A device and method for fusing NOTAM and AIP dynamic and static information analysis and display

Abstract

The application discloses a device and method for fusing NOTAM and AIP dynamic and static information information analysis and display, wherein the device is used for: batch analyzing a high safety correlation AIP PDF library according to an AIRAC cycle, structurally storing and completing position mapping; analyzing a chart PDF based on a Qwen3-VL multi-modal large model, and analyzing a text table PDF by using PaddleOCR-VL; accessing a NOTAM message in real time, analyzing the NOTAM message by using a self-developed large model, screening out key announcements of airports, runways and navigation stations endangering flight safety, matching the corresponding AIRAC cycle AIP document path and content point according to the announcement information, calling a target PDF through an interface, and marking and displaying announcement risk characteristics at corresponding elements. The application realizes automatic matching of dynamic and static aviation information, risk visualization, reduces manual operation, and improves intelligence and expansion capacity of information.

Description

Technical Field

[0001] This invention relates to the field of aviation intelligence information processing technology, and in particular to a device and method for analyzing and displaying dynamic and static intelligence information that integrates NOTAM and AIP. Background Technology

[0002] Aeronautical Information Processing (AIP) is a standardized aeronautical information document mandated by the International Civil Aviation Organization (ICAO). It contains static basic information such as airports, runways, navigation facilities, airspace, and routes, typically published in PDF format, and includes numerous charts, tables, and technical texts. Notifications to Airmen (NOTAMs), on the other hand, are dynamic, ad hoc information documents used to notify aviation users of changes in aviation facilities, services, procedures, or hazards, such as navigation beacon malfunctions or runway closures. Flight safety heavily relies on pilots and dispatchers' ability to obtain and understand this information promptly and accurately, requiring reference to both dynamic NOTAMs and static AIPs. However, AIP PDFs and NOTAMs still present parsing challenges: AIP charts contain complex graphic symbols (such as navigation beacons, runways, holding points) and text annotations; table structures are complex, and common issues like cross-row / column crossings and merged cells render traditional parsing tools ineffective; NOTAMs are published in natural language text format, with vague and variable descriptions, lacking a unified format, making it difficult for traditional keyword matching to accurately extract affected facilities and their status, and even more difficult to automatically correlate with the graphic locations in the AIP. This also makes:

[0003] Information fragmentation and heavy reliance on manual intervention: AIPs and NOTAMs are usually reviewed separately. Pilots or dispatchers need to manually locate the affected NOTAMs in the PDF version of the AIP and then mentally or manually mark them. This method is inefficient and prone to omissions or misplacements, especially when there are a large number of NOTAMs.

[0004] Lack of integrated display of dynamic and static information: Existing electronic flight bag (EFB) or aeronautical information systems usually display AIP and NOTAM separately, and cannot intuitively overlay NOTAM information on the corresponding AIP chart or table page, so users still need to compare them themselves.

[0005] Therefore, there is an urgent need for a method that can automatically parse AIP and NOTAM, accurately locate dynamic intelligence to the corresponding position in static data, and achieve integrated visualization display, so as to significantly reduce the workload of dispatchers or pilots, effectively avoid human error, and improve flight safety. Summary of the Invention

[0006] The main objective of this invention is to provide a device for analyzing and displaying dynamic and static intelligence information that integrates NOTAM and AIP.

[0007] Another objective of this invention is to propose a method for analyzing and displaying dynamic and static intelligence information from both NOTAM and AIP.

[0008] The third objective of this invention is to provide an electronic device.

[0009] A fourth objective of this invention is to provide a non-transitory computer-readable storage medium.

[0010] To achieve the above objectives, a first aspect of the present invention provides a device for analyzing and displaying dynamic and static intelligence information integrating NOTAM and AIP, comprising:

[0011] The AIP PDF parsing and secondary processing module is used to periodically parse and process AIP PDF sets that are highly related to flight safety according to the AIRAC cycle, and complete the structured result storage and location mapping. Among them, the Qwen3-VL multimodal large model is used to identify and parse aeronautical chart PDFs, and the PaddleOCR-VL tool is used to identify and parse table and text PDFs.

[0012] The real-time notam parsing module is used to receive notam messages in real time. It uses a self-trained large model to complete message parsing and filter out notams related to airports, runways, and navigation beacons that are of concern to airlines and pose a risk to flight safety.

[0013] The module that matches the parsing results of notam and AIP is used to determine the AIRAC cycle, target file, storage path and corresponding content location of the AIP to be matched based on the notam parsing results.

[0014] Displaying affected PDF marker units is used to load the AIP PDF files corresponding to the affected features of the notam in the system interface, and to complete the visualization marking and information display of the notam features at the corresponding feature positions in the PDF.

[0015] Optionally, the periodic AIP PDF parsing and secondary processing module includes:

[0016] The AIP PDF parsing scope definition unit is used to define the list of airports operated by the airline, the countries of AIPs to be used and the corresponding AIP versions, in combination with the actual operational needs of different airlines. It filters the target AIP PDF files to be parsed by using regular expressions that adapt to the AIP directories and naming rules of various countries.

[0017] AI tools are used to identify PDF units. During AIRAC cycle switching, based on the filtered PDF range, the system distinguishes between aeronautical chart PDFs and text / table PDFs. The system calls Qwen3-VL multimodal large model and PaddleOCR-VL tools to complete the identification and parsing, and outputs navigation station information, markdown format content and position mapping files according to the specified specifications.

[0018] The secondary processing unit for markdown recognition results is used to perform hierarchical error correction, chapter segmentation, and tag correction on the original markdown file output by PaddleOCR-VL. At the same time, it uses a table normalization processing algorithm to unify the parsing format of various irregular tables.

