KR102567618B1 - Apparatus for operating context information and method thereof - Google Patents

Abstract

The present invention includes a collection unit for collecting context information data from traps, a first data processing unit for extracting first data having a first form from context information data, and a second form different from the first form for the first data. A data conversion unit for generating second data by converting into a form, a metadata generation unit for generating metadata using the second data, and reproducing contextual information data, and reproducing contextual information data based on the metadata. Presenting a situation information operation device including a situation information display unit capable of selecting the playback time of the game, and a situation information operation method applied thereto, which can efficiently utilize situation information data collected from ships. do.

Description

Apparatus and method for operating context information {APPARATUS FOR OPERATING CONTEXT INFORMATION AND METHOD THEREOF}

The present invention relates to an apparatus and method for operating situation information, and more particularly, to an apparatus and method for operating situation information capable of efficiently utilizing situation information data collected from a ship.

During the operation of the ship, the combat system mounted on the ship transmits all information related to the operation of the ship, such as tactical information generated while operating the combat system and situational information such as video information received by the combat system, as video files and audio files. It is recorded. At this time, the contents recorded in the combat system can be used as very important basic data in various tasks such as ship crew evaluation and ship repair.

However, in the prior art, in order to analyze a specific event, when playing files recorded in a combat system, the user manipulates the time bar on the playback screen to directly scroll to the playback position where the event is recorded to adjust the playback time, or A method of increasing the playback speed of the file until the playback position where the event is recorded appears. Therefore, the longer the trap is operated, the more recent the event to be analyzed, the more time is consumed in adjusting the playback time until the analysis starts.

In addition, conventional combat systems do not support a rewind function when replaying a recorded file, so if an operation mistake goes beyond the recorded playback position, it is difficult to operate again from the beginning.

This conventional method causes inconvenience to a user who has to analyze combat system information based on a specific event, and has a problem of increasing the time required for information analysis.

The background technology of the present invention is published in the following patent documents.

KR 10-1523261 B1 KR 10-1779199 B1

The present invention provides an apparatus and method for operating contextual information capable of efficiently utilizing contextual information data collected from a ship.

A situation information operating apparatus according to an embodiment of the present invention includes a collection unit for collecting situation information data from a trap; a first data processing unit for extracting first data having a first form from the context information data; a data conversion unit configured to generate second data by converting the first data into a second form different from the first form; a metadata generator for generating metadata using the second data; and a context information display unit capable of reproducing the context information data and selecting a reproduction time of the context information data based on the metadata.

The collection unit is connected to the combat system of the ship and collects video data and sound data recorded in the combat system as the situation information data, and the first data processing unit collects the video data and the sound data. Any one of the data may be selected and extracted as the first data.

The data conversion unit may recognize the first data using a first data recognition model, convert the first data into a text format, and generate text data as the second data.

The metadata generator may analyze the meaning of the second data, divide the second data into meaningful units to generate an event, index the event according to time, and generate the metadata.

The situation information display unit receives a search word for searching for an event on the operation status of the trap, extracts an event matching the search word among events included in the metadata as a request event, and indexes the requested event. The situation information data can be reproduced from time.

A situation information management method according to an embodiment of the present invention includes a process of collecting situation information data from a trap; extracting first data having a first shape from the situation information data; generating second data by converting the first data into a second form different from the first form; generating metadata using the second data; and selecting a reproduction time point of the context information data based on the metadata and reproducing the context information data from the selected reproduction time point.

The process of collecting the situation information data may include: operating the ship using the combat system of the ship; Recording the operation status of the ship in the combat system in a plurality of forms; A process of collecting the data recorded in the combat system as the situation information data; may include.

The process of recording in the combat system may include: recording tactical information, radar detection information, visible light detection information, and infrared light detection information input and output from a tactical command system and a sensor system among subsystems of the combat system as image data; Among the subsystems of the combat system, a process of recording communication information input and output from the communication system as sound data; may be included.

The process of extracting the first data may include a process of selecting the image data or the sound data from among the context information data and extracting the first data.

The process of generating the second data may include recognizing the first data using a first data recognition model input in advance, converting the first data into text form, and generating the converted text data as the second data. can

The first data includes sound data, the first data recognition model includes a sound recognition model, and the process of generating the second data includes sound data having a volume equal to or greater than a preset reference volume among the sound data. selection process; identifying acoustic data corresponding to a crew member's voice from among the selected acoustic data using the acoustic recognition model, and generating voice-recognized texts by performing voice recognition; It may include a process of generating text data by arranging the generated voice recognition texts in order.

