Structured remote sensing report generation system, method and equipment based on syntax tree and medium

Abstract

The invention discloses a structured remote sensing report generation system, method and device based on a syntax tree and a medium, the system comprises a user input layer, a core processing layer and an output and verification layer, the system receives user demand parameters and input data, automatically analyzes the demand parameters and standardizes the input data; generating a syntax tree structure conforming to industry specifications, and driving the large model to obtain a first draft of a report under the constraint of the syntax tree; and performing automatic verification and manual intervention correction on the first draft of the report to form a professional remote sensing report. According to the method, the multi-dimensional syntax tree architecture is constructed, and the hierarchical syntax analysis algorithm and the domain knowledge graph are combined, so that semantic association of the multi-modal remote sensing data and accurate mapping of terminologies are realized, automatic generation of a long document report meeting remote sensing geographic industry specifications is supported, the structured degree and logic continuity of text generation are improved, and the text generation efficiency is improved. The method can be widely applied to the fields of geological monitoring analysis, environment evaluation report generation and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing report generation, and in particular to a syntax tree-based structured remote sensing report generation system, method, device and medium. Background Art

[0002] Large language models can significantly improve office efficiency by assisting with long document generation. However, remote sensing reports, as a type of report document with high professional and structural requirements, still face the limitations of traditional long document generation technology in practical applications. Current automated methods for remote sensing ecological research reports have significant shortcomings in terms of professional knowledge accuracy and structural rigor:

[0003] 1) The structure-content decoupling defect: a deep association mechanism between the document grammatical structure and professional knowledge has not been established, resulting in a mismatch between the chapter level and the ecological analysis granularity, and a lack of automatic verification of spatial description statements and GIS layer attributes. 2) The lack of dynamic adaptation capabilities: the existing system is unable to cope with the data-driven structural variation of remote sensing ecological research. When multi-source heterogeneous data is input, it is unable to autonomously generate fusion analysis chapters. 3) The verification system is imperfect and lacks an automated verification mechanism for professional documents. Specifically, term consistency verification relies solely on keyword matching and ignores contextual semantics; spatiotemporal logic error detection is missing, and multimodal feature fusion is insufficient; the generation processes of text, charts, spatial data, and remote sensing image features are independent of each other, and there is a lack of cross-modal consistency verification mechanism.

[0004] These shortcomings restrict the large-scale application of remote sensing technology in ecological assessment. Therefore, a structured remote sensing report generation system is needed. By building a syntax tree architecture, combining a hierarchical syntax parsing algorithm with a domain knowledge graph, this system can accurately map the semantic associations of multimodal remote sensing data with specialized terminology, and automatically generate remote sensing image report documents that comply with remote sensing and geographic industry standards. Summary of the Invention

[0005] In response to the above-mentioned problems, the purpose of the present invention is to provide a structured remote sensing report generation system, method, device and medium based on a syntax tree, which automatically generates report documents that comply with remote sensing geographic industry standards by constructing a multi-layer syntax tree architecture.

[0006] The embodiments of the present invention provide a syntax tree-based structured remote sensing report generation system, method, device and medium.

[0007] A first aspect: A structured remote sensing report generation system based on a syntax tree, comprising:

[0008] User input layer, used to receive and integrate user demand parameters and input data;

[0009] The core processing layer is used to associate input data with the knowledge base, generate a syntax tree structure that complies with industry standards based on required parameters, and drive the large model to complete report content filling and chart generation under the constraints of the syntax tree to obtain the first draft of the report;

[0010] The output and verification layer is used to automatically verify and manually modify the draft report to form a professional remote sensing report.

[0011] Optionally, the core processing layer includes:

[0012] Syntax tree construction module, used to build report chapter structure, chapter node expansion, semantic parsing and multimodal verification;

[0013] Dynamic parsing engine module, used for contextual reasoning and grammatical conflict resolution;

[0014] Intelligent generation module, used for large model prompt word generation and multimodal content generation.