[0019] The structured storage unit is used to classify and structure the AIP parsing data after secondary processing, storing text, tables, and navigation information into corresponding databases, and synchronously linking and storing chapter information, PDF positioning parameters, airport four codes, and all traceability information of file paths.

[0020] Optional, AIP resolution range definition unit, used for:

[0021] Define the airport dictionary for each airline based on their specific needs;

[0022] Define multiple AIP country codes for each airline based on their needs;

[0023] Based on the determined AIP country, match the corresponding AIP version as the AIP_version to be processed;

[0024] Configure dedicated regular expressions for different AIP_versions, and combine them with the AIP directory structure and file naming rules to filter out core PDF files related to airport, runway, arrival and departure, and approach procedures, while eliminating irrelevant files.

[0025] Optionally, the table normalization algorithm in the secondary processing unit of the markdown recognition result is used for:

[0026] Illustrate all tables in the document one by one according to the table labels;

[0027] Extract the title information from the top of the table and assign a unique title attribute to the table.

[0028] Identify tables containing merged cells, complete the cell labels, and standardize the table format;

[0029] Transpose a vertical header table into a standard horizontal header format;

[0030] Convert tables with multiple columns and repeated columns into a single-column field value format;

[0031] Remove redundant serial number columns from the table;

[0032] Identify duplicate tables spanning multiple pages, remove duplicate headers, and then merge the tables.

[0033] Optionally, the real-time Notam parsing module includes:

[0034] The Notam unit is filtered and parsed to receive Notam messages in real time. It uses Q codes to filter out core messages of airport, runway, and navigation station types. It uses a self-trained large model to perform structured parsing of the semantics of message D and E items and stores them in the database according to their categories.

[0035] The retained message unit is used to perform secondary filtering on the structured parsed messages, accurately identify risk messages related to airports, runways, and navigation beacons that are closed, open, or restricted, match the AIP association configuration table to determine the corresponding associated data category, and filter out invalid messages that have no impact on flight safety.

[0036] Optionally, the module for parsing and processing the matching notam and AIP includes:

[0037] The AIP periodic unit to be matched is determined based on the notam, which is used to extract the effective and expiration time of the notam message, match the corresponding AIRAC period through the periodic configuration table, and lock the target database schema of the AIP to be matched.

[0038] The content matching unit is used to prioritize matching the AIP structured data table based on the NOTAM message type, airport four codes, and runway / navigation beacon information. If the matching fails, it will perform a second matching of the text data and finally extract the associated PDF file path, page number, coordinate marker position, and full parsing fields.

[0039] Optionally, the unit for displaying the affected PDF tagging is used for:

[0040] Retrieve the corresponding target AIP PDF file based on the matching results;

[0041] Load PDF files into the system interface, and display the matching results of multiple files in a tabbed format.

[0042] Based on the matched page number and coordinate information, the interface is precisely positioned to the corresponding element location in the PDF;

[0043] Category-specific Notam icons are drawn at the corresponding locations of the marked positions to display the structured data and timely information of Notam in the form of cards. The Notam text can be manually shown or hidden, realizing a visual association between static AIP intelligence and dynamic NOTAM risks.

[0044] To achieve the above objectives, a second aspect of the present invention proposes a method for parsing and displaying dynamic and static intelligence information from both NOTAM and AIP, comprising:

[0045] The AIP PDF set, which is highly correlated with flight safety, is periodically parsed and processed according to the AIRAC cycle, and the structured results are stored and mapped. Among them, the Qwen3-VL multimodal large model is used to identify and parse aeronautical chart PDFs, and the PaddleOCR-VL tool is used to identify and parse table and text PDFs.

[0046] The system receives NOTAM messages in real time, uses a self-trained large model to parse the messages, and filters out NOTAMs related to airports, runways, and navigation beacons that are of concern to airlines and pose a risk to flight safety.

[0047] Based on the notam parsing results, determine the AIRAC cycle, target file, storage path, and corresponding content location of the AIP to be matched;

[0048] Load the AIP PDF file corresponding to the notam-affected features in the system interface, and complete the visualization marking and information display of notam features at the corresponding feature positions in the PDF.

[0049] To achieve the above objectives, a third aspect of this application provides an electronic device, including a processor and a memory; wherein the processor runs a program corresponding to the executable program code stored in the memory to implement the method described in the second aspect.

[0050] To achieve the above objectives, a fourth aspect of this application provides a non-transitory computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the method described in the second aspect.

[0051] The embodiments of this invention have the following beneficial effects: By automatically matching the real-time parsed NOTAMs with the AIP data stored in a structured manner according to the AIRAC cycle at the element level, combined with the accurate recognition of PDFs by a multimodal large model and the semantic parsing of messages by a self-trained large model, the difficulty of manual review and comparison is greatly reduced; at the same time, irrelevant notices are eliminated through a multi-level filtering mechanism, allowing operators to focus on key risk points, and the notice details are displayed intuitively in the form of icons and cards at the corresponding element positions on the AIP page, which significantly enhances situational awareness capabilities; the periodic archiving storage method provides a reliable foundation for historical tracing and operational auditing, shortens the overall dispatch and release preparation time, improves the intelligence level of aeronautical intelligence services, and the framework has good scalability, which can adapt to the future needs of integrating more types of intelligence. Attached Figure Description

[0052] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0053] Figure 1 This is a schematic diagram of the structure of a device for analyzing and displaying dynamic and static intelligence information integrating NOTAM and AIP, as proposed in an embodiment of the present invention.

[0054] Figure 2 This is a schematic diagram of the structure of the AIP PDF parsing and secondary processing module in an embodiment of the present invention;

[0055] Figure 3 This is a flowchart illustrating the workflow of the AIP parsing range definition unit in an embodiment of the present invention.

[0056] Figure 4 This is a flowchart illustrating the workflow of using AI tools to recognize PDF units in an embodiment of the present invention.

