Figure and activity traceability analysis method and system based on open source geospatial intelligence, and electronic equipment

Abstract

The invention discloses a character and activity traceability analysis method and system based on open source geospatial intelligence and electronic equipment, and the method comprises the steps: constructing an intelligence library and a character target library; obtaining to-be-analyzed target person information; searching according to the target person information to obtain intelligence information; performing event extraction on the intelligence information to obtain event information and construct an event library; according to the target person information, supplementing the information library to obtain supplemented target person information; the intelligence information is screened; performing space-time analysis and event analysis on the clustered intelligence information to obtain space-time information; according to the spatio-temporal information, event venation analysis is carried out, an intelligence result is formed, and character and activity traceability analysis based on open source geospatial intelligence is completed. According to the method, the problem that efficient, accurate and reliable traceability analysis on multi-mode and multi-source open-source geospatial information data is difficult to realize is solved.

Description

Technical Field

[0001] The present invention relates to a method, system and electronic device for tracing and analyzing people and activities, and in particular to a method, system and electronic device for tracing and analyzing people and activities based on open source geospatial intelligence. Background Art

[0002] This section merely provides background information related to the present disclosure and is not necessarily prior art.

[0003] With the rapid development of geographic information systems (GIS), social media, satellite images, and various open source databases, open source geospatial intelligence has become a powerful tool for acquiring, analyzing, and interpreting activities and events. However, these open source data are not only huge in volume, but also in various formats, including text, images, videos, and spatiotemporal data, making it difficult for traditional manual analysis methods to cope with them. At the same time, with the surge in various types of data, how to effectively extract evidence from geospatial data and make reliable inferences based on this evidence has become an urgent problem that needs to be solved.

[0004] There is no mature method to achieve efficient, accurate and reliable analysis of multi-modal and multi-source open source geospatial intelligence data. Traditional data fusion and reasoning methods often perform poorly in highly complex and dynamic environments.

[0005] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0006] Purpose of the invention: The technical problem to be solved by the present invention is to provide a method, system and electronic device for tracing and analyzing people and activities based on open source geospatial intelligence in response to the deficiencies of the prior art.

[0007] In order to solve the above technical problems, the present invention discloses a method, system and electronic device for tracing and analyzing people and activities based on open source geospatial intelligence, wherein the method comprises the following steps:

[0008] Step 1: Build an intelligence database and a character target database;

[0009] Step 2, obtaining the target person information to be analyzed;

[0010] Step 3, searching the intelligence database constructed in step 1 according to the target person information to obtain intelligence information;

[0011] Step 4, extracting events from the intelligence information obtained in step 3, obtaining event information and building an event library;

[0012] Step 5, supplementing the intelligence database according to the target person information to obtain supplemented target person information;

[0013] Step 6, screening the intelligence information based on the supplemented target person information and the event information in step 4 as the subject;

[0014] Step 7, performing spatiotemporal analysis and event analysis on the clustered intelligence information to obtain spatiotemporal information;

[0015] Step 8: Conduct event context analysis based on the spatiotemporal information to generate intelligence results, thereby completing the person and activity tracing analysis based on open source geospatial intelligence.

[0016] Furthermore, the construction of the intelligence database and the character target database described in step 1 specifically includes:

[0017] Step 1-1, building an intelligence database, and putting the open source geospatial intelligence into the intelligence database;

[0018] Step 1-2, constructing a character target library, in which the names, photos and locations of characters are pre-set; the characters include the target character and the characters related to the target character.

[0019] Further, the search described in step 3 specifically includes:

[0020] Step 3-1, constructing a knowledge graph related to the target person information as an underlying knowledge representation model to provide a search mode for semantic analysis and intent understanding;

[0021] Step 3-2, setting two rounds of relevance sorting rules to control the sorting method of search results;

[0022] Step 3-3, based on the large language model, perform a geospatial intelligence search in the intelligence database to obtain intelligence information.

[0023] Furthermore, the event extraction described in step 4 specifically includes:

[0024] Step 4-1, analyzing the intelligence information and determining its format, wherein the format includes: text format and image format;

[0025] Step 4-2, if the intelligence information is in text format, perform entity extraction and event extraction on the intelligence information to obtain text event information; if the text event information includes the name of a person, determine whether the name of the person is relevant to the target person; if so, determine whether the name of the person is in the person target library; if not, add the name, photo and location of the person to the person target library;

[0026] If the intelligence information is in image format, analyzing the image category of the intelligence information and executing step 4-3, the image category includes human image and building image;

[0027] Step 4-3, if the image category is a person image, after performing target recognition, target matching is performed in the person target library to obtain person image event information; it is determined whether the person image is relevant to the target person; if so, it is determined whether the person image is in the person target library; if not, the person image is added to the person target library as a photo;

[0028] If the image category is a building image, performing image-based search, comparison and analysis on it to obtain building image event information;

[0029] Step 4-4, constructing an event library according to the text event information described in step 4-2, the person image event information and the building image event information described in step 4-3.

[0030] Furthermore, the intelligence information is screened as described in step 6, that is, information related to the topic is screened out based on similarity.