[0015] Optionally, the core framework of the professional remote sensing report includes:

[0016] Metadata constraint unit, used to define basic report information and data boundaries to ensure content consistency;

[0017] Core chapter units are used to organize the main content of the report, covering the complete process from background to analysis;

[0018] The conclusion and recommendation unit is used to refine core findings and guide practice, thereby enhancing the practicality of the report.

[0019] The second aspect: a method for generating a structured remote sensing report based on a syntax tree, comprising:

[0020] Optionally, the structured remote sensing report generation method of the structured remote sensing report generation system comprises the steps of:

[0021] S1. Receive user requirement parameters and input data, automatically parse the requirement parameters and standardize the input data;

[0022] S2. Associate the input data with the knowledge base, generate a syntax tree structure that complies with industry standards based on the required parameters, and drive the large model to complete the report content filling and chart generation under the constraints of the syntax tree to obtain the first draft of the report;

[0023] S3. Carry out automated verification and manual intervention correction on the draft report to form a professional remote sensing report.

[0024] Optionally, the S1 includes:

[0025] S11, receiving user input data and performing standardization processing;

[0026] S12. Receive the user's submitted requirements, and the system automatically analyzes the requirement parameters;

[0027] The input data includes remote sensing images, vector data and / or auxiliary data, and the submission requirements include the research area, ecological indicators and / or analysis methods.

[0028] Optionally, the syntax tree adopts a four-layer nested structure of chapter layer, chapter layer, paragraph layer and sentence layer, supporting dynamic expansion of layer nodes and multi-scale nested analysis.

[0029] Optionally, the S2 includes:

[0030] S21, associating input data with the knowledge base through a dynamic parsing engine;

[0031] S22. Match the report framework according to the required parameters and dynamically expand the layer nodes;

[0032] S23. Analyze the correlation of demand parameter data and verify the spatiotemporal logic;

[0033] S24. Perform structured analysis on the demand parameters and intelligently generate a syntax tree structure that complies with industry standards;

[0034] S25. Drive the large model to automatically complete report content filling and chart generation under the constraints of the syntax tree to obtain the first draft of the report.

[0035] Optionally, the S3 includes:

[0036] S31. The system automatically verifies the report draft and outputs multi-format report documents;

[0037] S32. Manual intervention to modify the report document to form a professional remote sensing report that can be delivered directly;

[0038] Among them, multi-format report documents include standardized documents and / or interactive reports; automated verification includes structural integrity checking and data consistency verification; manual correction includes online revision and automatic synchronization to the syntax tree.

[0039] A third aspect: An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method provided in the second aspect are implemented.

[0040] A fourth aspect: A non-transitory computer-readable storage medium having a computer program stored thereon, which implements the steps of the method provided in the second aspect when executed by a processor.

[0041] Beneficial effects of the present invention:

[0042] 1. This invention builds a syntax tree architecture, combines a hierarchical syntax parsing algorithm with a domain knowledge graph, and realizes the semantic association of multimodal remote sensing data and precise mapping of professional terms. It supports the automatic generation of long document reports that comply with remote sensing geographic industry standards, improves the structured degree and logical coherence of text generation, and can be widely used in geological monitoring analysis, environmental assessment report generation and other fields.

[0043] 2. The syntax tree designed in this invention covers all levels from the overall report chapter to the specific sentence, and maintains flexibility to adapt to different analysis needs, realizes multi-scale analysis, supports dynamic expansion and nested structure, and combines professional knowledge in the field of remote sensing (such as NDVI calculation and land use classification) with natural language large model processing technology to ensure that the generated text is effectively combined with elements such as charts and spatial data, so that chart generation and text description are updated synchronously.

[0044] 3. During the report generation process, the present invention automatically adjusts the report content structure according to the input data and generates the corresponding report description. At the same time, it introduces a verification mechanism to check the consistency of terminology, spatiotemporal logic and data integrity to ensure that the generated text is effectively combined with elements such as charts and spatial data.