[0057] Figure 5 This is a flowchart of the secondary processing unit for Markdown recognition results in an embodiment of the present invention;

[0058] Figure 6 This is a flowchart of the table normalization processing algorithm according to an embodiment of the present invention;

[0059] Figure 7 This is a flowchart illustrating the workflow of the structured storage unit according to an embodiment of the present invention;

[0060] Figure 8 This is a flowchart illustrating the process of storing data into a table according to an embodiment of the present invention;

[0061] Figure 9 This is a flowchart illustrating the process of storing text into a data table according to an embodiment of the present invention;

[0062] Figure 10 This is a schematic diagram of the real-time parsing notam module structure in an embodiment of the present invention;

[0063] Figure 11 This is a flowchart illustrating the workflow of filtering and parsing the notam unit in an embodiment of the present invention.

[0064] Figure 12 The flowchart for retaining the focus message unit in the embodiments of the present invention is shown below;

[0065] Figure 13 This is a schematic diagram of the module structure for parsing and processing the matching of notam and AIP in an embodiment of the present invention;

[0066] Figure 14 This is a flowchart illustrating the process of determining the AIP cycle unit to be matched based on notam in an embodiment of the present invention.

[0067] Figure 15 This is a flowchart illustrating the workflow of the content matching unit in an embodiment of the present invention.

[0068] Figure 16 A flowchart illustrating the workflow of the affected PDF tagging unit is provided for an embodiment of the present invention. Detailed Implementation

[0069] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0070] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0071] The following description, with reference to the accompanying drawings, describes an apparatus and method for analyzing and displaying dynamic and static intelligence information that integrates NOTAM and AIP, according to an embodiment of the present invention.

[0072] In one embodiment of the present invention, such as Figure 1 As shown, a device for analyzing and displaying dynamic and static intelligence information integrating NOTAM and AIP is provided. Figure 1 As shown, the device includes: a module 1 for periodically parsing AIP PDFs and performing secondary processing; a module 2 for real-time parsing of NOTAMs; a module 3 for matching the parsing and processing results of NOTAMs and AIPs; and a unit 4 for displaying the affected PDF tags.

[0073] The detailed structure and functions of each module are explained below:

[0074] (1) In one embodiment of the present invention, the AIP PDF is periodically parsed and the secondary processing module 1 is used to periodically parse and process the AIP PDF set that is highly related to flight safety according to the AIRAC cycle, and complete the structured result storage and location mapping. The aeronautical chart PDF is identified and parsed by Qwen3-VL multimodal large model, and the table and text PDF is identified and parsed by PaddleOCR-VL tool.

[0075] Specifically Figure 2 This is a schematic diagram of the structure of the periodic AIP PDF parsing and secondary processing module proposed in an embodiment of the present invention. (Reference) Figure 2 The AIP PDF parsing and secondary processing module 1 also includes:

[0076] AIP PDF parsing scope definition unit 11 is used to define the list of airports operated by the airline, the country of AIP to be used and the corresponding AIP version, in combination with the actual operational needs of different airlines. By using regular expressions that adapt to the AIP catalog and naming rules of various countries, the target AIP PDF files to be parsed are accurately filtered.

[0077] The AI ​​tool is called to recognize PDF unit 12. When switching AIRAC cycles, it is used to distinguish between aeronautical charts and text / table PDFs based on the filtered PDF range. The Qwen3-VL multimodal large model and PaddleOCR-VL tools are called to complete the recognition and parsing respectively, and the navigation station information, markdown format content and position mapping file are output according to the specified specifications.

[0078] The markdown recognition result secondary processing unit 13 is used to perform hierarchical error correction, chapter segmentation, and tag correction on the original markdown file output by PaddleOCR-VL, and at the same time, it unifies the parsing format of various irregular tables through the table normalization processing algorithm.

[0079] The structured storage unit 14 is used to classify and structure the AIP parsing data after secondary processing, storing text, tables and navigation information into the corresponding databases respectively, and synchronously storing chapter information, PDF positioning parameters, airport four codes, and all traceability information of file paths.

[0080] In detail, the AIP PDF parsing scope definition unit 11 is used to adapt to different airline operation scenarios, customize the AIP parsing resources required by each airline, avoid invalid PDF parsing, and improve overall parsing efficiency; the AI ​​tool PDF recognition unit 12 is used to adapt to the parsing needs of both aviation charts, text, and tables by relying on the differentiated capabilities of multimodal large models and OCR tools, ensuring the recognition accuracy of different types of AIP data; the Markdown recognition result secondary processing unit 13 is used to correct problems such as hierarchical disorder, inconsistent table formats, and label recognition deviation caused by AI recognition, and standardize the output parsing file; the structured storage unit 14 is used to realize the structured, periodic, and traceable storage of AIP static intelligence, providing a standard data foundation for subsequent NOTAM dynamic intelligence matching.

[0081] In one embodiment of the present invention, the workflow diagram of the AIP pdf parsing range definition unit 11 is as follows: Figure 3 As shown. This unit's configuration rules only need to be defined once, until there are substantial changes to the AIP file organization structure and naming rules. Specific execution steps include:

[0082] Step a1: Define the airport dictionary for each airline based on their specific needs;

[0083] Step a2: Define multiple AIP country codes to be used by the airline according to its needs. For example, if the airline operates domestic airports, then configure the country codes to be used to include China.

[0084] Step a3: Based on the configured AIP country, match the corresponding AIP version as the AIP_version to be processed. For example, if the country includes China, the corresponding AIP version is NAIP.

[0085] Step a4: Configure specific regular expressions for different AIP_versions to accurately filter out core PDF files related to airport, runway, arrival / departure, and approach procedures based on the AIP directory structure and file naming rules of different countries, while eliminating irrelevant files. For domestic NAIP, first exclude PDFs outside the "Terminal" directory, and then filter target files such as departure and arrival diagrams based on file naming rules.

