A multi-scale grid remote sensing data partitioning method and a remote sensing data management method

Abstract

The application discloses a multi-scale grid remote sensing data profiling method and a remote sensing data management method.The profiling method comprises the following steps: obtaining original remote sensing data from a platform front-end data source and obtaining spatial position information corresponding to the original remote sensing data from a platform general control computer; performing grid division on the spatial position information corresponding to the original remote sensing data according to a fixed hierarchical grid to obtain a plurality of spatial grids; performing multi-scale grid coding on all the spatial grids to obtain an initial spatial grid coding set; determining a grid coding division mode according to preset grid hierarchical requirements and preset grid quantity requirements; and re-dividing all the spatial grid codes in the initial spatial grid coding set according to the grid coding division mode to obtain a target spatial grid coding set.The application realizes the representation of remote sensing data under multi-scale conditions, improves the representation precision of remote sensing data, and improves the efficiency, accuracy and precision of remote sensing data indexing and querying.

Description

Technical Field

[0001] This invention belongs to the field of remote sensing data processing technology, specifically relating to a multi-scale grid remote sensing data partitioning method and a remote sensing data management method. Background Technology

[0002] With the continuous development of Earth observation technologies and the constant improvement of spatial information acquisition capabilities, the amount of remote sensing data obtained is extremely abundant. This abundance of data is accompanied by a demand for rapid data services. In spatial information processing, the storage and management of remote sensing data is of paramount importance.

[0003] In spatial information processing, remote sensing data is always gridded as needed to form gridded data for storage and management. A core step in gridded satellite data storage and management methods is using spatial grid sets to describe satellite remote sensing data, also known as gridded partitioning of remote sensing data. This is a key technology and prerequisite for gridded satellite storage; gridded storage and management can only proceed after the satellite remote sensing data has been gridded. Currently, a common method for gridded remote sensing data is to use fixed-scale grids (temporal or spatial). This method is characterized by its simplicity, low computational requirements, and the use of spatial networks of the same scale to represent the remote sensing data.

[0004] However, current remote sensing data partitioning methods cannot adapt to remote sensing data of different volumes, and the representation accuracy of remote sensing data is low. That is, the grid coverage area is not close to the spatial data coverage area, which leads to low efficiency and accuracy of data indexing and querying. Summary of the Invention

[0005] To address the aforementioned problems in existing technologies, this invention provides a multi-scale grid remote sensing data partitioning method and a remote sensing data management method. The technical problem to be solved by this invention is achieved through the following technical solution:

[0006] In a first aspect, embodiments of the present invention provide a multi-scale grid remote sensing data partitioning method, including:

[0007] The platform obtains raw remote sensing data from the front-end data source and obtains the spatial location information corresponding to the raw remote sensing data from the platform's central control computer.

[0008] The spatial location information corresponding to the original remote sensing data is divided into several spatial grids according to a fixed-level grid; wherein, the fixed-level grid is determined according to preset grid level requirements and preset grid quantity requirements;

[0009] Multi-scale grid coding is performed on all spatial grids to obtain an initial spatial grid coding set;

[0010] The grid coding method is determined according to the preset grid level requirements and the preset grid quantity requirements. The target spatial grid coding set is obtained by re-dividing all spatial grid codes in the initial spatial grid coding set according to the grid coding method.

[0011] In one embodiment of the present invention, determining the fixed-level grid according to preset grid level requirements and preset grid quantity requirements includes:

[0012] When only the preset grid level requirement is met, the fixed-level grid is determined to be the minimum-scale grid;

[0013] When both the preset grid level requirement and the preset grid number are met, it is determined that the fixed-level grid is not the minimum-scale grid.

[0014] In one embodiment of the present invention, determining the grid coding division method according to the preset grid level requirements and the preset grid quantity requirements includes:

[0015] When only the preset grid level requirement is met, the grid coding partitioning method is determined to be an aggregation partitioning method; wherein, the aggregation partitioning method is a method of re-partitioning all spatial grid codes in the initial spatial grid coding set;

[0016] When both the preset grid level requirement and the preset grid quantity are met, the grid coding division method is determined to be either a classification division method or an aggregation division method; wherein, the classification division method is a method of re-dividing the grid topology relationship corresponding to all spatial grid codes in the initial spatial grid coding set.