[0031] Furthermore, the spatiotemporal analysis and event analysis described in step 7 specifically include:

[0032] If the clustered intelligence information is in text format, extracting location information and time information from the intelligence information;

[0033] If the clustered intelligence information is in image format, target recognition is performed on the intelligence information to obtain building information;

[0034] The location information, time information and building information are saved as a spatiotemporal information table.

[0035] Furthermore, the supplementation of the intelligence database in step 5 specifically includes:

[0036] Search the target person information to be analyzed in an open source person search engine to obtain the first person supplementary information;

[0037] The first person's supplementary information is added to the intelligence database.

[0038] Furthermore, the step of supplementing the intelligence database in step 5 further includes:

[0039] Searching in other search media according to the target person information to be analyzed to obtain second person supplementary information, wherein the other search media include search engines and social media;

[0040] Add the second person's supplementary information to the intelligence database.

[0041] The present invention also proposes a person and activity tracing analysis system based on open source geospatial intelligence, which is used to implement the above method. The system includes:

[0042] Construction module, used to build intelligence database and character target database;

[0043] An acquisition module is used to obtain the target person information to be analyzed;

[0044] A search module, used to search the intelligence database according to the target person information to obtain intelligence information;

[0045] An extraction module, used to extract events from the intelligence information according to the format of the intelligence information to obtain event information;

[0046] A clustering module, used to cluster the intelligence information based on the target person information and / or event information;

[0047] The analysis module is used to perform spatiotemporal analysis and event analysis on the clustered intelligence information to obtain spatiotemporal information;

[0048] The intelligence output module is used to perform event context analysis based on the spatiotemporal information to form intelligence results.

[0049] The present invention also provides an electronic device, comprising: a processor, a memory and a bus, for running the above system.

[0050] Beneficial effects:

[0051] The method for tracing and analyzing the sources of people and activities based on open source geospatial intelligence provided by the present invention is based on technologies such as geospatial intelligence retrieval, extraction and reasoning of multimodal knowledge, and spatiotemporal character activity analysis and event context analysis to achieve a complete process of tracing and analyzing the sources of character activities. It comprehensively uses advanced technologies such as natural language processing and knowledge graphs to conduct in-depth analysis and intelligent reasoning on massive open source geospatial intelligence data. In issues involving tracing the sources of character activities, it combines activity reports in text data, real-time updates in social media, geographic information in satellite images, and possibly even meteorological data, to conduct comprehensive analysis and reasoning. It solves the problem that the existing technology is difficult to achieve efficient, accurate, and reliable tracing analysis of multimodal and multi-source open source geospatial intelligence data. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments, and the above and / or other advantages of the present invention will become more clear.

[0053] Figure 1 The figure is a schematic diagram of the overall process of an embodiment.

[0054] Figure 2 The figure is a schematic diagram of an application flow of an embodiment.

[0055] Figure 3 FIG. 4 is a schematic diagram of a target person in an embodiment.

[0056] Figure 4 A schematic diagram of intelligence information in text format in an embodiment of the present invention.

[0057] Figure 5 Schematic diagram of intelligence information in image format in an embodiment of the present invention.

[0058] Figure 6 It is a schematic diagram of building comparison in an embodiment of the present invention.

[0059] Figure 7 Schematic diagram of first target recognition in an embodiment of the present invention.

[0060] Figure 8 Schematic diagram of second target recognition in an embodiment of the present invention.

[0061] Fig. 9 It is a schematic diagram of the architecture in an embodiment of the present invention.

[0062] Fig.10 This is a schematic diagram of the physical structure of the electronic device provided by the present invention.

[0063] The accompanying drawings are denoted as follows:

[0064] 10-building module, 20-acquisition module, 30-search module, 40-extraction module, 50-clustering module, 60-analysis module, 70-intelligence output module, 80-electronic device, 801-processor, 802-memory, 803-bus. DETAILED DESCRIPTION

[0065] The present invention provides a method, system and electronic device for tracing and analyzing people and activities based on open source geospatial intelligence, which is used to solve the problems in the prior art that the amount of geospatial intelligence data is huge and the formats are different, and the efficiency, accuracy and reliability of the existing analysis methods are low. The method includes:

[0066] Build intelligence database and character target database;

[0067] Obtain information about the target person to be analyzed;

[0068] According to the target person information, search in the intelligence database to obtain intelligence information;

[0069] Extracting events from the intelligence information according to the format of the intelligence information to obtain event information;

[0070] Clustering the intelligence information based on the target person information and / or event information;

[0071] Performing spatiotemporal analysis and event analysis on the clustered intelligence information to obtain spatiotemporal information;

[0072] Based on the spatiotemporal information, event context analysis is performed to form intelligence results.

[0073] Furthermore, in one implementation, the building of the intelligence database and the character target database includes:

[0074] Presetting the names, photos and locations of people in the person target library;

[0075] The characters include a target character and characters related to the target character.