[0045] 4. The method of the present invention has been tested and applied in multiple scenarios, and can shorten the remote sensing report preparation cycle by 82%. Relying on the improved efficiency of cross-departmental data collaboration in the report, the time required for data integration from meteorological, land, ecological and other departments can be shortened from 6.2 hours to 1.1 hours through the structured data anchors of the syntax tree. Through the structured constraints of the syntax tree and multimodal intelligent generation technology, it significantly solves the industry pain points of loose structure, disconnected data and high manual dependence in remote sensing professional reports, providing key technical support for the intelligent transformation of geographic information services. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a schematic diagram of the structure of the structured remote sensing report generation system of the present invention;

[0047] Figure 2 This is a schematic diagram of the core framework structure of the remote sensing report of the present invention;

[0048] Figure 3 Schematic diagram of the flow of the structured remote sensing report generation method of the present invention;

[0049] Figure 4 Schematic diagram of the structure of the electronic device of the present invention. DETAILED DESCRIPTION

[0050] The following describes embodiments of the present invention in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar symbols throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended only to explain the present invention and are not to be construed as limiting the present invention.

[0051] The current generation of long document reports for remote sensing images has shortcomings. The existing template method has a fixed hierarchy and cannot handle multi-scale analysis; professional knowledge in the remote sensing field (such as NDVI calculation and land use classification) is not sufficiently integrated with natural language processing technology; the content structure cannot be automatically adjusted according to the input data, and terminology consistency, spatiotemporal logic, and data integrity cannot be checked.

[0052] In order to solve the above problems, the present invention provides a structured remote sensing report generation system based on syntax tree. Figure 1 This is a schematic diagram of the structure of a structured remote sensing report generation system provided by an embodiment of the present invention. The system includes: a user input layer, a core processing layer, and an output and verification layer.

[0053] The user input layer includes a demand parameter module and a raw data module. The demand parameter module is used to receive and integrate user demand parameters, which include research areas and ecological indicators. The raw data module is used to receive and integrate user input data, which includes remote sensing images and vector data.

[0054] The core processing layer is used to associate input data with the knowledge base, generate a syntax tree structure that complies with industry standards based on required parameters, and drive the large model to complete report content filling and chart generation under the constraints of the syntax tree to obtain a draft report.

[0055] like Figure 1 As shown, the core processing layer includes a syntax tree building module, a dynamic parsing engine module and an intelligent generation module.

[0056] The syntax tree construction module is used to construct the report chapter structure, chapter node expansion, semantic parsing and multimodal verification.

[0057] The syntax tree can adopt a four-layer nested syntax tree structure at the chapter level, paragraph level, and sentence level, and supports dynamic expansion of layer nodes and multi-scale nested analysis (such as mixed analysis of regional-level ecological assessment and plot-level land use).

[0058] When constructing the syntax tree, industry standards are used to form a constraint generation mechanism, and remote sensing industry standards (such as "GB / T35645-2017 Specifications for Compilation of Geospatial Analysis Reports") are converted into syntax tree layer node attributes, and then the report generation process of the large language model (such as LLM) is constrained through structured prompt engineering.

[0059] The generated report implements a multimodal verification closed loop, designs cross-modal consistency verification rules, and realizes the collaborative generation and cross-verification of text, statistical tables, and spatial layers.

[0060] The dynamic parsing engine module is used for associative context reasoning and grammatical conflict resolution; the intelligent generation module is used for large model prompt word generation and multimodal content generation; the output and verification layer is used to automatically verify the draft report and manually intervene to correct it, forming a professional remote sensing report.

[0061] The constraints imposed by the syntax tree on large models include: metadata constraints for the entire report chapter, synchronized updates of figure generation and text analysis (corresponding figures should be automatically re-rendered based on the time range in the text), dynamic table generation, and the definition of statistical table column dimensions by the "Statistical Table" structure node attributes in the syntax tree results and analysis (for example, a "land use change matrix" must include three columns: transfer area, rate of change, and confidence interval). Table cell value ranges are checked (for example, the NDVI value range is limited to [-1, 1], with out-of-limit values triggering an alarm). Spatial descriptions are also bound to layers. For example, the "spatial distribution" constraint in the results analysis requires that all directional terms in the text (such as "southeast of the study area") must be associated with the spatial partitions of the GIS layer.

