A remote sensing image slice-free publishing system and method

Abstract

This invention relates to the field of remote sensing data publishing technology, specifically to a remote sensing image tile-free publishing system and method. The remote sensing image tile-free publishing system comprises a data support layer, a core business layer, and a web interface layer connected sequentially. The above solution provides a unified streaming read interface based on an adapter pattern for the local file system and the HDFS distributed file system through the data support layer. This allows the core business layer to complete metadata extraction, statistical information estimation, and thumbnail generation during the publishing phase without generating tile data. During the service phase, it adaptively selects the optimal streaming read strategy based on pixel organization to dynamically render and generate tile images. Simultaneously, it supports the virtual aggregation of multi-source images into a unified service for real-time synthesis, avoiding the problems of huge storage redundancy, high preprocessing costs, and insufficient flexibility caused by traditional pre-tiling methods.

Description

Technical Field

[0001] This invention relates to the field of remote sensing data publishing technology, and more specifically, to a remote sensing image tiling-free publishing system and method. Background Technology

[0002] Remote sensing imagery, as a crucial source of geospatial information, has wide applications in numerous fields such as surveying and mapping, environmental monitoring, urban planning, agriculture, and emergency response. With the rapid development of remote sensing technology, the amount of acquired remote sensing imagery data has exploded, data resolution has continuously improved, and data types have become increasingly diversified. How to efficiently store, manage, and publish this massive amount of remote sensing imagery data, and provide it to users in a service-oriented manner for on-demand access, has become a significant technical challenge facing the field of geographic information systems.

[0003] Traditional remote sensing image publishing methods primarily rely on pre-tiling technologies, such as WMTS (Web Map Tile Service) which follows the OGC (Open Geospatial Consortium) standard and the Google XYZ standard. These typically require pre-cutting the raw image data into tile pyramids of different levels according to a specified grid set. This pre-tiling model has several technical drawbacks: First, it results in massive storage redundancy, with the tile pyramid generating storage requirements approaching the size of the original data. Second, it is time-consuming and computationally expensive, as generating the tile pyramid is a computationally intensive process, which can be extremely lengthy for large-scale or frequently updated data sources. Third, it lacks flexibility, as the pre-generated tiles are fixed and cannot support users dynamically selecting band combinations, dynamically adjusting rendering parameters, or performing simple online analysis; all or some tiles need to be regenerated after data updates. Furthermore, existing technologies perform poorly when processing ultra-large-scale images stored on distributed file systems such as HDFS. They struggle to seamlessly and dynamically aggregate multiple image files scattered across heterogeneous, distributed data sources such as HDFS and local file systems into a unified virtual space service and enable efficient, slice-free access. They also lack a simple, loosely coupled extension mechanism when accessing new data sources, posing challenges to scalability. Summary of the Invention

[0004] The purpose of this invention is to provide a remote sensing image tiling-free publishing system and method to solve the above-mentioned technical problems.

[0005] In a first aspect, embodiments of the present invention provide a remote sensing image tile-free publishing system, comprising a data support layer, a core business layer, and a Web interface layer connected in sequence, wherein:

[0006] The data support layer is used to provide a unified streaming read interface based on the adapter pattern for image data in the local file system and HDFS distributed file system to support the upper-layer business logic, and to persistently manage system metadata information.

[0007] The core business layer is used to execute the entire process of image tile-free publishing and service. In the publishing stage, it realizes metadata extraction, statistical information estimation and thumbnail generation. In the service stage, it adaptively selects the optimal streaming reading strategy based on the pixel organization method to dynamically render and generate tile images, and supports the virtual aggregation of multi-source images into a unified service for real-time synthesis.

[0008] The Web interface layer is used to receive service publishing requests, service loading requests, and thumbnail retrieval requests from clients, and forward the requests to the core business layer.

[0009] Furthermore, the data support layer includes:

[0010] The data source abstraction layer is used to define standardized streaming read interfaces and dynamically instantiate corresponding adapters based on image data source type information using the factory pattern, so as to support the core business layer's unified streaming access to the local file system and HDFS distributed file system.

[0011] The information management module is used to persistently manage the system metadata information; wherein, the system metadata information includes the core metadata of the image, statistical information, service configuration information, and the association between the aggregated spatial service and its source image; the core metadata includes image size, bit depth, number of bands, pixel organization method, and georeferenced information.

[0012] Furthermore, the core business layer includes:

[0013] The workflow management module is used to standardize the core process of image publishing based on a template-based approach; the core process includes parsing metadata, calculating statistical information, storing information in the database, and generating thumbnails.

[0014] The metadata parsing module, connected to the publishing workflow management module and the data support layer, is used to extract core metadata from the original image files through streaming reading during the image publishing stage; wherein, the core metadata includes image size, bit depth, number of bands, pixel organization method and georeferenced information;

[0015] The statistical information estimation module, connected to the publishing workflow management module and the data support layer, is used to quickly estimate the statistical information of the image using a sparse sampling strategy during the image publishing stage; wherein, the statistical information includes the maximum value, minimum value and approximate percentile quantile of each band.

[0016] Furthermore, the release workflow management module is configured as follows:

[0017] Receive service publishing requests;

[0018] If the service publishing request is a single image publishing request, then the metadata parsing module and the statistical information estimation module are called in sequence to complete the metadata parsing and statistical information calculation, then a service entity is generated and stored in the information management module, and the thumbnail generation module is called to generate an image thumbnail, and finally the service address is returned;

[0019] If the service publishing request is a multi-image publishing request, the metadata parsing module and the statistical information estimation module are called in a loop to complete the metadata parsing and statistical information calculation for each image, generate each image service entity and store it in the information management module, and call the thumbnail generation module to generate a thumbnail for each image. After all images have been processed, an aggregated spatial service entity is generated and stored in the information management module, and the thumbnail generation module is called again to generate an aggregated service thumbnail based on the combination of each image thumbnail, and finally the aggregated service address is returned.

[0020] Furthermore, the core business layer also includes:

[0021] The dynamic tile generation and rendering module, connected to the data support layer, is used to respond to service loading requests during the image service phase. It calculates the corresponding original image pixel coordinate range in real time based on the tile coordinates and the core metadata, adaptively selects the optimal streaming reading strategy based on the pixel organization method to accurately obtain the target pixel data block, and dynamically renders the pixel data block to generate a tile image according to the rendering parameters specified by the user in real time or the service default.