[0076] Furthermore, in one implementation, extracting intelligence events from the intelligence information according to the format of the intelligence information to obtain event information includes:

[0077] Determining the format of the intelligence information, the format of the intelligence information includes: text format and image format;

[0078] If the intelligence information is in text format, perform entity extraction and event extraction on the intelligence information to obtain text event information;

[0079] If the intelligence information is in an image format, determining the image category of the intelligence information, the image category of the intelligence information including a person image and a building image;

[0080] If the intelligence information is a person image, target recognition is performed in the person target library according to the intelligence information to obtain person image event information;

[0081] If the intelligence information is a building image, an image-based search, comparison and analysis is performed on the intelligence information to obtain building image event information.

[0082] Furthermore, in one implementation, if the intelligence information is in text format, after performing entity extraction and event extraction on the intelligence information to obtain text event information, the method further includes:

[0083] Determining whether the intelligence information includes the name of the person;

[0084] If the intelligence information includes the name of a person, determining whether the name of the person is relevant to the target person;

[0085] If the name of the person is relevant to the target person, determining whether the name of the person is in the person target database;

[0086] If the name of the person is not in the person target library, the name, photo and location of the person are added to the person target library.

[0087] Furthermore, in one implementation, if the intelligence information is a person image, after performing target recognition in the person target library according to the intelligence information and obtaining the person image event information, the method further includes:

[0088] Determining whether the person image is relevant to the target person;

[0089] If the person image is relevant to the target person, determining whether the person image is in the person target library;

[0090] If the person image is not in the person target library, the person image is added to the person target library as a photo.

[0091] Furthermore, in one implementation, after extracting intelligence events from the intelligence information according to the format of the intelligence information and obtaining the event information, the method further includes:

[0092] Searching the target person information to be analyzed in an open source person search engine to obtain the first person supplementary information;

[0093] The first person's supplementary information is added to the intelligence database.

[0094] Furthermore, in one implementation, performing spatiotemporal analysis and event analysis on the clustered intelligence information to obtain spatiotemporal information includes:

[0095] If the clustered intelligence information is in text format, extracting location information and time information from the intelligence information;

[0096] If the clustered intelligence information is in image format, target recognition is performed on the intelligence information to obtain building information and other image information;

[0097] The location information, time information, building information and other image information are integrated into a spatiotemporal information table.

[0098] Furthermore, in one implementation, after extracting intelligence events from the intelligence information according to the format of the intelligence information and obtaining the event information, the method further includes:

[0099] An event library is constructed based on the event information obtained after event extraction.

[0100] Furthermore, in one implementation, after extracting events from the intelligence information according to the format of the intelligence information to obtain the event information, the method further includes:

[0101] Searching in other search media based on the target person information to obtain supplementary information of a second person, wherein the other search media includes search engines and social media;

[0102] Add the second person's supplementary information to the intelligence database.

[0103] Another aspect of the present invention provides a person and activity tracing and analysis system based on open source geospatial intelligence, the system comprising:

[0104] Construction module, used to build intelligence database and character target database;

[0105] An acquisition module is used to obtain the target person information to be analyzed;

[0106] A search module, used to search the intelligence database according to the target person information to obtain intelligence information;

[0107] An extraction module, used to extract events from the intelligence information according to the format of the intelligence information to obtain event information;

[0108] A clustering module, used to cluster the intelligence information based on the target person information and / or event information;

[0109] The analysis module is used to perform spatiotemporal analysis and event analysis on the clustered intelligence information to obtain spatiotemporal information;

[0110] The intelligence output module is used to perform event context analysis based on the spatiotemporal information to form intelligence results.

[0111] The present invention also provides an electronic device, comprising: a processor, a memory and a bus, for running the above system.

[0112] It can be seen from the above technical scheme that the present invention provides a person and activity tracing analysis method, system and electronic device based on open source geospatial intelligence, wherein the method includes: first constructing an intelligence database and a person target database, after obtaining the target person information to be analyzed, searching in the intelligence database according to the target person information to obtain intelligence information, extracting events according to the format of the intelligence information to obtain event information, clustering the intelligence information with the target person information and / or event information as the theme, and then performing spatiotemporal analysis and event analysis to obtain spatiotemporal information; finally, performing event context analysis to form intelligence results.

[0113] Embodiment 1:

[0114] like Figure 1 As shown, an embodiment of the present invention provides a method for tracing and analyzing people and activities based on open source geospatial intelligence, the method comprising:

[0115] S101, build intelligence database and character target database;

[0116] S102, obtaining target person information to be analyzed;

[0117] S103, searching the intelligence database according to the target person information to obtain intelligence information;

[0118] S104, extracting events from the intelligence information according to the format of the intelligence information to obtain event information;

[0119] S105, clustering the intelligence information based on the target person information and / or event information;

[0120] S106, performing spatiotemporal analysis and event analysis on the clustered intelligence information to obtain spatiotemporal information;

[0121] S107, performing event context analysis based on the spatiotemporal information to generate intelligence results.

[0122] In this embodiment, S101 is an offline step, and S102 to S107 are online steps;

[0123] The intelligence library constructed in S101 is a library set up for a specific intelligence analysis scenario. The library contains open source geospatial intelligence, including open source data from various data types such as geographic information systems, social media, and satellite images. The person target library constructed in S101 is a library pre-set for a specific intelligence analysis scenario. The library contains a collection of information on the target subject to be analyzed for person activity intelligence.

