claim 3 is characterized in that In the MBR that builds each element: For a point object, first get the coordinates (x, y) of the point, and then build a very small rectangle with the point as the center; For a line object, first obtain the coordinates of the two endpoints of the line segment (x1, y1) and (x2, y2); then find the minimum and maximum values ​​of x1 and x2 to determine the boundary of the MBR in the X direction, and find the minimum and maximum values ​​of y1 and y2 to determine the boundary of the MBR in the Y direction; finally, combine the above boundaries in the X and Y directions to construct the MBR of the line segment; For a polygon object, first obtain the coordinates of all vertices of the polygon; then traverse the X coordinates of all vertices, find the minimum and maximum values, determine the boundary of the MBR in the X direction, and traverse the Y coordinates of all vertices, find the minimum and maximum values, and determine the boundary of the MBR in the Y direction; finally, combine the above boundaries in the X and Y directions to construct the MBR of the polygon.

5. The method for efficiently retrieving large-scale spatial data of aeronautical geographic information according to claim 1, characterized in that The implementation process of step three is as follows: 3.1) Divide the large spatial data set into multiple small data subsets and construct a corresponding R-tree index for each data subset; 3.2) Assign a data subset and a corresponding R-tree index to each processing node; 3.3) All processing nodes retrieve their respective data subsets; 3.4) After all processing nodes complete the search, the search results of each node are gathered to the central node to merge and remove duplicate results.

6. The method for efficiently retrieving large-scale spatial data of aeronautical geographic information according to claim 5, characterized in that When allocating data subsets to each processing node, a dynamic scheduling strategy is used to dynamically allocate data subsets based on the real-time load of the node.

Images