A method and apparatus for optimizing remote sensing images based on spatially partitioned grids

Abstract

This invention discloses a method and apparatus for optimizing remote sensing imagery based on spatially partitioned grids. The method includes: acquiring a query region and a query remote sensing image set; determining grid partitioning information based on the query region; determining a gridded region set and a gridded remote sensing image set based on the grid partitioning information, the query region, and the query remote sensing image set; and determining a preferred remote sensing image set based on the gridded region set and the gridded remote sensing image set. Therefore, this application utilizes spatially partitioned grids to achieve automatic optimization of remote sensing image data, improving image optimization efficiency while ensuring image coverage.

Description

Technical Field

[0001] This invention belongs to the field of remote sensing image management and application technology, and particularly relates to a method and apparatus for optimizing remote sensing images based on spatially partitioned grids. Background Technology

[0002] With the rapid development of remote sensing technology, hundreds of satellites are currently in orbit globally conducting Earth observation missions simultaneously. The amount of remote sensing image data acquired daily has increased from gigabytes (GB) to petabytes (PB), showing a trend of massive data volume. This vast amount of remote sensing image data can meet the diverse needs of users and administrators, such as land cover change monitoring, surface water monitoring, and emergency mapping. However, traditional remote sensing image data retrieval methods based on attributes such as image data source, spatial location, and temporal range often result in significant spatial redundancy in the search results, increasing the workload of subsequent remote sensing image data processing. To address this spatial redundancy problem, some scholars have conducted research on optimizing remote sensing datasets. This involves selecting a smaller number of datasets from the search results that meet the criteria and provide the maximum coverage of the target area, thereby reducing the workload of data processing and improving the efficiency of remote sensing image data utilization.

[0003] The problem of optimizing remote sensing image sets needs to consider two aspects: the evaluation of multi-dimensional image attributes (such as resolution, imaging time, etc.) and image spatial coverage.

[0004] (1) Evaluation of multi-dimensional image attributes. The solutions to this problem are flexible and diverse, requiring the design of evaluation functions according to different application needs, and there is no unified measurement standard. The commonly used method is to construct an evaluation index system using the weighted average method to quantitatively evaluate the attributes of remote sensing images.

[0005] (2) Image Spatial Coverage Problem. This can be transformed into a Set Coverage Problem (SCP) to be solved. The Set Coverage Problem is defined as follows: Given two sets E and S, where E is the set of elements and S is the set of subsets of E, find a subset C of S such that the union of all sets in C equals E, and the minimum number of elements in C (minimum |C|) is obtained, which is the optimal image coverage set. In the solution process, the image data must first be spatially segmented, and then the segmented image coverage area is converted into a set of non-overlapping discrete spatial regions. Finally, a specific optimization algorithm is used to solve the problem. Among them, ① spatial segmentation methods are mainly divided into vector methods and grid methods. The vector method uses precise spatial coordinates to describe each irregular discrete spatial region. The advantage of this method is high accuracy and accurate segmentation, but the segmentation process involves a large number of simultaneous calculations, resulting in high complexity. The grid method performs regular grid subdivision of the entire study area and uses a grid set to represent the coverage area of ​​each image. The advantage of this method is that it is simple, easy to implement, and has low complexity, but there is a contradiction between the expression accuracy and the number of grids. ②The most widely used and efficient optimization algorithm for solving the SCP problem is the heuristic algorithm. Heuristic algorithms mainly include metaheuristic algorithms such as genetic algorithms and ant colony algorithms, and non-metaheuristic algorithms such as greedy search and Lagrange relaxation method.

[0006] In the problem of optimizing remote sensing image sets, the large number of candidate images and the typically large target area result in a significant consumption of storage space and computational resources, leading to low computational efficiency. Summary of the Invention

[0007] The technical problem to be solved by the present invention is to provide a method and apparatus for optimizing remote sensing images based on spatially partitioned grids, which achieves rapid and automatic optimization of image coverage sets through iterative optimization, thereby improving the efficiency of automatic coverage optimization of remote sensing images.

[0008] To address the aforementioned technical problems, a first aspect of this invention discloses a method for optimizing remote sensing images based on spatially partitioned grids, the method comprising:

[0009] S1, obtain the query area and the query remote sensing image set; the query remote sensing image set includes several query remote sensing images;

[0010] S2, determine the grid subdivision information based on the query area;

[0011] S3, based on the grid subdivision information, the query area, and the query remote sensing image set, determine the gridded area set and the gridded remote sensing image set;

[0012] The gridded remote sensing image set includes several gridded remote sensing image subsets of the queried remote sensing image;

[0013] S4, based on the gridded region set and the gridded remote sensing image set, a preferred remote sensing image set is determined.

[0014] As an optional implementation, in the first aspect of the present invention, the...

[0015] Based on the grid subdivision information, the query region, and the query remote sensing image set, a gridded region set and a gridded remote sensing image set are determined, including:

[0016] S31, parse the mesh subdivision information to obtain subdivision hierarchy information and mesh quantity information;

[0017] S32, using the segmentation hierarchy information and the grid quantity information, the query region is gridded to obtain a gridded region set;

[0018] S33, using the subdivision hierarchy information and the grid quantity information, the queried remote sensing image set is processed into a gridded remote sensing image set to obtain a gridded remote sensing image set.