[0022] Furthermore, the core business layer also includes:

[0023] The multi-image dynamic aggregation module, connected to the dynamic tile generation and rendering module and the data support layer, is used to virtually aggregate multiple image files scattered in the local file system and the HDFS distributed file system into a unified aggregation space service. When a tile request for the aggregation space service is received, the dynamic tile generation and rendering module is called in parallel to obtain the tile images corresponding to each source image, and the aggregated tile image is synthesized in real time according to the rendering order recorded in the association relationship.

[0024] Furthermore, the dynamic tile generation and rendering module and the multi-image dynamic aggregation module are configured collaboratively as follows:

[0025] Receive service loading request;

[0026] If the service loading request is a single image service loading request, then the image core metadata, statistical information and service configuration information stored in the information management module are obtained, the tile request is parsed to calculate the geographical range of the current request and the corresponding original image pixel coordinate range, and the optimal streaming reading strategy is adaptively selected based on the pixel organization method to accurately obtain the target pixel data block from the data source abstraction layer, and the pixel data block is dynamically rendered into a tile image of specified specifications and returned to the front end according to the rendering parameters specified by the user in real time or the default rendering parameters of the service.

[0027] If the service loading request is an aggregated spatial service loading request, then the association relationship between the aggregated spatial service and its source images stored in the information management module is obtained, the tile request is parsed to calculate the geographical range of the current request, one or more source images intersecting with the geographical range are determined based on the association relationship, the dynamic tile generation and rendering module is called in parallel to obtain the tile images corresponding to each source image, and the tile images are synthesized in real time onto the canvas of the requested tile size according to the rendering order recorded in the association relationship, and finally the aggregated tile image is rendered and returned to the front end.

[0028] Furthermore, the core business layer also includes:

[0029] A thumbnail generation module, connected to the publishing workflow management module and the data support layer, is used to generate thumbnails for single image services during the image publishing phase, and to calculate the maximum bounding rectangle of all source images and generate aggregate service thumbnails based on a hierarchical projection compositing strategy during the publishing of multi-image aggregation services; and

[0030] The image processing tools module provides low-level image processing functions, including rearranging pixel data according to band index, encoding pixel data blocks into a specified image file format, and storing the generated image file to a specified location.

[0031] Secondly, embodiments of the present invention provide a method for publishing remote sensing images without slicing, the method comprising:

[0032] Receive service publishing requests;

[0033] If the service publishing request is a single image publishing request, the core metadata is extracted from the original image file through streaming reading, the image statistics are quickly estimated using a sparse sampling strategy, a service entity is generated and persistently stored, an image thumbnail is generated, and finally the service address is returned.

[0034] If the service publishing request is a multi-image publishing request, then the above metadata extraction and statistical information estimation operations are executed in a loop to process each image one by one, generate each image service entity and store it persistently, and generate a thumbnail for each image one by one. After all images have been processed, an aggregated space service entity is generated and stored persistently, and an aggregated service thumbnail is generated based on the combination of each image thumbnail, and finally the aggregated service address is returned.

[0035] Furthermore, the method also includes:

[0036] Receive service loading request;

[0037] If the service loading request is a single image service loading request, then the persistently stored image core metadata, statistical information and service configuration information are obtained, the tile request is parsed to calculate the geographical range of the current request and the corresponding original image pixel coordinate range, the optimal streaming reading strategy is adaptively selected based on the pixel organization method in the core metadata to accurately obtain the target pixel data block, and the pixel data block is dynamically rendered into a tile image of specified specifications and returned to the front end according to the rendering parameters specified by the user in real time or the default rendering parameters of the service.

[0038] If the service loading request is an aggregated space service loading request, then the association between the persistently stored aggregated space service and its source images is obtained, the tile request is parsed to calculate the geographical range of the current request, one or more source images intersecting with the geographical range are determined based on the association, the tile images corresponding to each source image are obtained in parallel, and the tile images are synthesized in real time onto the canvas of the requested tile size according to the rendering order recorded in the association, and finally the aggregated tile image is rendered and returned to the front end.

[0039] The aforementioned solution provides a unified streaming interface based on the adapter pattern for the local file system and the HDFS distributed file system through a data support layer. This allows the core business layer to complete metadata extraction, statistical information estimation, and thumbnail generation during the deployment phase without generating tile data. During the service phase, it adaptively selects the optimal streaming strategy based on pixel organization to dynamically render and generate tile images. It also supports the virtual aggregation of multi-source images into a unified service for real-time synthesis, avoiding the problems of huge storage redundancy, high preprocessing costs, and insufficient flexibility caused by the traditional pre-tiling mode. On the other hand, the data support layer decouples business logic from the underlying data storage details by defining a standardized streaming interface FileReader and using the factory pattern to dynamically instantiate the corresponding adapter implementation. This allows the system to be extended to support new data source types without modifying the core business code. Furthermore, the core business layer uses a sparse sampling strategy to quickly estimate image statistical information during the deployment phase and adaptively selects the optimal streaming strategy based on pixel organization to accurately obtain target pixel data blocks during the service phase. It then dynamically renders the data based on user-specified or service-default rendering parameters, achieving on-demand reading and real-time rendering, thus improving service response speed and flexibility.

[0040] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0041] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments of the present invention will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a schematic diagram of the architecture of a remote sensing image tiling-free publishing system provided in an embodiment of the present invention;

[0043] Figure 2 This is a partial schematic diagram of the image service publishing process provided in an embodiment of the present invention;

[0044] Figure 3 This is another schematic diagram of the image service publishing process provided in an embodiment of the present invention;

[0045] Figure 4 This is a partial schematic diagram of the image service loading process provided in an embodiment of the present invention;

[0046] Figure 5 This is another schematic diagram of the image service loading process provided in an embodiment of the present invention;

[0047] Figure 6 This is a diagram illustrating the loading effect of HDFS image publishing services provided in an embodiment of the present invention.

[0048] Figure 7 This is a diagram illustrating the loading effect of another service after HDFS image publishing, provided in an embodiment of the present invention.

[0049] Figure 8 This is a diagram illustrating the loading effect of a local file system image after publication, as provided in an embodiment of the present invention.

[0050] Figure 9 This is an example of another service loading effect after the local file system image is published, as provided in an embodiment of the present invention.

[0051] Figure 10 An overview of the local file system storage images provided in this embodiment of the invention;

[0052] Figure 11 An overview of the HDFS distributed file system storage images provided in this embodiment of the invention;

[0053] Figure 12 This is a diagram illustrating the generation effect of the aggregated space service thumbnail provided in an embodiment of the present invention.