[0124] In the method for tracing and analyzing the source of people and activities based on open source geospatial intelligence described in this embodiment, the step S101, constructing an intelligence database and a person target database, includes:

[0125] S1011, presetting the name, photo and location of the person in the person target library;

[0126] S1012, the characters include a target character and characters related to the target character.

[0127] In this embodiment, it is necessary to pre-set a character target library in S101 according to a specific intelligence analysis scenario, that is, the target character information to be analyzed in the subsequent S102 is pre-set in the character target library;

[0128] Specifically, if the target person of the person activity intelligence analysis includes Zhang San, then Zhang San's name, photo and location are set in the person target library, and at the same time, the names, photos and locations of people related to Zhang San are also set in the person target library; the name of the person includes Zhang San's Chinese and English names; the photos include photos of the person that need to be paid attention to in a specific intelligence analysis scenario, such as photos of Zhang San participating in a specific activity; the location includes location information that needs to be paid attention to in a specific intelligence analysis scenario, such as the coordinates of a specific activity participated in by Zhang San;

[0129] Specifically, the people related to Zhang San include people who belong to the same field as Zhang San, people Zhang San meets in specific activities, and relatives of Zhang San.

[0130] In the person and activity tracing and analysis method based on open source geospatial intelligence described in this embodiment, the S103, searching in the intelligence database according to the target person information to obtain intelligence information, includes: searching in the intelligence database according to the target person information to obtain intelligence information; specifically, the search method of this step can adopt geospatial intelligence retrieval technology enhanced by a large language model, facing the intelligence retrieval problem, searching in the intelligence database according to the target task information through semantic retrieval technology, and obtaining intelligence and events related to the target person information;

[0131] In this embodiment, the steps involving search and retrieval can adopt the geospatial intelligence retrieval technology enhanced by the large language model, including: constructing a relevant knowledge graph as the underlying knowledge representation model to provide a search mode for semantic analysis and intent understanding, that is, when searching, linking the target entities in the knowledge graph and marking the target entities in the search input; at the same time, setting two rounds of relevance sorting rules to accurately control the sorting method of search results, including sorting rules based on full-text search and sorting rules based on vector search; and providing geospatial intelligence search based on the large language model, that is, assembling the data obtained from a large number of documents through search into the prompts required by the large language model in the form of Chat so that the large language model can quickly find the accurate answers to user questions and provide the source of the documents; its workflow includes: offline index construction, document vectorization based on the semantic vector model, online search, and generation of answers by the large language model.

[0132] In the method for tracing and analyzing the source of people and activities based on open source geospatial intelligence described in this embodiment, the step S104 extracts intelligence events from the intelligence information according to the format of the intelligence information to obtain event information, including:

[0133] S1041, determining the format of the intelligence information, the format of the intelligence information including: text format and image format;

[0134] S1042, if the intelligence information is in text format, perform entity extraction and event extraction on the intelligence information to obtain text event information;

[0135] S1043, if the intelligence information is in image format, determining the image category of the intelligence information, where the image category of the intelligence information includes human images and building images;

[0136] S1044, if the intelligence information is a person image, target recognition is performed in the person target library according to the intelligence information to obtain person image event information;

[0137] S1045: If the intelligence information is a building image, perform image-based search, comparison and analysis on the intelligence information to obtain building image event information.

[0138] Specifically, in this embodiment, the search and comparison analysis of the building image in S1045 can be performed by comparing and analyzing with search engine images, Google map street view images, etc., to obtain buildings similar to the building image, and then analyze and obtain information such as the shooting location of the image;

[0139] In this embodiment, through S1041 to S1045, different intelligence event extraction methods can be used to obtain event information according to the specific format of the intelligence information obtained after retrieval; in addition, if the format of the open source geospatial intelligence information also includes other formats in addition to text format and image format, it is necessary to use event extraction methods corresponding to other formats according to the specific circumstances of other formats.

[0140] In the method for tracing and analyzing people and activities based on open source geospatial intelligence described in this embodiment, in S1042, if the intelligence information is in text format, after performing entity extraction and event extraction on the intelligence information to obtain text event information, the method further includes:

[0141] S10421, determining whether the intelligence information includes the name of the person;

[0142] S10422, if the intelligence information includes the name of a person, determining whether the name of the person is relevant to the target person;

[0143] S10423, if the name of the person is relevant to the target person, determine whether the name of the person is in the person target library;

[0144] S10424: If the name of the person is not in the person target library, add the name, photo and location of the person to the person target library.

[0145] In this embodiment, if the intelligence information in S10422 includes the name of a person, it is determined whether the name of the person is relevant to the target person, including: it is necessary to pre-set judgment conditions according to specific intelligence analysis scenarios. For example, the name of the person and the target person appear in multiple specific intelligence analysis scenarios. For example, the judgment condition can be set to whether the person appears in three special meetings at the same time. If so, it can be determined that the person is relevant to the target person; through S10421 to S10424, the names in the character target library can be effectively supplemented, thereby expanding the scope of intelligence information acquisition.