[0019] As an optional implementation, in a first aspect of the present invention, determining a preferred remote sensing image set based on the gridded region set and the gridded remote sensing image set includes:

[0020] S41, using the gridded region set and the remote sensing image coverage evaluation model, the gridded remote sensing image set is processed to obtain a remote sensing image coverage result set;

[0021] S42, using the remote sensing image coverage result set and the optimal image selection model, traverse the gridded remote sensing image set to obtain the optimal remote sensing image set;

[0022] S43, perform redundancy removal processing on the optimal remote sensing image set to obtain a preferred remote sensing image set.

[0023] As an optional implementation, in the first aspect of the present invention, the expression for the remote sensing image coverage evaluation model is:

[0024]

[0025] Among them, W i Let i be the coverage result of the queried remote sensing image, and S be the queried remote sensing image. i For the gridded remote sensing image subset of the queried remote sensing image, |S i |For the gridded remote sensing image subset S i The number of elements in the middle. n1 is the maximum value of the discrete elements contained in the gridded remote sensing image subset; n2 is the number of effective coverage grids; n3 is the number of overlapping coverage grids.

[0026] The expression for the optimal image selection model is:

[0027]

[0028] Among them, W i The coverage result of the queried remote sensing image; S i g is a gridded subset of the queried remote sensing image; i is the target grid; g The optimal image for the target grid.

[0029] As an optional implementation, in the first aspect of the present invention, the step of processing the gridded remote sensing image set using the gridded region set and the remote sensing image coverage evaluation model to obtain a remote sensing image coverage result set includes:

[0030] S411, Traverse the set of gridded regions to obtain the target grid set;

[0031] The target grid set includes several target grids;

[0032] S412, obtain all target grids in the target grid set, traverse each query remote sensing image in the query remote sensing image set, and count the number of effective covered grids n1 and the number of duplicate covered grids n2 of each query remote sensing image.

[0033] The effective coverage grid expression is:

[0034] g∈U and g∈S i

[0035] Where g is the target mesh, U is the set of meshed regions, and S i This refers to a gridded subset of the remote sensing images being queried;

[0036] The expression for the repeating coverage grid is:

[0037] And g∈(U∩S) i )

[0038] Where g is the target grid, U is the gridded region set, i and j are two random query remote sensing images in the query remote sensing image set, and S i and S j For i and j, the gridded remote sensing image subsets are used.

[0039] S413, using the remote sensing image coverage evaluation model, process each of the query remote sensing images, the number of effective coverage grids n1, and the number of duplicate coverage grids n2 to obtain the query remote sensing image coverage result;

[0040] S414, all the query remote sensing image coverage results are combined to obtain the remote sensing image coverage result set. Each of the query remote sensing images...

[0041] As an optional implementation, in the first aspect of the present invention, the step of processing the gridded remote sensing image set using the remote sensing image coverage result set and the optimal image selection model to obtain the optimal remote sensing image set includes:

[0042] S421, traverse each target grid in the target grid set, match the gridded remote sensing image set and the corresponding remote sensing image coverage result set corresponding to each target grid to obtain the target gridded remote sensing image set and the target remote sensing image coverage result set;

[0043] S422, Traverse the target remote sensing image coverage result set to obtain the highest target remote sensing image coverage result;

[0044] S423, using the coverage result of the highest target remote sensing image, match the target gridded remote sensing image set to obtain the optimal remote sensing image of the target grid;

[0045] S424, Using the target grid optimal remote sensing image, match the query remote sensing image set to obtain the target optimal query image;

[0046] S425, the optimal combination of all the target query images is used to obtain the optimal remote sensing image set.

[0047] As an optional implementation, in the first aspect of the present invention, the step of performing redundancy removal processing on the optimal remote sensing image set to obtain a preferred remote sensing image set includes:

[0048] S431, Traverse the optimal remote sensing image set to obtain the highest target remote sensing image coverage result set corresponding to the optimal remote sensing image set;

[0049] S432, using the highest target remote sensing image coverage result set, determine whether the highest target remote sensing image coverage result of the optimal remote sensing image set is 1, and obtain the redundant image judgment result.

[0050] When the redundant image judgment result is negative, delete the optimal remote sensing image in the optimal remote sensing image set whose highest target remote sensing image coverage result is not 1, obtain the first optimal remote sensing image set, and execute S433.

[0051] When the redundancy image determination result is yes, the optimal remote sensing image set is determined as the preferred remote sensing image set;

[0052] S433, using the optimized remote sensing image coverage evaluation model, the first optimal remote sensing image set is processed to obtain the optimized remote sensing image coverage result set;

[0053] S434, The optimized remote sensing image coverage result set is processed using the optimized optimal image selection model to obtain the second optimal remote sensing image set;

[0054] S435, update the second optimal remote sensing image set to the optimal remote sensing image set, and execute S432.

[0055] As an optional implementation, in the first aspect of the present invention, the expression for the optimized remote sensing image coverage evaluation model is:

[0056]

[0057] Among them, W y Let y be the optimized remote sensing image coverage result of the first optimal remote sensing image, and S be the query remote sensing image. y For the gridded remote sensing image subset of the first optimal remote sensing image, |S y | is the gridded remote sensing image subset S of the first optimal remote sensing image. y The number of elements in the middle. m1 is the maximum value of discrete elements contained in the gridded remote sensing image subset of the first optimal remote sensing image; m2 is the number of effective coverage grids; m2 is the number of overlapping coverage grids.

[0058] The expression for the optimized image selection model is:

[0059]

[0060] Among them, W y The optimized remote sensing image coverage result for the first optimal remote sensing image; S y The first optimal remote sensing image is a gridded subset of remote sensing images; g is the target grid; i g The optimal image of the target grid.