[0062] like Figure 3 As shown in the figure, the syntax tree structure defines the core framework of a professional remote sensing report. The framework is divided into three modules: metadata constraint unit, core chapter unit, and conclusion and recommendation unit. Through a layered and nested design, the report content is ensured to comply with industry standards and have rigorous logic, while also supporting dynamic expansion and multimodal content binding.

[0063] The metadata constraint unit is used to define the basic information and data boundaries of the report to ensure content consistency.

[0064] The core chapter unit organizes the main content of the report, covering the complete process from background to analysis; the introduction includes the research background and technical route, automatically generates flow charts and instruction tables, and binds the analysis methods; data and methods include data sources and analysis models, dynamically inserts formulas and algorithm descriptions; result analysis includes spatial distribution and statistical tables, and the numerical range of table columns is automatically checked according to industry standards.

[0065] Conclusion and suggestion: The role of the unit is to refine core findings and guide practice, thereby enhancing the practicality of the report; core findings automatically extract key indicators and trigger logical verification; the countermeasure library matches the associated policy database and generates a targeted terminology manual to confirm mandatory terminology standardization and cite industry standard documents.

[0066] The output and verification layer includes a verification module, a document format module and a manual correction interface module. The verification module verifies the report structure and report text data; the document formatting is used to output reports in PDF, Word and PPT formats; and the manual correction interface module is used to manually revise reports.

[0067] The structured remote sensing report generation system of the present invention is adopted to realize the full-process automated generation of professional remote sensing reports by constructing a composite architecture that integrates domain knowledge, multimodal constraints and generative AI.

[0068] like Figure 2 As shown, based on the above system, the present invention also discloses a method for generating a structured remote sensing report based on a syntax tree, comprising the steps of:

[0069] S1. Receive user requirement parameters and input data, automatically parse the requirement parameters and standardize the input data.

[0070] This step mainly receives user input data and performs standardized processing. At the same time, it receives user-submitted requirements, and the system automatically parses the requirement parameters. After the user submits the requirements, the system automatically parses the requirement parameters and standardizes the input data (such as unified coordinate system and time series alignment) to provide structured input for subsequent processing.

[0071] Input data include remote sensing images (such as satellite / UAV data), vector data (such as administrative divisions, ecological protection zone boundaries) and / or auxiliary data (such as meteorological data, topographic data), etc. Submission requirements include study area (such as geographic boundaries, coordinate range), ecological indicators (such as vegetation cover, land use type) and / or analysis methods (such as trend analysis, classification model), etc.

[0072] S2. Associate the input data with the knowledge base, generate a syntax tree structure that complies with industry standards based on the required parameters, and drive the large model to complete the report content filling and chart generation under the constraints of the syntax tree to obtain the first draft of the report.

[0073] This step associates the input data with the knowledge base through a dynamic parsing engine, then matches the report framework (such as the ecological assessment template) according to the demand parameters, and dynamically expands the layer nodes (such as the newly added "multi-source data fusion analysis" module), parses the correlation of the demand parameter data, verifies the spatiotemporal logic, and constructs and expands the syntax tree.

[0074] Then, a structured analysis is performed on the required parameters, including semantic parsing and reasoning, analyzing data relevance (such as the relationship between vegetation index and climate factors), verifying spatiotemporal logic (avoiding time inversion or spatial range conflicts), and intelligently generating a syntax tree structure that complies with industry standards.

[0075] Finally, the large model is driven to call the associated knowledge base under the constraints of the syntax tree, automatically complete the report content filling and chart generation, and intelligently generate the first draft of the report, including the large model generating text content (generating professional descriptions under the constraints of the syntax tree), multimodal collaborative generation (automatically inserting matching charts and statistical tables), etc.

[0076] S3. Carry out automated verification and manual intervention correction on the draft report to form a professional remote sensing report.