[0054] Figure 13 This is a diagram illustrating the loading effect of the aggregated space service provided in an embodiment of the present invention.

[0055] Figure 14 Another loading effect diagram of the aggregated space service provided in this embodiment of the invention;

[0056] Figure 15 The loading effect diagram is shown for the image optimization linear stretching method provided in the embodiment of the present invention;

[0057] Figure 16 The loading effect diagram of the image extremum linear stretching method provided in the embodiment of the present invention;

[0058] Figure 17 This is a diagram illustrating the effect of combining false color bands in an image according to an embodiment of the present invention. Detailed Implementation

[0059] The embodiments of the technical solution of the present invention will be described in detail below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and are therefore only examples, and should not be used to limit the scope of protection of the present invention. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and the foregoing description of the drawings are intended to cover non-exclusive inclusion. In the description of the embodiments of the present invention, technical terms such as "first," "second," etc., are only used to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly indicating the number, specific order, or primary or secondary relationship of the indicated technical features. In the description of the embodiments of the present invention, "a plurality of" means two or more, unless otherwise explicitly specified. The reference to "embodiment" herein means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent 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.

[0060] The remote sensing image tiling-free publishing system provided in this invention aims to solve many problems existing in the publishing of remote sensing image services, especially for large-scale remote sensing image data stored on distributed file systems such as HDFS, which is difficult to achieve efficient tiling-free publishing (supporting service specifications such as OGC and XYZ); as well as the complexity and performance bottlenecks encountered when dynamically aggregating multiple image files scattered on heterogeneous and distributed data sources such as HDFS and local file systems into a unified virtual service; at the same time, existing systems generally have insufficient scalability when accessing new data source types.

[0061] like Figure 1 As shown, this embodiment of the invention provides a remote sensing image tiling-free publishing system, comprising: a data support layer 110, a core business layer 120, and a Web interface layer 130 connected in sequence, wherein:

[0062] The data support layer 110 is used to provide a unified streaming read interface based on the adapter pattern for image data in the local file system and HDFS distributed file system to support the upper layer business logic and persist system metadata information.

[0063] The core business layer 120 is used to execute the entire process of image tile-free publishing and service. In the publishing stage, it realizes metadata extraction, statistical information estimation and thumbnail generation. In the service stage, it adaptively selects the optimal streaming reading strategy based on the pixel organization method to dynamically render and generate tile images, and supports the virtual aggregation of multi-source images into a unified service for real-time synthesis.

[0064] Web interface layer 130 is used to receive service publishing requests, service loading requests and thumbnail retrieval requests from clients, and forward the requests to the core business layer.

[0065] The aforementioned data support layer 110 serves as the system's infrastructure layer. Its core function is to provide a standardized, unified streaming read interface for image data in the local file system and the HDFS distributed file system through the adapter pattern. This effectively isolates the upper-layer business logic from the underlying data storage details, allowing the core business layer to achieve unified access operations without needing to concern itself with the technical differences of specific data sources. At the same time, this layer is also responsible for the persistent management of system metadata information, providing a data foundation for the stable operation of the entire system.

[0066] The aforementioned core business layer 120 serves as the system's functional execution engine, responsible for executing the key business logic of image tile-free publishing and service throughout the entire process: During the publishing phase, this layer extracts core metadata from the original image files through streaming reading, uses a sparse sampling strategy to quickly estimate statistical information, and generates thumbnails required for previewing subsequent service calls; During the service phase, this layer adaptively selects the optimal streaming reading strategy based on the internal pixel organization of the image, dynamically renders and generates the tile image requested by the client, and supports the virtual aggregation of physically dispersed multi-source images into a unified spatial service and real-time synthesis, thereby completely avoiding the storage redundancy and lack of flexibility problems caused by the traditional pre-tiling mode.

[0067] The aforementioned Web interface layer 130 serves as the unified access layer for the system to the outside world. It is mainly responsible for receiving and forwarding client requests. Specifically, it is responsible for receiving service publishing requests, service loading requests, and thumbnail acquisition requests from clients, and accurately forwarding various requests to the core business layer for corresponding processing, so as to realize standardized interaction between the system and external applications.

[0068] Optionally, the aforementioned data support layer 110 includes:

[0069] The data source abstraction layer is used to define standardized streaming read interfaces and dynamically instantiate corresponding adapters based on image data source type information through the factory pattern to support the core business layer's unified streaming access to the local file system and HDFS distributed file system.

[0070] The information management module is used to persistently manage system metadata information. The system metadata information includes the core metadata of the image, statistical information, service configuration information, and the association between the aggregated spatial service and its source image. The core metadata includes image size, bit depth, number of bands, pixel organization method, and georeferenced information.

[0071] The aforementioned Data Source Abstraction Module is the core for achieving unified access to multi-source data. Based on the Adapter Pattern, it provides a unified access abstraction for different types of data storage systems (such as local file systems and HDFS distributed file systems). The Data Source Abstraction Module also defines a standardized streaming interface (FileReader), which encapsulates low-level read operations such as getInputStream(), seek(longoffset), and readFully(byte[] buffer, int off, int len). Upper-layer business modules (such as metadata parsers and dynamic tile generators) only need to rely on this unified interface and do not need to concern themselves with the API details of specific data sources. At runtime, the corresponding FileReader implementation is dynamically created and instantiated based on the image's data source type information using the factory pattern. This design greatly decouples business logic from data storage details. If new data sources need to be supported later, only the standard interface needs to be implemented and registered in the factory for seamless integration, without modifying the core publishing and reading logic, demonstrating high scalability.

[0072] The aforementioned Information Management Module serves as the system's central metadata and configuration repository. It is responsible for persistent management.

[0073] Image metadata includes, but is not limited to, width, height, data type, band information, pixel organization, georeference (projection, coordinate system), internal storage structure, statistical information (maximum and minimum values, quantiles), data source type, and storage path.

[0074] Service configuration information: Records the attributes of a single image service or a multi-image aggregation service, such as service name, description, list of included images (for aggregation services), overall geographic extent, thumbnail path, default rendering parameters (band combination, stretching method), NoData value definition, supported service types (such as WMS, WMTS, XYZ), tile parameters (format, size), etc.

[0075] Association: Maintain the mapping relationship between the multi-image aggregation service and the source images it comprises.