The process of generating the metadata may include analyzing the meaning of the voice recognition texts listed in order, and generating an event by dividing the voice recognition texts listed in the order into minimum units having meaning according to the analyzed meaning; Indexing the generated event according to time and generating the metadata; may include.

The process of reproducing the situation information data may include receiving a search word; extracting an event similar or identical to the search word from among the metadata as a request event; and reproducing the context data from the time indexed to the extracted request event.

According to an embodiment of the present invention, first data may be extracted from context information data, second data may be generated therefrom, metadata may be generated using the generated second data, and situation information may be extracted based on the metadata. The reproduction point of the information data can be selected. Accordingly, the situation information data can be reproduced from the selected reproduction time point.

In addition, according to an embodiment of the present invention, by generating text-type metadata from sound data among context information data, it is possible to smoothly select a playback time point through word search, which is a form familiar to users. Furthermore, it is possible to accurately select a playback time point through word search by using a word familiar and frequently used by ship crew members in communication as a search term. Accordingly, it is possible to accurately search for an event desired by the user, and immediately reproduce the searched event from the playback time, thereby providing convenience to a user who needs to analyze combat system information based on a specific event. In addition, the time required for information analysis can be effectively reduced. From this, the situation information data collected from the trap can be efficiently utilized by the user.

1 is a block diagram of a situation information operating device according to an embodiment of the present invention.
2 is a schematic diagram of a situation information operating device according to an embodiment of the present invention.
3 is a diagram exemplarily illustrating a method of generating metadata by a context information operating apparatus according to an embodiment of the present invention.
4 is a flowchart of a context information operating method according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, and will be implemented in a variety of different forms. Only the embodiments of the present invention are provided to complete the disclosure of the present invention and to fully inform those skilled in the art of the scope of the invention. In order to explain an embodiment of the present invention, the drawings may be exaggerated, and the same reference numerals in the drawings refer to the same elements.

The present invention relates to an apparatus and method for managing contextual information capable of efficiently utilizing contextual information data collected from a ship.

Hereinafter, an embodiment of the present invention will be described in detail by exemplifying a case in which a combat system is applied to a ship equipped with a combat system. Of course, the apparatus and method for operating situation information according to an embodiment of the present invention described below may be variously applied to trains, aircraft, and the like.

1. Context information operating device according to an embodiment of the present invention

1 is a block diagram of a context information operating device according to an embodiment of the present invention, FIG. 2 is a schematic diagram of a context information operating device according to an embodiment of the present invention, and FIG. 3 is a context information management device according to an embodiment of the present invention. It is a diagram exemplarily showing how the device generates metadata.

Referring to FIGS. 1 to 3 , a context information operating apparatus according to an embodiment of the present invention will be described.

The context information operating device 1000 according to an embodiment of the present invention includes a collection unit 100, a first data processing unit 200, a data conversion unit 300, a metadata generation unit 400, and a situation information display unit 500. ).

1.1. Trap(1)

The situation information operating device 1000 may be installed in the trap 1. At this time, the ship 1 may include a warship operated in the sea. The battle system 10 may be mounted on the ship 1. The vessel 1 can operate the mounted combat system 10 to navigate on the sea and perform assigned missions. When the situation information operating device 1000 is installed in the ship 1, it can be mounted in the combat system 10 or placed inside and outside the ship 1 in the form of various consoles or terminals. Meanwhile, the situation information operating device 1000 may be disposed outside the trap 1 and connected to the trap 1 through a tactical data link.

1.2. combat system (10)

The combat system 10 may include a tactical command system 11, a sensor system 12, a communication system 13, and the like. At this time, the tactical command system 11, the sensor system 12, the communication system 13, and the like may be referred to as sub-systems of the combat system 10. In addition, the types of subsystems included in the combat system 10 may vary. That is, the combat system 10 may further include an armed system, a navigation system, and the like as sub-systems. Meanwhile, the combat system 10 may be connected to a maritime, ground, or air joint control center through a tactical data link. Configurations of the tactical command system 11, the sensor system 12, and the communication system 13 may vary.

