A method and system for integrating maritime situation data and knowledge query

Abstract

The application provides a marine situation data and knowledge integrated query method and system, and belongs to the field of situation knowledge data management of a marine command control system. The method comprises the following steps: receiving a query request, judging whether a time range and a space range are contained in the query request elements, converting the time range and the space range into space-time grid codes if the time range and the space range are contained, obtaining a list of all space-time grid codes, obtaining a list of entity identifiers satisfying the space-time range conditions by accessing a space-time relation index table, obtaining corresponding entity elements in track type entities and knowledge type entities according to the entity identifiers in the list of entity identifiers, performing a keyword fuzzy query between the entity elements to obtain entity elements satisfying the keyword query conditions, and otherwise performing keyword query services of a full-text search engine and a graph database. Therefore, the application can effectively solve the problems of effective organization, integrated storage management and query of real-time updated situation data and situation knowledge in a highly dynamic environment.

Description

Technical Field

[0001] This invention belongs to the field of situational knowledge data management in maritime command and control systems, and more specifically, relates to an integrated query method and system for maritime situational data and knowledge. Background Technology

[0002] Situational assurance is a core and fundamental function of civilian command and control systems for maritime platforms. It involves comprehensively processing target data detected by the platform's sensors to generate comprehensive, complete, and reliable situational information, providing data support for command and decision-making. In practical applications, it's not only necessary to continuously track and monitor targets, acquiring their real-time location and status, but also to verify targets through cross-domain, multi-dimensional knowledge and target trajectory data. This cross-verification is crucial to effectively confirm target attributes, clarify relevant information, and ensure transparency of the operational area for oversight. However, there are significant differences between trajectory data and knowledge data. The former updates rapidly and in real-time, but has a small data volume per line, a single data modality, and is susceptible to uncertain errors. The latter updates less frequently, spans a longer timeframe, has a large data volume per line, rich data modalities, and higher reliability. Existing systems primarily store and manage these separately, hindering situational analysis applications and reducing efficiency. Therefore, effectively organizing, managing, and querying situational and knowledge data in a maritime environment has become a fundamental problem urgently needing to be solved in the field of maritime situational knowledge management. Summary of the Invention

[0003] In view of the shortcomings of the existing technology, the purpose of this invention is to provide a method and system for integrated query of maritime situation data and knowledge, which aims to solve the problem that the existing situation support system stores and manages time data and track data separately, resulting in low situation analysis efficiency.

[0004] To achieve the above objectives, on the one hand, the present invention provides an integrated query method for maritime situation data and knowledge, comprising the following steps:

[0005] D1: Receive the query request and determine whether the query request elements contain a time range and a spatial range. If they do, proceed to D2; otherwise, proceed to D3.

[0006] D2: Convert the time and space ranges into spatiotemporal grid codes, calculate a list of all spatiotemporal grid codes, and obtain a list of entity identifiers that meet the spatiotemporal range conditions by accessing the spatiotemporal relationship index table in the maritime situation knowledge graph; based on the entity identifiers in the entity identifier list, obtain the corresponding entity elements in track-type entities and knowledge-type entities; perform keyword fuzzy search among entity elements to obtain entity elements that meet the keyword search conditions, summarize and output the entity elements, stop execution, and wait for the next query request;

[0007] D3: Obtain entity query results through the full-text search engine and keyword query service of the graph database in the maritime situation knowledge graph, stop execution and wait for the next query request;

[0008] The method for constructing a maritime situation knowledge graph includes the following steps:

[0009] a. Construct a maritime situation knowledge graph entity set and relation set; wherein, the entity set includes track-type entities and knowledge-type entities; track-type entities include track description entities and track segment entities; the relation set includes ordinary relations and spatiotemporal relations; ordinary relations are relations in traditional knowledge graphs; spatiotemporal relations are the intersection of two or more graph entities with spatiotemporal attributes within a time range or spatial range;

[0010] b. Divide the Earth grid and simultaneously segment the time axis to form spatiotemporal grid units. Construct a spatiotemporal relationship index table by performing binary spatiotemporal grid encoding on the spatiotemporal grid units.

[0011] c. Knowledge entities and ordinary relationships are stored in a graph database. Track entities are retrieved by using track description entities. Track point data is stored and spatiotemporal relationships are associated through track segment entities and spatiotemporal index tables. Spatiotemporal element attributes in knowledge entities are mapped to the spatiotemporal index tables, and spatiotemporal grid cells store track segment entity identifiers and knowledge entity identifiers that have spatiotemporal relationships.

[0012] More preferably, the spatiotemporal grid coding is divided into spatial coding and temporal coding; the maximum length of both spatial coding and temporal coding is 32 bits.

[0013] The spatial encoding method includes the following steps:

[0014] The geographical coordinate space of the Earth, from 180 degrees east longitude to 180 degrees west longitude and from 90 degrees north latitude to 90 degrees south latitude, is regarded as a plane;

[0015] Given a degree span represented by a grid cell, divide the plane into several grid cells;

[0016] Each grid cell is numbered starting from 0 using a fill curve encoding method, and the grid cell number is the spatial code of the grid cell.

[0017] The time encoding method includes the following steps:

[0018] Divide the timeline into several time intervals and assign a unique 32-bit binary number to each time interval.

[0019] Spatial coding and temporal coding are combined to form spatiotemporal grid coding.

[0020] More preferably, the spatiotemporal relationship index table is composed of a grid cell index list, each grid cell index list including a grid cell summary and corresponding entity identifier list data; the grid cell summary includes the spatiotemporal grid code value, the number of entities in the grid, the statistical number of entity classifications in the grid, and the update timestamp.

[0021] More preferably, step c specifically includes the following steps:

[0022] c.1: Entity data preprocessing, specifically including the following steps:

[0023] c.1.1: Preprocessing of track-type entities, specifically including the following steps:

[0024] Remove abnormal data from track-type entities; abnormal data in c.1.1 includes incomplete key data elements, data values ​​exceeding the valid value range, and duplicate data;

[0025] Organize the track point entities into a track data classification summary table according to the order of their timestamps;

[0026] Generate track summary information based on the track data classification summary table. The track summary information includes the track start and end time, track spatial range, and number of track points.

[0027] c.1.2: Knowledge entity preprocessing, specifically including the following steps:

[0028] Abnormal data in knowledge entities will be removed. Abnormal data in c.1.2 includes incomplete key data elements, no title, no content, character set disorder, duplicate data, and more than 50% of image and video data are missing.