[0017] In one embodiment of the present invention, when the determined grid coding partitioning method is an aggregation partitioning method, all spatial grid codes in the initial spatial grid coding set are re-partitioned to obtain a target spatial grid coding set, including:

[0018] All spatial grid codes at level N in the initial spatial grid coding set are aggregated into several spatial grid codes at level N-1 according to a preset number of grids, until it is impossible to perform aggregation and partitioning operations at each level according to the preset number of grids. The target spatial grid coding set is obtained based on all spatial grid codes after aggregation and partitioning at each level; where N is the maximum number of grid levels corresponding to the selected grid partitioning and coding method; the preset number of grids is determined according to the selected grid partitioning and coding method.

[0019] In one embodiment of the present invention, the selected mesh partitioning and encoding method adopts a multi-scale encoding method based on GeoSOT mesh partitioning.

[0020] In one embodiment of the present invention, the multi-scale encoding method based on GeoSOT mesh partitioning encodes an inverted quadtree, and the predetermined number of meshes is 4.

[0021] In one embodiment of the present invention, when the determined grid coding partitioning method is a classification partitioning method, the spatial grid codes in the initial spatial grid coding set are re-partitioned to obtain a target spatial grid coding set, including:

[0022] Determine the grid topology relationships corresponding to all spatial grid codes in the initial spatial grid code set; the grid topology relationships include inclusion relationships and intersection relationships.

[0023] All spatial grid codes that are intersecting are further subdivided into smaller grid codes until there are no more spatial grid codes that are intersecting or the preset grid level requirements and preset grid number requirements are met. The target spatial grid code set is obtained based on all spatial grid codes that are inclusive.

[0024] Secondly, embodiments of the present invention provide a multi-scale grid remote sensing data management method, including:

[0025] The spatial location information corresponding to all data blocks is partitioned using any of the multi-scale grid remote sensing data partitioning methods described above to obtain a target spatial coding set; wherein, the data blocks are in the form of pre-stored original remote sensing data;

[0026] Establish an index table corresponding to the target spatial encoding set and the data block; wherein the target spatial encoding set includes several spatial encodings, and each spatial encoding and the data block satisfy a preset mapping relationship.

[0027] In one embodiment of the present invention, it further includes:

[0028] The time information corresponding to all the data blocks is encoded to obtain a time code;

[0029] The step of establishing an index table corresponding to the target spatial encoding set and the data block includes:

[0030] An index table is established corresponding to the time code, the target spatial code set, and the data block; wherein, there is a one-to-one mapping relationship between the time code and the data block; the target spatial code set includes several spatial codes, and each spatial code satisfies a preset mapping relationship with the data block.

[0031] In one embodiment of the present invention, the preset mapping relationship between each spatial code and data block includes at least one of the following: a one-to-one mapping between spatial codes and data blocks, a many-to-one mapping between spatial codes and data blocks, and a one-to-many mapping between spatial codes and data blocks.

[0032] The beneficial effects of this invention are:

[0033] The multi-scale grid remote sensing data partitioning method proposed in this invention breaks away from the traditional representation of remote sensing data. It no longer uses all the spatial grid codes after the initial partitioning to represent remote sensing data, and can adapt to different amounts of remote sensing data. According to user needs, different grid coding partitioning methods are used to re-partition all the spatial grid codes after the initial partitioning, realizing the representation of remote sensing data at multiple scales, improving the representation accuracy of remote sensing data, and thus improving the efficiency and accuracy of remote sensing data indexing and querying.

[0034] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0035] Figure 1 This is a flowchart illustrating a multi-scale grid remote sensing data partitioning method provided in an embodiment of the present invention;

[0036] Figure 2 This is a schematic diagram of spatial coding corresponding to an example of raw remote sensing data provided in an embodiment of the present invention;

[0037] Figure 3 This is a schematic diagram of the spatial coding of the original remote sensing data after the initial spatial grid coding set is aggregated and divided, as provided in an embodiment of the present invention.

[0038] Figure 4 This is a schematic diagram of the spatial coding of the original remote sensing data after classifying and dividing the initial spatial grid coding set, provided by an embodiment of the present invention.

[0039] Figure 5 This is a schematic diagram of the spatial coding of the original remote sensing data after detailed classification and division of the initial spatial grid coding set, provided by an embodiment of the present invention.