[0061] A second aspect of this invention discloses a remote sensing image optimization device based on spatially partitioned grids, used to perform some or all of the steps of the remote sensing image optimization method based on spatially partitioned grids disclosed in the first aspect of this invention. The device includes:

[0062] The acquisition module is used to acquire the query area and query the remote sensing image set;

[0063] The grid subdivision module is used to determine grid subdivision information based on the query area;

[0064] The gridding module is used to determine the gridded region set and the gridded remote sensing image set based on the grid subdivision information, the query region, and the query remote sensing image set.

[0065] The remote sensing image optimization module is used to determine a preferred remote sensing image set based on the gridded region set and the gridded remote sensing image set.

[0066] A third aspect of the present invention discloses another remote sensing image optimization device based on spatially partitioned grids, the device comprising:

[0067] Memory containing executable program code;

[0068] A processor coupled to the memory;

[0069] The processor calls the executable program code stored in the memory to execute some or all of the steps in the remote sensing image optimization method based on spatially partitioned grids disclosed in the first aspect of the present invention.

[0070] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0071] In this embodiment of the invention, spatially partitioned grids are used to achieve automatic optimization of remote sensing image data. While ensuring image coverage, the efficiency of image optimization is improved, which is a significant breakthrough in practicality and can be extended to related fields of remote sensing data management and application. Attached Figure Description

[0072] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0073] Figure 1 This is a flowchart illustrating a remote sensing image optimization method based on spatially partitioned grids disclosed in an embodiment of the present invention.

[0074] Figure 2 This is a schematic diagram of the structure of a remote sensing image optimization device based on spatially partitioned grids disclosed in an embodiment of the present invention;

[0075] Figure 3 This is a schematic diagram of another remote sensing image optimization device based on spatially partitioned grids disclosed in an embodiment of the present invention. Detailed Implementation

[0076] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0077] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0078] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0079] This invention discloses a method and apparatus for optimizing remote sensing imagery based on spatially partitioned grids. It utilizes spatially partitioned grids to automatically optimize remote sensing imagery data, improving optimization efficiency while ensuring image coverage. Detailed descriptions follow.

[0080] Example 1

[0081] Please see Figure 1 , Figure 1 This is a flowchart illustrating a remote sensing image optimization method based on spatially partitioned grids, as disclosed in an embodiment of the present invention. Figure 1 The described method for optimizing remote sensing images based on spatially partitioned grids is applicable to remote sensing image management and application systems, such as local servers or cloud servers used for remote sensing image management. This invention does not limit the application of this method. Figure 1 As shown, the remote sensing image optimization method based on spatially partitioned grids may include the following operations:

[0082] S1, retrieve the query area and the query remote sensing image set;

[0083] It should be noted that the query remote sensing image set includes several query remote sensing images; each query remote sensing image corresponds to a unique image ID number;

[0084] S2, determine the grid subdivision information based on the query area;

[0085] It should be noted that the mesh subdivision information includes the subdivision level and the number of meshes;

[0086] Taking Beijing as the query area as an example, searching the Natural Resources Satellite Remote Sensing Cloud Service Platform (http: / / www.sasclouds.com / chinese / normal / ) for Gaofen-6 remote sensing images taken between August 2022 and August 2023 with cloud cover less than 20% yielded 41 images (specific images omitted). Based on the query area, querying the "GB / T 40087-2021 Earth Spatial Grid Coding Rules" yielded a 13-level subdivision with 5578 grids.

[0087] S3, based on the grid subdivision information, the query area and the query remote sensing image set, the gridded area set and the gridded remote sensing image set are determined;

[0088] The gridded remote sensing image set includes several gridded remote sensing image subsets of the queried remote sensing image;

[0089] S4. Based on the gridded region set and the gridded remote sensing image set, a preferred remote sensing image set is determined.

[0090] As can be seen, the remote sensing image optimization method based on spatially partitioned grids described in the embodiments of the present invention utilizes spatially partitioned grids to achieve automatic optimization of remote sensing image data, thereby improving image optimization efficiency while ensuring image coverage.

[0091] Optionally, determining the gridded region set and the gridded remote sensing image set based on the grid subdivision information, the query region, and the query remote sensing image set includes:

[0092] S31, parse the mesh subdivision information to obtain subdivision hierarchy information and mesh quantity information;

[0093] S32, using the segmentation hierarchy information and the grid quantity information, the query region is gridded to obtain a gridded region set;

[0094] S33, using the subdivision hierarchy information and the grid quantity information, the queried remote sensing image set is processed into a gridded remote sensing image set to obtain a gridded remote sensing image set;

[0095] It should be noted that several segmentation methods can be used for remote sensing images, such as equidistant, equal latitude difference, regular hexahedron, and GeoSOT, and this invention does not impose any limitations.

[0096] As can be seen, the remote sensing image optimization method based on spatially partitioned grids described in the embodiments of the present invention utilizes spatially partitioned grids to achieve automatic optimization of remote sensing image data, thereby improving image optimization efficiency while ensuring image coverage.

[0097] Optionally, determining the preferred remote sensing image set based on the gridded region set and the gridded remote sensing image set includes:

[0098] S41, using the gridded region set and the remote sensing image coverage evaluation model, the gridded remote sensing image set is processed to obtain a remote sensing image coverage result set;

[0099] S42, using the remote sensing image coverage result set and the optimal image selection model, the gridded remote sensing image set is processed to obtain the optimal remote sensing image set;

[0100] S43, perform redundancy removal processing on the optimal remote sensing image set to obtain a preferred remote sensing image set.