[0146] In the method for tracing and analyzing people and activities based on open source geospatial intelligence described in this embodiment, in S1044, if the intelligence information is a person image, target recognition is performed in the person target library according to the intelligence information to obtain person image event information, and the method further includes:

[0147] S10441, determining whether the person image is relevant to the target person;

[0148] S10442, if the person image is relevant to the target person, determine whether the person image is in a person target library;

[0149] S10443: If the person image is not in the person target library, add the person image as a photo to the person target library.

[0150] In this embodiment, the judgment method in S10441 needs to pre-set judgment conditions according to specific intelligence analysis scenarios. For example, a certain person image similarity threshold can be pre-set. If the person image reaches the similarity threshold after target recognition with the target person image, it is determined that the person image is relevant to the target person. If the person image is not in the person target library, it is supplemented. Therefore, through S10441 to S10443, the photos in the person target library can be effectively supplemented, thereby expanding the scope of intelligence information acquisition.

[0151] In the method for tracing and analyzing the source of people and activities based on open source geospatial intelligence described in this embodiment, the step S104 extracts intelligence events from the intelligence information according to the format of the intelligence, and after obtaining the event information, further includes:

[0152] S1046: construct an event library according to the event information obtained after event extraction.

[0153] Specifically, in this embodiment, an event library is constructed based on event information, which can integrate the event information to a certain extent, so as to facilitate operations such as clustering of topics in subsequent steps.

[0154] In the method for tracing and analyzing the source of people and activities based on open source geospatial intelligence described in this embodiment, the step S106 performs spatiotemporal analysis and event analysis on the clustered intelligence information to obtain spatiotemporal information, including:

[0155] S1061, if the clustered intelligence information is in text format, extracting location information and time information from the intelligence information;

[0156] S1062, if the clustered intelligence information is in image format, perform target recognition on the intelligence information to obtain building information and other image information;

[0157] S1063, integrating the location information, time information, building information and other image information into a spatiotemporal information table.

[0158] In this embodiment, by integrating the spatiotemporal information table through S1061~S1063, it is possible to clearly and accurately represent the correlated spatiotemporal information and event information, and integrate multiple key events according to time and place, so that the target person's activities can be decomposed into several continuous behaviors, and the current behavioral intentions of the target person can be inferred from dimensions such as behavior type and occurrence order, thereby breaking through the difficulty of character activity analysis in the spatiotemporal dimension; character activity tracing analysis needs to pay attention to specific events and more specific information about character-related events, such as the origin of the event, the character's participation in the development of the event, and the development direction of the event; this embodiment uses event tracing and evolution analysis, with characters or specific events as clues, to break through the event time sequence relationship mining technology in the character activity tracing analysis scenario.

[0159] In addition, in addition to the spatiotemporal analysis and event analysis methods provided in S106 of this embodiment, this method can also obtain the spatiotemporal information table as a prerequisite, combined with the character activity analysis method based on spatiotemporal features, to further infer the intentions and actions of the target person, for example: on the basis of the knowledge graph association expression, use rules and reasoning technology to conduct regular analysis of the behavior sequence; based on the behavior sequence and results of multiple actual activity events, summarize and analyze the regularity of the character's behavior and intention, and use the knowledge graph reasoning technology to reason about the behavior pattern; in the various behavior patterns of the character, set the behavior process and process constraints of the activity to form specific reasoning rules, such as abnormal calls and abnormal transactions in a certain activity. By analyzing the existing event activities in the graph based on the rules, the intention of the target person and the possible next action can be predicted through fast pattern reasoning.

[0160] In addition, in addition to the above-mentioned character activity analysis method combined with spatiotemporal characteristics, this method can also be combined with the event context analysis method of technical spatiotemporal data, targeting the character's target activity trajectory, communication behavior, activities and other multimodal data, based on the character intelligence knowledge graph data, combined with the hierarchical complex relationship network, based on the deep network association reasoning technology, discover the sequence, cause and effect, association, cycle and other patterns among various elements in the space of people, things, places, objects and organizations, and use the association analysis algorithm to mine and predict the relationship between different event elements, phenomena or events.

[0161] In the method for tracing and analyzing the source of people and activities based on open source geospatial intelligence described in this embodiment, the step S104 extracts intelligence events from the intelligence information according to the format of the intelligence, and after obtaining the event information, further includes:

[0162] S1047, searching the target person information to be analyzed in an open source person search engine to obtain first person supplementary information;

[0163] S1048, adding the first person's supplementary information to the intelligence database.

[0164] Specifically, in this embodiment, the open source person search can be selected according to a specific intelligence analysis scenario; through S1047 to S1048, multimodal geospatial intelligence related to the target person can be obtained in a targeted manner, further expanding the intelligence retrieval scope of the target person information.

[0165] In the method for tracing and analyzing people and activities based on open source geospatial intelligence described in this embodiment, in S104, after extracting events from the intelligence information according to the format of the intelligence information and obtaining the event information, the method further includes:

[0166] S1049, searching in other search media according to the target person information to obtain second person supplementary information, wherein the other search media includes search engines and social media;

[0167] S10410, adding the second person's supplementary information to the intelligence database.