[0040] Figure 6 This is a flowchart illustrating a multi-scale grid remote sensing data management method provided in an embodiment of the present invention;

[0041] Figure 7 a~ Figure 7 b is a schematic diagram illustrating the mapping relationship between spatial coding and data blocks in the multi-scale grid remote sensing data management method provided in this embodiment of the invention;

[0042] Figure 8This is a flowchart illustrating another multi-scale grid remote sensing data management method provided in an embodiment of the present invention;

[0043] Figure 9 a to 9b are schematic diagrams illustrating the mapping relationship between spatial encoding and data blocks in another management method based on grid collection objects provided in this embodiment of the invention. Detailed Implementation

[0044] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0045] Example 1

[0046] Firstly, to improve the representation accuracy of gridded remote sensing data, please refer to... Figure 1 This invention provides a multi-scale grid remote sensing data partitioning method, comprising the following steps:

[0047] S101. Obtain raw remote sensing data from the platform's front-end data source and obtain the spatial location information corresponding to the raw remote sensing data from the platform's central control computer.

[0048] Specifically, embodiments of the present invention acquire raw remote sensing data through a platform front-end data source. For example, the platform front-end data source can be an onboard payload camera that acquires raw remote sensing data and stores it on the satellite. However, the acquisition method is not limited to the onboard payload camera; raw remote sensing data can also be acquired through drones, aerospace equipment, etc. Correspondingly, auxiliary data is acquired through the platform's central control computer. For example, auxiliary information can be acquired through a satellite-based computer. Auxiliary data may include the satellite's attitude, orbit, time, etc. Based on the satellite's orbit, attitude, time, etc., combined with the optical path design of the sensors in the imaging subsystem, the positioning processing of the remote sensing data is completed to obtain the spatial location information corresponding to the remote sensing data. The auxiliary data varies depending on the scenario and is determined according to the actual situation of the scenario. There are many solutions for the onboard imaging subsystem, such as optical remote sensing imaging systems, radar remote sensing imaging systems, and infrared remote sensing imaging systems, and the specific imaging method is not limited.

[0049] S102. The spatial location information corresponding to the original remote sensing data is divided into several spatial grids according to a fixed-level grid; wherein, the fixed-level grid is determined according to the preset grid level requirements and the preset grid quantity requirements.

[0050] Specifically, in this embodiment of the invention, the spatial location information corresponding to the original remote sensing data is divided into several spatial grids, for example... Figure 2The grids, numbered 1 to 13, represent the raw remote sensing data. During grid partitioning, different satellites require different grid partitioning and encoding methods. Common methods include GeoSOT, GeoHash, and GJB8896-2017, but are not limited to these three. Therefore, users will specify different requirements based on the satellites from which they acquire the raw remote sensing data, such as grid level requirements and grid quantity requirements. Different requirements correspond to different fixed-level grid requirements. For example, the multi-scale encoding method based on GeoSOT grid partitioning requires dividing the grid to obtain a fixed-level grid at level 31. This level 31 fixed-level grid is a single-scale grid, and further processing of the remote sensing data is then performed based on this level 31 fixed-level grid.

[0051] S103. Perform multi-scale grid coding on all spatial grids to obtain the initial spatial grid coding set.

[0052] Specifically, in this embodiment of the invention, all spatial grids are encoded using a multi-scale spatial grid encoding method to obtain an initial spatial grid encoding set. For example, a multi-scale encoding method based on GeoSOT grid partitioning yields, as shown below. Figure 3 The 31st level of grid coding in the middle, and the initial spatial grid coding set is formed by all grid codes at this level. Among them, Figure 3 The numbers involved are corresponding Figure 2 Spatial coding of the medium grid.

[0053] The main problem with current remote sensing data partitioning methods is the low accuracy of remote sensing data representation, meaning the grid coverage area is not very close to the spatial data coverage area, resulting in low efficiency and accuracy in remote sensing data indexing and querying. Figure 2 As shown, the representation precision of traditional remote sensing data can be expressed as:

[0054]

[0055] As can be seen, traditional remote sensing data is directly represented by all the initial spatial grids that intersect or contain the remote sensing data in terms of topological relationships. The standard for judging the representation accuracy of the original remote sensing data is: the closer the "accuracy" value is to 1, the higher the representation accuracy of the remote sensing data, that is, the more accurately the original remote sensing data is represented. Among them, all the grids used for traditional remote sensing data representation are of the same scale.

[0056] In grid-based data organization and indexing, the number of grids used to describe spatial data is crucial to its efficiency. Reducing the number of grids can improve data organization and indexing efficiency, but it can also reduce the accuracy of remote sensing data representation and spatial query accuracy based on grid-based data models. Therefore, this invention proposes a multi-scale grid remote sensing data partitioning method to achieve multi-scale grid filling, improve the accuracy of multi-scale grid filling results, and ensure improved accuracy of spatial query results without compromising data organization and indexing efficiency.

