Visual narrative method and system for typical war operations

Abstract

The present invention belongs to the field of information visualization technology, and discloses a method and system for visual narration of typical war actions, the method comprising: step 1: using information extraction technology in intelligent language processing, extracting various aspects of information describing the war from a large amount of text, and obtaining fields corresponding to different combat forces; step 2: processing the obtained fields corresponding to different combat forces, obtaining common data model fields, and completing the construction of a visual narrative data model; step 3: constructing a visual narrative model based on the visual narrative data model, and visually expressing the war background, war process, war results and impact. The present invention fully explores the characteristics of different stages of war by conducting narrative visualization research on the action data of typical wars, and studies different visualization methods according to different characteristics, so that war visualization can better display narrative, help users understand the war background, grasp the war process, and clarify the war impact.

Description

Technical Field

[0001] The present invention belongs to the technical field of information visualization, and in particular, relates to a visualization narrative method and system for typical war actions. Background Art

[0002] With the development of information technology, data is widely used in all aspects of the military field, becoming the main force driving the development of military science and technology and a key factor in determining the outcome of war. Knowing yourself and the enemy, you can fight a hundred battles without danger. In modern warfare, it is particularly important to study and analyze the mode of the opponent's combat operations. By collecting and processing the historical data of the opponent's combat operations, we can obtain the evolution of different stages of the war and provide a reference for predicting future combat operations.

[0003] Visual narrative analysis [Zhu Mengze, Zhao Haiying. A review of narrative visualization [J]. Journal of Computer-Aided Design & Graphics, 2019, 31(10): 1719-1727.] has incomparable advantages over traditional war analysis. It can quickly understand a war, including analyzing the course of the war and the main tactics in the war. It is widely used in studying the characteristics of the opponent's combat laws and popularizing war science education.

[0004] There is no universally accepted definition of visual narrative in the field of visualization. Wita Wojtkowski and W. Gregory Wojtkowski believe that narrative visualization should be information visualization with the following three characteristics: ① Containing and presenting effective information; ② Being as detailed and fascinating as a good story; ③ Making the information more quickly and efficiently understood by readers. Among them, how to tell the visual information like a story is the difficulty of visual narrative analysis.

[0005] At present, the research on narrative visualization mainly focuses on news narrative visualization, that is, information visualization; however, there are few studies on narrative methods and visualization of war, and the research content is lacking. The existing visualization research on war mainly focuses on the battlefield environment, enemy situation, and our situation, and there is a gap in the research on the visualization narrative method of the overall process of war.

[0006] For historical combat operations, the recording materials are mainly in the form of text and images, and some are visualized in the form of infographics. There are relatively few results in visual narrative analysis. Describing wars in text has a wide range of coverage and can describe the entire war in a relatively comprehensive manner. However, due to its pure text characteristics, the expression effect and efficiency are poor, and it is often unable to dynamically display the dynamic changes of the war process, unable to highlight important information such as personnel and weapons in the event, and difficult to fully display the historical picture of combat operations; visual analysis of wars in animation overcomes the shortcomings of poor visualization of pure text analysis and can more intuitively visualize the entire war, but due to its non-interactive characteristics, it is not conducive to the audience's analysis of the war based on a certain aspect; analyzing wars in the traditional visual analysis method, most of its analysis can describe the war in stages through the interactive method of the timeline, and the description of each stage of the war is relatively clear, but due to its limited coverage, most of the visual analysis can only be general descriptions and cannot show details. Summary of the invention

[0007] The present invention aims to solve the problems that the description of war in text form has poor expression effect and efficiency; the visual analysis of war in animation form cannot be interactive, which is not conducive to the audience's analysis of the war based on a certain aspect; and the analysis of war in traditional visual analysis mode cannot show details. A visual narrative method and system for typical war actions are proposed.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions:

[0009] On one hand, the present invention provides a method for visual narration of typical war actions, comprising:

[0010] Step 1: Using information extraction technology in intelligent language processing, extract various aspects of information describing the war from a large amount of text to obtain fields corresponding to different combat forces;

[0011] Step 2: Process the fields corresponding to the different combat forces to obtain the common data model fields and complete the construction of the visual narrative data model;

[0012] Step 3: Build a visual narrative model based on the visual narrative data model to visualize the war background, war process, war results and impact.