Meanwhile, in the process of operating the ship 1 using the battle system 10 of the ship 1, various messages and various signals are generated from the tactical command system 11, sensor system 12, and communication system 13. can be created and output. The tactical command system 11 may generate an internal topic message. The sensor system 12 may generate an Interface Control Document (ICD) message. At this time, the internal topic message and the ICD message may include tactical information about the operating situation of the trap 1. The sensor system 12 may generate a radar signal, a TV signal, and an IR signal. At this time, the TV signal may be for generating a search result as a visible ray image, and the IR signal may generate a search result as an infrared (IR) image. It may be for generating, and the radar signal may be for generating a search result as a radar image. On the other hand, the radar signal may include radar detection information about the operation situation of the ship 1, the TV signal may include visible light detection information about the operation situation of the ship 1, and the IR signal may include the ship (1). 1) may include infrared detection information about the operation situation. The communication system 13 can generate an audio signal. At this time, the audio signal may include communication information about the operating situation of the trap 1.

1.3. Collecting unit (100)

The collection unit 100 serves to collect situational information data from traps. The collection unit 100 may be connected to the combat system 10 of the trap 1. The collection unit 100 may exchange data with the combat system 10 . In addition, the collection unit 100 may collect video data and sound data recorded in the combat system 10 as situation information data.

1.4. First data processor 200

The first data processing unit 200 serves to extract first data having a first form from context information data. The first data processing unit 200 is connected to the collecting unit 100 and can exchange data with the collecting unit 100 . In addition, the first data processing unit 200 may select any one of image data and sound data and extract it as first data. That is, the first data processing unit 200 may select sound data and extract it as first data, or select image data and extract it as first data.

1.5. Data conversion unit 300

The data conversion unit 300 serves to convert the first data into a second form different from the first form to generate second data. The data converter 300 may be connected to the first data processor 200 to exchange data. Accordingly, the data conversion unit 300 may receive the first data from the first data processing unit 200 . In addition, the data conversion unit 300 may recognize the first data using the first data recognition model, convert it into text form, and generate text data as the second data. Also, the data converter 300 may output text data to the metadata generator 400 .

1.6. Metadata generator 400

The metadata generating unit 400 serves to generate metadata using the second data. The metadata generation unit 400 may receive second data from the data conversion unit 300 . In this case, the second data may be text data generated from sound data. Also, the second data may be text data generated from image data. In addition, the metadata generator 400 analyzes the meaning of the second data, divides the second data into meaningful units to generate events, and indexes the events according to time to generate metadata. . That is, an event generated in a test form from sound data or video data may be associated with time information of each data and stored as metadata.

1.7. Situation information display unit (500)

The context information display unit 500 plays a role of reproducing context information data and selects a playback time of context information data based on metadata. The situation information display unit 500 is connected to the combat system 10 to receive situation information data, and is connected to the metadata generation unit 400 to receive metadata. In addition, the situation information display unit 500 may include an input device, a data processor, and a display screen. The input device may receive an input of a search word for searching for an event related to the operating situation of the trap 1. The data processor extracts an event matching a search term from among events included in the metadata as a request event, and manipulates the display screen to reproduce context data from the time indexed in the request event. Thus, on the display screen, context information data can be reproduced from the time indexed to the requested event.

2. Context information operating method according to an embodiment of the present invention

4 is a flowchart of a context information operating method according to an embodiment of the present invention.

Hereinafter, a context information operating method according to an embodiment of the present invention will be described in detail with reference to FIGS. 1 to 4 .

The method for operating contextual information according to an embodiment of the present invention includes a process of collecting contextual information data from a trap 1 (S110), a process of extracting first data having a first form from contextual information data (S120), , The process of converting the first data into a second form different from the first form to generate second data (S130), the process of generating metadata using the second data (S140), and based on the metadata A process of selecting a reproduction time point of context information data and reproducing context information data from the selected reproduction time point (S150) is included.

2.1. Process of collecting situation information data from trap (1) (S110)

First, a process of collecting situation information data from the trap 1 (S110) is performed. At this time, in the process of collecting situation information data from the trap 1 (S110), the process of operating the trap 1 using the combat system 10 of the trap 1 and the operating situation of the trap 1 are described in multiple numbers. It may include a process of recording in the combat system 10 in the form of, and a process of collecting the data recorded in the combat system 10 as situation information data.