[0057] Next, we will introduce in detail the multi-scale remote sensing data partitioning method based on the initial spatial grid coding set proposed in the embodiments of the present invention.

[0058] S104. Determine the grid coding division method according to the preset grid level requirements and preset grid quantity requirements, and re-divide all spatial grid codes in the initial spatial grid coding set according to the grid coding division method to obtain the target spatial grid coding set.

[0059] Specifically, this invention proposes two multi-scale grid remote sensing data partitioning methods. Different multi-scale grid remote sensing data partitioning methods can be adopted for different application scenarios. The process begins with a fixed-level grid as the initial state, and grid partitioning is performed based on this initial state. This invention also provides an optional scheme, determining the fixed-level grid as the initial state according to preset grid level requirements and preset grid quantity requirements, specifically including:

[0060] When only the preset grid level requirement is met, the fixed-level grid is determined as the minimum-scale grid. For example, if a user requests that the preset grid level requirement be 20-31, then the 31st level grid is determined as the fixed-level grid. In this case, the 31st level grid corresponds to the minimum-scale grid, i.e., a single-scale grid, and cannot be further subdivided into smaller scales. For instance, in grid partitioning, the original remote sensing data may include grid data at multiple scales, such as township, county, city, and province. In such scenarios, the grid data corresponding to the township is selected as the minimum-scale grid for subsequent grid partitioning.

[0061] When both the preset grid level requirements and the preset grid number are met, the fixed-level grid is determined not to be the smallest-scale grid, i.e., it is a multi-scale grid. For example, if a user requires a preset grid level of 20-31 and a maximum of 10 grids, then the grids corresponding to all levels except level 31 are determined to be fixed-level grids. That is, grids outside level 31 are not the smallest-scale grids, and grids outside level 31 are multi-scale grids. For instance, the original remote sensing data in the grid partitioning may include grid data at multiple scales, such as township, county, city, and province. In such scenarios, grid data corresponding to any of the following scales (excluding township) are selected for subsequent grid partitioning: county, city, and province.

[0062] Furthermore, in this embodiment of the invention, the fixed-level grid is selected as either a minimum-scale grid or a non-minimum-scale grid, corresponding to different grid coding methods. This embodiment of the invention provides an optional scheme, determining the grid coding method based on preset grid level requirements and preset grid quantity requirements, specifically including:

[0063] When only the preset grid level requirement is met, i.e., the fixed-level grid is the minimum scale grid, the grid coding method is determined to be the aggregation partitioning method; wherein, the aggregation partitioning method is a method of re-partitioning all spatial grid codes in the initial spatial grid coding set. An embodiment of the present invention provides an optional solution: when the determined grid coding method is the aggregation partitioning method, the target spatial grid coding set is obtained by re-partitioning all spatial grid codes in the initial spatial grid coding set, including: aggregating all spatial grid codes of the Nth level in the initial spatial grid coding set into several spatial grid codes of the N-1th level according to a preset number of grids, until it is impossible to perform aggregation partitioning operations at each level according to the preset number of grids; and obtaining the target spatial grid coding set based on all spatial grid codes after aggregation partitioning at each level; wherein, N is the maximum number of grid levels corresponding to the selected grid partitioning and coding method; the preset number of grids is determined according to the selected grid partitioning and coding method.

[0064] Preferably, the selected mesh partitioning and encoding method adopts a multi-scale encoding method based on GeoSOT mesh partitioning.

[0065] Preferably, a multi-scale encoding method based on GeoSOT grid partitioning is used to encode an inverted quadtree, with a predetermined number of grids of 4.

[0066] by Figure 2 Taking the remote sensing data shown as an example, Figure 2 In the diagram, 1-13 represent the remote sensing image data represented by these grid data. The spatial grid coding set obtained by encoding using a multi-scale coding method based on GeoSOT grid partitioning is as follows: Figure 3The spatial grid coding set at level 31 shown is... Figure 3 The numbers at level 31 represent the grid codes corresponding to the remote sensing data, which can be represented by these level 31 grid codes. The set of spatial grid codes at level 31 is used as the initial spatial grid code set. At this initial set, the spatial grid codes correspond to the smallest scale grids. Therefore, an aggregation partitioning method is used to perform aggregation partitioning on all spatial grid codes at level 31. The preset grid size is 4. The specific aggregation partitioning process is as follows:

[0067] The spatial grid codes of level 31 are aggregated and divided into level 30 spatial grid code sets by averaging every four unique values. For example, the values ​​0, 2, 4, and 6 of level 31 are grouped together and averaged to form level 30. The grid coverage area corresponding to grid code 3 in level 30 is equal to the sum of the grid coverage areas corresponding to grid codes 0, 2, 4, and 6 in level 31. This process is repeated to complete the aggregation and division of all spatial grid codes in level 31. Similarly, a similar aggregation and division operation is performed on all spatial grid codes in level 30 to obtain the spatial grid code set of level 29. This process is repeated for each new level until the current level can no longer be aggregated by averaging every four values. The final spatial grid codes for each level form the following set: Figure 3 The diagram shows an inverted quadtree, which serves as the target spatial grid encoding set. This target spatial grid encoding set represents a multi-scale grid. It is evident that the clustering method proposed in this embodiment transforms the single-scale representation of remote sensing images into scale-based grid filling, achieving accurate representation of remote sensing data at multiple scales, thereby improving the efficiency and accuracy of remote sensing data indexing and querying.

[0068] An optional solution provided by this invention involves determining the grid coding partitioning method as either a classification partitioning method or an aggregation partitioning method when both a preset grid level requirement and a preset grid number are simultaneously met. The aggregation partitioning method employs the aforementioned aggregation partitioning approach. The classification partitioning method involves re-partitioning the grid topology relationships corresponding to all spatial grid codes in the initial spatial grid coding set. Another optional solution provided by this invention involves re-partitioning all spatial grid codes in the initial spatial grid coding set to obtain a target spatial grid coding set when the determined grid coding partitioning method is a classification partitioning method. This includes: determining the grid topology relationships corresponding to all spatial grid codes in the initial spatial grid coding set; wherein the grid topology relationships include inclusion relationships and intersection relationships; further subdividing all spatial grid codes with intersection relationships into smaller grid codes until no spatial grid codes with intersection relationships exist, or the preset grid level requirement and preset grid number requirement are met, i.e., achieving the minimum grid partitioning or a grid partitioning that meets user requirements; and obtaining the target spatial grid coding set based on all spatial grid codes with inclusion relationships.

[0069] Similarly Figure 2 Taking the remote sensing data shown as an example, we can see Figure 2 Grids 1-13 not only indicate that the remote sensing data can be represented using these grids, but also reflect the topological relationship between each grid and the remote sensing data. Figure 4 It can be seen that each grid has two grid topology relationships with the remote sensing data, including such as Figure 4 The inclusion relationship of grids 6 and 7, and the intersection relationship of grids 1, 2, 3, 4, 5, 8, 9, 10, 11, 12, and 13 are obtained in advance. In other words, for all spatial grid codes in the initial spatial grid code set, the grid topology relationship corresponding to the spatial grid code can be obtained in advance, and then the grids with intersecting topology relationships are classified and divided. Figure 4 The example shown is merely a sample classification result for grid 2. Other intersecting grids will undergo a similar classification process. Specifically, the classification process for grid 2 is as follows:

[0070] Please see Figure 5Grid 2 is a rectangular structure. By connecting the midpoints of opposite sides, grid 2 is divided into four equal parts, resulting in grids 201, 202, 203, and 204. Grid 203 is contained within grid 203, while grids 201, 202, and 204 are intersecting. Therefore, it is not necessary to classify grid 203, which is contained within grid 203. Only grids 201, 202, and 204, which are intersecting, need to be further classified. Taking grid 204 as an example, the same classification method as for grid 2 is used to classify grid 204, resulting in grids 2041, 2042, 2043, and 2044. Grids 2043 and 2044 are in an inclusive relationship, while grids 2041 and 2042 are in an intersecting relationship. Therefore, it is not necessary to classify grids 2043 and 2044, which are in an inclusive relationship, but only grids 2041 and 2042, which are in an intersecting relationship. This process continues until there are no grids in an intersecting relationship. Thus, grid 2 can be represented by networks such as grids 203, 2043, and 2044, which are in an inclusive relationship, to represent the remote sensing data corresponding to grid 2. Similarly, the same classification operation is performed on grids 1, 3, 4, 5, 8, 9, 10, 11, 12, and 13, which are in an intersecting relationship. The resulting spatial grid codes corresponding to all grids in an inclusive relationship are finally used as the target spatial grid code set in this embodiment of the invention. At this time, the target spatial grid code set is a multi-scale grid representation. As can be seen, for the classification method, the grids ultimately representing remote sensing data are all grids that contain the remote sensing data, which is different from the traditional method. Figure 2 The representation proposed in this embodiment of the invention can more accurately represent remote sensing data using grids, and the closer the representation accuracy of remote sensing data is to 1, the more accurate the representation of remote sensing data can be at multiple scales, thereby improving the efficiency and accuracy of remote sensing data indexing and querying; moreover, this division through grid topology also intuitively reflects the relationship between grids and remote sensing images.