[0029] Entities are classified according to their source, and entities from the same source are grouped together to form a set of knowledge entity categories.

[0030] c.2: Extract entity data, specifically including the following steps:

[0031] c.2.1: Extraction and updating of track-type entity data, specifically including the following steps:

[0032] Generate a track description entity from track summary information;

[0033] The track category entities in the track data classification summary table are divided into several track segment entities, and the track segment entity identifiers are added to the track segment list attribute of the track description entity.

[0034] Iterate through all track points of the track to which the track description entity belongs, calculate the spatiotemporal code based on the track point position, put all track points with the same spatiotemporal code into the track point list of the same track segment entity, and add the track segment entity identifier to the track segment list attribute of the track description entity.

[0035] c.2.2: Extraction and updating of knowledge entity data, specifically:

[0036] Knowledge entities are extracted using keyword matching methods, and information is filled into entity elements according to the specific description of the entity.

[0037] c.3: Extraction and updating of spatiotemporal relationships, specifically:

[0038] Calculate the time code based on the time range, calculate the spatial code based on the spatial range, combine the time code and spatial code, and add the entity identification information to the entity identification list in the corresponding spatiotemporal relationship index table;

[0039] c.4: Extract ordinary relations using traditional knowledge graph relations.

[0040] More preferably, the spatiotemporal relationship includes temporal inclusion, temporal intersection, spatial inclusion, spatial intersection, spatiotemporal intersection, and spatiotemporal inclusion.

[0041] On the other hand, the present invention provides an integrated query system for maritime situation data and knowledge, including: a query request receiving module, a first query module, a second query module, an entity set construction module, a relationship set construction module, a spatiotemporal relationship index table construction module, and a situation knowledge graph construction module;

[0042] The query request receiving module is used to receive query requests and determine whether the query request elements contain a time range and a spatial range. If they do, the first query module is activated; otherwise, the second query module is activated.

[0043] The first query module converts the time and space ranges into spatiotemporal grid codes, calculates a list of all spatiotemporal grid codes, accesses the spatiotemporal relationship index table in the maritime situation knowledge graph to obtain a list of entity identifiers that meet the spatiotemporal range conditions, retrieves the corresponding entity elements from track-type entities and knowledge-type entities based on the entity identifiers in the entity identifier list, performs a keyword fuzzy search among the entity elements to obtain entity elements that meet the keyword search conditions, summarizes and outputs the entity elements, stops execution, and waits for the next query request.

[0044] The second query module is used to obtain entity query results through the full-text search engine and graph database keyword query service in the maritime situation knowledge graph, and then stop execution to wait for the next query request.

[0045] The entity set construction module is used to construct the entity set of the maritime situation knowledge graph; the entity set includes track-type entities and knowledge-type entities; track-type entities include track description entities and track segment entities;

[0046] The relation set construction module is used to construct relation sets for the maritime situation knowledge graph; among them, relation sets include ordinary relations and spatiotemporal relations; ordinary relations are relations in traditional knowledge graphs; spatiotemporal relations are the intersection of two or more graph entities with spatiotemporal attributes within a time range or a spatial range;

[0047] The spatiotemporal relation index table construction module is used to divide the Earth grid and simultaneously segment the time axis to form spatiotemporal grid units. By performing binary spatiotemporal grid encoding on the spatiotemporal grid units, a spatiotemporal relation index table is constructed.

[0048] The situational knowledge graph construction module is used to store knowledge entities and ordinary relationships in a graph database. Track entities are retrieved by summarizing track description entities. Track point data is stored and spatiotemporal relationships are associated through track segment entities and spatiotemporal index tables. Spatiotemporal element attributes in knowledge entities are mapped to the spatiotemporal index tables, and spatiotemporal grid cells store track segment entity identifiers and knowledge entity identifiers that have spatiotemporal relationships.

[0049] More preferably, the spatiotemporal relationship index table construction module includes a spatial coding unit, a temporal coding unit, and a coding merging unit;

[0050] The spatiotemporal grid coding is divided into spatial coding and temporal coding; the maximum length of both spatial coding and temporal coding is 32 bits.

[0051] The spatial coding unit is used to treat the geographic coordinate space from 180 degrees east longitude to 180 degrees west longitude and from 90 degrees north latitude to 90 degrees south latitude as a plane; given the degree span represented by a grid unit, the plane is divided into several grid units; each grid unit is numbered starting from 0 using a filled curve code, and the grid unit number is the spatial code of the grid unit;

[0052] The time encoding unit is used to divide the time axis into several time intervals, and assigns a 32-bit binary number to each time interval for unique identification.

[0053] The encoding merging unit is used to combine spatial encoding and temporal encoding to form a spatiotemporal grid encoding.

[0054] More preferably, the spatiotemporal relationship index table is composed of a grid cell index list, each grid cell index list including a grid cell summary and corresponding entity identifier list data; the grid cell summary includes the spatiotemporal grid code value, the number of entities in the grid, the statistical number of entity classifications in the grid, and the update timestamp.

[0055] More preferably, the situational knowledge graph construction module includes a track-type entity preprocessing unit, a knowledge-type entity preprocessing unit, a track-type entity extraction and update unit, a knowledge-type entity extraction and update unit, a spatiotemporal relationship extraction and update unit, and a general relationship extraction and update unit;

[0056] The track entity preprocessing unit is used to remove abnormal data from track entities; abnormal data includes incomplete key data elements, data values ​​exceeding the valid value range, and duplicate data; track point entities are organized into a track data classification summary table according to the order of timestamps; track summary information is generated based on the track data classification summary table, and the track summary information includes track start and end time, track spatial range, and number of track points;

[0057] The knowledge entity preprocessing unit is used to remove abnormal data from knowledge entities. Abnormal data includes incomplete key data elements, no title, no content, character set disorder, duplicate data, and more than 50% of image and video data are missing. The knowledge entities are classified according to their source, and entities from the same source are grouped together to form a knowledge entity classification set.