[0076] The information management module provides data query and retrieval services for other modules and is the foundation for data discovery, request scheduling and dynamic aggregation. It can be implemented based on a relational database (PostgreSQL with PostGIS extension).

[0077] Optionally, the aforementioned core business layer 120 includes:

[0078] The workflow management module is used to standardize the core process of image publishing based on template methods and patterns. The core process includes parsing metadata, calculating statistical information, storing information in the database, and generating thumbnails.

[0079] The metadata parsing module, connected to the publishing workflow management module and the data support layer, is used to extract core metadata from the original image files through streaming during the image publishing stage. The core metadata includes image size, bit depth, number of bands, pixel organization method, and georeferenced information.

[0080] The statistical information estimation module, connected to the publishing workflow management module and the data support layer, is used to quickly estimate the statistical information of images during the image publishing stage using a sparse sampling strategy. The statistical information includes the maximum and minimum values ​​of each band and the approximate percentile.

[0081] The aforementioned Metadata Parsing Module is responsible for accurately and efficiently extracting core metadata from remote sensing imagery. It selects the appropriate metadata parser using a factory pattern, considering common TIFF file structures (including standard TIFF and BigTIFF variants) and different byte orders (such as Little-Endian and Big-Endian). The parser employs a streaming approach, directly reading key parts such as the file header and image file directory (IFD) from the data stream obtained from the data source abstraction layer, and extracting important parameters such as image width, height, data bit depth, number of bands, pixel organization (such as PlanarConfiguration), georeferenced information, and internal tile or stripe layout. After parsing, the metadata is persistently stored and indexed through the information management module.

[0082] Considering the enormous time overhead of directly traversing large remote sensing images to calculate pixel statistics (such as maximum / minimum values ​​and quantiles for each band), the aforementioned Rapid Statistics Estimation Module employs a fast estimation strategy based on sparse sampling. It efficiently reads sample pixel values ​​from the image at configurable sampling intervals (e.g., acquiring a sample point every 16x16 pixels in two-dimensional space). Subsequently, a fast sorting algorithm is applied to the set of sample values ​​collected for each band to quickly estimate the maximum, minimum, and approximate upper and lower percentile (e.g., 2%) quantiles of that band, serving as the basis for image stretching and analysis. The estimation results are also stored through the information management module. This method reduces the time cost of statistical information acquisition while ensuring representativeness.

[0083] Optionally, the above-mentioned publishing workflow management module is configured to: receive service publishing requests; if the service publishing request is a single image publishing request, then sequentially call the metadata parsing module and the statistical information estimation module to complete the metadata parsing and statistical information calculation, generate a service entity and store it in the information management module, and call the thumbnail generation module to generate an image thumbnail, finally returning the service address; if the service publishing request is a multi-image publishing request, then cyclically call the metadata parsing module and the statistical information estimation module to complete the metadata parsing and statistical information calculation for each image, generate each image service entity and store it in the information management module, and call the thumbnail generation module to generate a thumbnail for each image, after all images have been processed, generate an aggregated spatial service entity and store it in the information management module, and call the thumbnail generation module again to generate an aggregated service thumbnail based on the combination of each image thumbnail, finally returning the aggregated service address. This implementation example is as follows: Figure 2 and Figure 3As shown, when the Web interface layer forwards the service publishing request submitted by the client to the publishing workflow management module, the module first determines the request type: if the received request is for a single image, it strictly follows the process skeleton defined by the template method pattern, first calling the metadata parsing module to extract core metadata from the original image file through the unified streaming reading interface provided by the data source abstraction layer, then calling the statistical information estimation module to quickly estimate the image statistical information using a sparse sampling strategy. After the metadata parsing and statistical calculation are completed, the corresponding service entity is generated and stored in the information database of the information management module, then the thumbnail generation module is called to generate the thumbnail of the image, and finally the service address is returned to the user; if the received request is for multiple images, a loop processing mechanism is started, and the aforementioned metadata parsing, statistical information estimation, service entity storage and thumbnail generation operations are repeated for each image until all images are processed. Then, an aggregated spatial service entity is generated and stored in the information management module, and the thumbnail generation module is called again to generate an aggregated service thumbnail based on the location information of each image thumbnail through hierarchical projection combination, and finally the aggregated service address is returned to the user, thereby realizing the standardized and process-oriented management of single and multiple image publishing.

[0084] Understandably, the aforementioned image publishing workflow management module employs the Template Method Pattern to standardize and manage the core process of image publishing. The module can define an abstract base publisher class containing an immutable `publish()` template method. This method calls a series of abstract or overridable step methods in a fixed order, such as `parseMetadata()`, `calculateStatistics()`, `storeInformation()`, and `generateThumbnail()`. Specific publisher subclasses can selectively call the concrete implementation methods of each node to implement specific logic, while maintaining consistency in the overall publishing process. This design ensures the standardization and consistency of the publishing process. Furthermore, when adjustments to the process are needed (such as adding a verification step or removing thumbnail generation), only the template method itself needs modification, reducing coupling to the specific step implementations and improving maintainability.

[0085] Optionally, the aforementioned core business layer 120 also includes: a dynamic tile generation and rendering module, which is connected to the data support layer. It is used to respond to service loading requests during the image service stage, calculate the corresponding original image pixel coordinate range in real time based on tile coordinates and core metadata, adaptively select the optimal streaming reading strategy based on pixel organization method to accurately obtain target pixel data blocks, and dynamically render the pixel data blocks to generate tile images according to the rendering parameters specified by the user in real time or the service default.

[0086] The aforementioned Dynamic Tiling and Rendering Module is the core execution unit for implementing the tile-free service, responsible for responding to client tile requests in real time. Based on the tile coordinates requested by the client (such as XYZ numbers) and the service's coordinate system / projection information, it accurately calculates the range of pixel coordinates of the tile on the target image (or the virtual canvas of the aggregation service). Crucially, this module adaptively selects the optimal streaming read strategy: based on the internal storage structure recorded in the image metadata (e.g., whether to store striped tiles as rows or tiled tiles as blocks), it employs different read algorithms to obtain data from the data source abstraction layer, minimizing I / O operations and data transfer. After obtaining the pixel data, this module applies the specified band combination and pixel value stretching method (such as linear stretching, standard deviation stretching, percentage truncation stretching, etc.) in real time for rendering, according to user requests or service default settings. Here, the Strategy Pattern and Factory Pattern are used to manage different stretching algorithms, allowing the addition of new stretching methods to simply implement predefined strategy interfaces, improving the scalability of the rendering logic. The final rendered result is returned as an image file encoded in the specified format.