[0086] In one embodiment of the present invention, the workflow diagram of calling AI tools to recognize PDF unit 12 is as follows: Figure 4 As shown, the specific execution steps include:

[0087] Step b1: Whenever the AIRAC 28-day cycle changes, perform a full PDF file parsing job according to the AIP PDF range obtained by the above steps;

[0088] Step b2: Determine the file type one by one according to the PDF file name. If it is a flight chart PDF, skip to step b3. If it is a text / table PDF, skip to step b4.

[0089] Step b3: Select to call the Qwen3-VL multimodal large model to recognize and parse aeronautical chart PDFs, then proceed to the next step;

[0090] Step b4: Select to use the PaddleOCR-VL tool to recognize and parse text / table PDFs, then proceed to the next step;

[0091] Step b5: Identify navigation information from the PDF using customized prompts, output the navigation category, name, PDF page number, and coordinates of the marked boxes in the specified format, and store the recognition results as a map.json file. After completion, proceed to... Figure 7 Step d12;

[0092] Step b6: Recognize the PDF content and generate the original content file original.md in Markdown format, along with a corresponding position mapping file positions.json. Automatically remove PDF watermarks, headers, and footers, and recognize and generate multi-level chapters, headings, tables, and free text elements. After completion, redirect to... Figure 5Step c1 involves secondary processing.

[0093] Step b1: Whenever the AIRAC 28-day cycle changes, perform a full PDF file parsing job according to the AIP PDF range obtained by the above steps;

[0094] Step b2: Determine the file type one by one according to the PDF file name. If it is a flight chart PDF, skip to step b3. If it is a text / table PDF, skip to step b4.

[0095] Step b3: Use the Qwen3-VL multimodal large model to recognize and parse aeronautical chart PDFs;

[0096] Step b4: Use the PaddleOCR-VL tool to recognize and parse text / table PDFs;

[0097] Step b5: Identify navigation information from the PDF using customized prompts, output the navigation category, name, PDF page number, and coordinates of the marked boxes in the specified format, and store the recognition results as a map.json file. After completion, proceed to... Figure 7 Step d12;

[0098] Step b6: Recognize the PDF content and generate the original content file original.md in Markdown format, along with a corresponding position mapping file positions.json. Automatically remove PDF watermarks, headers, and footers, and recognize and generate multi-level chapters, headings, tables, and free text elements. After completion, redirect to... Figure 5 Step c1 involves secondary processing.

[0099] In one embodiment of the present invention, the workflow diagram of the markdown recognition result secondary processing unit 13 is as follows: Figure 5 As shown, the specific execution steps include:

[0100] Step c1: Perform a comprehensive check on all heading # levels in the original.md file, correct any incorrectly identified heading levels, and proceed to the next step;

[0101] Step c2: Divide the Markdown file into chapter blocks based on the lowest level headings, then proceed to the next step;

[0102] Step c3: Check whether each leaf chapter contains a div tag. If it does, proceed to step c4; otherwise, proceed to step c7.

[0103] Step c4: Determine whether the content of the div tag and the content of the current leaf chapter are at the same level. If so, proceed to step c6; otherwise, proceed to step c7.

[0104] Step c5: Convert the content of the sibling div tags into parallel leaf chapter #heading format, then proceed to the next step;

[0105] Step c6: Execute the table normalization algorithm to standardize the format of all tables in the file. See details for the process. Figure 6 Proceed to the next step;

[0106] Step c7: Save the completed secondary processing content as a standardized modified.md file and proceed to the next stage.

[0107] Step c1: Perform a comprehensive check on all heading # levels in the original.md file and correct any incorrectly identified heading levels;

[0108] Step c2: Divide the Markdown file into chapter blocks based on the lowest level headings in the file;

[0109] Step c3: Check whether each leaf chapter contains a div tag. If it does, proceed to step c4; otherwise, proceed to step c7.

[0110] Step c4: Determine whether the content of the div tag and the content of the current leaf chapter are at the same level. If so, proceed to step c6; otherwise, proceed to step c7.

[0111] Step c5: Convert the content of sibling div tags into parallel leaf chapter #heading format;

[0112] Step c6: Execute the table normalization algorithm to standardize the format of all tables in the file. See details for the process. Figure 6 ;

[0113] Step c7: Save the completed secondary processing content as a standardized modified.md file.

[0114] In one embodiment of the present invention, the flowchart of the table normalization processing algorithm in the Markdown recognition result secondary processing unit 13 is as follows: Figure 6 As shown, the specific execution steps include:

[0115] Step c61: Iterate through all instances of the markdown file.

[0116]

[0117]

[0118]

[0119]

[0120]

[0121]

[0122] Figure 7

[0123]

[0124]

[0125] The steps are as follows: Step 1: Select the tables to be processed one by one; Step 2: Extract the #title, div title, or text title immediately above each table and assign a unique title attribute to the table; Step 3: Identify tables with merged cells, add tr and td tags according to the minimum cell size, and restore them to the standard general table format; Step 4: Identify tables with vertically arranged headers and convert them to a horizontal header format using a matrix transpose algorithm; Step 5: Identify tables with multiple columns of repeated arrangement and convert them to a single-column field-key value standard format; Step 66: Identify redundant sequence number columns in the table and remove them in batches; Step 67: Compare the titles of all tables in the same file, determine the cross-page tables with duplicate titles as tables from the same source, remove duplicate headers, and complete the table merging. In one embodiment of the present invention, the workflow diagram of the structured storage unit 14 is shown below. The specific execution steps include: Step d1: Traverse all leaf chapters in the modified.md file obtained after processing text / table type PDFs, and proceed to the next step; Step d2: Record the chapter # content and level in real time, update the last complete chapter lastTitle, form a multi-level chapter tracing path, and proceed to the next step; Step d3: Determine whether the current leaf chapter contains