[0077] This step is mainly to generate standardized documents and ensure quality. After the first draft is generated, the system automatically performs logic and data verification, outputs multi-format documents for user review, and allows manual intervention and correction, ultimately forming a professional remote sensing report that can be delivered directly.

[0078] Among them, multi-format report documents include standardized documents (PDF, Word, PPT) and / or interactive reports (supporting map interaction and data capture); automated verification includes structural integrity checks (chapter title hierarchy, numbering rules) and data consistency checks (matching of text descriptions and chart values); manual corrections include support for expert online revisions (marking errors, supplementary explanations) and automatic synchronization of revised content to the syntax tree (to ensure the uniformity of subsequent versions).

[0079] The method of the present invention realizes a closed loop from the demand input end to the verification output end, covering the entire process of report generation from demand input to verification output; it is driven by industry specifications, and the syntax tree is deeply bound to industry specifications to ensure professionalism; at the same time, it realizes human-computer collaboration, seamlessly connects automatic generation and manual correction, balances efficiency and accuracy, and significantly improves the standardization and reliability of remote sensing report generation, making it suitable for various scenarios such as ecological monitoring and disaster assessment.

[0080] Application example: Generation of the "Yangtze River Delta Vegetation Change Monitoring Report".

[0081] 1. Input parameters:

[0082] {

[0083] "Study Area": "Yangtze River Delta Urban Agglomeration",

[0084] "Data Source": ["Landsat8 2020-2023 NDVI", "MODIS Land Surface Temperature"],

[0085] "Analysis method": ["Sen slope estimation", "MK trend test"]

[0086] }

[0087] 2. Implementation steps:

[0088] Syntax tree initialization:

[0089] Load the "Vegetation Change Monitoring" template, and the root node will automatically bind to the GeoJSON boundary of the Yangtze River Delta;

[0090] Dynamically expand the "Surface Temperature-NDVI Correlation Analysis" subchapter (due to detection of MODIS data);

[0091] Constrained generation:

[0092] When LLM generates the "Result Analysis" paragraph, the system intercepts the following error:

[0093] [Error] The text "2022 NDVI peak 0.85" conflicts with the calculated result 0.812 and should be automatically replaced with "0.81(±0.02)";

[0094] Insert statistical table TB_001, including columns: year / NDVI mean / change slope / p value;

[0095] Multimodal output:

[0096] Generate heat map HT_003 (spatial distribution of NDVI changes), and automatically match the color scheme of the Jungle color system;

[0097] The text "Hangzhou Bay area is significantly degraded" is associated with the ROI area of HT_003 (spatial matching degree 99.3%);

[0098] Verification and correction:

[0099] If the system detects that "data coverage in 2020 is less than 95%", a yellow warning should be triggered;

[0100] Automatic supplementary explanation: "Affected by cloud coverage, valid data in 2020 accounted for 93.2%";

[0101] Output:

[0102] Generate a 62-page PDF report (including 12 statistical tables and 9 thematic maps) and pass the "HJ 1152-2020" standard review;

[0103] Manual correction takes only 8 minutes (traditional methods take 2.5 hours).

[0104] This invention significantly solves the industry pain points of loose structure, disconnected data, and high manual dependence in remote sensing professional reports through the structured constraints of syntax trees and multimodal intelligent generation technology, providing key technical support for the intelligent transformation of geographic information services.

[0105] The present invention also provides an electronic device, Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 4As shown, the electronic device may include: a processor, a communications interface, a memory, and a communication bus, wherein the processor, the communications interface, and the memory communicate with each other via the communication bus. The processor may call logic instructions in the memory, for example, to execute the following method:

[0106] S1. Receive user requirement parameters and input data, automatically parse the requirement parameters and standardize the input data;

[0107] S2. Associate the input data with the knowledge base, generate a syntax tree structure that complies with industry standards based on the required parameters, and drive the large model to complete the report content filling and chart generation under the constraints of the syntax tree to obtain the first draft of the report;

[0108] S3. Carry out automated verification and manual intervention correction on the draft report to form a professional remote sensing report.