[0087] Optionally, the aforementioned core business layer 120 also includes: a multi-image dynamic aggregation module, which is connected to the dynamic tile generation and rendering module and the data support layer. It is used to virtually aggregate multiple image files scattered in the local file system and the HDFS distributed file system into a unified aggregation space service. When a tile request for the aggregation space service is received, the dynamic tile generation and rendering module is called in parallel to obtain the tile images corresponding to each source image, and the aggregated tile image is synthesized in real time according to the rendering order recorded in the association relationship.

[0088] The aforementioned Multi-Image Dynamic Aggregation Module enables the dynamic, tile-free aggregation of multiple independent image files from different data sources (accessed via a data source abstraction layer) into a single virtual spatial service. The process is roughly as follows: First, each source image is published individually, and its metadata and statistics are obtained. Then, a new aggregation service definition is created, the overall spatial extent of the aggregation service (the union of the extents of all source images) is calculated and stored, and the association between the aggregation service and its contained source images is recorded in the information management module. When a client requests tile data from this aggregation service, the system queries the information management module based on the requested tile number and spatial extent to determine one or more source images intersecting with the tile. The dynamic tile generation and rendering module is invoked in parallel to retrieve and read the tile images within the tile extent of these intersecting images. Finally, all generated tiles are composited onto a canvas with the same size as the requested tile, in the display order specified by the user. The final composited canvas is rendered as an image and returned to the client. The generated tiles can be cached (e.g., in memory, local files, or a columnar database) to accelerate subsequent access. In addition, the aforementioned multi-image dynamic aggregation module allows the client to dynamically specify rendering parameters when making a request, providing a high degree of interactivity and flexibility.

[0089] Optionally, the aforementioned dynamic tile generation and rendering module and multi-image dynamic aggregation module are configured collaboratively to: receive service loading requests; if the service loading request is a single-image service loading request, obtain the image core metadata, statistical information, and service configuration information stored in the information management module, parse the tile request to calculate the geographic range of the current request and the corresponding original image pixel coordinate range, and adaptively select the optimal streaming reading strategy based on the pixel organization method to accurately obtain the target pixel data block from the data source abstraction layer, and dynamically render the pixel data block into a tile image of the specified size according to the rendering parameters specified by the user in real time or the service default and return it to the front end; if the service loading request is an aggregated spatial service loading request, obtain the association relationship between the aggregated spatial service and its source images stored in the information management module, parse the tile request to calculate the geographic range of the current request, determine one or more source images intersecting with the geographic range based on the association relationship, call the dynamic tile generation and rendering module in parallel to obtain the tile images corresponding to each source image, and synthesize the tile images into a canvas of the requested tile size in real time according to the rendering order recorded in the association relationship, and finally render and generate an aggregated tile image and return it to the front end. This implementation example is as follows: Figure 4 and Figure 5As shown, when the Web interface layer forwards the single image service loading request initiated by the client to the dynamic tile generation and rendering module, this module first obtains the persistent core metadata, statistical information, and service configuration information of the image from the information management module. Then, it parses the tile request to accurately calculate the geographic range corresponding to the current request, and converts this geographic range into the pixel coordinate range of the original image based on the image's geographic reference information. Subsequently, the dynamic tile generation and rendering module adaptively selects the optimal streaming reading strategy according to the pixel organization method (RowStrip or Tiled) recorded in the core metadata. It accurately obtains pixel data blocks within the target pixel coordinate range through the data source abstraction layer and dynamically renders the data according to the rendering parameters specified by the user in real time or the service's default parameters (including band combination, stretching method, etc.). The system generates tile images that meet the requested specifications and returns them to the front end for loading. If the received service loading request is an aggregated spatial service loading request, it is handled collaboratively by the multi-image dynamic aggregation module: First, it obtains the association relationship between the aggregated spatial service and its source images stored in the information management module, parses the tile request to calculate the geographical range of the current request, quickly determines one or more source images that intersect with the geographical range based on the association relationship, and calls the dynamic tile generation and rendering module in parallel to obtain the tile images corresponding to each source image. Then, according to the rendering order pre-recorded in the association relationship, it synthesizes the obtained tile images in real time onto a canvas with the same size as the requested tile, and finally renders the aggregated tile image and returns it to the front end, thereby realizing efficient aggregation loading and dynamic rendering of multiple images in heterogeneous distributed data sources.

[0090] Optionally, the aforementioned core business layer 120 also includes:

[0091] The thumbnail generation module, connected to the release workflow management module and data support layer, is used to generate thumbnails for single image services during the image release phase, and to calculate the maximum bounding rectangle of all source images and generate aggregate service thumbnails based on a hierarchical projection compositing strategy when releasing multi-image aggregation services; and

[0092] The image processing tools module provides low-level image processing functions, including rearranging pixel data according to band index, encoding pixel data blocks into a specified image file format, and storing the generated image file to a specified location.

[0093] The aforementioned thumbnail generation module is responsible for generating preview images for spatial services. This mainly includes: (1) For a single image service: Based on the user-specified or system-default thumbnail size (e.g., 2048x2048), combined with the image's metadata (resolution, range) and preset or default image stretching methods, the sampling position of each pixel in the thumbnail corresponding to the original image is calculated. Efficient streaming reading is performed through the data source abstraction layer (possibly utilizing an overview image within the image or on-demand resampling) to obtain the required pixel data. Finally, the image processing module is called to render and encode the pixel data into a thumbnail file in a specified format (e.g., PNG, JPEG). (2) For a multi-image aggregation service: First, the union (maximum bounding rectangle) of all image ranges contained in the aggregation service is calculated. A proportionally scaled canvas (configurable size) is created based on this range. Then, each image in the aggregation service is traversed, and its geographic coordinates are projected onto the corresponding pixel area on the canvas. The image's own thumbnail is read (or dynamically generated on demand), and its pixels are drawn to the corresponding positions on the canvas in a specified order. After all image processing is complete, the final canvas will be exported as a service thumbnail.

[0094] The aforementioned Image Processing Utility Module provides a range of low-level image processing functions for other modules to call. These functions include, but are not limited to: generating in-memory image objects in different channel modes (such as RGB, RGBA, and grayscale) by rearranging pixel data according to a specified band index and order; encoding pixel data blocks (byte arrays) into various common image file formats (such as PNG, JPEG, and WebP); and storing the generated image files to a specified location, supporting local file systems, HDFS, or object storage.