[0058] The track entity extraction and update unit is used to generate track description entities from track summary information; divide the track entities in the track data classification summary table into several track segment entities, add the track segment entity identifier to the track segment list attribute of the track description entity; traverse all track points of the track to which the track description entity belongs, calculate the spatiotemporal code based on the track point position, put all track points with the same spatiotemporal code into the track point list of the same track segment entity, and add the track segment entity identifier to the track segment list attribute of the track description entity;

[0059] The knowledge entity extraction and update unit is used to extract knowledge entities using keyword matching methods, and fill the information into the entity elements according to the specific description of the entity.

[0060] The spatiotemporal relationship extraction and update unit is used to calculate the time code based on the time range, calculate the spatial code based on the spatial range, combine the time code and the spatial code, and add the entity identification information to the entity identification list in the corresponding spatiotemporal relationship index table;

[0061] The ordinary relation extraction and update unit is used to extract ordinary relations using traditional knowledge graph relations.

[0062] More preferably, the spatiotemporal relationship includes temporal inclusion, temporal intersection, spatial inclusion, spatial intersection, spatiotemporal intersection, and spatiotemporal inclusion.

[0063] In summary, compared with the prior art, the above-described technical solutions conceived by this invention have the following advantages:

[0064] Beneficial effects:

[0065] 1. This invention provides an integrated query method and system for maritime situation data and knowledge. The relationship set in the maritime situation knowledge graph includes ordinary relationships and spatiotemporal relationships. A spatiotemporal relationship is the intersection of two or more graph entities with spatiotemporal attributes within a time or spatial range. Knowledge entities and ordinary relationships are stored in a graph database. Track entities are retrieved through track description entities. Track point data is stored and associated with spatiotemporal relationships through track segment entities and a spatiotemporal index table. Spatiotemporal element attributes in knowledge entities are mapped to the spatiotemporal index table, and spatiotemporal grid cells store the identifiers of track segment entities and knowledge entities with spatiotemporal relationships. Upon receiving a query request, it is determined whether the query request elements contain a time range and a spatial range. If a time range and a spatial range exist, the entity elements can be obtained using the time range, spatial range, and keywords. Therefore, this invention can effectively solve the problem of effectively organizing, storing, managing, and querying real-time updated situation data and situation knowledge in a highly dynamic environment, providing timely and reliable situation graph data support for intelligent maritime situation cognition and analysis applications.

[0066] 2. This invention provides a method and system for integrated querying of maritime situation data and knowledge. It realizes the integrated organization and querying of massive target track data and massive situation knowledge data. It can efficiently retrieve situation knowledge based on triples, and can also efficiently retrieve related target track data and situation knowledge based on spatiotemporal relationships. The retrieval results are provided to applications through a unified interface for convenient use.

[0067] 3. This invention provides an integrated query method and system for maritime situation data and knowledge. Knowledge entities and general relationships are stored using a graph database. Track entities are retrieved via track description entities. Track point data is stored and associated with spatiotemporal relationships through track segment entities and a spatiotemporal index table. Spatiotemporal element attributes in knowledge entities are mapped to the spatiotemporal index table. Furthermore, spatiotemporal grid cells store the identifiers of track segment entities and knowledge entities with spatiotemporal relationships (see c.1–c.3 for details). This demonstrates that this invention supports both static updates of situation knowledge and dynamic updates of real-time track data and situation knowledge.

[0068] 4. This invention provides an integrated query method for maritime situational data and knowledge. The spatial encoding method involves treating the geographic coordinate space from 180°E to 180°W and from 90°N to 90°S as a plane. Given a grid cell representing a degree span (m,m), the plane is divided into several grid cells. Each grid cell is numbered starting from 0 using a filled curve encoding method; this grid cell number is the spatial code of the grid cell. The temporal encoding method involves dividing the time axis into several time intervals and assigning a unique 32-bit binary number to each time interval. By setting different values ​​for m, grid cells of different granularities can be obtained. The lower the value of m, the smaller the geographic spatial area represented by the grid cell, indicating higher encoding accuracy. When m approaches 0, the accuracy of the grid cell is infinitely close to the original continuous latitude and longitude coordinate space. Therefore, this invention supports multi-scale spatiotemporal encoding methods and can dynamically adjust the spatiotemporal grid granularity to meet the needs of different applications.

[0069] 5. This invention provides an integrated query method for maritime situational data and knowledge. Overall, the process is simple and clear, easy to implement, and can improve the reliability and stability of the implementation. At the same time, the query computational complexity is low, parameter adjustment is simple, and parallel processing is easy, further improving query efficiency. Attached Figure Description

[0070] Figure 1 This is a schematic diagram of entity relationships in the maritime situation knowledge graph provided in an embodiment of the present invention;

[0071] Figure 2This is a schematic diagram of the integrated organizational structure of maritime situation data and knowledge provided in an embodiment of the present invention;

[0072] Figure 3 This is a schematic diagram of the spatiotemporal relationship index table structure provided in an embodiment of the present invention;

[0073] Figure 4 This is a flowchart of the integrated query process for maritime situation data and knowledge provided in an embodiment of the present invention. Detailed Implementation

[0074] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0075] On the one hand, this invention provides an integrated query method for maritime situation data and knowledge, comprising the following steps:

[0076] Step 1: Construct a set of entities for the maritime situation knowledge graph: Based on the characteristics of track data and knowledge, the entities are divided into two main categories: track-type entities and knowledge-type entities;

[0077] A track entity consists of a track description entity and a track data segment entity. The track description entity describes the tagging and summary information of the track and consists of two parts: basic track entity elements and extended track entity elements. The basic track entity elements include track start time, track end time, track point count, track batch number, and target identifier. The extended track entity elements consist of a list of multiple element description tuples, which include: extended element number, extended element type, and extended element value. The track data segment entity elements include: track identifier, data segment start and end timestamps, data segment spatiotemporal code, and track data.