[0126] Figure 8

[0127]

[0128]

[0129]

[0130]

[0131] Figure 9

[0132]

[0133]

[0134] Figure 4

[0135]

[0136]

[0137]

[0138]

[0139] If the table label is present, proceed to step d4; otherwise, proceed to step d6. Step d4: Execute the process of storing the table into the data table. See details for the process; proceed to the next step. Step d5: Determine if there is any text content other than the table content in the current chapter. If so, proceed to step d6; otherwise, return to step d1 to process the next chapter. Step d6: Use regular expressions to determine if the chapter text contains subheading numbers such as 1, 1.1, etc. If such numbers exist, proceed to step d10; otherwise, proceed to step d7. Step d7: Set the current text field `field` to the `lastTitle` content; proceed to the next step. Step d8: Encapsulate the entire text content into a single storage record; proceed to the next step. Step d9: Execute the text... The process of storing data in the data table is detailed in the following steps; proceed to the next step. Step d10: Set the current text field `field` to the identified subheading number, proceed to the next step. Step d11: Split the text content by subheading, generate multiple structured text records, and then proceed to step d9. Step d12: Receive the `map.json` file output from step b5, and structure and store the navigation station category, name, page number, coordinates, and other information into the navigation station data table, proceed to the next step. Step d13: Attach the corresponding airport four-code association to each navigation station data table record, proceed to the next step. Step d14: Attach the PDF file name, `map.json` file path, and other traceability information to each navigation station data table record, proceed to the next stage. Step d1: Traverse all leaf chapters in the `modified.md` file obtained after processing the text / table PDF. Step d2: Record the chapter # content and level in real time, update the most recent complete chapter `lastTitle`, and form a multi-level chapter traceability path. Step d3: Determine whether the current leaf chapter contains...

[0140] Figure 8

[0141]

[0142]

[0143]

[0144]

[0145] Figure 9

[0146]

[0147]

[0148] Figure 4

[0149]

[0150]

[0151]

[0152] Figure 10 Figure 10

[0153]

[0154]

[0155]

[0156] Figure 11

[0157]

[0158]

[0159]

[0160]

[0161]

[0162]

[0163]

[0164]

[0165] Figure 12

[0166]

[0167]

[0168]

[0169]

[0170] Figure 13

[0171]

[0172]

[0173]

[0174]

[0175]

[0176]

[0177] Figure 13

[0178]

[0179]

[0180]

[0181] Figure 13 Figure 13

[0182]

[0183]

[0184]

[0185] Figure 14

[0186]

[0187]

[0188]

[0189]

[0190]

[0191]

[0192]

[0193]

[0194] Figure 15

[0195]

[0196]

[0197]

[0198]

[0199]

[0200]

[0201]

[0202]

[0203]

[0204]

[0205]

[0206]

[0207]

[0208]

[0209]

[0210]

[0211]

[0212] Figure 16

[0213]

[0214]

[0215]

[0216]

[0217]

[0218]

[0219]

[0220]

[0221]

[0222]

[0223]

[0224]

[0225]

[0226]