[0109] In addition, the logical instructions in the above-mentioned memory can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0110] An embodiment of the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method provided in each of the above embodiments is implemented, for example, including:

[0111] S1. Receive user requirement parameters and input data, automatically parse the requirement parameters and standardize the input data;

[0112] S2. Associate the input data with the knowledge base, generate a syntax tree structure that complies with industry standards based on the required parameters, and drive the large model to complete the report content filling and chart generation under the constraints of the syntax tree to obtain the first draft of the report;

[0113] S3. Carry out automated verification and manual intervention correction on the draft report to form a professional remote sensing report.

[0114] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0115] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0116] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A structured remote sensing report generation system based on syntax tree, characterized in that: include: User input layer, used to receive and integrate user demand parameters and input data; The core processing layer is used to associate input data with the knowledge base, generate a syntax tree structure that complies with industry standards based on required parameters, and drive the large model to complete report content filling and chart generation under the constraints of the syntax tree to obtain the first draft of the report; The output and verification layer is used to automatically verify and manually modify the draft report to form a professional remote sensing report.

2. The structured remote sensing report generation system according to claim 1, characterized in that: The core processing layer includes: Syntax tree construction module, used to build report chapter structure, chapter node expansion, semantic parsing and multimodal verification; Dynamic parsing engine module, used for contextual reasoning and grammatical conflict resolution; Intelligent generation module, used for large model prompt word generation and multimodal content generation.

3. The structured remote sensing report generation system according to claim 1, characterized in that: The core framework of the professional remote sensing report includes: Metadata constraint unit, used to define basic report information and data boundaries to ensure content consistency; Core chapter units are used to organize the main content of the report, covering the complete process from background to analysis; The conclusion and recommendation unit is used to refine core findings and guide practice, thereby enhancing the practicality of the report.

4. The structured remote sensing report generation system according to any one of claims 1 to 3, characterized in that: The structured remote sensing report generation method of the structured remote sensing report generation system comprises the following steps: S1, receiving user demand parameters and input data, automatically parsing demand parameters and standardizing input data; S2. Associate the input data with the knowledge base, generate a syntax tree structure that complies with industry standards based on the required parameters, and drive the large model to complete the report content filling and chart generation under the constraints of the syntax tree to obtain the first draft of the report; S3. Carry out automated verification and manual intervention correction on the draft report to form a professional remote sensing report.

5. The method for generating a structured remote sensing report according to claim 4, wherein: Said S1 comprises: S11, receiving user input data and performing standardization processing; S12. Receive the user's submitted requirements, and the system automatically analyzes the requirement parameters; The input data includes remote sensing images, vector data and / or auxiliary data, and the submission requirements include the research area, ecological indicators and / or analysis methods.

6. The method for generating a structured remote sensing report according to claim 4, wherein: The grammatical tree adopts a four-layer nested structure of chapter layer, chapter layer, paragraph layer and sentence layer, and supports dynamic expansion of layer nodes and multi-scale nested analysis.

7. The method for generating a structured remote sensing report according to claim 6, wherein: The S2 includes: S21, associating input data with the knowledge base through a dynamic parsing engine; S22. Match the report framework according to the required parameters and dynamically expand the layer nodes; S23. Analyze the correlation of demand parameter data and verify the spatiotemporal logic; S24. Perform structured analysis on the demand parameters and intelligently generate a syntax tree structure that complies with industry standards; S25. Drive the large model to automatically complete report content filling and chart generation under the constraints of the syntax tree to obtain the first draft of the report.

8. The method for generating a structured remote sensing report according to claim 4, wherein: The S3 includes: S31. The system automatically verifies the report draft and outputs multi-format report documents; S32. Manual intervention to modify the report document to form a professional remote sensing report that can be delivered directly; Among them, multi-format report documents include standardized documents and / or interactive reports; automated verification includes structural integrity checking and data consistency verification; manual correction includes online revision and automatic synchronization to the syntax tree.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the structured remote sensing report generation method according to any one of claims 4 to 8 are implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the structured remote sensing report generation method according to any one of claims 4 to 8 are implemented.

Images