[0013] Furthermore, the step 1 comprises:

[0014] In the information extraction stage, data fields related to the combat process are extracted through a rule-based matching algorithm: matching rules are designed to integrate the fields existing in the data model into the matching rules; matching rules are used to obtain paragraphs related to the description of the combat process, and data item fields are extracted from the paragraphs; by traversing the text and analyzing the topic of each paragraph, the paragraphs related to the combat process are determined, and then the data item fields are extracted based on information extraction technology.

[0015] Furthermore, the step 2 comprises:

[0016] Fill in missing data: Based on the matching of place names, use the place name and address matching algorithm to fill in the locations where the longitude and latitude information of the combat sites exists in the place name database. If the longitude and latitude information of the combat sites does not exist in the place name database, then use the relevant open source information on the Internet and the relevant descriptions of the combat sites during the combat operations phase to determine the longitude and latitude information of the combat sites through the water injection software.

[0017] Furthermore, the general data model fields include: combat time, combat name, strike dimension, equipment type, equipment name, number of weapons, starting point, destination, starting point longitude, starting point latitude, destination longitude, destination latitude, duration, participating services, combat purpose, action scale, number of annihilations, number of casualties, war impact, results, and introduction.

[0018] Furthermore, the construction process of the visual narrative model includes:

[0019] In the war background visualization narrative model construction phase, the evolution process before the outbreak of the war is presented in the form of a timeline; while presenting the war background and information, the war background is presented through corresponding videos and texts in combination with pattern analysis; according to the video progress or text and pattern browsing situation, the content corresponding to the current time node event is supplemented with visualization charts;

[0020] In the stage of constructing a visual narrative model of the war process, a flow map is used to display the starting point and destination of the air strike. The thickness and color of the trajectory line reflect the number and type of aircraft departing from the starting point. A timeline is used to describe the progress of the air strike, and the same air strike is placed at the same time node. With time as the main line, the route changes of the ground offensive forces and related situations as the war progresses are displayed, and the number of casualties, equipment names, and the number of weapons are added to the ground offensive route map.

[0021] During the construction phase of the visual narrative model of the war’s results and impacts, a visual expression based on statistical charts was adopted, with different statistical charts used to demonstrate the impacts of different aspects: in terms of politics, text was used to express it, supplemented by images to enhance the expressiveness; in terms of economy, bar charts and line charts were used to demonstrate the changes in domestic GDP, employment, and per capita consumption levels of both sides after the war; in terms of society, rose charts, fan charts, and regional distribution charts were used to display domestic public opinion and poll information after the war.

[0022] Another aspect of the present invention provides a visual narrative system for typical war actions, comprising:

[0023] The information extraction module is used to extract various information describing the war from a large amount of text using information extraction technology in intelligent language processing, and obtain fields corresponding to different combat forces;

[0024] A visual narrative data model construction module is used to process the fields corresponding to different combat forces, obtain the common data model fields, and complete the visual narrative data model construction;

[0025] The visual narrative model construction and visual expression module is used to build a visual narrative model based on the visual narrative data model, and to visually express the war background, war process, war results and impact.

[0026] Furthermore, the information extraction module is specifically used for:

[0027] In the information extraction stage, data fields related to the combat process are extracted through a rule-based matching algorithm: matching rules are designed to integrate the fields existing in the data model into the matching rules; matching rules are used to obtain paragraphs related to the description of the combat process, and data item fields are extracted from the paragraphs; by traversing the text and analyzing the topic of each paragraph, the paragraphs related to the combat process are determined, and then the data item fields are extracted based on information extraction technology.

[0028] Furthermore, the visual narrative data model construction module is specifically used for:

[0029] Fill in missing data: Based on the matching of place names, use the place name and address matching algorithm to fill in the locations where the longitude and latitude information of the combat sites exists in the place name database. If the longitude and latitude information of the combat sites does not exist in the place name database, then use the relevant open source information on the Internet and the relevant descriptions of the combat sites during the combat operations phase to determine the longitude and latitude information of the combat sites through the water injection software.

[0030] Furthermore, the general data model fields include: combat time, combat name, strike dimension, equipment type, equipment name, number of weapons, starting point, destination, starting point longitude, starting point latitude, destination longitude, destination latitude, duration, participating services, combat purpose, action scale, number of annihilations, number of casualties, war impact, results, and introduction.