2.1.A. The process of operating the trap (1)

In the process of operating the ship (1) using the battle system (10) of the ship (1), the battle system (10) mounted on the ship (1) is operated to navigate the ship (1) at sea, and the ship (1) ) can perform the assigned tasks. Here, missions may vary, such as training, support, reconnaissance, transport, security, engagement and defense.

The combat system 10 may include a tactical command system 11, a sensor system 12, a communication system 13, and the like. At this time, the tactical command system 11, the sensor system 12, the communication system 13, and the like may be referred to as sub-systems of the combat system 10. In addition, the types of subsystems included in the combat system 10 may vary. That is, the combat system 10 may further include an armed system, a navigation system, and the like as sub-systems. Meanwhile, the combat system 10 may be connected to a maritime, ground, or air joint control center through a tactical data link.

The tactical command system 11 may include a multifunction console (MFC) provided in the control room of the ship 1. The sensor system 12 is a search radar (SRS), tracking radar (TRS), electro-optical tracking equipment (EOTS) and infrared detection and tracking equipment (IRST) installed in a plurality of locations of the ship 10 and a plurality of interlocking stages ( SRS ICU, TRS ICU, EOTS ICU, IRST ICU). The communication system 13 may include an integrated communication system (ICS). Of course, other configurations of the tactical command system 11, the sensor system 12, and the communication system 13 may vary.

Meanwhile, in the process of operating the ship 1 using the battle system 10 of the ship 1, various messages and various signals are generated from the tactical command system 11, sensor system 12, and communication system 13. can be created and output.

For example, the multifunction console (MFC) of the tactical command system 11 may generate an internal topic message. Search radar (SRS), tracking radar (TRS), electro-optical tracking equipment (EOTS) and infrared detection and tracking equipment (IRST) of the sensor system 12 may generate an Interface Control Document (ICD) message. At this time, the internal topic message and the ICD message may include tactical information about the operating situation of the trap 1.

In addition, the search radar (SRS) of the sensor system 12 can generate a radar signal, the tracking radar (TRS) can generate a TV signal, and the electro-optical tracking equipment (EOTS) can generate a TV signal and an IR signal. In addition, infrared detection and tracking equipment (IRST) can generate TV signals and IR signals. In this case, the TV signal may be for generating a search result as a visible ray image, the IR signal may be for generating a search result as an infrared (IR) image, and the radar signal may be for generating a search result as a radar image. it could be On the other hand, the radar signal may include radar detection information about the operation situation of the ship 1, the TV signal may include visible light detection information about the operation situation of the ship 1, and the IR signal may include the ship (1). 1) may include infrared detection information about the operation situation.

In addition, the integrated communication system (ICS) of the communication system 13 may generate an audio signal. At this time, the audio signal may include communication information about the operating situation of the trap 1.

2.1.B. The process of recording in the combat system (10)

In the process of recording the operational status of the ship (1) in the combat system (10) in a plurality of forms, the tactical information input and output from the tactical command system (11) among sub-systems of the combat system (10) is converted into image data for the combat system (10) can be recorded. In addition, radar detection information, visible light detection information, and infrared detection information input and output from the sensor system 12 among subsystems of the combat system 10 may be recorded in the combat system 10 as image data. In addition, communication information input and output from the communication system 13 among subsystems of the combat system 10 may be recorded in the combat system 10 as sound data. At this time, these data may be recorded in the server of the combat system 10.

2.1.C. The process of collecting situational information data

In the process of collecting the data recorded in the combat system 10 as situation information data, the same video data and sound data as the above-mentioned video data and sound data recorded in the combat system 10 are collected connected to the tactical system 10. It can be recorded in the information storage device and image processing device of the unit 100 in the same way. At this time, tactical information may be recorded as video data in the information storage device. In addition, radar detection information, visible light detection information, and infrared detection information may be recorded as image data in the image processing device, and communication information may be recorded as sound data. Thus, the situation information data may have tactical information, radar detection information, visible light detection information, and infrared light detection information in the form of images, and may have communication information in the form of sound. Meanwhile, a process of collecting data recorded in the combat system 10 as situation information data may be performed in various ways.

2.2. Process of extracting first data having a first form (S120)

Thereafter, a process of extracting first data having a first form from context information data (S120) is performed. Here, the process of extracting the first data having the first form from the context information data (S120) may include a process of selecting video data or audio data from the context information data and extracting the first data.