[0095] It is understood that the remote sensing image tile-free publishing system provided in this embodiment of the invention can support multiple data sources and is highly scalable, mainly by adopting the following mechanisms:

[0096] (1) Adaptive on-demand streaming image reading and rendering mechanism;

[0097] The remote sensing image tiling-free publishing system provided in this embodiment of the invention abandons the traditional preprocessing tiling mode and adopts a more efficient and flexible "on-demand service" strategy.

[0098] The publishing phase employs lightweight metadata extraction: During image data publishing, the system utilizes streaming technology to efficiently parse necessary metadata from the original image files (standard TIFF or BigTIFF), including but not limited to image size (width, height), bit depth, number of bands, pixel organization (such as strip or tile storage), and georeferenced information. Simultaneously, it can combine rapidly estimated image statistics (such as extreme values) and user-configured service parameters (default band combination, NoData value, preferred stretching method, target service type such as WMS / WMTS / XYZ, etc.) to persist this key information to a database for unified management. This process does not require the generation of tile data.

[0099] The service phase involves dynamic and adaptive reading and rendering: When a client requests a map tile with a specified tile number, the system responds in real time: Based on the requested tile number and service definition (tile matrix set, etc.), it accurately calculates the original image pixel coordinate range covered by the tile; crucially, the system can adaptively select the optimal streaming reading strategy based on pre-stored image metadata (especially the internal storage structure: RowStrip or Tile), and read data blocks within the target pixel range on demand and accurately through the data source abstraction layer to minimize I / O overhead and data transfer volume; after obtaining the pixel data within the target range, it can dynamically render the final tile image on the server side according to parameters specified by the user in real time or defaulted by the service (such as band combination, pixel value stretching method, etc.); finally, the rendered tile image is returned to the front end.

[0100] The aforementioned on-demand, view window-range-based, and adaptive image internal storage structure streaming and dynamic rendering mechanism avoids the huge overhead of preprocessing, significantly shortens data uptime, improves the flexibility of service updates, and can efficiently respond to user requests, thereby enhancing the service's interactive experience and system throughput.

[0101] (2) A highly scalable service architecture based on design patterns;

[0102] To ensure the long-term maintainability and scalability to future changes, the remote sensing image tile-free publishing system provided in this embodiment of the invention widely applies mature design patterns in its architecture design:

[0103] Data source access decoupling (Adapter Pattern): By defining a unified streaming read interface (e.g., FileReader) and implementing specific adapters for each supported data source (e.g., local disk, HDFS), the details of upper-layer business logic and underlying data storage are completely isolated. Upper-layer modules (e.g., publishers, tile readers) only aggregate this interface. In the future, when integrating new data sources (e.g., cloud storage), only the new adapter needs to be implemented and registered with the factory; no modification to core business code is required, greatly improving the efficiency of data source integration and the system's openness.

[0104] Strategy Pattern: For behaviors in a system that are volatile or have multiple implementations (such as byte parsing and image stretching algorithms), the strategy pattern is used for encapsulation. For example, BitDepthParsingStrategy (including 8-bit, 16-bit, etc. implementations) and StretchStrategy (including extreme value stretching, percentage truncation, etc. implementations) are provided. The system dynamically selects and applies the specific strategy based on configuration or context at runtime. Adding new parsing methods or stretching algorithms only requires implementing the corresponding strategy interface; there is no need to modify the client code using these strategies, enhancing the pluggability of the algorithm and the flexibility of the system.

[0105] Workflow standardization (Template Method Pattern): The core image publishing process (parsing metadata → calculating statistical information → data storage → generating thumbnails) is solidified into a template method in an abstract base class. This template method defines the skeleton of the process, while specific steps can be customized or overridden by subclasses (such as publishers for specific sensors). This ensures the consistency and standardization of the publishing process, and when adjustments to the main process are needed (adding or deleting steps), only the template method needs to be modified, reducing the complexity of process changes and the coupling between modules.

[0106] The combined application of these design patterns has created a software architecture that is loosely coupled, highly cohesive, and easy to extend and maintain.

[0107] (3) Mechanism for dynamic aggregation and publishing of multiple images from heterogeneous data sources;

[0108] The remote sensing image tile-free publishing system provided in this invention overcomes the limitations of a single data source, supporting the virtual aggregation of multiple image files physically scattered across different types and locations (such as image A on HDFS and image B on a local disk) into a unified spatial service for external publication. The system records the association relationships and metadata (such as spatial extent, data source path, priority / rendering order, etc.) between the aggregation service and its constituent source images through an information management center. When processing client requests for tiles from the aggregation service, the system first queries the association relationships based on the requested range to determine the source images intersecting with the tile's spatial extent. It then uses a data source abstraction layer to read pixel data from corresponding regions of each relevant source image in parallel and in a streaming manner. On the server side, according to the user-specified order and rendering parameters, the pixel data of the corresponding range of the corresponding image is dynamically "mounted" onto a temporary canvas in real time. Finally, the canvas is rendered into a tile image and returned. Cache optimization of the generated tiles is supported.

[0109] This mechanism shields users from the heterogeneity and distributed details of the underlying data storage, providing a unified and seamless access experience and supporting highly flexible real-time rendering control (band combination, stretching, ordering, etc.), greatly expanding the application scenarios and value of the service.

[0110] (4) Automatic generation component for aggregated service thumbnails;

[0111] To facilitate users' quick preview of complex aggregated service content (supporting multiple images distributed across different data sources), the remote sensing image tile-free publishing system provided in this embodiment of the invention also offers a targeted thumbnail generation method: Calculate the maximum bounding rectangle of all images included in the aggregated service; create a canvas (the size is also configurable) with the width and height of the maximum bounding rectangle scaled proportionally based on this range; traverse each image in the aggregated service and project its thumbnail (automatically generated by the system) onto the corresponding position on the canvas according to its geographic coordinates; finally, export the synthesized canvas as the overall thumbnail of the aggregated service and store it in the information management center.

[0112] The above components effectively solve the problem of generating intuitive preview images for virtual, dynamically aggregated datasets.

[0113] To verify the application effect of the aforementioned remote sensing image tiling-free publishing system, multiple image datasets stored on the local file system and HDFS were selected to test the method from multiple perspectives. This test was based on a single host node (12 cores, 64GB) and limited the maximum memory space that the service could be allocated to to 4GB.