[0031] Furthermore, the construction process of the visual narrative model includes:

[0032] In the war background visualization narrative model construction phase, the evolution process before the outbreak of the war is presented in the form of a timeline; while presenting the war background and information, the war background is presented through corresponding videos and texts in combination with pattern analysis; according to the video progress or text and pattern browsing situation, the content corresponding to the current time node event is supplemented with visualization charts;

[0033] In the stage of constructing a visual narrative model of the war process, a flow map is used to display the starting point and destination of the air strike. The thickness and color of the trajectory line reflect the number and type of aircraft departing from the starting point. A timeline is used to describe the progress of the air strike, and the same air strike is placed at the same time node. With time as the main line, the route changes of the ground offensive forces and related situations as the war progresses are displayed, and the number of casualties, equipment names, and the number of weapons are added to the ground offensive route map.

[0034] During the construction phase of the visual narrative model of the war’s results and impacts, a visual expression based on statistical charts was adopted, with different statistical charts used to demonstrate the impacts of different aspects: in terms of politics, text was used to express it, supplemented by images to enhance the expressiveness; in terms of economy, bar charts and line charts were used to demonstrate the changes in domestic GDP, employment, and per capita consumption levels of both sides after the war; in terms of society, rose charts, fan charts, and regional distribution charts were used to display domestic public opinion and poll information after the war.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] This invention conducts narrative visualization research on action data of typical wars, fully explores the characteristics of different stages of war, and studies different visualization methods according to different characteristics, so that war visualization can better display narrative, help users understand the background of war, grasp the progress of war, and clarify the impact of war. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A basic flow chart of a method for visual narration of typical war actions according to an embodiment of the present invention;

[0038] Figure 2This is a diagram showing some data examples of a visual narrative data model for a typical war action visual narrative method according to an embodiment of the present invention;

[0039] Figure 3 A schematic diagram of a visualization narrative model architecture of a visualization narrative method for typical war actions according to an embodiment of the present invention;

[0040] Figure 4 This is an example diagram of a visualized narrative method for a typical war action according to an embodiment of the present invention;

[0041] Figure 5 This is an example diagram of a war background visualization narrative method for a typical war action visualization narrative method according to an embodiment of the present invention;

[0042] Figure 6 This is an example diagram of a war process visualization narrative method for a typical war action visualization narrative method according to an embodiment of the present invention;

[0043] Figure 7 This is an example diagram of a base situation for a typical war action visualization narrative method according to an embodiment of the present invention;

[0044] Figure 8 This is an example diagram of a visual narrative of war results and impacts for a typical war action visual narrative method according to an embodiment of the present invention;

[0045] Fig. 9 The figure is a schematic diagram of the architecture of a visual narrative system for typical war actions according to an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The present invention will be further explained below with reference to the accompanying drawings and specific embodiments:

[0047] Since most of the relevant data at this stage still exist in the form of plain text, the information extraction technology in intelligent language processing is used in the data preparation stage to extract various information describing the war from a large amount of text. In the war text data, the location information is mainly described in the form of place names, and the visualization of location data is based on longitude and latitude. For objects that do not exist in the place name database, their longitude and latitude cannot be determined. Therefore, it is necessary to complete the data item. The place name address matching algorithm is used to improve the data item in combination with human-computer interaction. Since there are too many data items extracted about the description of the war, and there are also a large number of invalid or even vacant data items, the dynamic feature selection algorithm is used to simplify them and finally generate a data model. Based on this data model, the visual narrative analysis method is used to explore and analyze the actions of typical wars (Gulf War, Kosovo War, Afghanistan War, Iraq War).

[0048] Specifically, Figure 1 As shown in the figure, a visual narrative method for typical war operations includes:

[0049] Step S101: using information extraction technology in intelligent language processing to extract various aspects of information describing the war from a large amount of text, and obtaining fields corresponding to different combat forces;

[0050] Step S102: Process the fields corresponding to the different combat forces to obtain the common data model fields, and complete the construction of the visual narrative data model;

[0051] Step S103: construct a visual narrative model based on the visual narrative data model to visually express the war background, war process, war results and impact.

[0052] Specifically, due to the particularity of war data, it is impossible to obtain all the data describing a typical war through a single channel, and most of the existing relevant materials are text data, and there is no structured data that can be used directly, which brings more difficult challenges to the data collection stage.

[0053] The data used can be divided into two parts: open source data on the Internet and text data in books. Books are the main source of data. Structured data is formed through the text descriptions of typical wars through various stages in the books, and then a general data model is designed at the data level based on project exploration.

[0054] Taking into account all the data required in the visualization design stage, the typical war operation process is studied. For the war visualization narrative direction, more attention is paid to the entire war operation process and the combat force and combat style used in each process. Then, on the premise of ensuring that the data model is universal and highly scalable, the data model is divided into three stages: air strike force, ground combat force, and surface combat force according to the order of participation of various forces in the combat process. All data fields are shown in Table 1.