[0227] If the table label is present, proceed to step d4; otherwise, proceed to step d6. Step d4: Execute the process of storing the table into the data table; see details for the process. Step d5: Determine if there is any text content other than the table content in the current chapter. If so, proceed to step d6; otherwise, return to step d1 to process the next chapter. Step d6: Use regular expressions to determine if the chapter text contains subheading numbers such as 1, 1.1, etc. If such numbers exist, proceed to step d10; otherwise, proceed to step d7. Step d7: Set the current text field `field` to the `lastTitle` content. Step d8: Encapsulate the entire text content into a single storage record. Step d9: : Execute the text storage data table process, see details of the process; Step d10: Set the current text field field to the identified subheading number; Step d11: Split the text content according to the subheading, generate multiple structured text records and then jump to step d9; Step d12: Receive the map.json file output by step b5, and store the navigation station category, name, page number, coordinates and other information in the navigation station data table in a structured manner; Step d13: Attach the corresponding airport four-code association to each navigation station data table record; Step d14: Attach the PDF file name, map.json file path and other traceability information to each navigation station data table record. (2) In one embodiment of the present invention, the real-time parsing notam module 2 is used to receive notam messages in real time, and uses a self-trained large model to complete message parsing, and filters out the airport, runway and navigation station notams that are of concern to the airline and pose a risk to flight safety. Specifically, it is a schematic diagram of the structure of the real-time parsing notam module proposed in the embodiment of the present invention. For reference, the real-time notam parsing module 2 also includes: a notam filtering and parsing unit 21, used to receive notam messages in real time, filter out core messages of airport, runway, and navigation beacon categories using Q codes, perform structured parsing of message D and E semantics using a self-trained large model, and store the information in a database; and a message retention unit 22, used to perform secondary filtering on the structured parsed messages, accurately identify airport, runway, and navigation beacon-related closed, open, and restricted risk messages, match the AIP association configuration table to determine the corresponding associated data category, and filter invalid messages that have no impact on flight safety. Specifically, the notam filtering and parsing unit 21 is used to perform initial message screening based on the NOTAM standard Q code, eliminating irrelevant messages, and deeply analyzes the message field semantics using a self-developed large model to achieve the conversion of unstructured messages into structured data; the message retention unit 22 is used to perform risk-level screening on the parsed messages based on airline flight safety requirements, accurately identify high-risk messages of concern, and establish a mapping relationship between NOTAM dynamic messages and AIP static data.In one embodiment of the present invention, the workflow diagram of the filtering and parsing NotAM unit 21 is shown below. The specific execution steps include: Step e1: continuously acquiring the next NotAM message received by the system in real time, and proceeding to the next step; Step e2: determining the message type based on the message Q code, filtering out airport, runway, and navigation station type messages. If the type is correct, proceed to step e3; otherwise, return to step e1 to continue receiving the next message; Step e3: calling the self-trained large model to perform deep parsing of the semantics of message D and E, generating differentiated structured parsing fields according to the message type, and proceeding to the next step; Step e4: classifying and storing the structured parsing results into a relational data table according to the message type, and proceeding to the next stage. Step e1: Continuously acquire the next NOTAM message received by the system in real time; Step e2: Determine the message type based on the message Q code, and filter out airport, runway, and navigation beacon messages. If the type matches, proceed to step e3; otherwise, return to step e1 to continue receiving the next message; Step e3: Call the self-trained large model to perform deep analysis of the semantics of message D and E, and generate differentiated structured parsing fields according to the message type; Step e4: Classify and store the structured parsing results into a relational data table according to the message type. In one embodiment of the present invention, the workflow diagram of the retained concern message unit 22 is shown below. The specific execution steps include: Step f1: Determine whether the message parsing result is a risk message of airport closure, airport opening, airport restriction, runway closure, runway opening, runway restriction, or navigation beacon unavailability. If yes, proceed to step f2; otherwise, determine that it is a message with no safety impact and return to step e1. Step f2: Determine whether the current message is an airport or runway restriction message. If yes, proceed to step f3; otherwise, proceed to step f4. Step f3: Determine whether the message subcategory includes scenarios of concern to airlines such as airport strike, power supply unavailability, take-off and landing restrictions, no alternative landing, aircraft type restrictions, flight restrictions, abnormal fuel supply, fire level changes, runway take-off and landing restrictions, aircraft type restrictions, published distance changes, and PCN changes, and match them accordingly. If successful, proceed to step f6; otherwise, return to step e1. Step f4: Determine if the current message is an airport / runway open / closed message. If yes, proceed to step f7; otherwise, determine if it is a navigation station unavailable message and proceed to step f5. Step f5: Extract the navigation station name from the message, search the critical navigation station dictionary table, and determine if it is a critical navigation station. If yes, proceed to step 31 to start the matching process; otherwise, return to step e1. Step f6: Using the message restriction subcategory as the search condition, query the AIP association configuration table to obtain the target association AIP data table name corresponding to the current message, and proceed to the next step. Step f7: Using the airport / runway open / closed category corresponding to the message as the search condition, query the AIP association configuration table to obtain the corresponding target association AIP data table name, and proceed to the next stage.Step f1: Determine if the message parsing result is a risk message related to airport closure, airport opening, airport restrictions, runway closure, runway opening, runway restrictions, or navigation beacon unavailability. If yes, proceed to step f2; otherwise, determine it as a message with no safety impact and return to step e1. Step f2: Determine if the current message is an airport or runway restriction message. If yes, proceed to step f3; otherwise, proceed to step f4. Step f3: Determine if the message subcategory includes scenarios of concern to airlines, such as airport strikes, power outages, restricted takeoffs and landings, inability to divert, aircraft type restrictions, flight restrictions, abnormal fuel supply, changes in fire safety ratings, runway restrictions, aircraft type restrictions, changes in published distances, and changes in PCN. If a match is found, proceed to step f6; otherwise, return to step e1. Step e1; Step f4: Determine whether the current message is an airport / runway open / closed message. If yes, jump to step f7; otherwise, determine that it is a navigation station unavailable message and jump to step f5; Step f5: Extract the navigation station name in the message, search the key navigation station dictionary table, and determine whether it is a key navigation station. If yes, jump to step 31 to start the matching process; otherwise, return to step e1; Step f6: Use the message restriction sub-category as the search condition to query the AIP association configuration table and match the target association AIP data table name corresponding to the current message; Step f7: Use the airport / runway open / closed category corresponding to the message as the search condition to query the AIP association configuration table and match the corresponding target association AIP data table name. (3) In one embodiment of the present invention, the notam and AIP parsing processing result module 3 is used to determine the AIRAC cycle, target file, storage path and corresponding content location of the AIP to be matched based on the notam parsing result. Specifically, it is a schematic diagram of the structure of the notam and AIP parsing processing result module proposed in the embodiment of the present invention. For reference, module 3, which matches the parsing results of NOTAM and AIP, also includes: a notarized AIP period determination unit 31, used to extract the effective and expiration times of the notarized message, match the corresponding AIRAC period through the period configuration table, and lock the target database schema of the AIP to be matched; and a content matching unit 32, used to prioritize matching the AIP structured data table based on the notarized message type, airport four codes, and runway / navigation beacon information, and to perform a secondary matching of text data if the matching fails, ultimately extracting the associated PDF file path, page number, coordinate marker position, and full parsing fields. In detail, the notarized AIP period determination unit 31 is used to accurately locate the corresponding version of AIP static data by combining the notarized message's timeliness, solving the data version mismatch problem caused by AIP period iteration; and the content matching unit 32 is used to adopt a hierarchical matching strategy, prioritizing