[0168] Specifically, in this embodiment, the search engine and social media can be selected according to a specific intelligence analysis scenario. Through S1049 to S10410, the multimodal geospatial intelligence from the perspectives of person, time, place, etc. involved in ordinary search engines and social media can further expand the intelligence retrieval scope of the target person information.

[0169] Based on the method described in the above embodiment, the embodiment of the present invention also provides a person and activity tracing analysis system based on open source geospatial intelligence, such as Fig. 9 As shown, the system comprises:

[0170] Construction module 10, used to construct an intelligence database and a character target database;

[0171] An acquisition module 20 is used to acquire target person information to be analyzed;

[0172] A search module 30, used to search the intelligence database according to the target person information to obtain intelligence information;

[0173] An extraction module 40, configured to extract events from the intelligence information according to the format of the intelligence information to obtain event information;

[0174] A clustering module 50, used to cluster the intelligence information based on the target person information and / or event information;

[0175] An analysis module 60 is used to perform spatiotemporal analysis and event analysis on the clustered intelligence information to obtain spatiotemporal information;

[0176] The intelligence output module 70 is used to perform event context analysis based on the spatiotemporal information to generate intelligence results.

[0177] Embodiment 2:

[0178] like Figure 2 As shown in FIG. 1 , it is a schematic diagram of the application process of this embodiment; first, an intelligence database and a character target database are constructed; in this embodiment, the intelligence analysis task is carried out around the key person Zhang San, so the names of the people to be analyzed in the character target database are pre-set to include Zhang San (SanZhang), and the photos of Zhang San include Figure 3 As shown, Zhang San's location includes City A, and the specific coordinates are (X1, Y1);

[0179] Obtain information of a target person to be analyzed. In this embodiment, the target person to be analyzed is set to Zhang San according to the requirements of the intelligence analysis task;

[0180] According to Zhang San's name, photo and location, search in the intelligence database to obtain intelligence information; Figure 4 and Figure 5 Shown are two pieces of intelligence information obtained from the search;

[0181] like Figure 4 The intelligence information shown is in text format. Entity extraction and event extraction are performed on the intelligence information to obtain text event information: time (August 2), location (Airport A, City A), person (Zhang San), etc.

[0182] like Figure 5 The intelligence information shown is in image format, and target recognition is performed based on the constructed human target library; the recognition result is as follows Figure 5 As shown in the middle rectangular wireframe, "SanZhang (Zhang San)" can be obtained as the target person;

[0183] For Figure 5 the buildings contained in, through image-based search and comparison analysis, including comparison analysis with search engine pictures, Google Maps street view pictures, etc., similar buildings in the photo such as Figure 6 shown, which is a picture of the "XXX" building, the shooting location information of the picture can be analyzed;

[0184] Such as Figure 5 in the intelligence information shown, the target recognition result also includes "SiZhao (Zhao Si)", then the photo of Zhao Si is subjected to target recognition in the person target library, as Figure 7 shown, it can be determined that Zhao Si and Zhang San appeared simultaneously in multiple special meetings. At the same time, Zhao Si's person image is not in the person target library, then Zhao Si's photo, name, and location information as Figure 8 shown are all supplemented to the person target library; in addition, through target recognition, it is confirmed that Figure 5 the other people in do not reach the preset person image similarity threshold, so there is no need to supplement for other people;

[0185] Use open-source person search engines (TruthFinder, PeopleFinders, etc.) to search for Zhang San and Zhao Si to further supplement and confirm person target information; for example, through the image target person recognition algorithm, relevant pictures and relevant news reports are found, and the target person "Zhao Si" is extracted from the news. And in the basic knowledge graph of intelligence analysis, it shows that "Zhang San" and "Zhao Si" are relatives, and the occupation of "Zhao Si" is XX, which attracts the attention of intelligence analysts; to further analyze why "Zhao Si" appeared in Zhang San's trip to City A, and whether Zhao Si has other intentions, next, a traceability analysis of the person target "Zhao Si" is further carried out.

[0186] The process of activity traceability analysis for the target person "Zhao Si" is as follows:

[0187] Conduct an intelligence retrieval with "Zhao Si" as the condition to obtain relevant intelligence documents. Example news such as "Zhang San actually secretly brought his relatives when going to City A and was suspected of engaging in illegal activities!" and "X media reported that Zhang San's relatives used the trip to City A to engage in illegal activities";

[0188] Perform event extraction on the relevant intelligence texts to obtain events such as "A mysterious man appeared in the photos of the occasion of visiting Country B, and the relevant personnel said he was Zhao Si, Zhang San's relative" and "The report on X website seriously questioned Zhao Si's appearance in the City A area, and it was thought that he wanted to engage in illegal activities for his family by selling influence";

[0189] Identify people in relevant news photos, etc.; use a person search engine to search, confirm their identity information, and obtain their information on social media;

[0190] Cluster the acquired information to obtain all the relevant information, events, and activities of Zhao Si during Zhang San's trip to City Jia;

[0191] Use social media searches and people search engines to further expand intelligence information;

[0192] Through event analysis and spatiotemporal analysis, we extracted from the relevant intelligence news texts the occasions where "Zhao Si" appeared in Country B, Country C, and City A. We performed target recognition on the relevant pictures and identified the national flags of Country B and Country C (and Country D) from the pictures, which shows that "Zhao Si" appeared in Country B and Country C. However, we did not extract or identify the intelligence about him being in Country E from the news texts and pictures, which means that he "disappeared" during his time in Country E.