2.2.A. Process of selecting sound data and extracting it as first data

In the process of selecting sound data from context information data and extracting it as first data, the first data processor 200 may select and extract sound data from data recorded in the image processing device of the collection unit 100. . That is, the first data may be sound data.

2.2.B. Process of selecting image data and extracting it as first data

In the process of selecting image data from context information data and extracting it as first data, the first data processing unit 200 may select and extract image data from data recorded in the information processing device of the collection unit 100. . That is, the first data may be image data. In this case, specifically, the image data to be selected and extracted may be tactical information image data, radar detection information image data, visible light detection information image data, or infrared detection information image data.

2.3. Process of generating second data (S130)

Thereafter, a process of generating second data by converting the first data into a second form different from the first form (S130) is performed. The first data may be sound data, and the first form may be a sound form. Of course, the first data may be image data, and the first form may be an image form. The second form may be a text form, and the second data may be text data. When the first data is sound data and the first form is the sound form, the process of converting the first data into a second form different from the first form to generate second data (S130) is a pre-input sound recognition model. It may include a process of recognizing sound data and converting it into a text form using , and generating the converted text data as second data. In addition, when the first data is image data and the first form is an image form, the process of converting the first data into a second form different from the first form to generate second data (S130) is a pre-input image The method may include recognizing image data using a recognition model, converting the image data into text form, and generating the converted text data as second data.

2.3.A. Process of converting sound data into text form and generating second data

In the process of converting the sound data into text form and generating the second data by the data conversion unit 300, sound data having a volume equal to or greater than a preset reference volume may be selected from among the sound data. In this case, the preset reference volume may be a volume within a predetermined range for distinguishing between the voice of the ship crew and the noise excluding it. That is, when the ship crew communicates using the integrated communication system (ICS), various noises may be input and recorded together with the ship crew's voice. At this time, the noise may be various, such as the electronic driving sound of the equipment, the noise of the control room, the vibration sound of the trap 1, and the like. Therefore, by selecting sound data having a volume equal to or higher than the reference volume among the sound data, it is possible to select sound data in which the voice of the ship's crew is recorded.

In addition, the data conversion unit 300 may use the acoustic recognition model to identify acoustic data corresponding to the crew member's voice among the selected acoustic data, perform voice recognition, and generate voice recognition texts. Here, the acoustic recognition model may be implemented in various configurations using commercial technologies, that is, known technologies. In the embodiment of the present invention, this configuration is not particularly limited. Similarly, when voice recognition is performed to generate voice recognition texts, voice recognition may be performed in various ways and voice recognition texts may be generated using known technologies. Meanwhile, since the construction of a sound recognition model, a method of performing voice recognition, and generating voice recognition texts are widely known in the field of artificial intelligence such as deep learning, a detailed description thereof will be omitted.

In addition, text data may be generated by sequentially arranging voice recognition texts generated from sound data using the data converter 300 . In this case, text data may be generated by sequentially arranging speech recognition texts generated from the sound data according to a reproduction time of the sound data. In this case, the text data may be generated to be classified for each crew member of the ship. That is, when there are a plurality of ship crew members, their voices may be classified and voice recognition texts classified for each ship crew member may be sequentially arranged according to time to generate text data.

2.3.B. The process of converting image data into text form and generating second data

The data conversion unit 300 may select an image frame from the image data at predetermined time intervals in the process of converting the image data into text form and generating the second data. Also, among the video frames listed in chronological order, when the previous video frame and the next video frame are compared, an image frame in which an image change is detected or a significant image change can be selected. In this way, by reducing the number of video frames based on time or by selecting meaningful video frames, the amount of data processing can be effectively reduced in the following process.

In addition, the data conversion unit 300 may use an image recognition model to identify objects within each selected image frame, perform object recognition, and generate object recognition texts. Here, the object recognition text may be texts for explaining the type of recognized object, and may be, for example, texts expressing the name of an object, a description of the object, and the number of objects. In addition, the object recognition text may be texts for explaining the situation of the image frame determined according to the type of the recognized object. For example, when a non-friend target, such as an aptitude target, is detected as an object in an image frame selected from image data of radar information, the object recognition text may be generated as text with an identification code of the target, "first detection of aptitude target", It may be generated in the form of a predetermined sentence that can explain the situation with respect to the video frame, such as "engagement start", "armed channel 1 operation", and the like.

