Tile data compact storage format design and application method containing time information

Abstract

The invention belongs to the technical field of remote sensing image application, and provides a compact storage format design and application method for tile data containing time information in order to facilitate quick loading and display of multi-phase remote sensing images, which comprises the following steps: S1, setting a file format of tile data containing multi-temporal remote sensing images as a compact tile data file format; s2, initializing a compact tile data file; s3, saving the tile data to a corresponding compact tile data file; s4, merging compact tile data files with the same name; s5, performing image supplement processing on the missing tile data; and S6, database generation and query of multi-temporal remote sensing image tile data. According to the method, time information is included in various metadata information in the compact tile data file for storing the remote sensing image tile data, the complete process of tile data storage, merging, image complementing, construction and browsing is explained, and the method has practical application value.

Description

Technical Field

[0001] The invention belongs to the technical field of remote sensing image application. Background Art

[0002] The amount of remote sensing image data is large, especially for remote sensing images covering the entire world. The amount of original remote sensing image data has reached the PB (petabyte) level. There are great difficulties in displaying, processing and transmitting massive remote sensing image data, involving challenges such as slow loading speed, low data processing efficiency and high network transmission cost.

[0003] Processing remote sensing images into tile data is a common way for GIS systems (Geographic Information System) to display and process massive remote sensing data. Tile data is a spatial data format based on raw remote sensing images. It cuts map data into small tiles to facilitate network transmission and display. Tile data uses a block encoding mode. When the client obtains it, it only returns the tile map of the requested area and the corresponding level. When displayed on the front end, the tile data can be directly called for drawing.

[0004] The tile data compact storage format is the mainstream tile format currently widely used. It is easy to migrate, faster to create and reduces storage space. It uses a certain compression and encryption mechanism for the original slice file. In terms of structure, a set of files is used to replace the tile mechanism of the original image format. At this time, the tile data in folders of different scales will be presented as a set of files. The set of files includes two parts: index data source and tile data. The tile data structure uses a set of files to replace the original image format, and the tile data in folders of different scales will be stored in compact files.

[0005] However, due to the lack of time information in tile data, multi-temporal remote sensing images of the same area cannot be displayed in time series after being processed into tile data, which is not convenient for fast loading and display of multi-temporal remote sensing images. Summary of the invention

[0006] In order to facilitate the rapid loading and display of multi-period remote sensing images, the present invention proposes a "compact storage format design and application method for tile data containing time information", such as Figure 1 As shown, including:

[0007] S1. Setting the file format of the multi-temporal remote sensing image tile data to a compact tile data file format, wherein the compact tile data file format includes file attribute data, tile attribute data and tile data:

[0008] The file attribute data contains 4 fields: level, row, col, and depth, each of which occupies 4 bytes;

[0009] The tile attribute data contains four fields: position, length, datetime, and other, each of which occupies 4 bytes;

[0010] Tile data is used to save the actual tile data;

[0011] The meaning of each field is shown in the following table:

[0012]

[0013] S2, reading multi-temporal remote sensing image tile data, initializing the compact tile data file size, and writing the file attribute data;

[0014] S3, according to the slice level and grid number corresponding to the tile data, save it to the corresponding compact tile data file, and write the tile attribute data;

[0015] S4, merging compact tile data files with the same name;

[0016] S5, using historical tile data to perform image patching of missing tile data of current remote sensing images;

[0017] S6. Database generation and query of multi-temporal remote sensing image tile data:

[0018] The multi-temporal remote sensing image tile data are grouped by date, and each group of data generates a compact tile data file carrying the time information of the multi-temporal remote sensing image and merges them, and saves them in a folder named with the milliseconds corresponding to the current time.

[0019] Establish a database, traverse the compact tile data files in each folder in turn, and store the tile level, row and column number, time information, and folder name in the database.

[0020] When accessing multi-temporal remote sensing image tile data according to time series, query the database according to the time information and the tile level, row and column number of the current request. If the query result is not empty, execute the next step, otherwise it means that the tile data does not exist.

[0021] According to the folder name in the query result, read the tile data file in the folder and obtain the corresponding tile data. Technical effect:

[0022] The present invention includes time information in a variety of metadata information in a compact tile data file that stores remote sensing image tile data, thereby solving the technical problem that multi-temporal remote sensing images are not easy to load and display quickly. When multi-temporal remote sensing image tile data are stored in a database, the folder name plays a role in grouping and saving, so the database can save tile data information of the same area at different shooting times. When querying tile data of any area, all tile data information of the corresponding area can be directly retrieved. Since the tile data shooting time is included, the following can be achieved in the actual processing process: Figure 3-5 As shown in the figure, the timeline is constructed with the tile shooting time as the node, and the tile data is automatically switched in chronological order or the corresponding tile data can be manually viewed by clicking any time node. This is not only convenient for establishing and querying the existing multi-temporal remote sensing image tile data, but also for dynamically entering and utilizing the new multi-temporal remote sensing image tile data read later in the database. This illustrates the complete process of remote sensing image tile data storage, merging, infilling, construction and browsing, which has practical application value.