[0101] As can be seen, the remote sensing image optimization method based on spatially partitioned grids described in the embodiments of the present invention automatically optimizes remote sensing image data by utilizing spatially partitioned grids. This improves image optimization efficiency while ensuring image coverage, thereby enhancing the automation level and work efficiency of remote sensing image management and application.

[0102] Optionally, the expression for the remote sensing image coverage evaluation model is:

[0103]

[0104] Among them, W i Let i be the coverage result of the queried remote sensing image, and S be the queried remote sensing image. i For the gridded remote sensing image subset of the queried remote sensing image, |S i |For the gridded remote sensing image subset S i The number of elements in the middle. n1 is the maximum value of the discrete elements contained in the gridded remote sensing image subset; n2 is the number of effective coverage grids; n3 is the number of overlapping coverage grids.

[0105] The higher the coverage score of the queried remote sensing image, the better the coverage effect of the queried remote sensing image on the selected grid area is considered. Taking Beijing as the query area, the coverage score of 41 remote sensing images is shown in the table below:

[0106]

[0107]

[0108] In step S42 above, the expression for the optimal image selection model is:

[0109]

[0110] Among them, W i The coverage result of the queried remote sensing image; S i g is a gridded subset of the queried remote sensing image; i is the target grid; g The optimal image for the target grid.

[0111] Based on the coverage score of the queried remote sensing images, the queried remote sensing images are filtered, and images with a coverage score of 0 are removed, resulting in a filtered result of 13 images.

[0112] Image ID First rating Image ID First rating 1420334295 0.131329 1120247362 0.426487 1420321084 0.087486 1120247086 0.306035 1420276908 0.164722 1120247087 0.0898927 1420267143 0.203332 1120243540 0.0887028 1420264903 0.0527977 1120243583 0.39806 1420264894 0.227935 1120243547 0.089267 1120252186 0.22712

[0113] As can be seen, the remote sensing image optimization method based on spatially partitioned grids described in the embodiments of the present invention automatically optimizes remote sensing image data by utilizing spatially partitioned grids. This improves image optimization efficiency while ensuring image coverage, thereby enhancing the automation level and work efficiency of remote sensing image management and application.

[0114] Optionally, the step of processing the gridded remote sensing image set using the gridded region set and the remote sensing image coverage evaluation model to obtain a remote sensing image coverage result set includes:

[0115] S411, Traverse the set of gridded regions to obtain the target grid set;

[0116] The target grid set includes several target grids;

[0117] S412, obtain all target grids in the target grid set, traverse each query remote sensing image in the query remote sensing image set, and count the number of effective covered grids n1 and the number of duplicate covered grids n2 of each query remote sensing image.

[0118] The effective coverage grid expression is:

[0119] g∈U and g∈S i

[0120] Where g is the target mesh, U is the set of meshed regions, and S i This refers to a gridded subset of the remote sensing images being queried;

[0121] The expression for the repeating coverage grid is:

[0122] And g∈(U∩S) i )

[0123] Where g is the target grid, U is the gridded region set, i and j are two random query remote sensing images in the query remote sensing image set, and S i and S j For i and j, the gridded remote sensing image subsets are used.

[0124] S413, using the remote sensing image coverage evaluation model, process each of the query remote sensing images, the number of effective coverage grids n1, and the number of duplicate coverage grids n2 to obtain the query remote sensing image coverage result;

[0125] S414, all the query remote sensing image coverage results are combined to obtain the remote sensing image coverage result set. Each of the query remote sensing images...

[0126] As can be seen, the remote sensing image optimization method based on spatially partitioned grids described in the embodiments of the present invention automatically optimizes remote sensing image data by utilizing spatially partitioned grids. This improves image optimization efficiency while ensuring image coverage, thereby enhancing the automation level and work efficiency of remote sensing image management and application.

[0127] Optionally, the step of processing the gridded remote sensing image set using the remote sensing image coverage result set and the optimal image selection model to obtain the optimal remote sensing image set includes:

[0128] S421, traverse each target grid in the target grid set, match the gridded remote sensing image set and the corresponding remote sensing image coverage result set corresponding to each target grid to obtain the target gridded remote sensing image set and the target remote sensing image coverage result set;

[0129] S422, Traverse the target remote sensing image coverage result set to obtain the highest target remote sensing image coverage result;

[0130] S423, using the coverage result of the highest target remote sensing image, match the target gridded remote sensing image set to obtain the optimal remote sensing image of the target grid;

[0131] S424, Using the target grid optimal remote sensing image, match the query remote sensing image set to obtain the target optimal query image;

[0132] S425, the optimal combination of all the target query images is used to obtain the optimal remote sensing image set.

[0133] As can be seen, the remote sensing image optimization method based on spatially partitioned grids described in the embodiments of the present invention automatically optimizes remote sensing image data by utilizing spatially partitioned grids. This improves image optimization efficiency while ensuring image coverage, thereby enhancing the automation level and work efficiency of remote sensing image management and application.

[0134] Optionally, the step of performing redundancy removal processing on the optimal remote sensing image set to obtain a preferred remote sensing image set includes:

[0135] S431, Traverse the optimal remote sensing image set to obtain the highest target remote sensing image coverage result set corresponding to the optimal remote sensing image set;

[0136] S432, using the highest target remote sensing image coverage result set, determine whether the highest target remote sensing image coverage result of the optimal remote sensing image set is 1, and obtain the redundant image judgment result.