[0078] like Figure 1 As shown, knowledge entities are divided into 15 categories, and the element information of each category is as follows:

[0079] Situational Entities: Entity identifier, basic information, spatiotemporal range, target list, event list, attachment list, public opinion list, and display settings;

[0080] Platform Entities: Entity Identifier, Basic Information, Performance Parameters, Sensor List, Equipment List, Development Process, Event History, Attachment List, and Public Opinion List;

[0081] Sensor entity: entity identifier, basic information, performance parameters, development process, event history, attachment list and public opinion list;

[0082] Equipment entity: entity identification, basic information, performance parameters, development process, event history, list of attachments and list of public opinion;

[0083] Individual entity: entity identifier, basic information, education process, training process, employment process, reward process, event history, attachment list and public opinion list;

[0084] National and regional entities: entity identification, basic information, customs and ethnicity, economic situation;

[0085] Organizational Entity: Entity identifier, basic information, personnel structure, deployment information, equipment list, logistics list, training status, incident history, attachment list, and public opinion list;

[0086] Event Entity: Entity Identifier, Basic Information, Spatiotemporal Scope, Participating Organizations, Event Process, Event Report, Public Opinion List, and Attachment List;

[0087] Public opinion entity: entity identifier, public opinion title, basic information, source, update time, attachment content, and list of related public opinion;

[0088] Facility entity: entity identification, basic information, technical parameters, structural features, engineering process, event associations, attachment list, public opinion list, associated organizations and personnel;

[0089] Attachment Entity: Entity Identifier, Attachment Title, Attachment Type, Basic Information, Source, Update Time, and File Access Address;

[0090] Target entity: entity identifier, basic information, identification status, identification process, list of associated track entities, event list, public opinion list, and attachment list;

[0091] Regional Entities: Entity Identifier, Basic Information, Spatiotemporal Range, Height Range, and Regional Type;

[0092] Environmental entities: entity identifier, spatiotemporal range, meteorological data, hydrological data, electromagnetic environment, public opinion list, and attachment list;

[0093] Common knowledge entity: entity identifier, basic information, spatiotemporal scope, attachment list, and public opinion list;

[0094] Entity identifiers are represented by 64-bit hash strings, while all other elements are described using JSON strings; attributes can be further divided within the JSON strings as needed.

[0095] Step 2: Construct a knowledge graph set of maritime situational awareness relationships;

[0096] Based on the aforementioned entity composition, the relationship set in the maritime situation knowledge graph includes two main categories: ordinary relationships and spatiotemporal relationships. Ordinary relationships are represented using similar definitions to those in traditional knowledge graphs, establishing associations through entity identifiers. For example, the public opinion list within each entity is represented as a string of public opinion entity identifiers, establishing associations between entities and public opinion entities through this string. Spatiotemporal relationships are defined as follows: if two or more graph entities with spatiotemporal attributes intersect within a time or space range, a spatiotemporal relationship exists between them. Spatiotemporal relationships include temporal containment, temporal intersection, spatial containment, spatial intersection, spatiotemporal intersection, and spatiotemporal containment.

[0097] Time inclusion: If the time range of Entity 1 is within the time range of Entity 2, then Entity 2 is said to time include Entity 1.

[0098] Temporal intersection: If the time range of entity 1 intersects with the time range of entity 2, then entity 1 and entity 2 are said to have a temporal intersection relationship, and the temporal relationship is commutative;

[0099] Spatial inclusion: If the spatial extent of entity 1 is within the spatial extent of entity 2, then entity 2 is said to spatially include entity 1.

[0100] Spatial intersection: If the spatial extent of entity 1 intersects with the spatial extent of entity 2, then entity 1 and entity 2 are said to have a spatial intersection relationship, and the spatial intersection relationship is commutative;

[0101] Spatiotemporal intersection: If entity 1 and entity 2 intersect in both time and space, then entity 1 and entity 2 are said to have a spatiotemporal intersection relationship.

[0102] Spatiotemporal inclusion: If entity 1 and entity 2 are both temporally and spatially included, then entity 1 and entity 2 are said to have a spatiotemporal inclusion relationship.

[0103] Ordinary relationships are described and managed using existing knowledge graph relation methods, and can be stored and accessed through graph databases, such as NEO4J, to store ordinary entities and relationships; spatiotemporal relationships are managed, stored, and accessed through a spatiotemporal index table defined in this method.

[0104] Step 3: Construct a maritime situation knowledge graph;

[0105] The maritime situation knowledge graph combines maritime situation data (track-type entity data) with knowledge data (knowledge-type entities) according to... Figure 2The system unifies the organization of maritime situation data and knowledge data. Traditional knowledge entities and general relationships are stored and retrieved using a graph database. Track entity data is retrieved through track description entities. Track point data is stored and associated with spatiotemporal relationships through track segment entities and a spatiotemporal index table. The spatiotemporal element attributes in knowledge entities are also mapped to the spatiotemporal grid cells of the spatiotemporal index table. The spatiotemporal grid cells store the identifiers of track segment entities and knowledge entities that have spatiotemporal relationships. Queries based on spatiotemporal relationships first retrieve the spatiotemporal index table to obtain the identifier information of relevant entities, forming a candidate result set. Then, the traditional keyword-based retrieval method is used to search for entity information matching the search conditions in the candidate result set, thereby realizing the integrated organization and querying of maritime situation data and knowledge data.

[0106] S3.1: Data Preprocessing

[0107] Data preprocessing mainly consists of two parts: preprocessing of track-type entity data and preprocessing of knowledge-type entity data. The purpose of data preprocessing is to remove outlier data, classify the data, and provide cleaned data support for the extraction of entity and relation information. Based on the differences between track-type entity data and knowledge-type entity data, track-type entity data preprocessing removes outlier data from the track, classifies and summarizes track point entity data according to target batch number, and generates track summary information, among other functions. Knowledge data preprocessing, on the other hand, removes outlier data, classifies entities, and determines the validity of multimodal data, among other functions.

[0108] Preprocessing of track-type entities:

[0109] Remove abnormal flight track data: discard data with incomplete key data elements, data values ​​exceeding the valid range, and duplicate data;

[0110] Based on the target batch number, the track point data is organized into a track data classification summary table according to the chronological order of the timestamps;

[0111] Generate track summary information based on the track data classification summary table, including track start and end times, track spatial range, and number of track points;

[0112] Knowledge-based entity anomaly removal rules:

[0113] Remove abnormal knowledge data: Discard data with incomplete key data elements, no title, no content, chaotic character set, more than 50% missing image or video data, and completely duplicate data.

[0114] Entities are classified according to their source, and entities from the same source are grouped together to form a set of knowledge entity categories.

[0115] For the data in the knowledge entity classification set, its validity is further determined according to the knowledge entity validity judgment rules. The specific data validity rules can be set according to the application needs.

[0116] S3.2: Extract and update entity information

[0117] Extraction of track-type entities:

[0118] Generate track description entities using preprocessed track summary information;

[0119] The track data classification summary table is divided into several track segments. Each track segment generates a track segment entity. The track segment entity identifier is added to the track segment list attribute of the track description entity.