[0193] We reasoned and analyzed why Zhao Si disappeared in Country E. We obtained the following intelligence information from other intelligence: Zhao Si is currently the president of AB Company. The company was originally going to sign a cooperation agreement with a company in Country E in February 2023, but the agreement was postponed after Zhao Si was investigated by XX in January. From the above information, we can speculate whether he was completing an agreement that had not been signed before. At the same time, relevant people analyzed from the search engine that "he reached a business deal in Country E";

[0194] Generation of analysis results: The above intelligence retrieval and analysis ultimately form an analysis result report.

[0195] It can be seen from the above technical solutions that the present invention provides a person and activity tracing analysis method based on open source geospatial intelligence. First, an intelligence database and a person target database are constructed. After obtaining the target person information to be analyzed, the intelligence database is searched according to the target person information to obtain intelligence information. Events are extracted according to the format of the intelligence information to obtain event information. The intelligence information is clustered with the target person information and / or event information as the theme, and then spatiotemporal analysis and event analysis are performed to obtain spatiotemporal information. Finally, event context analysis is performed to form intelligence results. Compared with the prior art, the present invention has the following advantages:

[0196] The present invention provides a method for tracing and analyzing the sources of people and activities based on open source geospatial intelligence, based on technologies such as geospatial intelligence retrieval, extraction and reasoning of multimodal knowledge, and spatiotemporal character activity analysis and event context analysis, to achieve a complete process for tracing and analyzing the sources of character activities. Advanced technologies such as natural language processing and knowledge graphs are used in combination to conduct in-depth analysis and intelligent reasoning on massive open source geospatial intelligence data. In issues involving tracing the sources of character activities, comprehensive analysis and reasoning are performed in combination with activity reports in text data, real-time updates in social media, geographic information in satellite images, and possibly even meteorological data. This solves the problem that existing technologies are difficult to achieve efficient, accurate, and reliable tracing and analysis of multimodal and multi-source open source geospatial intelligence data.

[0197] Fig.10 A schematic diagram of the physical structure of an electronic device provided in an embodiment of the present invention, such as Fig.10 As shown, the electronic device 80 includes: a processor 801 (processor), a memory 802 (memory) and a bus 803;

[0198] Among them, the processing 801 and the memory 802 communicate with each other through the bus 803;

[0199] Processor 801 is used to call program instructions in memory 802 to execute the methods provided by the above-mentioned method embodiments, for example, including: constructing an intelligence library and a character target library; obtaining target character information to be analyzed; searching in the intelligence library based on the target character information to obtain intelligence information; performing event extraction on the intelligence information according to the format of the intelligence information to obtain event information; clustering the intelligence information with the target character information and / or event information as the theme; performing spatiotemporal analysis and event analysis on the clustered intelligence information to obtain spatiotemporal information; performing event context analysis based on the spatiotemporal information to form intelligence results.

[0200] The present embodiment also provides a non-transitory computer-readable storage medium, which stores computer instructions. The computer instructions enable a computer to execute the methods provided by the above-mentioned method embodiments, for example, including: constructing an intelligence library and a character target library; obtaining target character information to be analyzed; searching the intelligence library based on the target character information to obtain intelligence information; extracting events from the intelligence information in the format of the intelligence information to obtain event information; clustering the intelligence information with the target character information and / or event information as the theme; performing spatiotemporal analysis and event analysis on the clustered intelligence information to obtain spatiotemporal information; performing event context analysis based on the spatiotemporal information to form intelligence results.

[0201] In a specific implementation, the present application provides a computer storage medium and a corresponding data processing unit, wherein the computer storage medium can store a computer program, and when the computer program is executed by the data processing unit, the invention content of a method, system and electronic device for tracing and analyzing people and activities based on open source geospatial intelligence provided by the present invention, as well as some or all steps in each embodiment. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0202] Those skilled in the art can clearly understand that the technical solutions in the embodiments of the present invention can be implemented by means of computer programs and their corresponding general hardware platforms. Based on such an understanding, the technical solutions in the embodiments of the present invention can be essentially or partly contributed to the prior art in the form of computer programs, i.e., software products, which can be stored in a storage medium and include several instructions for enabling a device including a data processing unit (which can be a personal computer, a server, a single-chip microcomputer, an MCU or a network device, etc.) to execute the methods described in various embodiments of the present invention or certain parts of the embodiments.