[0055] Table 1 Fields corresponding to different combat forces

[0056]

[0057]

[0058] Through the research and analysis of data types and combined with the needs of the next step, the dynamic feature selection algorithm can be used to further optimize the data fields, so that the data field dimensions are reduced. For the attribute items of casualties of the own side, enemy casualties, participating troops and duration, due to the large number of vacancies, they are deleted; for the impact of the war, due to the inconsistency of the relevant descriptions in various materials, it is difficult to obtain generally recognized results, so they are also deleted; at the same time, according to the division of combat forces, the types of weapons and equipment of the three combat forces can be merged to generate two new data items of strike dimension and equipment type; for the description of the two data items of destination and starting point, their usefulness in the spatiotemporal visualization stage is far less than the longitude and latitude information of the destination and starting point, so the longitude and latitude attribute items are supplemented through post-processing; for the description of the combat results, due to the complexity of its description, the density of its available information is too low, and victory and failure are determined as two values ​​of the result attribute through information extraction. Finally, the data model designed for this project is formed, and the fields contained in the final data model are shown in Table 2.

[0059] Table 2 Common data model fields

[0060] Operation time Battle Name Attack Dimension Equipment Type Equipment Name Number of weapons Starting point destination Longitude of origin Latitude of starting point Destination longitude Destination Latitude Duration Participating Armed Forces Operational Purpose Scale of action Number of people killed Casualties Impact of war result Introduction

[0061] The combat time, battle name, and introduction are used to describe each independent action, introduce the significance and role of each action, and its impact on subsequent combat operations; the strike dimension is used to distinguish the air strike phase, ground combat phase, and surface support phase, and then in the subsequent visualization display, it can be divided into three stages for visual narration, or the three can be combined for visualization analysis to describe the entire combat operation; the equipment type, equipment name, and number of weapons can be used to explore and analyze the use of weapons and equipment of each service; the starting point, destination, starting point longitude, starting point latitude, destination longitude, destination latitude and other related attribute items can be used for spatial display to form an action trajectory. The war visualization method without a specific trajectory is adopted. Some wars only occur in a fixed one or several locations without a specific movement trajectory. The spatial characteristics need to be reflected by marking the location of occurrence on the map. Cluster analysis of these points can be used to perform statistics on the frequency of occurrence, obtain implicit information, and better display the combat operations; finally, the combat time field is added to all attribute fields to form a timeline, and the timeline can be used to dynamically narrate the war visualization.

[0062] Furthermore, the step S101 includes:

[0063] In the information extraction stage, data fields related to the combat process are extracted through a rule-based matching algorithm: matching rules are designed to integrate the fields existing in the data model into the matching rules; matching rules are used to obtain paragraphs related to the description of the combat process, and data item fields are extracted from the paragraphs; by traversing the text and analyzing the topic of each paragraph, the paragraphs related to the combat process are determined, and then the data item fields are extracted based on information extraction technology.

[0064] Specifically, the data used in the data model mostly comes from text. In the text, the combat process data item is mostly represented in an unstructured form. Due to the characteristics of unstructured text, it needs to be preprocessed to provide data support for subsequent visualization. In the data entry stage, the paper text data is digitized to realize OCR technology and human-computer integration technology to improve accuracy. In the information extraction stage, the data fields related to the combat process are extracted through a rule-based matching algorithm. First, the matching rules are designed: when a field of user concern in the data model appears in a paragraph, the paragraph is determined as a candidate paragraph; the above matching method is used to obtain paragraphs related to the description of the combat process, and data item fields are extracted from these paragraphs. By traversing the text and analyzing the theme of each paragraph, the paragraphs related to the combat process are determined, and then the data item fields are extracted based on the information extraction technology. The matching rules greatly reduce the redundancy of the data in the input stage of the information extraction algorithm, improve the efficiency of information extraction, and can obtain most of the field information. Then, the missing data values ​​are supplemented through open source data and manual processing.

[0065] Furthermore, the step S102 includes:

[0066] Fill in missing data: Based on the matching of place names, use the place name and address matching algorithm to fill in the locations where the longitude and latitude information of the combat sites exists in the place name database. If the longitude and latitude information of the combat sites does not exist in the place name database, then use the relevant open source information on the Internet and the relevant descriptions of the combat sites during the combat operations phase to determine the longitude and latitude information of the combat sites through the water injection software.