[0071] In summary, the multi-scale grid remote sensing data partitioning method proposed in this invention breaks away from the traditional representation of remote sensing data. It no longer uses all the spatial grid codes after the initial partitioning to represent remote sensing data, and it can adapt to different amounts of remote sensing data. According to user needs, different grid coding partitioning methods are used to re-partition all the spatial grid codes after the initial partitioning, realizing the representation of remote sensing data at multiple scales, improving the representation accuracy of remote sensing data, and thus improving the efficiency and accuracy of remote sensing data indexing and querying.

[0072] Because current on-board remote sensing data storage and management strategies do not establish a structured relationship between current remote sensing data and multiple scales such as geographic location, space, and time, on-board remote sensing data applications lack real-time capability, resulting in insufficient utilization of satellite resources. Regarding this issue, the second aspect is discussed below. Figure 6This invention also proposes a multi-scale grid remote sensing data management method. The grid partitioning during the management process uses the aforementioned multi-scale grid remote sensing data partitioning method embodiment. This remote sensing data management method includes the following steps:

[0073] S201. Using the multi-scale grid remote sensing data partitioning method described in S101 to S104, the spatial location information corresponding to all data blocks is partitioned to obtain the target spatial coding set; wherein, the data blocks are in the form of pre-stored original remote sensing data.

[0074] Specifically, in this embodiment of the invention, the spatial location information corresponding to all data blocks is pre-processed into a target spatial encoding set by performing grid subdivision processing on the multi-scale grid remote sensing data subdivision method of the first aspect. Specifically, in step S101, the original remote sensing image and its corresponding spatial location information are acquired, and the acquired original remote sensing data is stored in the form of data blocks.

[0075] S202. Establish an index table corresponding to the target spatial encoding set and the data block; wherein, the target spatial encoding set includes several spatial encodings, and each spatial encoding and the data block satisfy a preset mapping relationship.

[0076] Specifically, the target spatial encoding set of the present invention includes several spatial encodings, and the preset mapping relationship between each spatial encoding and data block includes at least one of the following: a one-to-one mapping between spatial encoding and data block, a many-to-one mapping between spatial encoding and data block, and a one-to-many mapping between spatial encoding and data block. Figure 7 a~ Figure 7 b clearly demonstrates several mapping relationships between spatial encoding and data blocks, such as Figure 7 In file a, spatial encoding 1, spatial encoding 2, spatial encoding 3, and spatial encoding 4 in files 1, 2, and 3 have a one-to-one mapping relationship with data block 1, data block 2, data block 3, and data block 4, respectively. Figure 7 In file b, spatial code 1 in file 1 has a one-to-many mapping relationship with data block 1, data block 2, and data block 3, while spatial code 2, spatial code 3, and spatial code 4 have a many-to-one mapping relationship with data block 4. Within the same file, these three mapping relationships can be combined arbitrarily.

[0077] There are no restrictions on the method used to establish the index table corresponding to the target space encoding set and the data block. It can be established using a file-based method or a non-file-based method. Figure 7 a~ Figure 7 Example b provides an index table that establishes the target space encoding set and the corresponding data block based on a file-based approach, thereby realizing gridded storage.

[0078] The multi-scale grid remote sensing data management method proposed in this invention establishes a structured relationship between current remote sensing data and geographical location, space, etc., during on-board management, ensuring the real-time application of satellite remote sensing data and improving the utilization rate of satellite resources. This gridded storage method breaks away from the traditional ground-based storage method that requires standard storage equipment, directly achieving storage mapping of on-board storage media through an index table. Simultaneously, during subsequent data retrieval, data is directly read on-board through the index table. Gridded data storage mapping management is closer to the underlying storage medium, resulting in shorter data retrieval paths and higher efficiency. The combination of this invention with a file system is more suitable for the actual situation of on-orbit operation. The management method of this invention allows multiple satellites to use the same grid coding method, effectively improving the accuracy of satellite resource retrieval, reducing the pressure on the ground-to-ground data downlink channels and ground processing stations, and in the future, it can support the realization of an "on-board data storage center" and provide "virtual remote sensing satellite services."