[0137] When the redundant image judgment result is negative, delete the optimal remote sensing image in the optimal remote sensing image set whose highest target remote sensing image coverage result is not 1, obtain the first optimal remote sensing image set, and execute S433.

[0138] It should be noted that the first optimal remote sensing image set includes several first optimal remote sensing images;

[0139] When the redundancy image determination result is yes, the optimal remote sensing image set is determined as the preferred remote sensing image set;

[0140] S433, using the optimized remote sensing image coverage evaluation model, the first optimal remote sensing image set is processed to obtain the optimized remote sensing image coverage result set;

[0141] S434, The optimized remote sensing image coverage result set is processed using the optimized optimal image selection model to obtain the second optimal remote sensing image set;

[0142] S435, update the second optimal remote sensing image set to the optimal remote sensing image set, and execute S432.

[0143] In yet another optional embodiment, the expression for the optimized remote sensing image coverage evaluation model is:

[0144]

[0145] Among them, W y Let y be the optimized remote sensing image coverage result of the first optimal remote sensing image, and S be the query remote sensing image. y For the gridded remote sensing image subset of the first optimal remote sensing image, |S y| is the gridded remote sensing image subset S of the first optimal remote sensing image. y The number of elements in the middle. m1 is the maximum value of discrete elements contained in the gridded remote sensing image subset of the first optimal remote sensing image; m2 is the number of effective coverage grids; m2 is the number of overlapping coverage grids.

[0146] It should be noted that the effective coverage grid expression is:

[0147] g∈U and g∈S y

[0148] Where g is the target grid, U is the set of gridded regions, and S y This is a gridded subset of the first optimal remote sensing image;

[0149] It should be noted that the expression for the repeating coverage mesh is:

[0150] And g∈(U∩S) y )

[0151] Where g is the target grid, U is the set of gridded regions, x and y are two randomly selected query remote sensing images from the first optimal remote sensing image, and S x and S y The gridded remote sensing image subsets corresponding to the first optimal remote sensing image x and the first optimal remote sensing image y;

[0152] The expression for the optimized image selection model is:

[0153]

[0154] Among them, W y The optimized remote sensing image coverage result for the first optimal remote sensing image; S y The first optimal remote sensing image is a gridded subset of remote sensing images; g is the target grid; i g The optimal image for the target grid.

[0155] Based on the results of the screening, it can be seen that the filtered remote sensing image set has redundancy, that is, the score is not 1, indicating that there are still multiple images covering the same area. After the first optimization of the filtered remote sensing image set, the results are shown in the table below:

[0156] Image ID Second rating Image ID Second rating 1420334295 0.131329 1120247362 0.426487 1420321084 1 1120247086 1 1420276908 1 1120247087 1 1420267143 0.203332 1120243540 0.0887028 1420264903 1 1120243583 0.39806 1420264894 0.227935 1120243547 1 1120252186 0.22712

[0157] Based on the scoring results after the first optimization, there are still scores that are not 1, indicating that multiple images still cover the same area. After a second optimization of the remote sensing image set after the first optimization, the results are shown in the table below:

[0158] Image ID Second rating Image ID Second rating 1420321084 1 1120247086 1 1420276908 1 1120247087 1 1420264903 1 1120243547 1 1120247362 0.426487

[0159] Based on the scoring results after the second optimization, the scores are still not 1, indicating that multiple images still cover the same area. After a third optimization of the remote sensing image set after the second optimization, the results are shown in the table below:

[0160] Image ID Third score Image ID Third score 1420321084 1 1120247086 1 1420276908 1 1120247087 1 1420264903 1 1120243547 1 1120247362 1

[0161] Based on the results of the third optimization scoring, all 7 images received a score of 1, indicating that each image has a unique coverage area and cannot be removed again, thus obtaining the optimized remote sensing image set.

[0162] As can be seen, the remote sensing image optimization method based on spatially partitioned grids described in the embodiments of the present invention automatically optimizes remote sensing image data by utilizing spatially partitioned grids. This improves image optimization efficiency while ensuring image coverage, thereby enhancing the automation level and work efficiency of remote sensing image management and application.

[0163] Example 2

[0164] Please see Figure 2 , Figure 2 This is a schematic diagram of a remote sensing image optimization device based on spatially partitioned grids, as disclosed in an embodiment of the present invention. Figure 2 The described apparatus can be applied to remote sensing image management and application systems, such as local servers or cloud servers for remote sensing image management, and the embodiments of the present invention are not limited thereto. Figure 2 As shown, the device may include:

[0165] The acquisition module 201 is used to acquire the query area and the query remote sensing image set; the query remote sensing image set includes several query remote sensing images; each query remote sensing image corresponds to a unique image ID number;

[0166] The grid subdivision module 202 is used to determine grid subdivision information based on the query area;

[0167] The gridding module 203 is used to determine the gridded region set and the gridded remote sensing image set based on the grid subdivision information, the query region, and the query remote sensing image set;

[0168] The remote sensing image optimization module 204 is used to determine a preferred remote sensing image set based on the gridded region set and the gridded remote sensing image set.

[0169] It is evident that implementation Figure 2The described remote sensing image optimization device based on spatial grid partitioning utilizes spatial grid partitioning to automatically optimize remote sensing image data. While ensuring image coverage, it improves image optimization efficiency, thereby enhancing the automation level and work efficiency of remote sensing image management and application.

[0170] In another alternative embodiment, such as Figure 2 As shown, the meshing module 203 is used to determine mesh subdivision information based on the query area, including:

[0171] According to the "GB / T 40087-2021 Earth Spatial Grid Coding Rules" and actual needs, the target area and candidate images are divided by specifying the subdivision level to obtain subdivision level information and grid quantity information.