[0023] By merging compact tile data files with the same name, it is more convenient to transmit and browse, because the target area is often composed of multiple remote sensing images. When divided according to grids, there is a situation where the same grid contains multiple images. Finally, multiple compact tile data files with the same name are generated and merged into one file. The historical tile data is also used to supplement the current missing data, which solves the problem that the remote sensing image data taken at the current time fails to cover the target area 100%, making it easier to display the complete target area image. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a schematic diagram of the overall process of the present invention.

[0025] Figure 2 A diagram showing a list of compact tile data files.

[0026] Figure 3 This is a schematic diagram of the results of reading tile data of a single remote sensing image.

[0027] Figure 4 This is a schematic diagram of the stitching effect after reading multiple remote sensing image tile data of the target area.

[0028] Figure 5 This is a schematic diagram of the effect of browsing multi-temporal remote sensing image tile data along the time axis. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by ordinary technicians in the field without creative work by adopting the embodiments of the present invention are within the scope of protection of the present invention.

[0030] The S2 is specifically as follows: according to the level division rule of the earth space grid, the remote sensing image is divided into n blocks according to the grid, and the corresponding tile data file is created according to the number of the grid where each image is located, and the initial level of the slice is calculated as level, and the slice depth is depth. The total number of tiles produced is total, and the file attribute data is written into the tile data file. The tile data file size is initialized to 16+total*16 to ensure that the tile data file reserves tile attribute data storage space for each tile data.

[0031] The step S3 specifically includes: opening the corresponding tile data file according to the slice level and grid number corresponding to the tile data, appending the tile data to the end of the tile data file, and recording the starting position and the number of bytes of the tile data saved in the tile data file, calculating the storage position of the tile attribute data in the tile data file according to the slice level and grid number corresponding to the tile data, and writing the tile attribute data;

[0032] The S4 is specifically as follows: mFile1 and mFile2 are two tile data files with the same name generated according to S1 and S2, each tile attribute data is read from mFile2 in turn, and according to the storage location information in the tile attribute data, the tile data is read from mFile2, the tile data of mFile2 is appended to mFile1, and the tile attribute data of mFile1 is updated.

[0033] The specific calculation process of each step of the embodiment of the present invention is described below.

[0034] The tile attribute data of each file in S1 contains the tile attribute data of multiple tile data. The tile index is stored in the order of X number priority.

[0035] tile(1, 1)-tile(x_max, 1), tile(1, 2)-tile(x_max, 2)...tile(1, y_max)-tile(x_max, y_max), the tile data part is composed of several tile data, the order of tile data depends on the order in which tiles are written. When obtaining tile data, you need to first calculate the values ​​of the position field and length field of the target tile in the attribute data, and then read the correct data in the tile data part.

[0036] S2 reads multi-temporal remote sensing image tile data, initializes the compact tile data file size, and writes the file attribute data: optional,

[0037] S2.1 Input remote sensing image path, slice initial level level1, slice depth depth1;

[0038] S2.2 read remote sensing image data;

[0039] S2.3 Calculate the latitude and longitude of the upper left corner (left_lon, left_lat) and the latitude and longitude of the lower right corner (right_lon, right_lat) of the original remote sensing image according to the affine transformation parameters of the remote sensing image;

[0040] S2.4 According to the earth space grid division rule, calculate the row number and column number (left_row, left_col) of the grid where the upper left corner of the remote sensing image to be processed is located, and the row number and column number (right_row, right_col) of the grid where the lower right corner of the remote sensing image to be processed is located at level 1;

[0041] S2.5 divides the remote sensing image to be processed into n blocks according to the grid to which each pixel belongs.

[0042] n=(right_row-left_row+1)*(right_col-left_col+1);

[0043] S2.6 Let (rowi, coli) be the grid code to which the i-th image belongs, 1≤i≤n, i∈N + ;

[0044] S2.7 Create a new tile data file world_level1_rowi-coli.pyr, recorded as File;

[0045] S2.8 writes 16 bytes of file attribute data to File in sequence, namely level1, rowi, coli, depth1;

[0046] S2.9 calculates the total number of tiles,

[0047] total=pow(4,0)+pow(4,1)+...+pow(4,depth1-1)+pow(4,depth1);

[0048] S2.10 initializes the file size to 16+total*16 bytes;

[0049] S2.11 repeats S2.6 to S2.10 to initialize n tile data files in total. The generated tile data files are as follows: Figure 2 As shown, each pyr file stores all tile data from 10 to 18 of a ten-level grid, 10 is the tile level represented by the grid,

[0050] The last two numbers are the row and column values ​​of the grid.

[0051] In S3, according to the slice level and grid number corresponding to the tile data, the tile data is saved to the corresponding compact tile data file, and the tile attribute data is written: optional,

[0052] S3.1: cFile is the tile data file to be written, clevel is the tile level to be written, crow is the row number of the tile grid number, ccol is the column number of the tile grid number, cdatetime is the time information of the tile, and csize is the number of bytes of the tile data.