[0067] Specifically, the information about the combat phase description obtained based on the text, especially the geographical location information, especially the geographical location latitude and longitude information, usually does not appear in the text, so post-processing is required to fill in the missing data problem. First, based on the matching of place names, the place name address matching algorithm (Ai Jiahao. Research on Place Name Address Matching Algorithm [D]. Kunming University of Science and Technology, 2019.) is used to fill in the location where the longitude and latitude information of the place exists in the place name database. However, since the locations where combat operations take place are mostly small locations in small and medium-sized countries in the world, most of the geographical location information is not stored in the database. Then, according to the relevant open source information on the Internet, the relevant description of the combat location during the combat operation phase is used, and the location information such as the longitude and latitude of the combat location is manually determined through the water injection software.

[0068] The required data items are obtained through the above data processing technology. For the attribute items of friendly casualties, enemy casualties, participating troops and duration, due to the sensitivity of this data item, there are few relevant descriptions in the open source data and text materials on the Internet, and there are too many missing data values ​​and no way to supplement them, so these attribute items are deleted; for the impact of the war, since the relevant descriptions in various materials are not unified and it is difficult to obtain a generally recognized result, it is also deleted. The final result is as follows Figure 2 However, through subsequent processing, it was found that the missing rate of some data items was about 40%, and the number of gaps was still relatively large. Afterwards, manual methods were used to complete some of the data that could be collected through various channels.

[0069] Furthermore, the construction process of the visual narrative model includes:

[0070] In the war background visualization narrative model construction phase, the evolution process before the outbreak of the war is presented in the form of a timeline; while presenting the war background and information, the war background is presented through corresponding videos and texts in combination with pattern analysis; according to the video progress or text and pattern browsing situation, the content corresponding to the current time node event is supplemented with visualization charts;

[0071] In the stage of constructing a visual narrative model of the war process, a flow map is used to display the starting point and destination of the air strike. The thickness and color of the trajectory line reflect the number and type of aircraft departing from the starting point. A timeline is used to describe the progress of the air strike, and the same air strike is placed at the same time node. With time as the main line, the route changes of the ground offensive forces and related situations as the war progresses are displayed, and the number of casualties, equipment names, and the number of weapons are added to the ground offensive route map.

[0072] During the construction phase of the visual narrative model of the war’s results and impacts, a visual expression based on statistical charts was adopted, with different statistical charts used to demonstrate the impacts of different aspects: in terms of politics, text was used to express it, supplemented by images to enhance the expressiveness; in terms of economy, bar charts and line charts were used to demonstrate the changes in domestic GDP, employment, and per capita consumption levels of both sides after the war; in terms of society, rose charts, fan charts, and regional distribution charts were used to display domestic public opinion and poll information after the war.

[0073] Specifically, war visualization narrative adds the particularity of war events themselves on the basis of narrative visualization. War events are usually composed of elements such as time and place, tasks, attributes (type of army, strength, weapons, results, etc.) and development process. Visual narrative models such as Figure 3 The construction process of the visual narrative model specifically includes:

[0074] (1) Visualization of the war background

[0075] The war background has complex components, including political, economic, cultural, military and other factors. Since there is no spatial correlation between the various components, but they have a certain connection in time, we choose to use a timeline to show the evolution process before the outbreak of the war.

[0076] When displaying the war background and information, corresponding videos and texts are needed to cooperate with pattern analysis to correctly and in detail display the war background; according to the progress of the video or the browsing of the text and pattern, the content corresponding to the current time node event is supplemented by visual charts. Videos, text, and image information are used to display detailed information of key events, while visual charts express the impact of corresponding events, extract text information into more perceptible chart information, and enhance readers' perception ability; the two are combined with the timeline to integrate time information, description information, and image information together, making complex war background information easier to recognize; such as Figure 4 As shown, the lower left is the timeline, showing time information and the names of key events; the upper left is text and picture information, showing detailed information of the event, including the time, location, personnel, and process of the event; on the right is a visual chart for additional explanation, and the content of the chart depends on the specific circumstances of the event.

[0077] (2) Visualization of the War Process

[0078] a. Visualization of the air raid phase

[0079] When visualizing an airstrike, a flow map is used to display the starting point and destination of the airstrike. The thickness and color of the trajectory line can reflect the number and type of aircraft departing from the starting point, ensuring a clear expression of the attack route. Due to the time correlation of airstrikes and the brevity of a single action, a traditional timeline is used to describe the progress of the airstrike. All airstrikes in the same action are placed at the same time node, and each time node stores situations of different dates.