[0120] Entity extraction method for track segments:

[0121] Iterate through all track points of the track to which the track description entity belongs, calculate the spatiotemporal code based on the track point position, put all track points with the same spatiotemporal code into the track point list of the same track segment entity, and add the track segment entity identifier to the track segment list attribute of the track description entity.

[0122] Knowledge entity extraction:

[0123] Knowledge entities can be extracted using general knowledge graph entity extraction methods, such as keyword matching. Based on the specific description of the entity, information is filled into the entity elements. In particular, the following extraction methods are used for spatiotemporal range elements.

[0124] Calculate the time code based on the time range, calculate the spatial code based on the spatial range, combine the time code and spatial code, and add the entity identification information to the entity identification list in the corresponding spatiotemporal relationship index table;

[0125] S3.3: Extract and update relationship information

[0126] Spatiotemporal relationship extraction: Spatiotemporal relationships are mainly established through spatiotemporal coding. First, the spatiotemporal range is converted into a spatiotemporal coding representation, and then the relevant entity identifier is added to the index list of all spatiotemporal grids related to the entity.

[0127] General Relationship Extraction: A general knowledge graph relationship extraction method is adopted. In particular, the association relationship between this entity and other entities is established between the identifier information of other entities referenced in the entity element. These association relationships do not need to be stored separately, but can be stored together with the entity element.

[0128] S3.4: Update the spatiotemporal relation index table

[0129] The structure of the spatiotemporal relation index table is as follows: Figure 2 As shown, in order to achieve integrated organization of maritime situation data and knowledge, this invention constructs a spatiotemporal relationship index table to achieve efficient organization of spatiotemporal relationships. The spatiotemporal relationship index table uses spatiotemporal grid encoding to convert the time and latitude / longitude positions in the spatiotemporal relationships into codes with a length not exceeding 64 bits. Specifically, the spatiotemporal grid encoding adopts a unified encoding method, which is divided into two parts: spatial encoding and time encoding. The maximum length of both parts is 32 bits, which together make up a 64-bit binary number.

[0130] Spatial encoding methods:

[0131] The geographical coordinate space of the Earth from 180 degrees east longitude to 180 degrees west longitude and from 90 degrees north latitude to 90 degrees south latitude is regarded as a plane. The X-axis represents the longitude direction coordinate from -180 to 180, with positive numbers representing east longitude. The Y-axis represents the latitude direction, with Y-axis coordinates from -90 degrees to 90 degrees, and positive numbers representing north latitude.

[0132] Given a (m,m) value representing the degree span of a grid cell, for example, (1,1) means that the above plane is divided into grids every 1 degree of longitude and every 1 degree of latitude, which can be divided into 360*180 grid cells. Each grid cell represents a geographic spatial area with a length of 1 degree of longitude and a width of 1 degree of latitude.

[0133] Each grid cell is numbered starting from 0 using fill curve encoding; this number is the spatial code of the grid cell. The encoding method varies depending on the fill curve used, and this method does not impose any restrictions, but it is recommended to use the Hilbert fill curve encoding method. This spatial code can uniquely identify a grid cell in grid coordinates.

[0134] By setting different values ​​of m, grid cells of different granularities can be obtained. The lower the value of m, the smaller the geographic space represented by the grid cell, which means that the coding accuracy is higher. When m approaches 0, the accuracy of the grid cell is infinitely close to the original continuous latitude and longitude coordinate space.

[0135] Time encoding methods:

[0136] Using a similar idea to spatial coding, the time axis is divided into several time intervals, and each time interval is assigned a 32-bit binary number as a unique identifier.

[0137] The spatial and temporal codes described above are combined to form a spatiotemporal grid code. Each spatiotemporal grid code uniquely corresponds to a grid cell in the spatiotemporal grid. If two spatiotemporal grid codes are equal, it indicates that the two spatiotemporal codes point to the same spatiotemporal grid cell. If the first N bits of the spatial code and temporal code of the spatiotemporal grid code are the same, it indicates that the spatiotemporal grid cells pointed to by the two spatiotemporal codes are also adjacent in time and space. Any spatiotemporal grid coding method that satisfies the above conditions can be adopted by this invention. In this embodiment, the GEOSOT standard coding method is used to generate spatiotemporal grid code values ​​for grid cells.

[0138] The structure of the spatiotemporal relation index table is as follows: Figure 3 As shown, the spatiotemporal relationship index table consists of a grid cell index list. Each grid cell index list consists of a grid cell summary and a corresponding entity identifier list. The grid cell summary includes information elements such as the spatiotemporal grid code value corresponding to this grid, the number of entities in the grid, the statistical number of entity classifications in the grid, and the update timestamp.

[0139] The entity identifiers in the entity identifier list are the identifier information of the maritime situation knowledge graph entities involved in step one. An entity can be uniquely identified through this identifier information.

[0140] The entity grid cell is located by HASH search of spatiotemporal grid encoding. The spatiotemporal relation index table stores the spatiotemporal grid cell index information of the entity data, which can reduce the index table size and speed up the retrieval efficiency.

[0141] Step 4: The process of the integrated query method for maritime situation data and knowledge based on the maritime situation knowledge graph is as follows: Figure 4 As shown;

[0142] First, the system receives and analyzes query requests. Query requests are received through the query interface and include three elements: query keywords, represented by strings; query time range, identified by start and end times; and query spatial range, represented by longitude, latitude, and altitude ranges. The system then parses the query request to determine whether it contains both time and spatial ranges.

[0143] If both time and space ranges are included, the time and space ranges are converted into spatiotemporal network coding representations. A list of all spatiotemporal grid codes is obtained through the spatiotemporal grid coding calculation method. By accessing the spatiotemporal relationship index table, a list of entity identifiers that meet the spatiotemporal range conditions is obtained. Based on the entity identifiers in the entity identifier list, the corresponding entity elements are obtained from track-type entities and knowledge-type entities. Keyword fuzzy search is performed among the elements to obtain the entity element content that meets the keyword search conditions. The entity elements are then summarized and output.

[0144] If the time and space ranges are not included, the entity query results are obtained directly through the keyword query service of the full-text search engine and graph database, and the entity elements are summarized and output.