[0079] In data storage, simple auxiliary methods are often used, such as timestamps, strings, and time counts. These methods are scattered across file systems, database systems, and various programming languages. However, in reality, the time information in big data is diverse, with varying scales and precision. Some data uses instantaneous counts, some uses precise time periods, and some uses fuzzy time ranges. Currently, this time information is not fully utilized. This invention, while considering the aforementioned empty location information, also incorporates the time information of remote sensing data during the storage mapping process. Please refer to [link to relevant documentation]. Figure 8 An alternative solution is proposed, which, based on S201, also includes:

[0080] S203. Encode the time information corresponding to all data blocks to obtain the time code.

[0081] Specifically, through S101, the time information corresponding to the remote sensing data can also be obtained. In this embodiment of the invention, each data block can be encoded according to the corresponding time information by using a single-scale time grid coding method or a multi-scale time grid coding method to obtain the time code. The specific method of using the single-scale time grid coding method or the multi-scale time grid coding method is not limited.

[0082] Since the temporal information of the remote sensing data has been taken into account, correspondingly, S202 establishes an index table corresponding to the target spatial coding set and data blocks, which is updated to S204:

[0083] S204. Establish an index table corresponding to the time code, the target spatial code set, and the data block; wherein, there is a one-to-one mapping relationship between the time code and the data block; the target spatial code set includes several spatial codes, and each spatial code satisfies a preset mapping relationship with the data block.

[0084] Specifically, the spatial location information is processed in the same way as in S202, and the target spatial code set also includes several spatial codes. The preset mapping relationship between each spatial code and the data block includes at least one of the following: a one-to-one mapping between the spatial code and the data block, a many-to-one mapping between the spatial code and the data block, and a one-to-many mapping between the spatial code and the data block. Figure 7 a~ Figure 7 b clearly demonstrates the mapping relationship between spatial encoding and data blocks. Building upon this, temporal encoding is further considered, and an index table is established corresponding to the temporal encoding, the target spatial encoding set, and the data blocks. Please refer to [link to relevant documentation]. Figure 9 a~ Figure 9 b. This index table shows the index relationship between file encoding, target spatial encoding set and time encoding, thereby realizing gridded storage. The storage takes into account both spatial location information and time information, and establishes a structured relationship between current remote sensing data and geographical location, space, time, etc., so as to enable the real-time application of satellite remote sensing data and make full use of satellite resources.

[0085] The multi-scale gridded remote sensing data management method proposed in this invention establishes a structured relationship between current remote sensing data and geographical location, space, and time during the on-board storage mapping process, ensuring the real-time application of satellite remote sensing data and improving the utilization rate of satellite resources. This gridded storage method breaks away from the traditional ground-based storage method that requires standard storage equipment, directly achieving storage mapping of the on-board storage medium through an index table. Furthermore, during subsequent data retrieval, data is directly read on-board through the index table. Gridded data storage mapping management is closer to the underlying storage medium, resulting in shorter data retrieval paths and higher efficiency. The integration of this invention with a file system is more suitable for the actual situation of on-orbit operation. The management method of this invention allows multiple satellites to use the same grid coding method, effectively improving the accuracy of satellite resource retrieval, reducing the pressure on the ground-to-ground data downlink channels and ground processing stations, and in the future, it can support the realization of an "on-board data storage center" and provide "virtual remote sensing satellite services."

[0086] It should be noted that the embodiments of the multi-scale grid remote sensing data management method proposed in this invention adopt the embodiments of the multi-scale grid remote sensing data subdivision method of the first aspect described above, and all of them can achieve the same or similar beneficial effects, which will not be repeated here.

[0087] In the description of this invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0088] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. In addition, those skilled in the art can combine and integrate the different embodiments or examples described in this specification.

[0089] Although this application has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings, disclosure, and appended claims, will understand and implement other variations of the disclosed embodiments in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce good results.