[0172] In another alternative embodiment, such as Figure 2 As shown, the gridding module 203 is used to determine a gridded region set and a gridded remote sensing image set based on the grid subdivision information, the query region, and the query remote sensing image set, including:

[0173] The mesh subdivision information is parsed to obtain subdivision hierarchy information and mesh quantity information;

[0174] Using the hierarchical information and the number of grids, the query region is gridded to obtain a set of gridded regions;

[0175] Using the hierarchical information and the number of grids, the queried remote sensing image set is gridded to obtain a gridded remote sensing image set.

[0176] It is evident that implementation Figure 2 The described remote sensing image optimization device based on spatial grid partitioning utilizes spatial grid partitioning to automatically optimize remote sensing image data. While ensuring image coverage, it improves image optimization efficiency, thereby enhancing the automation level and work efficiency of remote sensing image management and application.

[0177] In another alternative embodiment, such as Figure 2 As shown, the remote sensing image optimization module 204 is used to determine a preferred remote sensing image set based on the gridded region set and the gridded remote sensing image set, including:

[0178] Using the gridded region set and the remote sensing image coverage evaluation model, the gridded remote sensing image set is processed to obtain a remote sensing image coverage result set;

[0179] Using the remote sensing image coverage result set and the optimal image selection model, the gridded remote sensing image set is traversed to obtain the optimal remote sensing image set;

[0180] The optimal remote sensing image set is subjected to redundancy removal processing to obtain a preferred remote sensing image set.

[0181] It is evident that implementation Figure 2 The described remote sensing image optimization method based on spatial grid partitioning utilizes spatial grid partitioning to automatically optimize remote sensing image data. While ensuring image coverage, it improves image optimization efficiency, thereby enhancing the automation level and work efficiency of remote sensing image management and application.

[0182] In another alternative embodiment, such as Figure 2 As shown, the expression for the remote sensing image coverage evaluation model of the remote sensing image optimization module 204 is:

[0183]

[0184] Among them, W i Let i be the coverage result of the queried remote sensing image, and S be the queried remote sensing image. i For the gridded remote sensing image subset of the queried remote sensing image, |S i |For the gridded remote sensing image set S i The number of elements in the middle. n1 is the maximum value of the discrete elements contained in the gridded remote sensing image set; n2 is the number of effective coverage grids; n3 is the number of overlapping coverage grids.

[0185] In another alternative embodiment, such as Figure 2 As shown, the expression for the optimal image selection model in the remote sensing image optimization module 204 is:

[0186]

[0187] Among them, W i The coverage result of the queried remote sensing image; S i i is a gridded subset of the queried remote sensing image; g is the optimal image of the target grid; g is the target grid.

[0188] It is evident that implementation Figure 2 The described remote sensing image optimization method based on spatial grid partitioning utilizes spatial grid partitioning to automatically optimize remote sensing image data. While ensuring image coverage, it improves image optimization efficiency, thereby enhancing the automation level and work efficiency of remote sensing image management and application.

[0189] In another alternative embodiment, such as Figure 2 As shown, the remote sensing image optimization module 204 uses the gridded region set and the remote sensing image coverage evaluation model to process the gridded remote sensing image set, obtaining a remote sensing image coverage result set, including:

[0190] Traverse the set of gridded regions to obtain the target grid set;

[0191] The target grid set includes several target grids;

[0192] Obtain all target grids in the target grid set, traverse each query remote sensing image in the query remote sensing image set, and count the number of effective covered grids n1 and the number of duplicate covered grids n2 for each query remote sensing image.

[0193] The effective coverage grid expression is:

[0194] g∈U and g∈S i

[0195] Where g is the target grid, U is the set of gridded regions, and S i This refers to a gridded subset of the remote sensing images being queried;

[0196] The expression for the repeating coverage grid is:

[0197] And g∈(U∩S) i )

[0198] Where g is the target grid, U is the set of gridded regions, i and j are two random remote sensing images from the query remote sensing image set, and S i and S j For i and j, the gridded remote sensing image subsets are used.

[0199] Using the remote sensing image coverage evaluation model, each query remote sensing image, the number of effective coverage grids n1, and the number of duplicate coverage grids n2 are processed to obtain the query remote sensing image coverage result;

[0200] All the query results of remote sensing image coverage are combined to obtain the remote sensing image coverage result set.

[0201] It is evident that implementation Figure 2 The described remote sensing image optimization method based on spatial grid partitioning utilizes spatial grid partitioning to automatically optimize remote sensing image data. While ensuring image coverage, it improves image optimization efficiency, thereby enhancing the automation level and work efficiency of remote sensing image management and application.

[0202] In another alternative embodiment, such as Figure 2 As shown, the remote sensing image optimization module 204 uses the remote sensing image coverage result set and the optimal image selection model to process the gridded remote sensing image set to obtain the optimal remote sensing image set, including:

[0203] Traverse each target grid in the target grid set, match the gridded remote sensing image set and the corresponding remote sensing image coverage result set for each target grid to obtain the target gridded remote sensing image set and the target remote sensing image coverage result set;

[0204] Traverse the target remote sensing image coverage result set to obtain the highest target remote sensing image coverage result;

[0205] Using the highest target remote sensing image coverage result, the target gridded remote sensing image set is matched to obtain the target grid optimal remote sensing image;

[0206] Using the target grid's optimal remote sensing image, match the query remote sensing image set to obtain the target's optimal query image;

[0207] The optimal combination of all the target query images yields the optimal remote sensing image set.