[0114] I. Test Results;

[0115] (1) Publish images stored on HDFS:

[0116] Image name: GF1_PMS2_E119.2_N26.0_20190919_L1A0004254423_FUSION_GEO.tiff.

[0117] Image size: 1.8GB.

[0118] The post took 1.05 seconds to publish.

[0119] The loading effect of the service (XYZ service) after publication is as follows: Figure 6 and Figure 7 As shown, each tile takes an average of 500ms.

[0120] (2) Publish images stored on the local file system;

[0121] File name: ZZ.tiff, file size: 9.8GB, publication time: 284ms.

[0122] The loading effect of the service (XYZ service) after publication is as follows: Figure 8 and Figure 9 As shown, each tile takes an average of 300ms.

[0123] (3) Publish two images, one stored in the local file system and the other in the HDFS file system, as a single service;

[0124] Image names: GF1_PMS1_E119.2_N26.1_20180408_L1A0003109713_FUSION_GEO.tiff (local), GF1_PMS2_E119.2_N26.0_20190919_L1A0004254423_FUSION_GEO.tiff (HDFS). Image size: 1.8GB each. Release time: 967ms. Overview images of the two images are shown below. Figure 10 and Figure 11 As shown. The system automatically generates a thumbnail for this service after publication, as shown below. Figure 12 As shown, the generation time was 601ms. The loading effect of the service (XYZ service) after deployment is as follows. Figure 13 and Figure 14 As shown, each tile takes an average of 400ms.

[0125] (4) Publish images stored on HDFS;

[0126] Image name: SVN3-01_20241025_0000049420_L1B0000124491_06-MUX.tiff, image size: 2.0GB. The effect of setting the stretching method to optimized linear stretching during loading after publishing is shown in the image below. Figure 15As shown in the image, the effect of setting the stretching method to maximum / minimum linear stretching during loading after publishing is as follows: Figure 16 As shown, after publishing, when loading, the band combination is set to false color (bands 2, 3, and 4), and the stretching method is optimized linear stretching. The resulting image is shown below. Figure 17 As shown.

[0127] II. Timeliness Analysis: Statistical results show that the remote sensing image tile-free publishing system provided in this embodiment of the invention can efficiently publish image data on local file systems and HDFS, and supports combining images from different data sources into a single service. The published service supports WMTS, Google XYZ, and other specifications. The published service can load quickly in the browser (the average loading time for tiles of different images is less than 1 second). Combined with the "2-5-10" principle of web application performance testing, the service published by this patented method can save time and space requirements.

[0128] III. Stability Analysis: For data from various data sources including local file systems and HDFS, under continuous drag-and-drop loading scenarios, this system experienced 0 crashes and 0 exceptions during execution. The success rate was 100%, demonstrating excellent system stability.

[0129] Based on the same inventive concept, embodiments of the present invention also provide a method for publishing remote sensing images without slicing, including:

[0130] Receive service publishing requests;

[0131] If the service publishing request is a single image publishing request, the core metadata is extracted from the original image file through streaming reading, the image statistics are quickly estimated using a sparse sampling strategy, the service entity is generated and persistently stored, an image thumbnail is generated, and finally the service address is returned.

[0132] If the service publishing request is for multiple images, the above metadata extraction and statistical information estimation operations are executed in a loop to process each image one by one, generate each image service entity and store it persistently, and generate a thumbnail for each image one by one. After all images have been processed, an aggregated space service entity is generated and stored persistently, and an aggregated service thumbnail is generated based on the combination of each image thumbnail, and finally the aggregated service address is returned.

[0133] Optionally, the above-mentioned method for publishing remote sensing images without tiling may further include: receiving a service loading request; if the service loading request is a single image service loading request, then obtaining persistently stored image core metadata, statistical information, and service configuration information, parsing the tile request to calculate the geographic range of the current request and the corresponding original image pixel coordinate range, adaptively selecting the optimal streaming reading strategy based on the pixel organization method in the core metadata to accurately obtain the target pixel data block, and dynamically rendering the pixel data block into a tile image of a specified size and returning it to the front end according to the rendering parameters specified by the user in real time or the service default; if the service loading request is an aggregated spatial service loading request, then obtaining the association relationship between the persistently stored aggregated spatial service and its source images, parsing the tile request to calculate the geographic range of the current request, determining one or more source images intersecting with the geographic range based on the association relationship, obtaining the tile images corresponding to each source image in parallel, and synthesizing the tile images into a canvas of the requested tile size in real time according to the rendering order recorded in the association relationship, and finally rendering and generating an aggregated tile image and returning it to the front end.

[0134] It is understood that the remote sensing image tiling-free publishing method provided in the embodiments of the present invention can achieve any function of the remote sensing image tiling-free publishing system described above. For the way the remote sensing image tiling-free publishing method implements each function, please refer to the remote sensing image tiling-free publishing system provided in the embodiments of the present invention. The method embodiment section will not be repeated here.

[0135] In the embodiments provided by this invention, it should be understood that the disclosed systems and methods can be implemented in other ways. The system embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the shown or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces. Indirect couplings or communication connections between systems or units may be electrical, mechanical, or other forms.

[0136] Furthermore, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0137] Furthermore, the functional modules in the various embodiments of the present invention can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0138] The above description is merely an embodiment of the present invention and is not intended to limit the scope of protection of the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A remote sensing image tiling-free publishing system, characterized in that, It includes a data support layer, a core business layer, and a web interface layer that are connected in sequence, wherein: The data support layer is used to provide a unified streaming read interface based on the adapter pattern for image data in the local file system and HDFS distributed file system to support the upper-layer business logic, and to persistently manage system metadata information. The core business layer is used to execute the entire process of image tile-free publishing and service. In the publishing stage, it realizes metadata extraction, statistical information estimation and thumbnail generation. In the service stage, it adaptively selects the optimal streaming reading strategy based on the pixel organization method to dynamically render and generate tile images, and supports the virtual aggregation of multi-source images into a unified service for real-time synthesis. The Web interface layer is used to receive service publishing requests, service loading requests, and thumbnail retrieval requests from clients, and forward the requests to the core business layer.

2. The remote sensing image tile-free publishing system according to claim 1, characterized in that, The data support layer includes: The data source abstraction layer is used to define standardized streaming read interfaces and dynamically instantiate corresponding adapters based on image data source type information using the factory pattern, so as to support the core business layer's unified streaming access to the local file system and HDFS distributed file system. The information management module is used to persistently manage the system metadata information; wherein, the system metadata information includes the core metadata of the image, statistical information, service configuration information, and the association between the aggregated spatial service and its source image; the core metadata includes image size, bit depth, number of bands, pixel organization method, and georeferenced information.