[0090] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A multi-scale grid remote sensing data management method, characterized in that, include: The raw remote sensing data obtained from the platform's front-end data source is pre-stored on the satellite in data blocks, and the spatial location information corresponding to the raw remote sensing data is obtained from the platform's central control computer. The platform's front-end data source includes the on-board payload camera; the platform's central control computer includes the satellite's operational computer; auxiliary data is acquired through the platform's central control computer, including the satellite's attitude, orbit, and time. Based on the satellite's orbit, attitude, and time, and combined with the optical path design of the sensors in the imaging subsystem, the positioning processing of the raw remote sensing data is completed, forming the spatial location information corresponding to the raw remote sensing data. For each data block on the satellite, several spatial grids are obtained by dividing the data into grids according to the corresponding spatial location information and a fixed-level grid; wherein, the fixed-level grid is determined according to preset grid level requirements and preset grid quantity requirements; Multi-scale grid coding is performed on all spatial grids to obtain an initial spatial grid coding set; The grid coding partitioning method is determined based on the preset grid level requirements and the preset grid quantity requirements. Then, all spatial grid codes in the initial spatial grid code set are re-partitioned according to the grid coding partitioning method to obtain the target spatial grid code set. Specifically, determining the grid coding partitioning method based on the preset grid level requirements and the preset grid quantity requirements includes: when only the preset grid level requirements are met, the grid coding partitioning method is determined to be an aggregation partitioning method; wherein, the aggregation partitioning method is a method of re-partitioning all spatial grid codes in the initial spatial grid code set; when both the preset grid level requirements and the preset grid quantity requirements are met, the grid coding partitioning method is determined to be a classification partitioning method or an aggregation partitioning method; wherein, the classification partitioning method is a method of re-partitioning the grid topology relationships corresponding to all spatial grid codes in the initial spatial grid code set. An index table is established based on a file-based approach to correspond to the target spatial grid coding set and the data block. During on-board management, the index table establishes a structured relationship between the original remote sensing data and the geographical location and space. The index table enables storage mapping of the on-board storage medium and direct data reading from the satellite via the index table. The gridded data storage mapping management is close to the underlying storage medium, resulting in a short data reading path and high efficiency. The target spatial grid coding set includes several spatial codes, and each spatial code satisfies a preset mapping relationship with the data block.

2. The multi-scale grid remote sensing data management method according to claim 1, characterized in that, The fixed-level grid is determined according to the preset grid hierarchy requirements and the preset grid quantity requirements, including: When only the preset grid level requirement is met, the fixed-level grid is determined to be the minimum-scale grid; When both the preset grid level requirement and the preset grid number are met, it is determined that the fixed-level grid is not the minimum-scale grid.

3. The multi-scale grid remote sensing data management method according to claim 1, characterized in that, When the determined grid coding partitioning method is an aggregation partitioning method, all spatial grid codes in the initial spatial grid coding set are re-partitioned to obtain the target spatial grid coding set, including: All spatial grid codes at level N in the initial spatial grid coding set are aggregated into several spatial grid codes at level N-1 according to a preset number of grids, until it is impossible to perform aggregation and partitioning operations at each level according to the preset number of grids. The target spatial grid coding set is obtained based on all spatial grid codes after aggregation and partitioning at each level; where N is the maximum number of grid levels corresponding to the selected grid partitioning and coding method; the preset number of grids is determined according to the selected grid partitioning and coding method.

4. The multi-scale grid remote sensing data management method according to claim 3, characterized in that, The selected mesh partitioning and encoding method adopts a multi-scale encoding method based on GeoSOT mesh partitioning.

5. The multi-scale grid remote sensing data management method according to claim 4, characterized in that, The multi-scale encoding method based on GeoSOT grid partitioning encodes an inverted quadtree, and the predetermined number of grids is 4.

6. The multi-scale grid remote sensing data management method according to claim 1, characterized in that, When the determined grid coding partitioning method is a classification partitioning method, the spatial grid codes in the initial spatial grid coding set are re-partitioned to obtain the target spatial grid coding set, including: Determine the grid topology relationships corresponding to all spatial grid codes in the initial spatial grid code set; the grid topology relationships include inclusion relationships and intersection relationships. All spatial grid codes that are intersecting are further subdivided into smaller grid codes until there are no more spatial grid codes that are intersecting or the preset grid level requirements and preset grid number requirements are met. The target spatial grid code set is obtained based on all spatial grid codes that are inclusive.

7. The multi-scale grid remote sensing data management method according to claim 1, characterized in that, Also includes: The time information corresponding to all the data blocks is encoded to obtain a time code; The step of establishing an index table corresponding to the target spatial grid coding set and the data block includes: An index table is established corresponding to the time code, the target spatial grid code set, and the data block; wherein, there is a one-to-one mapping relationship between the time code and the data block; the target spatial grid code set includes several spatial codes, and each spatial code satisfies a preset mapping relationship with the data block.

8. The multi-scale grid remote sensing data management method according to any one of claims 1 or 7, characterized in that, The pre-defined mapping relationship between each spatial code and data block includes at least one of the following: a one-to-one mapping between spatial codes and data blocks, a many-to-one mapping between spatial codes and data blocks, and a one-to-many mapping between spatial codes and data blocks.

Images