[0203] The present invention provides a method, system and electronic device for tracing and analyzing people and activities based on open source geospatial intelligence. There are many methods and ways to implement the technical solution. The above is only a preferred implementation of the present invention. It should be pointed out that for ordinary technicians in this technical field, without departing from the principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the scope of protection of the present invention. All components not specified in this embodiment can be implemented by existing technologies.

Claims

1. A method for tracing and analyzing people and activities based on open source geospatial intelligence, characterized in that: The method comprises the following steps: Step 1: Build an intelligence database and a character target database; Step 2, obtaining the target person information to be analyzed; Step 3, searching the intelligence database constructed in step 1 according to the target person information to obtain intelligence information; Step 4, extracting events from the intelligence information obtained in step 3, obtaining event information and building an event library; Step 5, supplementing the intelligence database according to the target person information to obtain supplemented target person information; Step 6, screening the intelligence information based on the supplemented target person information and the event information in step 4 as the subject; Step 7, performing spatiotemporal analysis and event analysis on the clustered intelligence information to obtain spatiotemporal information; Step 8: Conduct event context analysis based on the spatiotemporal information to generate intelligence results, thereby completing the person and activity tracing analysis based on open source geospatial intelligence.

2. According to claim 1, a method for tracing and analyzing people and activities based on open source geospatial intelligence is characterized in that: The construction of the intelligence library and character target library described in step 1 specifically includes: Step 1-1, building an intelligence database, and putting the open source geospatial intelligence into the intelligence database; Step 1-2, constructing a character target library, in which the names, photos and locations of characters are pre-set; the characters include the target character and the characters related to the target character.

3. According to claim 2, a method for tracing and analyzing people and activities based on open source geospatial intelligence is characterized in that: Perform the search described in step 3, including: Step 3-1, constructing a knowledge graph related to the target person information as an underlying knowledge representation model to provide a search mode for semantic analysis and intent understanding; Step 3-2, setting two rounds of relevance sorting rules to control the sorting method of search results; Step 3-3, based on the large language model, perform a geospatial intelligence search in the intelligence database to obtain intelligence information.

4. According to claim 3, a method for tracing and analyzing people and activities based on open source geospatial intelligence is characterized in that: The event extraction described in step 4 specifically includes: Step 4-1, analyzing the intelligence information and determining its format, wherein the format includes: text format and image format; Step 4-2, if the intelligence information is in text format, perform entity extraction and event extraction on the intelligence information to obtain text event information; if the text event information includes the name of a person, determine whether the name of the person is relevant to the target person; if so, determine whether the name of the person is in the person target library; if not, add the name, photo and location of the person to the person target library; If the intelligence information is in image format, analyzing the image category of the intelligence information and executing step 4-3, the image category includes human image and building image; Step 4-3, if the image category is a person image, after performing target recognition, target matching is performed in the person target library to obtain person image event information; it is determined whether the person image is relevant to the target person; if so, it is determined whether the person image is in the person target library; if not, the person image is added to the person target library as a photo; If the image category is a building image, performing image-based search, comparison and analysis on it to obtain building image event information; Step 4-4, constructing an event library according to the text event information described in step 4-2, the person image event information and the building image event information described in step 4-3.

5. According to claim 4, a method for tracing and analyzing people and activities based on open source geospatial intelligence is characterized in that: The intelligence information is screened in step 6, that is, information related to the topic is screened out based on similarity.

6. The method for tracing and analyzing people and activities based on open source geospatial intelligence according to claim 5 is characterized in that: The spatiotemporal analysis and event analysis described in step 7 specifically includes: If the clustered intelligence information is in text format, extracting location information and time information from the intelligence information; If the clustered intelligence information is in image format, target recognition is performed on the intelligence information to obtain building information; The location information, time information and building information are saved as a spatiotemporal information table.

7. The method for tracing and analyzing people and activities based on open source geospatial intelligence according to claim 6 is characterized in that: The supplementation of the intelligence database described in step 5 specifically includes: Search the target person information to be analyzed in an open source person search engine to obtain the first person supplementary information; The first person's supplementary information is added to the intelligence database.

8. The method for tracing and analyzing people and activities based on open source geospatial intelligence according to claim 7 is characterized in that: The step 5 of supplementing the intelligence base further includes: Searching in other search media according to the target person information to be analyzed to obtain second person supplementary information, wherein the other search media include search engines and social media; Add the second person's supplementary information to the intelligence database.

9. A person and activity tracing analysis system based on open source geospatial intelligence, used to implement the method described in any one of claims 1 to 8, characterized in that: The system comprises: Construction module, used to build intelligence database and character target database; An acquisition module is used to obtain the target person information to be analyzed; A search module, used to search the intelligence database according to the target person information to obtain intelligence information; An extraction module, used to extract events from the intelligence information according to the format of the intelligence information to obtain event information; A clustering module, used to cluster the intelligence information based on the target person information and / or event information; The analysis module is used to perform spatiotemporal analysis and event analysis on the clustered intelligence information to obtain spatiotemporal information; The intelligence output module is used to perform event context analysis based on the spatiotemporal information to form intelligence results.

10. An electronic device, characterized in that: include: A processor, a memory and a bus, for running the system as claimed in claim 9.

Images