[0145] On the other hand, the present invention provides an integrated query system for maritime situation data and knowledge, including: a query request receiving module, a first query module, a second query module, an entity set construction module, a relationship set construction module, a spatiotemporal relationship index table construction module, and a situation knowledge graph construction module;

[0146] The query request receiving module is used to receive query requests and determine whether the query request elements contain a time range and a spatial range. If they do, the first query module is activated; otherwise, the second query module is activated.

[0147] The first query module converts the time and space ranges into spatiotemporal grid codes, calculates a list of all spatiotemporal grid codes, accesses the spatiotemporal relationship index table in the maritime situation knowledge graph to obtain a list of entity identifiers that meet the spatiotemporal range conditions, retrieves the corresponding entity elements from track-type entities and knowledge-type entities based on the entity identifiers in the entity identifier list, performs a keyword fuzzy search among the entity elements to obtain entity elements that meet the keyword search conditions, summarizes and outputs the entity elements, stops execution, and waits for the next query request.

[0148] The second query module is used to obtain entity query results through the full-text search engine and graph database keyword query service in the maritime situation knowledge graph, and then stop execution to wait for the next query request.

[0149] The entity set construction module is used to construct the entity set of the maritime situation knowledge graph; the entity set includes track-type entities and knowledge-type entities; track-type entities include track description entities and track segment entities;

[0150] The relation set construction module is used to construct relation sets for the maritime situation knowledge graph; among them, relation sets include ordinary relations and spatiotemporal relations; ordinary relations are relations in traditional knowledge graphs; spatiotemporal relations are the intersection of two or more graph entities with spatiotemporal attributes within a time range or a spatial range;

[0151] The spatiotemporal relation index table construction module is used to divide the Earth grid and simultaneously segment the time axis to form spatiotemporal grid units. By performing binary spatiotemporal grid encoding on the spatiotemporal grid units, a spatiotemporal relation index table is constructed.

[0152] The situational knowledge graph construction module is used to store knowledge entities and ordinary relationships in a graph database. Track entities are retrieved by summarizing track description entities. Track point data is stored and spatiotemporal relationships are associated through track segment entities and spatiotemporal index tables. Spatiotemporal element attributes in knowledge entities are mapped to the spatiotemporal index tables, and spatiotemporal grid cells store track segment entity identifiers and knowledge entity identifiers that have spatiotemporal relationships.

[0153] More preferably, the spatiotemporal relationship index table construction module includes a spatial coding unit, a temporal coding unit, and a coding merging unit;

[0154] The spatiotemporal grid coding is divided into spatial coding and temporal coding; the maximum length of both spatial coding and temporal coding is 32 bits.

[0155] The spatial coding unit is used to treat the geographic coordinate space from 180 degrees east longitude to 180 degrees west longitude and from 90 degrees north latitude to 90 degrees south latitude as a plane; given the degree span represented by a grid unit, the plane is divided into several grid units; each grid unit is numbered starting from 0 using a filled curve code, and the grid unit number is the spatial code of the grid unit;

[0156] The time encoding unit is used to divide the time axis into several time intervals, and assigns a 32-bit binary number to each time interval for unique identification.

[0157] The encoding merging unit is used to combine spatial encoding and temporal encoding to form a spatiotemporal grid encoding.

[0158] More preferably, the spatiotemporal relationship index table is composed of a grid cell index list, each grid cell index list including a grid cell summary and corresponding entity identifier list data; the grid cell summary includes the spatiotemporal grid code value, the number of entities in the grid, the statistical number of entity classifications in the grid, and the update timestamp.

[0159] More preferably, the situational knowledge graph construction module includes a track-type entity preprocessing unit, a knowledge-type entity preprocessing unit, a track-type entity extraction and update unit, a knowledge-type entity extraction and update unit, a spatiotemporal relationship extraction and update unit, and a general relationship extraction and update unit;

[0160] The track entity preprocessing unit is used to remove abnormal data from track entities; abnormal data includes incomplete key data elements, data values ​​exceeding the valid value range, and duplicate data; track point entities are organized into a track data classification summary table according to the order of timestamps; track summary information is generated based on the track data classification summary table, and the track summary information includes track start and end time, track spatial range, and number of track points;

[0161] The knowledge entity preprocessing unit is used to remove abnormal data from knowledge entities. Abnormal data includes incomplete key data elements, no title, no content, character set disorder, duplicate data, and more than 50% of image and video data are missing. The knowledge entities are classified according to their source, and entities from the same source are grouped together to form a knowledge entity classification set.

[0162] The track entity extraction and update unit is used to generate track description entities from track summary information; divide the track entities in the track data classification summary table into several track segment entities, add the track segment entity identifier to the track segment list attribute of the track description entity; traverse all track points of the track to which the track description entity belongs, calculate the spatiotemporal code based on the track point position, put all track points with the same spatiotemporal code into the track point list of the same track segment entity, and add the track segment entity identifier to the track segment list attribute of the track description entity;

[0163] The knowledge entity extraction and update unit is used to extract knowledge entities using keyword matching methods, and fill the information into the entity elements according to the specific description of the entity.

[0164] The spatiotemporal relationship extraction and update unit is used to calculate the time code based on the time range, calculate the spatial code based on the spatial range, combine the time code and the spatial code, and add the entity identification information to the entity identification list in the corresponding spatiotemporal relationship index table;

[0165] The ordinary relation extraction and update unit is used to extract ordinary relations using traditional knowledge graph relations.

[0166] More preferably, the spatiotemporal relationship includes temporal inclusion, temporal intersection, spatial inclusion, spatial intersection, spatiotemporal intersection, and spatiotemporal inclusion.

[0167] In practical applications, this invention also provides an integrated maritime situation data and knowledge query system, including:

[0168] The sensor data receiving and processing module is used to receive real-time target detection data sent by the sensor, select different processing methods according to the classification of the detection data, and complete the correlation, fusion and contraction of the target detection data to form target trajectory data;

[0169] The situation knowledge graph construction module is used to classify situation knowledge data, extract entities and relationships from each type of situation knowledge data according to entity definition and attribute element definition, and form a situation knowledge graph.

[0170] The integrated situational data management module is used to establish associations based on the spatiotemporal attributes of target trajectory data and situational knowledge graph data using a spatiotemporal relationship grid. It aligns and merges target trajectory data points with entities in the situational knowledge graph, and synchronously updates the spatiotemporal relationship grid according to the update status of situational data and situational knowledge graph, and maintains the grid index.