accurate matching of structured data tables and providing fallback matching of text data, maximizing the integrity of the association and matching between NOTAM and AIP intelligence.In one embodiment of the present invention, the flowchart of determining the AIP cycle unit 31 to be matched based on NOTAM is shown below. The specific execution steps include: Step g1: Detect whether the NOTAM parsing result contains effective time and expiration time information. If it exists, jump to step g2; otherwise, jump to step g3. Step g2: Search the cycle configuration table according to the message effective time and expiration time, match the corresponding AIRAC iteration cycle as the message impact cycle, and proceed to the next step. Step g3: If there is no clear time information, the latest AIRAC cycle is selected as the impact cycle by default, and proceed to the next step. Step g4: According to the determined affected AIRAC cycle, locate the AIP target database schema of the corresponding cycle, lock the static data source to be matched, and enter the next stage. Step g1: Check if the NOTAM parsing result contains effective time and expiration time information. If it exists, proceed to step g2; otherwise, proceed to step g3. Step g2: Search the period configuration table based on the message's effective and expiration times to match the corresponding AIRAC iteration period, which is used as the message's impact period. Step g3: If there is no explicit time information, the latest AIRAC period is selected as the impact period by default. Step g4: Based on the determined affected AIRAC period, locate the corresponding AIP target database schema and lock the static data source to be matched. In one embodiment of the present invention, the workflow diagram of the content matching unit 32 is shown below. The specific execution steps include: Step h1: Determine whether the current NOTAM message type is a navigation station type. If yes, proceed to step h2; otherwise, proceed to step h3. Step h2: Extract the navigation station name and airport four-code fields of the message, match them with the navigation station data table records under the corresponding period schema, and proceed to the next step. Step h3: Extract the runway name, airport four-code, message type, and subtype fields of the message, match them with the airport and runway type structured data tables (excluding text tables) under the corresponding period schema, and proceed to the next step. Step h4: Determine whether the number of structured data table matching results is greater than or equal to 1. If the match is successful, proceed to the next step. Proceed to step h7; otherwise, proceed to step h5. Step h5: Activate the fallback matching strategy, match the AIP text data table under the corresponding period schema, prioritize matching field names, then match field content, and proceed to the next step. Step h6: Determine if the number of matching results in the text table is greater than or equal to 1. If a match is successful, proceed to step h7; otherwise, return to step e1 to process the next message. Step h7: Extract the PDF file path, PDF page number, and marker box coordinates corresponding to all successfully matched records. If there are multiple matching records, retain them all and proceed to the next step. Step h8: Extract the original text, D-item content, E-item content, and all structured parsing fields of the current NOTAM message, complete the matching data summary, and proceed to the next stage.Step h1: Determine if the current NOTAM message type is a navigation station type. If yes, proceed to step h2; otherwise, proceed to step h3. Step h2: Extract the navigation station name and airport 4-code fields from the message and match them with the navigation station data table records under the corresponding period schema. Step h3: Extract the runway name, airport 4-code, message type, and subtype fields from the message and match them with the airport and runway type structured data tables (excluding text tables) under the corresponding period schema. Step h4: Determine if the number of matching results in the structured data tables is greater than or equal to 1. If a match is successful, proceed to step h7; otherwise, proceed to step h5. Step h5: Start the fallback matching strategy, match the AIP text data table under the corresponding period schema, prioritize matching field names, and then match field content; Step h6: Determine whether the number of matching results in the text table is greater than or equal to 1. If the match is successful, jump to step h7; otherwise, return to step e1 to process the next message; Step h7: Extract the PDF file path, PDF page number, and marker box coordinate position corresponding to all successfully matched records. All matching records are retained; Step h8: Extract the original text, D item content, E item content, and all structured parsing fields of the current NOTAM message to complete the matching data summary. (4) In one embodiment of the present invention, the affected PDF marker unit 4 is displayed to load the AIP PDF file corresponding to the affected elements of NOTAM in the system interface and complete the visualization marking and information display of NOTAM features at the corresponding element positions in the PDF. In one embodiment of the present invention, the workflow diagram of the affected PDF marking unit 4 is shown as follows. The specific execution steps include: Step 41: Locate and retrieve the corresponding target AIP PDF file according to the matched PDF file path; Step 42: Load the PDF file in the system visualization interface. When there are multiple matching PDFs, display all files in the form of tab switching; Step 43: Accurately position the interface view to the corresponding element position of the PDF according to the matched PDF page number and the coordinate information of the marking box; Step 44: Draw the corresponding type of NOTAM icon at the preset offset position in the upper right corner of the marking box position. At the same time, display the full structured data of NOTAM, the effective and invalidation time information in the form of a card pop-up window. Users can manually show or hide the original text of NOTAM, realizing the visual association display of AIP static flight intelligence and NOTAM dynamic risk intelligence.This invention also provides a method for parsing and displaying dynamic and static intelligence information from NOTAM and AIP. The method includes the following steps: periodically parsing and processing AIP PDF sets highly correlated with flight safety according to the AIRAC cycle, completing structured result storage and location mapping, wherein the Qwen3-VL multimodal large model is used to identify and parse aeronautical chart PDFs, and the PaddleOCR-VL tool is used to identify and parse table and text PDFs; receiving NOTAM messages in real time, using a self-trained large model to complete message parsing, and filtering out airport, runway, and navigation beacon-type NOTAMs that airlines are concerned about and that pose a risk to flight safety; determining the AIRAC cycle, target file, storage path, and corresponding content location of the AIP to be matched based on the NOTAM parsing results; loading the AIP PDF file corresponding to the affected elements of the NOTAM on the system interface, and completing the visualization marking and information display of NOTAM features at the corresponding element locations in the PDF. Regarding the method in the above embodiments, the specific execution methods of each step have been described in detail in the embodiments related to the device, and will not be elaborated further here. To implement the methods of the above embodiments, the present invention also provides an electronic device, which includes a memory and a processor; wherein the processor runs a program corresponding to the executable program code stored in the memory to implement the various steps of the methods described above. To implement the above embodiments, this application also proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the methods described in the foregoing embodiments. The above descriptions are merely preferred embodiments of the present invention and are not intended to limit 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 protection scope of the present invention. In the description of this specification, the reference to the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., means that the specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. Additionally, those skilled in the art can combine and integrate the different embodiments or examples described herein, as well as the features of those different embodiments or examples, without contradiction. Moreover, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature.In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A device for analyzing and displaying dynamic and static intelligence information integrating NOTAM and AIP, characterized in that, include: The AIP PDF parsing and secondary processing module is used to periodically parse and process AIP PDF sets that are highly related to flight safety according to the AIRAC cycle, and complete the structured result storage and location mapping. Among them, the Qwen3-VL multimodal large model is used to identify and parse aeronautical chart PDFs, and the PaddleOCR-VL tool is used to identify and parse table and text PDFs. The real-time notam parsing module is used to receive notam messages in real time. It uses a self-trained large model to complete message parsing and filter out notams related to airports, runways, and navigation beacons that are of concern to airlines and pose a risk to flight safety. The module that matches the parsing results of notam and AIP is used to determine the AIRAC cycle, target file, storage path and corresponding content location of the AIP to be matched based on the notam parsing results. Displaying affected PDF marker units is used to load the AIP PDF files corresponding to the affected features of the notam in the system interface, and to complete the visualization marking and information display of the notam features at the corresponding feature positions in the PDF.