[0080] b. Visualization of the ground attack phase

[0081] The ground offensive phase is mainly based on ground offensive forces, supplemented by air support. Ground offensive forces are relatively mobile and more flexible in space; therefore, the visual expression is based on time, showing the changes in the route of ground offensive forces and related situations as the war progresses; information such as the number of casualties, equipment names, and the number of weapons can be added to the route map to increase the amount of chart information and enrich the display content.

[0082] (3) Visualization of war outcomes and impacts

[0083] The impact of war is similar to the background of war, involving many aspects, including politics, economy, military (including military strength, national defense strength, etc.), culture, social factors, etc.; the difference is that the impact of war is usually not temporally related, and each part is often independent of each other. Compared with the expression form of video images, visualization charts have more advantages, such as interactive analysis. Visualization charts display the content of each element and link them through related attributes, which has obvious advantages in explanation and analysis.

[0084] There are usually four ways for war to end: ① One side is destroyed or defeated, and the other side achieves complete victory. This often occurs in total wars. ② One side is not completely defeated, but partially destroyed, and the other side also achieves limited war goals, and the war ends. This often occurs in smaller local wars. ③ Due to various reasons, the war is in a stalemate, and no one can achieve a decisive victory, and both sides end the war in the form of compromise. ④ Although one side is not destroyed or defeated, it agrees to accept the other side's conditions and end the war due to considerations of the cost of the war or other reasons. The four different ways also represent different visualization methods. The first way needs to emphasize the defeated or destroyed party, the second situation needs a detailed description of the purpose of the war, the third situation focuses on analyzing the reasons for the compromise of both parties, and the fourth situation focuses on the motivation and consumption of the defeated party to end the war.

[0085] In order to verify the effect of the present invention, the following application example analysis is carried out:

[0086] Taking the Afghan War as an example, the main initiator of the war is called A, and all the experiences of A in the war are visualized. In the model design, different visualizations are performed according to different stages of the war; before the war starts (i.e., the war background description stage), it is presented in the form of a timeline plus descriptive information and visual charts; at the outbreak stage of the war, exploratory narrative analysis is adopted, and exploratory visual charts are added on the basis of the flow map. Through interactivity and explorability, users are allowed to decide the amount of information to obtain, realizing multi-angle narrative; at the end of the war, i.e., the stage of displaying the results and impact of the war, statistical charts are used to display information.

[0087] (1) Visualization of War Background Narratives

[0088] Visual expressions for war backgrounds, such as Figure 5 exhibit. Figure 5 The upper left part of the middle screen shows the details before the war broke out, including pictures and text information; the lower left part is a timeline, which contains the dates and names of key events before the war broke out. Users can learn about background information by selecting a time point or event name; on the right is a visual chart that displays supplementary information about the event. The chart shows the number of terrorist attacks in different regions, helping users to enhance their understanding of global terrorist attacks.

[0089] (2) Visualization of War Process Narrative

[0090] The war process is divided into the air strike phase and the ground attack phase. In the air strike phase, the flow map plus timeline is adopted, such as Figure 6 The time axis shows the time of each air strike, and the flow map shows the starting point of the air strike and the bombing location. The thickness of the air strike route line indicates the amount of air force dispatched, the shape of the aircraft indicates the largest number of aircraft in the air force dispatched, and the flashing dots indicate the bombed locations.

[0091] The advantage of the flow map is that it can dynamically display the attack route, clearly showing the starting and end points as well as the number of participating forces.

[0092] At the same time, label all of A's air force bases. Click on the label to display the air force that departed from this base on that day. Figure 7 shown. Figure 7 The middle is a comparison of the types and numbers of aircraft of the aircraft carrier formation departing near the waters of Bay A on October 7, 2001. This is an important part of the explorable narrative. Users can choose whether to explore the base information according to their own needs. Different information acquisition will lead to different understandings of the war process.

[0093] During the ground attack phase, the route of A's ground attack needs to be displayed. The thickness of the route line indicates the current number of troops. At the same time, information such as the number of casualties is added to it through labels. Since air power also assists the attack during the ground attack phase, the flow map of the air strike needs to be displayed at the same time.

[0094] (3) Visualization of war outcomes and impact narratives

[0095] Due to the diversity of factors, the results and impacts of the war need to be presented through different statistical charts. In terms of politics, there is a lot of textual description involved, so text is used to express it, supplemented by images to enhance the expression; in terms of economy, the main analysis is the changes in GDP, employment, per capita consumption level and other factors of both sides after the war, and bar charts, line charts and other charts are used to display them; in terms of social factors, rose charts, fan charts, regional distribution charts and other information are used to display domestic public opinion, polls and other information after the war. Figure 8 shown.