[0171] The integrated query module for situational data and knowledge is used to receive application query requests, convert the query requests into query semantic tuples, divide the query requests into several subqueries based on the semantic tuples, start a query thread for each subquery, obtain situational data and related situational knowledge through the spatiotemporal relation grid index, merge the indexes to form links, and feed them back to the application.

[0172] The log management module is used to record the timestamps of operations such as data reception, entity classification, entity extraction, relationship extraction, graph construction and updating, calculate processing latency, and issue alarms for timeouts.

[0173] In summary, compared with the prior art, the present invention has the following advantages:

[0174] This invention provides an integrated query method and system for maritime situation data and knowledge. The maritime situation knowledge graph includes general relationships and spatiotemporal relationships. A spatiotemporal relationship is the intersection of two or more graph entities with spatiotemporal attributes within a time or spatial range. Knowledge entities and general relationships are stored in a graph database. Track entities are retrieved via track description entities. Track point data is stored and associated with spatiotemporal relationships through track segment entities and a spatiotemporal index table. The spatiotemporal element attributes of knowledge entities are mapped to the spatiotemporal index table, and the spatiotemporal grid cells store the identifiers of track segment entities and knowledge entities with spatiotemporal relationships. Upon receiving a query request, it is determined whether the query request elements contain a time range and a spatial range. If a time range and a spatial range exist, the entity elements can be obtained using the time range, spatial range, and keywords. Therefore, this invention effectively solves the problem of effectively organizing, storing, managing, and querying real-time updated situation data and knowledge in highly dynamic environments, providing timely and reliable situation graph data support for intelligent maritime situation cognition and analysis applications.

[0175] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for integrated querying of maritime situation data and knowledge, characterized in that, Includes the following steps: D1: Receive the query request and determine whether the query request elements contain a time range and a spatial range. If they do, proceed to D2; otherwise, proceed to D3. D2: Convert the time and space ranges into spatiotemporal grid codes, calculate a list of all spatiotemporal grid codes through the spatiotemporal grid codes, and obtain a list of entity identifiers that meet the spatiotemporal range conditions by accessing the spatiotemporal relationship index table in the maritime situation knowledge graph. Based on the entity identifiers in the entity identifier list, retrieve the corresponding entity elements from track-type entities and knowledge-type entities; perform a keyword fuzzy search among the entity elements to retrieve the entity elements that meet the keyword search conditions, summarize and output the entity elements, stop execution, and wait for the next query request; D3: Obtain entity query results through the full-text search engine and graph database keyword query service in the maritime situation knowledge graph, stop execution and wait for the next query request; The method for constructing a maritime situation knowledge graph includes the following steps: a. Construct a maritime situation knowledge graph entity set and relation set; wherein, the entity set includes track-type entities and knowledge-type entities; track-type entities include track description entities and track segment entities; the relation set includes ordinary relations and spatiotemporal relations; ordinary relations are relations in traditional knowledge graphs; spatiotemporal relations are the intersection of two or more graph entities with spatiotemporal attributes within a time range or spatial range; b. Divide the Earth grid and simultaneously segment the time axis to form spatiotemporal grid units. Construct a spatiotemporal relationship index table by performing binary spatiotemporal grid encoding on the spatiotemporal grid units. c. Knowledge entities and ordinary relationships are stored in a graph database. Track entities are retrieved by using track description entities. Track point data is stored and spatiotemporal relationships are associated through track segment entities and spatiotemporal index tables. Spatiotemporal element attributes in knowledge entities are mapped to the spatiotemporal index tables, and spatiotemporal grid cells store track segment entity identifiers and knowledge entity identifiers that have spatiotemporal relationships.

2. The method for integrated querying of maritime situation data and knowledge according to claim 1, characterized in that, Spatiotemporal grid coding is divided into spatial coding and temporal coding; the maximum length of both spatial coding and temporal coding is 32 bits. The spatial encoding method includes the following steps: The geographical coordinate space of the Earth, from 180 degrees east longitude to 180 degrees west longitude and from 90 degrees north latitude to 90 degrees south latitude, is regarded as a plane; Given a degree span represented by a grid cell, divide the plane into several grid cells; Each grid cell is numbered starting from 0 using a fill curve encoding method, and the grid cell number is the spatial code of the grid cell. The time encoding method includes the following steps: Divide the timeline into several time intervals and assign a unique 32-bit binary number to each time interval. Spatial coding and temporal coding are combined to form spatiotemporal grid coding.

3. The method for integrated querying of maritime situation data and knowledge according to claim 2, characterized in that, The spatiotemporal relationship index table consists of a grid cell index list. Each grid cell index list includes a grid cell summary and corresponding entity identifier list data. The grid cell summary includes the spatiotemporal grid code value, the number of entities in the grid, the statistical number of entity classifications in the grid, and the update timestamp.

4. The method for integrated querying of maritime situation data and knowledge according to claim 1, characterized in that, Step c specifically includes the following steps: c.1: Entity data preprocessing, specifically including the following steps: c.1.1: Preprocessing of track-type entities, specifically including the following steps: Remove abnormal data from track-type entities; abnormal data includes incomplete key data elements, data values ​​exceeding the valid value range, and duplicate data; Organize the track point entities into a track data classification summary table according to the order of their timestamps; Generate track summary information based on the track data classification summary table. The track summary information includes the track start and end time, track spatial range, and number of track points. c.1.2: Knowledge entity preprocessing, specifically including the following steps: Abnormal data in knowledge entities will be removed. Abnormal data includes incomplete key data elements, no title, no content, character set disorder, duplicate data, and more than 50% of image and video data are missing. Entities are classified according to their source, and entities from the same source are grouped together to form a set of knowledge entity categories. c.2 Extract entity data, specifically including the following steps: c.2.1: Extraction and updating of track-type entity data, specifically including the following steps: Generate a track description entity from track summary information; The track category entities in the track data classification summary table are divided into several track segment entities, and the track segment entity identifiers are added to the track segment list attribute of the track description entity. Iterate through all track points of the track to which the track description entity belongs, calculate the spatiotemporal code based on the track point position, put all track points with the same spatiotemporal code into the track point list of the same track segment entity, and add the track segment entity identifier to the track segment list attribute of the track description entity. c.2.2: Extraction and updating of knowledge entity data: Knowledge entities are extracted using keyword matching, and information is filled into entity elements according to the specific description of the entity. c.3: Extraction and updating of spatiotemporal relationships: Calculate the time code based on the time range, calculate the spatial code based on the spatial range, combine the time code and spatial code, and add the entity identification information to the entity identification list in the corresponding spatiotemporal relationship index table; c.4: Extract ordinary relations using traditional knowledge graph relations.