2. The apparatus according to claim 1, characterized in that, The periodic AIP PDF parsing and secondary processing module includes: The AIP PDF parsing scope definition unit is used to define the list of airports operated by the airline, the countries of AIPs to be used and the corresponding AIP versions, in combination with the actual operational needs of different airlines. It filters the target AIP PDF files to be parsed by using regular expressions that adapt to the AIP directories and naming rules of various countries. AI tools are used to identify PDF units. During AIRAC cycle switching, based on the filtered PDF range, the system distinguishes between aeronautical chart PDFs and text / table PDFs. The system calls Qwen3-VL multimodal large model and PaddleOCR-VL tools to complete the identification and parsing, and outputs navigation station information, markdown format content and position mapping files according to the specified specifications. The secondary processing unit for markdown recognition results is used to perform hierarchical error correction, chapter segmentation, and tag correction on the original markdown file output by PaddleOCR-VL. At the same time, it uses a table normalization processing algorithm to unify the parsing format of various irregular tables. The structured storage unit is used to classify and structure the AIP parsing data after secondary processing, storing text, tables, and navigation information into corresponding databases, and synchronously linking and storing chapter information, PDF positioning parameters, airport four codes, and all traceability information of file paths.

3. The apparatus according to claim 2, characterized in that, AIP resolution range definition unit, used for: Define the airport dictionary for each airline based on their specific needs; Define multiple AIP country codes for each airline based on their needs; Based on the determined AIP country, match the corresponding AIP version as the AIP_version to be processed; Configure dedicated regular expressions for different AIP_versions, and combine them with the AIP directory structure and file naming rules to filter out core PDF files related to airport, runway, arrival and departure, and approach procedures, while eliminating irrelevant files.

4. The apparatus according to claim 3, characterized in that, The table normalization algorithm in the secondary processing unit of the Markdown recognition result is used for: Illustrate all tables in the document one by one according to the table labels; Extract the title information from the top of the table and assign a unique title attribute to the table. Identify tables containing merged cells, complete the cell labels, and standardize the table format; Transpose a vertical header table into a standard horizontal header format; Convert tables with multiple columns and repeated columns into a single-column field value format; Remove redundant serial number columns from the table; Identify duplicate tables spanning multiple pages, remove duplicate headers, and then merge the tables.

5. The apparatus according to claim 4, characterized in that, The real-time notam parsing module includes: The Notam unit is filtered and parsed to receive Notam messages in real time. It uses Q codes to filter out core messages of airport, runway, and navigation station types. It uses a self-trained large model to perform structured parsing of the semantics of message D and E items and stores them in the database according to their categories. The retained message unit is used to perform secondary filtering on the structured parsed messages, accurately identify risk messages related to airports, runways, and navigation beacons that are closed, open, or restricted, match the AIP association configuration table to determine the corresponding associated data category, and filter out invalid messages that have no impact on flight safety.

6. The apparatus according to claim 5, characterized in that, The module for parsing and processing the results of matching notam and AIP includes: The AIP periodic unit to be matched is determined based on the notam, which is used to extract the effective and expiration time of the notam message, match the corresponding AIRAC period through the periodic configuration table, and lock the target database schema of the AIP to be matched. The content matching unit is used to prioritize matching the AIP structured data table based on the NOTAM message type, airport four codes, and runway / navigation beacon information. If the matching fails, it will perform a second matching of the text data and finally extract the associated PDF file path, page number, coordinate marker position, and full parsing fields.

7. The apparatus according to claim 6, characterized in that, The unit for displaying affected PDF tags is used for: Retrieve the corresponding target AIP PDF file based on the matching results; Load PDF files into the system interface, and display the matching results of multiple files in a tabbed format. Based on the matched page number and coordinate information, the interface is precisely positioned to the corresponding element location in the PDF; Category-specific Notam icons are drawn at the corresponding locations of the marked positions to display the structured data and timely information of Notam in the form of cards. The Notam text can be manually shown or hidden, realizing a visual association between static AIP intelligence and dynamic NOTAM risks.

8. A method for parsing and displaying dynamic and static intelligence information integrating NOTAM and AIP, characterized in that, include: The AIP PDF set, which is highly correlated with flight safety, is periodically parsed and processed according to the AIRAC cycle, and the structured results are stored and mapped. Among them, the Qwen3-VL multimodal large model is used to identify and parse aeronautical chart PDFs, and the PaddleOCR-VL tool is used to identify and parse table and text PDFs. The system receives NOTAM messages in real time, uses a self-trained large model to parse the messages, and filters out NOTAMs related to airports, runways, and navigation beacons that are of concern to airlines and pose a risk to flight safety. Based on the notam parsing results, determine the AIRAC cycle, target file, storage path, and corresponding content location of the AIP to be matched; Load the AIP PDF file corresponding to the notam-affected features in the system interface, and complete the visualization marking and information display of notam features at the corresponding feature positions in the PDF.

9. An electronic device, characterized in that, Including processor and memory; The processor reads executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the method as described in claim 8.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in claim 8.

Images