[0053] S3.2 Let row_num be the number of tiles per row in the global grid at the clevel slicing level, row_num = pow(2,clevel), let col_num be the number of tiles per column in the global grid at the clevel slicing level,

[0054] col_num = pow(2,clevel);

[0055] S3.3 calculates the size of cFile as size bytes;

[0056] S3.4 Tile data is written to the end of cFile in an appended manner;

[0057] S3.5 calculates the starting position of the tile attribute data stored in cFile,

[0058] cindex=pow(4,0)+pow(4,1)+...+pow(4,clevel-1)+(crow-1)*row_num+col;

[0059] S3.6 writes size+1, size+csize, and cdatetime in sequence at position cindex+1 of cFile.

[0060] Merge compact tile data files with the same name in S4: optional,

[0061] S4.1: mFile1 and mFile2 are two tile data files with the same name generated according to S2 and S3;

[0062] S4.2 starts from the 29th byte of mFile2 and reads 16 bytes for judgment.

[0063] (1) If all bytes are null, continue reading;

[0064] (2) If the first 12 bytes are not empty and bytes 13 to 16 are empty, execute S4.3;

[0065] (3) If all 16 bytes are not empty, the tile file has been read and the file is closed;

[0066] S4.3 records the starting position of the tile data from bytes 1 to 4 as ms_index, the ending position of the tile data from bytes 5 to 8 as me_index, and the time information of the tile data from bytes 9 to 12 as mdatetime;

[0067] S4.4 reads the data from ms_index to me_index from mFile2, and writes the data to mFile1 according to S3.4-S3.6.

[0068] S4.5 repeats S4.2-S4.4 until all tile data in mFile2 are read.

[0069] In S5, the missing data of the current remote sensing image is supplemented by historical tile data: optional.

[0070] The missing tile data of S5.1 is recorded as (dlevel, drow, dcol), where dlevel is the tile slice level, drow is the tile row number, and dcol is the tile column number;

[0071] S5.2 searches the historical tile data file for tile data with a slice level of dlevel, a row number of drow, and a column number of dcol, and writes the searched tile data into the tile data file that needs to be infilled according to S3.

[0072] The preferred embodiments of the present invention are described in detail above in conjunction with the accompanying drawings. However, the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, a variety of simple modifications can be made to the technical solution of the present invention, and these simple modifications all belong to the protection scope of the present invention.

[0073] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

Claims

1. A compact storage format design and application method for tile data containing time information, characterized in that: include: S1. Setting the file format of the multi-temporal remote sensing image tile data to a compact tile data file format, wherein the compact tile data file format includes file attribute data, tile attribute data and tile data. The file attribute data contains 4 fields: level, row, col, and depth, each of which occupies 4 bytes; The tile attribute data contains four fields: position, length, datetime, and other, each of which occupies 4 bytes; Tile data is used to save the actual tile data; The meaning of each field is shown in the following table: S2, reading multi-temporal remote sensing image tile data, initializing the compact tile data file size, and writing the file attribute data; S3, according to the slice level and grid number corresponding to the tile data, save it to the corresponding compact tile data file, and write the tile attribute data; S4, merging compact tile data files with the same name; S5. Use historical tile data to process missing data of current remote sensing images; S6. Database generation and query of multi-temporal remote sensing image tile data: The multi-temporal remote sensing image tile data are grouped by date, and each group of data generates a compact tile data file carrying the time information of the multi-temporal remote sensing image and merges them, and saves them in a folder named with the milliseconds corresponding to the current time. Establish a database, traverse the compact tile data files in each folder in turn, and store the tile level, row and column number, time information, and folder name in the database. When accessing multi-temporal remote sensing image tile data according to time series, query the database according to the time information and the tile level, row and column number of the current request. If the query result is not empty, execute the next step, otherwise it means that the tile data does not exist. According to the folder name in the query result, read the tile data file in the folder to obtain the corresponding tile data.

2. The design and application method of compact storage format of tile data containing time information according to claim 1 is characterized in that: The S2 is specifically as follows: according to the level division rule of the earth space grid, the remote sensing image is divided into n blocks according to the grid, and the corresponding tile data file is created according to the number of the grid where each image is located, and the initial level of the slice is calculated as level, and the slice depth is depth. The total number of tiles produced is total, and the file attribute data is written into the tile data file. The tile data file size is initialized to 16+total*16 to ensure that the tile data file reserves tile attribute data storage space for each tile data.

3. The design and application method of compact storage format of tile data containing time information according to claim 1, characterized in that: The S3 is specifically as follows: according to the slice level and grid number corresponding to the tile data, the corresponding tile data file is opened, the tile data is appended to the end of the tile data file, and the starting position and the number of tile data bytes saved in the tile data file are recorded; according to the slice level and grid number corresponding to the tile data, the storage position of the tile attribute data in the tile data file is calculated, and the tile attribute data is written.

4. The design and application method of compact storage format of tile data containing time information according to claim 1, characterized in that: The S4 is specifically as follows: mFile1 and mFile2 are two tile data files with the same name generated according to S2 and S3, each tile attribute data is read from mFile2 in turn, and according to the storage location information in the tile attribute data, the tile data is read from mFile2, the tile data of mFile2 is appended to mFile1, and the tile attribute data of mFile1 is updated.

Images