[0096] Based on the above embodiments, Fig. 9 As shown, the present invention also proposes a visual narrative system for typical war actions, including:

[0097] The information extraction module is used to extract various information describing the war from a large amount of text using information extraction technology in intelligent language processing, and obtain fields corresponding to different combat forces;

[0098] A visual narrative data model construction module is used to process the fields corresponding to different combat forces, obtain the common data model fields, and complete the visual narrative data model construction;

[0099] The visual narrative model construction and visual expression module is used to build a visual narrative model based on the visual narrative data model, and to visually express the war background, war process, war results and impact.

[0100] Furthermore, the information extraction module is specifically used for:

[0101] In the information extraction stage, data fields related to the combat process are extracted through a rule-based matching algorithm: matching rules are designed to integrate the fields existing in the data model into the matching rules; matching rules are used to obtain paragraphs related to the description of the combat process, and data item fields are extracted from the paragraphs; by traversing the text and analyzing the topic of each paragraph, the paragraphs related to the combat process are determined, and then the data item fields are extracted based on information extraction technology.

[0102] Furthermore, the visual narrative data model construction module is specifically used for:

[0103] Fill in missing data: Based on the matching of place names, use the place name and address matching algorithm to fill in the locations where the longitude and latitude information of the combat sites exists in the place name database. If the longitude and latitude information of the combat sites does not exist in the place name database, then use the relevant open source information on the Internet and the relevant descriptions of the combat sites during the combat operations phase to determine the longitude and latitude information of the combat sites through the water injection software.

[0104] Furthermore, the general data model fields include: combat time, combat name, strike dimension, equipment type, equipment name, number of weapons, starting point, destination, starting point longitude, starting point latitude, destination longitude, destination latitude, duration, participating services, combat purpose, action scale, number of annihilations, number of casualties, war impact, results, and introduction.

[0105] Furthermore, the construction process of the visual narrative model includes:

[0106] In the war background visualization narrative model construction phase, the evolution process before the outbreak of the war is presented in the form of a timeline; while presenting the war background and information, the war background is presented through corresponding videos and texts in combination with pattern analysis; according to the video progress or text and pattern browsing situation, the content corresponding to the current time node event is supplemented with visualization charts;

[0107] In the stage of constructing a visual narrative model of the war process, a flow map is used to display the starting point and destination of the air strike. The thickness and color of the trajectory line reflect the number and type of aircraft departing from the starting point. A timeline is used to describe the progress of the air strike, and the same air strike is placed at the same time node. With time as the main line, the route changes of the ground offensive forces and related situations as the war progresses are displayed, and the number of casualties, equipment names, and the number of weapons are added to the ground offensive route map.

[0108] During the construction phase of the visual narrative model of the war’s results and impacts, a visual expression based on statistical charts was adopted, with different statistical charts used to demonstrate the impacts of different aspects: in terms of politics, text was used to express it, supplemented by images to enhance the expressiveness; in terms of economy, bar charts and line charts were used to demonstrate the changes in domestic GDP, employment, and per capita consumption levels of both sides after the war; in terms of society, rose charts, fan charts, and regional distribution charts were used to display domestic public opinion and poll information after the war.

[0109] In summary, this paper has fully explored the characteristics of different stages of war by conducting narrative visualization research on action data of typical wars, and has developed different visualization methods based on different characteristics, so that war visualization can better display narrative, help users understand the background of the war, grasp the progress of the war, and clarify the impact of the war.