[0208] It is evident that implementation Figure 2 The described remote sensing image optimization method based on spatial grid partitioning utilizes spatial grid partitioning to automatically optimize remote sensing image data. While ensuring image coverage, it improves image optimization efficiency, thereby enhancing the automation level and work efficiency of remote sensing image management and application.

[0209] In another alternative embodiment, such as Figure 2 As shown, the remote sensing image optimization module 204 performs redundancy removal processing on the optimal remote sensing image set to obtain a preferred remote sensing image set, including:

[0210] S431, Traverse the optimal remote sensing image set to obtain the highest target remote sensing image coverage result set corresponding to the optimal remote sensing image set;

[0211] S432, using the highest target remote sensing image coverage result set, determine whether the highest target remote sensing image coverage result of the optimal remote sensing image set is 1, and obtain the redundant image judgment result.

[0212] When the redundant image judgment result is negative, delete the optimal remote sensing image in the optimal remote sensing image set whose highest target remote sensing image coverage result is not 1, obtain the first optimal remote sensing image set, and execute S433.

[0213] When the redundancy image determination result is yes, the optimal remote sensing image set is determined as the preferred remote sensing image set;

[0214] S433, using the optimized remote sensing image coverage evaluation model, the first optimal remote sensing image set is processed to obtain the optimized remote sensing image coverage result set;

[0215] S434, The optimized remote sensing image coverage result set is processed using the optimized optimal image selection model to obtain the second optimal remote sensing image set;

[0216] S435, update the second optimal remote sensing image set to the optimal remote sensing image set, and execute S432.

[0217] In another alternative embodiment, such as Figure 2 As shown, the expression for the optimized remote sensing image coverage evaluation model described in the remote sensing image optimization module 204 is:

[0218]

[0219] Among them, W y Let y be the optimized remote sensing image coverage result of the first optimal remote sensing image, and S be the query remote sensing image. y For the gridded remote sensing image subset of the first optimal remote sensing image, |S y | is the gridded remote sensing image subset S of the first optimal remote sensing image. y The number of elements in the middle. m1 is the maximum value of discrete elements contained in the gridded remote sensing image subset of the first optimal remote sensing image; m2 is the number of effective coverage grids; m2 is the number of overlapping coverage grids.

[0220] In another alternative embodiment, such as Figure 2 As shown, the effective coverage grid expression of the remote sensing image optimization module 204 is:

[0221] g∈U and g∈S y

[0222] Where g is the target grid, U is the set of gridded regions, and S y This is a gridded subset of the first optimal remote sensing image;

[0223] In another alternative embodiment, such as Figure 2 As shown, the expression for the repeated coverage grid in the remote sensing image optimization module 204 is:

[0224] And g∈(U∩S) y )

[0225] Where g is the target grid, U is the set of gridded regions, x and y are two randomly selected query remote sensing images from the first optimal remote sensing image, and S x and S y The gridded remote sensing image subsets corresponding to the first optimal remote sensing image x and the first optimal remote sensing image y;

[0226] The expression for the optimized image selection model is:

[0227]

[0228] Among them, W y The optimized remote sensing image coverage result for the first optimal remote sensing image; S y The first optimal remote sensing image is a gridded subset of remote sensing images; g is the target grid; i g The optimal image for the target grid.

[0229] It is evident that implementation Figure 2 The described remote sensing image optimization method based on spatial grid partitioning utilizes spatial grid partitioning to automatically optimize remote sensing image data. While ensuring image coverage, it improves image optimization efficiency, thereby enhancing the automation level and work efficiency of remote sensing image management and application.

[0230] Example 3

[0231] Please see Figure 3 , Figure 3 This is a schematic diagram of another remote sensing image optimization device based on spatially partitioned grids disclosed in an embodiment of the present invention. Figure 3 The described apparatus can be applied to remote sensing image management and application systems based on spatially partitioned grids, such as local servers or cloud servers for remote sensing image management. This invention is not limited to these specific applications. Figure 3 As shown, the device may include:

[0232] Memory 301 storing executable program code;

[0233] Processor 302 coupled to memory 301;

[0234] The processor 302 calls the executable program code stored in the memory 301 to execute the steps in the outbound order splitting method described in Embodiment 1.