3. The remote sensing image tile-free publishing system according to claim 2, characterized in that, The core business layer includes: The workflow management module is used to standardize the core process of image publishing based on a template-based approach; the core process includes parsing metadata, calculating statistical information, storing information in the database, and generating thumbnails. The metadata parsing module, connected to the publishing workflow management module and the data support layer, is used to extract core metadata from the original image files through streaming reading during the image publishing stage; wherein, the core metadata includes image size, bit depth, number of bands, pixel organization method and georeferenced information; The statistical information estimation module, connected to the publishing workflow management module and the data support layer, is used to quickly estimate the statistical information of the image using a sparse sampling strategy during the image publishing stage; wherein, the statistical information includes the maximum value, minimum value and approximate percentile quantile of each band.

4. The remote sensing image tile-free publishing system according to claim 3, characterized in that, The release workflow management module is configured as follows: Receive service publishing requests; If the service publishing request is a single image publishing request, then the metadata parsing module and the statistical information estimation module are called in sequence to complete the metadata parsing and statistical information calculation, then a service entity is generated and stored in the information management module, and the thumbnail generation module is called to generate an image thumbnail, and finally the service address is returned; If the service publishing request is a multi-image publishing request, the metadata parsing module and the statistical information estimation module are called in a loop to complete the metadata parsing and statistical information calculation for each image, generate each image service entity and store it in the information management module, and call the thumbnail generation module to generate a thumbnail for each image. After all images have been processed, an aggregated spatial service entity is generated and stored in the information management module, and the thumbnail generation module is called again to generate an aggregated service thumbnail based on the combination of each image thumbnail, and finally the aggregated service address is returned.

5. The remote sensing image tile-free publishing system according to claim 3, characterized in that, The core business layer also includes: The dynamic tile generation and rendering module, connected to the data support layer, is used to respond to service loading requests during the image service phase. It calculates the corresponding original image pixel coordinate range in real time based on the tile coordinates and the core metadata, adaptively selects the optimal streaming reading strategy based on the pixel organization method to accurately obtain the target pixel data block, and dynamically renders the pixel data block to generate a tile image according to the rendering parameters specified by the user in real time or the service default.

6. The remote sensing image tile-free publishing system according to claim 5, characterized in that, The core business layer also includes: The multi-image dynamic aggregation module, connected to the dynamic tile generation and rendering module and the data support layer, is used to virtually aggregate multiple image files scattered in the local file system and the HDFS distributed file system into a unified aggregation space service. When a tile request for the aggregation space service is received, the dynamic tile generation and rendering module is called in parallel to obtain the tile images corresponding to each source image, and the aggregated tile image is synthesized in real time according to the rendering order recorded in the association relationship.

7. The remote sensing image tile-free publishing system according to claim 6, characterized in that, The dynamic tile generation and rendering module and the multi-image dynamic aggregation module are configured collaboratively as follows: Receive service loading request; If the service loading request is a single image service loading request, then the image core metadata, statistical information and service configuration information stored in the information management module are obtained, the tile request is parsed to calculate the geographical range of the current request and the corresponding original image pixel coordinate range, and the optimal streaming reading strategy is adaptively selected based on the pixel organization method to accurately obtain the target pixel data block from the data source abstraction layer, and the pixel data block is dynamically rendered into a tile image of specified specifications and returned to the front end according to the rendering parameters specified by the user in real time or the default rendering parameters of the service. If the service loading request is an aggregated spatial service loading request, then the association relationship between the aggregated spatial service and its source images stored in the information management module is obtained, the tile request is parsed to calculate the geographical range of the current request, one or more source images intersecting with the geographical range are determined based on the association relationship, the dynamic tile generation and rendering module is called in parallel to obtain the tile images corresponding to each source image, and the tile images are synthesized in real time onto the canvas of the requested tile size according to the rendering order recorded in the association relationship, and finally the aggregated tile image is rendered and returned to the front end.

8. The remote sensing image tile-free publishing system according to claim 3, characterized in that, The core business layer also includes: A thumbnail generation module, connected to the publishing workflow management module and the data support layer, is used to generate thumbnails for single image services during the image publishing phase, and to calculate the maximum bounding rectangle of all source images and generate aggregate service thumbnails based on a hierarchical projection compositing strategy during the publishing of multi-image aggregation services; and The image processing tools module provides low-level image processing functions, including rearranging pixel data according to band index, encoding pixel data blocks into a specified image file format, and storing the generated image file to a specified location.

9. A method for publishing remote sensing images without tiling in a system as described in any one of claims 1 to 8, characterized in that, The method includes: Receive service publishing requests; If the service publishing request is a single image publishing request, the core metadata is extracted from the original image file through streaming reading, the image statistics are quickly estimated using a sparse sampling strategy, a service entity is generated and persistently stored, an image thumbnail is generated, and finally the service address is returned. If the service publishing request is a multi-image publishing request, then the above metadata extraction and statistical information estimation operations are executed in a loop to process each image one by one, generate each image service entity and store it persistently, and generate a thumbnail for each image one by one. After all images have been processed, an aggregated space service entity is generated and stored persistently, and an aggregated service thumbnail is generated based on the combination of each image thumbnail, and finally the aggregated service address is returned.

10. The method for publishing remote sensing images without tiling according to claim 9, characterized in that, The method further includes: Receive service loading request; If the service loading request is a single image service loading request, then the persistently stored image core metadata, statistical information and service configuration information are obtained, the tile request is parsed to calculate the geographical range of the current request and the corresponding original image pixel coordinate range, the optimal streaming reading strategy is adaptively selected based on the pixel organization method in the core metadata to accurately obtain the target pixel data block, and the pixel data block is dynamically rendered into a tile image of specified specifications and returned to the front end according to the rendering parameters specified by the user in real time or the default rendering parameters of the service. If the service loading request is an aggregated space service loading request, then the association between the persistently stored aggregated space service and its source images is obtained, the tile request is parsed to calculate the geographical range of the current request, one or more source images intersecting with the geographical range are determined based on the association, the tile images corresponding to each source image are obtained in parallel, and the tile images are synthesized in real time onto the canvas of the requested tile size according to the rendering order recorded in the association, and finally the aggregated tile image is rendered and returned to the front end.

Images