Meanwhile, as in the case of sound recognition, the image recognition model may be implemented in various configurations using commercial technologies, that is, known technologies. In the embodiment of the present invention, this configuration is not particularly limited. Also, when object recognition is performed to generate object recognition texts, object recognition may be performed and object recognition texts may be generated in various ways using known technologies. Since the configuration of the image recognition model, the method of performing object recognition, and generating object recognition texts are widely known in the field of artificial intelligence such as deep learning, a detailed description thereof will be omitted.

In addition, text data may be generated by sequentially arranging object recognition texts generated from image data using the data conversion unit 300 . In this case, text data may be generated by sequentially arranging object recognition texts generated from the image data according to the reproduction time of the image data. Meanwhile, object recognition text may also be referred to as image recognition text.

2.4. Process of generating metadata (S140)

Thereafter, a process of generating metadata using the second data (S140) is performed. At this time, in the process of generating metadata using the second data (S140), the meaning of the voice recognition texts listed in order is analyzed, and according to the analyzed meaning, the voice recognition texts listed in the order are divided into minimum units having meaning. It may include a process of generating an event by dividing it, and a process of generating metadata by indexing the generated event according to time. These processes may be performed in the metadata generator 400.

2.4.A. The process of creating an event

In the process of generating an event, the meaning of voice recognition texts listed in order may be analyzed. For example, words, word segments, and phrases may be recognized from voice recognition texts arranged in order, and the meaning of the recognized words, word segments, and phrases may be analyzed. Meanwhile, in order to analyze the meaning, a vast amount of basic data for natural language may be learned by the metadata generating unit 400 . Of course, if there is a new type of word that has not been learned, the user may be requested to be given a characteristic for it. Meanwhile, after analyzing the meaning, it is possible to analyze the correlation between the analyzed meanings, and based on the result, a sentence may be generated from the recognized words, and a paragraph may also be created. In addition, an event may be generated by dividing voice recognition texts arranged in order according to the analyzed meaning into minimum units having meaning. For example, when the word “control one” is recognized, the meaning is analyzed to generate a “call command” event, and when the word “318 notes” is recognized, a “do command” event can be generated. Of course, an event of "one control" can be generated as the word "one control" is. Similarly, a recognized word may be generated as an event. For this process, the same or similar method to that used in the semantic web, for example, may be applied. Of course, there may be various methods of generating an event by analyzing the meaning of voice recognition texts and dividing them into minimum meaningful units.

2.4.B. The process of creating metadata

The process of indexing generated events according to time and generating metadata includes voice recognition texts arranged in order using time information included in audio data (or video data) used to generate second data. Each time, corresponding speech recognition texts may be indexed and generated as metadata.

2.5. A process of selecting a playback time point and reproducing situation information data from the selected playback time point (S150)

Thereafter, a playback time of context information data is selected based on metadata, and a process of reproducing context data from the selected playback time (S150) is performed. At this time, the process of selecting the reproduction time of the context information data based on the metadata and reproducing the context information data from the selected reproduction time is the process of receiving a search word, and an event similar or identical to the search word among the metadata as a request event. It may include a process of extracting, and a process of reproducing context data from the time indexed to the extracted request event.

2.5.A. The process of entering a search term

In the process of inputting a search word, the search word may be input through an input device provided in the context information display unit 500 . In this case, the search word may be a natural language. In addition, the search word may be a search word generated by the user to search for an event related to the operating situation of the trap 1.

2.5.B. The process of extracting events similar or identical to the search term among metadata as request events

In the process of extracting an event similar or identical to a search word from among metadata as a request event, the context information display unit 500 compares the event list of metadata and the entered search word using a predetermined algorithm, so that an event similar to the entered search word is displayed. , or select the same event as the entered search term, and extract the selected event as a request event.

2.5.C. The process of reproducing context data from the time indexed in the extracted request event

In the process of reproducing context data from the time indexed to the extracted request event, the context information display unit 500 checks the time indexed to the extracted request event, and creates a playback plot to reproduce the context data from the checked time. It is created, and contextual information data can be reproduced according to the created playback plot. For example, if the input search term is "checking the direction of a trap" and the requested event extracted accordingly is "direction", situational information data may be reproduced from the playback time indexed to the "direction" event.