[0110] The above is only a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A visual narrative method for typical war actions, characterized in that: include: Step 1: Using information extraction technology in intelligent language processing, extract various aspects of information describing the war from a large amount of text to obtain fields corresponding to different combat forces; The step 1 comprises: In the information extraction stage, data fields related to the combat process are extracted through a rule-based matching algorithm: matching rules are designed to integrate the fields existing in the data model into the matching rules; paragraphs related to the description of the combat process are obtained using the matching rules, and data item fields are extracted from the paragraphs; the text is traversed, the theme of each paragraph is analyzed, the paragraphs related to the combat process are determined, and then the data item fields are extracted based on information extraction technology; Step 2: Process the fields corresponding to the different combat forces to obtain the common data model fields and complete the construction of the visual narrative data model; The step 2 comprises: Fill in missing data: Based on the matching of place names, use the place name and address matching algorithm to fill in the locations where the longitude and latitude information of the combat sites exists in the place name database. If the longitude and latitude information of the combat sites does not exist in the place name database, then use the relevant open source information on the Internet and the relevant descriptions of the combat sites during the combat action phase, and then use the water injection software to determine the longitude and latitude information of the combat sites; Step 3: Build a visual narrative model based on the visual narrative data model to visualize the war background, war process, war results and impact; The construction process of the visual narrative model includes: In the war background visualization narrative model construction phase, the evolution process before the outbreak of the war is presented in the form of a timeline; while presenting the war background and information, the war background is presented through corresponding videos and texts in combination with pattern analysis; according to the video progress or text and pattern browsing situation, the content corresponding to the current time node event is supplemented with visualization charts; In the stage of constructing a visual narrative model of the war process, a flow map is used to display the starting point and destination of the air strike. The thickness and color of the trajectory line reflect the number and type of aircraft departing from the starting point. A timeline is used to describe the progress of the air strike, and the same air strike is placed at the same time node. With time as the main line, the route changes of the ground offensive forces and related situations as the war progresses are displayed, and the number of casualties, equipment names, and the number of weapons are added to the ground offensive route map. During the construction phase of the visual narrative model of the war’s results and impacts, a visual expression based on statistical charts was adopted, with different statistical charts used to demonstrate the impacts of different aspects: in terms of politics, text was used to express it, supplemented by images to enhance the expressiveness; in terms of economy, bar charts and line charts were used to demonstrate the changes in domestic GDP, employment, and per capita consumption levels of both sides after the war; in terms of society, rose charts, fan charts, and regional distribution charts were used to display domestic public opinion and poll information after the war.

2. The method for visual narration of typical war actions according to claim 1 is characterized in that: The general data model fields include: combat time, combat name, strike dimension, equipment type, equipment name, number of weapons, starting point, destination, starting point longitude, starting point latitude, destination longitude, destination latitude, duration, participating military branches, combat purpose, action scale, number of annihilations, number of casualties, war impact, results, and introduction.

3. A visual narrative system for typical war actions, characterized by: include: The information extraction module is used to extract various information describing the war from a large amount of text using information extraction technology in intelligent language processing, and obtain fields corresponding to different combat forces; The information extraction module is specifically used to extract data fields related to the combat process through a rule-based matching algorithm during the information extraction phase: design matching rules and integrate the fields existing in the data model into the matching rules; Using matching rules to obtain paragraphs related to the description of the combat process, extracting data item fields from the paragraphs; by traversing the text and analyzing the theme of each paragraph, determining the paragraphs related to the combat process, and then extracting the data item fields based on information extraction technology; A visual narrative data model construction module is used to process the fields corresponding to different combat forces, obtain the common data model fields, and complete the visual narrative data model construction; The visualization narrative data model construction module is specifically used to fill in missing data: based on the matching of place names, the place name and address matching algorithm is used to fill in the location where the longitude and latitude information of the combat site exists in the place name database. If the longitude and latitude information of the combat site does not exist in the place name database, then the longitude and latitude information of the combat site is determined by the water injection software according to the relevant open source information on the Internet and the relevant description of the combat site during the combat action phase. Visual narrative model construction and visual expression module, which is used to build a visual narrative model based on the visual narrative data model, and to visually express the war background, war process, war results and impact; The construction process of the visual narrative model includes: In the war background visualization narrative model construction phase, the evolution process before the outbreak of the war is presented in the form of a timeline; while presenting the war background and information, the war background is presented through corresponding videos and texts in combination with pattern analysis; according to the video progress or text and pattern browsing situation, the content corresponding to the current time node event is supplemented with visualization charts; In the stage of constructing a visual narrative model of the war process, a flow map is used to display the starting point and destination of the air strike. The thickness and color of the trajectory line reflect the number and type of aircraft departing from the starting point. A timeline is used to describe the progress of the air strike, and the same air strike is placed at the same time node. With time as the main line, the route changes of the ground offensive forces and related situations as the war progresses are displayed, and the number of casualties, equipment names, and the number of weapons are added to the ground offensive route map. During the construction phase of the visual narrative model of the war’s results and impacts, a visual expression based on statistical charts was adopted, with different statistical charts used to demonstrate the impacts of different aspects: in terms of politics, text was used to express it, supplemented by images to enhance the expressiveness; in terms of economy, bar charts and line charts were used to demonstrate the changes in domestic GDP, employment, and per capita consumption levels of both sides after the war; in terms of society, rose charts, fan charts, and regional distribution charts were used to display domestic public opinion and poll information after the war.

Images