5. The method for integrated querying of maritime situation data and knowledge according to any one of claims 1 to 4, characterized in that, The spatiotemporal relationships include temporal inclusion, temporal intersection, spatial inclusion, spatial intersection, spatiotemporal intersection, and spatiotemporal inclusion.

6. A maritime situation data and knowledge integrated query system, characterized in that, include: The system includes a query request receiving module, a first query module, a second query module, an entity set construction module, a relation set construction module, a spatiotemporal relation index table construction module, and a situational knowledge graph construction module. The query request receiving module is used to receive query requests and determine whether the query request elements contain a time range and a spatial range. If they do, the first query module is activated; otherwise, the second query module is activated. The first query module is used to convert the time range and spatial range into spatiotemporal grid codes, calculate a list of all spatiotemporal grid codes through the spatiotemporal grid codes, and obtain a list of entity identifiers that meet the spatiotemporal range conditions by accessing the spatiotemporal relationship index table in the maritime situation knowledge graph. Based on the entity identifiers in the entity identifier list, retrieve the corresponding entity elements from track-type entities and knowledge-type entities; perform a keyword fuzzy search among the entity elements to retrieve the entity elements that meet the keyword search conditions, summarize and output the entity elements, stop execution, and wait for the next query request; The second query module is used to obtain entity query results through the full-text search engine and keyword query service of the graph database in the maritime situation knowledge graph, and then stop execution and wait for the next query request; The entity set construction module is used to construct a maritime situation knowledge graph entity set; wherein, the entity set includes track-type entities and knowledge-type entities; track-type entities include track description entities and track segment entities; The relation set construction module is used to construct a relation set for the maritime situation knowledge graph; wherein, the relation set includes ordinary relations and spatiotemporal relations; ordinary relations are relations in traditional knowledge graphs; spatiotemporal relations are the intersection of two or more graph entities with spatiotemporal attributes within a time range or a spatial range; The spatiotemporal relationship index table construction module is used to divide the Earth grid and simultaneously segment the time axis to form spatiotemporal grid units. By performing binary spatiotemporal grid encoding on the spatiotemporal grid units, a spatiotemporal relationship index table is constructed. The situational knowledge graph construction module is used to store knowledge entities and ordinary relationships in a graph database. Track entities are retrieved by summarizing track description entities. Track point data is stored and spatiotemporal relationships are associated through track segment entities and spatiotemporal index tables. Spatiotemporal element attributes in knowledge entities are mapped to spatiotemporal index tables, and spatiotemporal grid cells store track segment entity identifiers and knowledge entity identifiers that have spatiotemporal relationships.

7. The integrated maritime situation data and knowledge query system according to claim 6, characterized in that, The spatiotemporal relationship index table construction module includes a spatial coding unit, a temporal coding unit, and a coding merging unit; The spatiotemporal grid coding is divided into spatial coding and temporal coding; the maximum length of both spatial coding and temporal coding is 32 bits. The spatial coding unit is used to treat the geographic coordinate space from 180 degrees east longitude to 180 degrees west longitude and from 90 degrees north latitude to 90 degrees south latitude as a plane; given the degree span represented by a grid unit, the plane is divided into several grid units; each grid unit is numbered starting from 0 using a filled curve code, and the grid unit number is the spatial code of the grid unit; The time encoding unit is used to divide the time axis into several time intervals, and assigns a 32-bit binary number to each time interval for unique identification. The encoding merging unit is used to combine spatial encoding and temporal encoding to form a spatiotemporal grid encoding.

8. The integrated maritime situation data and knowledge query system according to claim 7, characterized in that, The spatiotemporal relationship index table consists of a grid cell index list. Each grid cell index list includes a grid cell summary and corresponding entity identifier list data. The grid cell summary includes the spatiotemporal grid code value, the number of entities in the grid, the statistical number of entity classifications in the grid, and the update timestamp.

9. The integrated maritime situation data and knowledge query system according to any one of claims 6 to 8, characterized in that, The situational knowledge graph construction module includes a track-type entity preprocessing unit, a knowledge-type entity preprocessing unit, a track-type entity extraction and update unit, a knowledge-type entity extraction and update unit, a spatiotemporal relationship extraction and update unit, and a general relationship extraction and update unit. The track entity preprocessing unit is used to remove abnormal data from track entities; abnormal data includes incomplete key data elements, data values ​​exceeding the valid value range, and duplicate data; the track point entities are organized into a track data classification summary table according to the order of timestamps; track summary information is generated based on the track data classification summary table, and the track summary information includes track start and end time, track spatial range, and number of track points; The knowledge entity preprocessing unit is used to remove abnormal data from knowledge entities. Abnormal data includes incomplete key data elements, no title, no content, character set disorder, duplicate data, and more than 50% of image and video data are missing. The knowledge entities are classified according to their source, and entities from the same source are grouped together to form a knowledge entity classification set. The track-type entity extraction and update unit is used to generate track description entities from track summary information; divide the track-type entities in the track data classification summary table into several track segment entities, add the track segment entity identifier to the track segment list attribute of the track description entity; traverse all track points of the track to which the track description entity belongs, calculate the spatiotemporal code based on the track point position, put all track points with the same spatiotemporal code into the track point list of the same track segment entity, and add the track segment entity identifier to the track segment list attribute of the track description entity; The knowledge entity extraction and update unit is used to extract knowledge entities through keyword matching method, and fill information into entity elements according to the specific description of the entity. The spatiotemporal relationship extraction and update unit is used to calculate the time code based on the time range, calculate the spatial code based on the spatial range, combine the time code and the spatial code, and add the entity identification information to the entity identification list in the corresponding spatiotemporal relationship index table. The ordinary relation extraction and update unit is used to extract ordinary relations using traditional knowledge graph relations.

10. The integrated maritime situation data and knowledge query system according to claim 6, characterized in that, The spatiotemporal relationships include temporal inclusion, temporal intersection, spatial inclusion, spatial intersection, spatiotemporal intersection, and spatiotemporal inclusion.

Images