As described above, according to an embodiment of the present invention, metadata in the form of text indexed according to time is generated using video data or sound data among context information data, and from this, when context information data is reproduced, reproduction time is determined. Context information data can be reproduced immediately from the searched reproduction point. Accordingly, since an event point in time desired to be reproduced can be retrieved and reproduced easily, the situation information data does not need to be reproduced from the beginning, and can be repeatedly reproduced from a desired point in time.

The above embodiments of the present invention are for explanation of the present invention and are not intended to limit the present invention. It should be noted that the configurations and methods disclosed in the above embodiments of the present invention may be combined and modified in various forms by combining or crossing each other, and variations thereof may also be considered within the scope of the present invention. That is, the present invention will be implemented in a variety of different forms within the scope of the claims and equivalent technical ideas, and various embodiments are possible within the scope of the technical idea of the present invention by those skilled in the art to which the present invention pertains. will be able to understand

1: Trap 10: Combat System
11: tactical command system 12: sensor system
13: communication system 100: collection unit
200: first data processing unit 300: data conversion unit
400: metadata generation unit 500: situation information display unit

Claims (13)

a collection unit for collecting context information data from traps;
a first data processing unit for extracting first data having a first form from the context information data;
a data conversion unit configured to generate second data by converting the first data into a second form different from the first form;
a metadata generator for generating metadata using the second data; and
a context information display unit capable of reproducing the context information data and selecting a playback time of the context information data based on the metadata;
The metadata generation unit analyzes the meaning of the second data, divides the second data into meaningful units to generate events, indexes the events according to time, and operates contextual information to generate the metadata. Device. The method of claim 1,
The collection unit is connected to the combat system of the ship and collects video data and sound data recorded in the combat system as the situation information data,
The first data processing unit selects one of the image data and the sound data and extracts it as the first data. The method of claim 1,
The data conversion unit recognizes the first data using a first data recognition model, converts the first data into text form, and generates text data as the second data. delete a collection unit for collecting context information data from traps;
a first data processing unit for extracting first data having a first form from the context information data;
a data conversion unit configured to generate second data by converting the first data into a second form different from the first form;
a metadata generator for generating metadata using the second data; and
a context information display unit capable of reproducing the context information data and selecting a playback time of the context information data based on the metadata;
The situation information display unit receives a search word for searching for an event on the operation status of the trap, extracts an event matching the search word among events included in the metadata as a request event, and indexes the requested event. A context information operating device for reproducing the context information data from time. The process of collecting situational information data from traps;
extracting first data having a first shape from the situation information data;
generating second data by converting the first data into a second form different from the first form;
generating metadata using the second data;
Selecting a reproduction time point of the context information data based on the metadata and reproducing the context information data from the selected reproduction time point;
The process of collecting the context information data,
A process of operating the ship using the combat system of the ship;
Recording the operation status of the ship in the combat system in a plurality of forms;
and collecting the data recorded in the combat system as the situation information data. delete The method of claim 6,
The process of recording in the combat system,
Recording tactical information, radar detection information, visible light detection information, and infrared light detection information input and output from a tactical command system and a sensor system among subsystems of the combat system as image data;
Among the subsystems of the combat system, a process of recording communication information input and output from the communication system as sound data; The method of claim 8,
The process of extracting the first data,
and selecting the video data or the sound data from among the context information data and extracting the first data. The method of any one of claims 6 and 8 to 9,
The process of generating the second data,
and recognizing the first data using a first data recognition model input in advance, converting the first data into text form, and generating the converted text data as the second data. The method of claim 10,
The first data includes sound data, the first data recognition model includes a sound recognition model,
The process of generating the second data,
selecting sound data having a volume greater than or equal to a preset reference volume among the sound data;
identifying acoustic data corresponding to a crew member's voice from among the selected acoustic data using the acoustic recognition model, and generating voice-recognized texts by performing voice recognition;
A method of operating contextual information, comprising: generating text data by arranging the generated voice recognition texts in order. The method of claim 11,
The process of generating the metadata,
analyzing the meaning of the voice recognition texts listed in order, and generating an event by dividing the voice recognition texts listed in the order into minimum units having meaning according to the analyzed meaning;
A method of managing contextual information that includes indexing generated events according to time and generating the metadata. The method of claim 11,
The process of reproducing the situation information data,
The process of receiving a search term input;
extracting an event similar or identical to the search word from among the metadata as a request event;
and reproducing the context data from the time indexed to the extracted request event.

Images