[0235] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0236] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0237] Finally, it should be noted that the remote sensing image optimization method and apparatus based on spatial grid partitioning disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for optimal selection of remote sensing images based on spatially partitioned grid, characterized in that, The method comprises: S1, acquiring a query area and a query remote sensing image set; the query remote sensing image set comprises a plurality of query remote sensing images; S2, determining grid subdivision information according to the query area; S3, determining a grid region set and a grid remote sensing image set based on the grid subdivision information, the query area and the query remote sensing image set; The grid remote sensing image set comprises a plurality of grid remote sensing image subsets of the query remote sensing images; S4, determining an optimal remote sensing image set based on the grid region set and the grid remote sensing image set; Wherein, the determination of the optimal remote sensing image set based on the grid region set and the grid remote sensing image set comprises: S41, processing the grid remote sensing image set by using the grid region set and a remote sensing image coverage evaluation model to obtain a remote sensing image coverage result set; S42, processing the grid remote sensing image set by using the remote sensing image coverage result set and an optimal image screening model to obtain an optimal remote sensing image set; S43, performing a redundancy removal processing on the optimal remote sensing image set to obtain the optimal remote sensing image set; Wherein, the remote sensing image coverage evaluation model expression is: wherein, a number of the query remote sensing image coverage results, i the query remote sensing image, a subset of the query remote sensing image, the subset of the query remote sensing image a number of elements in the subset of the query remote sensing image, a maximum value of discrete elements contained in the subset of the query remote sensing image; a number of effective coverage grids; a number of repeated coverage grids; The optimal image screening model expression is: wherein, a coverage result for the query remote sensing image; a gridded remote sensing image subset for the query remote sensing image; g is a target grid; an optimal image for the target grid; Wherein, the processing of the grid remote sensing image set by using the grid region set and the remote sensing image coverage evaluation model to obtain the remote sensing image coverage result set comprises: S411, traversing the grid region set to obtain a target grid set; The target grid set comprises a plurality of target grids; S412, acquire all target grids in the target grid set, traverse each query remote sensing image in the query remote sensing image set, and count the number of effective coverage grids of each query remote sensing image and the number of repetitive coverage grids of each of said query remote sensing images ; The effective coverage grid expression is: wherein g is the target grid, U is the set of gridded regions, a subset of gridded remote sensing images for the query remote sensing image; The repeated coverage grid expression is: wherein g is the target grid, U is the set of gridded regions, i and j are two random remote sensing images in the set of query remote sensing images, and is the subset of gridded remote sensing images corresponding to i and j. S413, processing each of the query remote sensing images and the number of the effective coverage grids by using the remote sensing image coverage evaluation model to obtain a query remote sensing image coverage result. the number of repeated coverage grids ​ S414, combining all the query remote sensing image coverage results to obtain the remote sensing image coverage result set.

2. The method of claim 1, wherein, The determination of the grid region set and the grid remote sensing image set based on the grid subdivision information, the query area and the query remote sensing image set comprises: S31, analyzing the grid subdivision information to obtain subdivision level information and grid quantity information; S32, performing a grid processing on the query area by using the subdivision level information and the grid quantity information to obtain the grid region set; S33, performing a grid processing on the query remote sensing image set by using the subdivision level information and the grid quantity information to obtain the grid remote sensing image set.

3. The method of claim 1, wherein the method further comprises: The processing of the grid remote sensing image set by using the remote sensing image coverage result set and the optimal image screening model to obtain the optimal remote sensing image set comprises: S421, traversing each target grid in the target grid set, matching the grid remote sensing image set corresponding to each target grid and the remote sensing image coverage result set corresponding to each target grid to obtain a target grid remote sensing image set and a target remote sensing image coverage result set; S422, traversing the target remote sensing image coverage result set to obtain a highest target remote sensing image coverage result; S423, matching the target grid remote sensing image set by using the highest target remote sensing image coverage result to obtain an optimal remote sensing image of the target grid. S424, matching the query remote sensing image set by using the target grid optimal remote sensing image to obtain a target optimal query image; S425, combining all the target optimal query images to obtain an optimal remote sensing image set.

4. The method of claim 1, wherein the method further comprises: The de-redundancy processing of the optimal remote sensing image set to obtain an optimal remote sensing image set comprises: S431, traversing the optimal remote sensing image set to obtain a highest target remote sensing image coverage result set corresponding to the optimal remote sensing image set; S432, using the highest target remote sensing image coverage result set to determine whether the highest target remote sensing image coverage result of the optimal remote sensing image set is 1 to obtain a redundant image judgment result; When the redundant image judgment result is no, deleting the optimal remote sensing image in the optimal remote sensing image set whose highest target remote sensing image coverage result is not 1 to obtain a first optimal remote sensing image set, and executing S433; When the redundant image judgment result is yes, determining that the optimal remote sensing image set is an optimal remote sensing image set; S433, using an optimized remote sensing image coverage evaluation model to process the first optimal remote sensing image set to obtain an optimized remote sensing image coverage result set; S434, using an optimized optimal image screening model to process the optimized remote sensing image coverage result set to obtain a second optimal remote sensing image set; S435, updating the second optimal remote sensing image set to the optimal remote sensing image set, and executing S432.

5. The method of claim 4, wherein the method further comprises: The expression of the optimized remote sensing image coverage evaluation model is: wherein, is an optimized remote sensing image coverage result of the first optimal remote sensing image, y is the query remote sensing image, is a gridded remote sensing image subset of the first optimal remote sensing image, is the gridded remote sensing image subset is the number of elements in the subset, is the maximum value of discrete elements contained in the gridded remote sensing image subset of the first optimal remote sensing image; is the number of effectively covered grids; is the number of repeatedly covered grids. The expression of the optimized optimal image screening model is: wherein, is an optimized remote sensing image coverage result of the first optimal remote sensing image; is a gridded remote sensing image subset of the first optimal remote sensing image; g is a target grid; is a sufficient image of the target grid.

6. A spatially partitioned grid-based remote sensing image selection device for performing the spatially partitioned grid-based remote sensing image selection method according to any one of claims 1 to 5, characterized in that, The device comprises: An acquisition module is configured to acquire a query region and a query remote sensing image set; A grid division module is configured to determine grid division information according to the query region; A gridding module is configured to determine a gridded region set and a gridded remote sensing image set based on the grid division information, the query region and the query remote sensing image set; A remote sensing image optimization module is configured to determine an optimal remote sensing image set based on the gridded region set and the gridded remote sensing image set.

7. A remote sensing image optimization device based on spatially dissected grid, characterized in that, The device comprises: A memory storing executable program codes; A processor coupled with the memory; The processor invokes the executable program codes stored in the memory to execute the spatially divided grid-based remote sensing image optimization method according to any one of claims 1-5.

Images