system

Abstract

The system according to this embodiment aims to analyze surveillance camera footage in real time and respond quickly. [Solution] The system according to the embodiment comprises a collection unit, an analysis unit, an alert unit, and a notification unit. The collection unit collects video from surveillance cameras. The analysis unit analyzes the video collected by the collection unit. The alert unit issues an alert based on the results of the analysis by the analysis unit. The notification unit notifies the police based on the alert issued by the alert unit.

Description

Technical Field

[0001] The technology of the present disclosure relates to a system.

Background Art

[0002] Patent Document 1 discloses a persona chatbot control method performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the conventional technology, there is a problem that it is difficult to analyze the video of a surveillance camera in real time and respond quickly.

[0005] <00所示的系统旨在实时分析监控摄像头的视频并迅速做出响应。

Means for Solving the Problems

[0006] The system according to this embodiment comprises a collection unit, an analysis unit, an alert unit, and a notification unit. The collection unit collects video footage from surveillance cameras. The analysis unit analyzes the video footage collected by the collection unit. The alert unit issues an alert based on the results of the analysis by the analysis unit. The notification unit notifies the police based on the alert issued by the alert unit. [Effects of the Invention]

[0007] The system according to this embodiment can analyze surveillance camera footage in real time and respond quickly. [Brief explanation of the drawing]

[0008] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Modes for carrying out the invention]

[0009] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0010] First, let's explain the terminology used in the following explanation.

[0011] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), APU (Accelerated Processing Unit), or TPU (Tensor Processing Unit).

[0012] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0013] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0014] In the following embodiments, the labeled communication I / F (Interface) is an interface including a communication processor, an antenna, and the like. The communication I / F controls communication between a plurality of computers. Examples of communication standards applied to the communication I / F include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0015] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B". That is, "A and / or B" means that it may be only A, only B, or a combination of A and B. Also, in this specification, when expressing three or more matters connected by "and / or", the same concept as "A and / or B" is applied.

[0016] [First Embodiment] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0017] As shown in FIG. 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0018] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. Also, the database 24 and the communication I / F 26 are connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0019] The smart device 14 comprises a computer 36, a receiving device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The receiving device 38, output device 40, and camera 42 are also connected to the bus 52.

[0020] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, and accepts user input. The touch panel 38A accepts user input via touch by detecting contact with an object (e.g., a pen or finger). The microphone 38B accepts user input via voice by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 (see Figure 2) acquires the data indicating the user input.

[0021] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user by outputting the data in a form perceptible to the user (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0022] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0023] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0024] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0025] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0026] In the smart device 14, specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The specific processing program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as a control unit 46A according to the specific processing program 60 executed on the RAM 48. The smart device 14 also has a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0027] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device having the data generation model 58. The data processing device 12 may also be a server device or a terminal device owned by a user (e.g., a mobile phone, robot, home appliance, etc.). Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example of form 1) The surveillance system according to an embodiment of the present invention is a system that uses AI to analyze crime records and pre-crime footage that may lead to future crimes recorded by surveillance cameras in real time, and issues alerts and notifies the police. The surveillance system detects risk from "footage of the few minutes before and after an incident or accident" captured by surveillance cameras, and automatically notifies the police in near real time only when a risk is actually detected. Specifically, it identifies "the physical characteristics of the suspect" and "actions at a scene where an incident or accident is likely" from the image, performs the following matching / inference using a generating AI that has been fed past crime records, and automatically generates the following information and notifies / recommends the police at the same time as transmitting the above image to the police in real time. First, the AI ​​analyzes the footage recorded by the surveillance camera in real time. At this time, the AI ​​identifies the physical characteristics and actions of people in the footage and compares them with past crime records. For example, it detects actions such as an intruder trying to break glass or a suspicious person trying to force their way through an office gate. If it is determined that there is a high risk of crime, the AI ​​automatically transmits the footage to the police in real time. Next, the generating AI infers the suspect's next actions (such as specific escape methods) based on similar past cases. For example, it presents a ranking of escape routes and destinations chosen by suspects in similar past cases. This allows the police to respond quickly. Furthermore, the generating AI infers the next actions the police should take in real time based on similar past cases. For example, it suggests locations where the suspect is likely to escape or where the police should rush. This enables the police to make an effective initial response. This system enables a higher level of incident and accident resolution and crime deterrence than existing security providers, leading to a reduction in crime rates. In addition, automated analysis and reporting by AI reduces labor costs, making it possible to provide high-performance security services at a low price. As a result, the monitoring system can detect crime risks in real time and quickly report them to the police.

[0029] The surveillance system according to this embodiment comprises a collection unit, an analysis unit, an alert unit, and a notification unit. The collection unit collects video from surveillance cameras. For example, the collection unit acquires video data from surveillance cameras in real time. The collection unit can also store past video data and use it for analysis as needed. For example, the collection unit stores video from surveillance cameras in cloud storage so that the analysis unit can access it. Furthermore, the collection unit can perform noise reduction and image correction to improve the quality of the video data. The analysis unit analyzes the video collected by the collection unit. For example, the analysis unit identifies people and objects in the video using an image recognition algorithm. The analysis unit can also detect suspicious movements using motion detection technology. For example, the analysis unit recognizes the faces of people in the video and compares them with past crime records. The analysis unit can also analyze movement patterns in the video and detect abnormal movements. The alert unit issues alerts based on the results of the analysis performed by the analysis unit. For example, the alert unit issues an audio alert to draw the attention of people in the vicinity. Furthermore, the alert unit can send alerts to relevant parties via email or app notifications. For example, the alert unit sends alerts to the police or security personnel based on the analysis results. The reporting unit reports to the police based on the alerts issued by the alert unit. The reporting unit can, for example, automatically transmit video to the police in real time. The reporting unit can also send the report content to the police as text data. For example, the reporting unit sends detailed report content to the police based on the analysis results. As a result, the surveillance system according to this embodiment can analyze surveillance camera footage in real time, detect the risk of crime, and report it to the police.

[0030] The data collection unit collects video footage from surveillance cameras. For example, the data collection unit acquires video data from surveillance cameras in real time. Specifically, surveillance cameras capture high-resolution video, and this video data is transmitted to a central server via a network. The data collection unit receives this video data and prepares it for real-time processing. The data collection unit can also store past video data and use it for analysis as needed. For example, the data collection unit can store surveillance camera footage in cloud storage, making it accessible to the analysis unit. Cloud storage can efficiently store large amounts of data, enabling long-term storage of video data. Furthermore, the data collection unit can perform noise reduction and image correction to improve the quality of the video data. Noise reduction removes unwanted noise from the video, while image correction adjusts brightness and contrast to improve the visibility of the video. This allows the data collection unit to collect high-quality video data from surveillance cameras, providing a foundation for accurate analysis by the analysis unit. Additionally, the data collection unit can integrate video data from multiple surveillance cameras to achieve wide-area surveillance. For example, video from multiple cameras can be displayed on a single screen to grasp the overall situation. This allows the data collection unit to collect data efficiently and effectively, improving the overall performance of the system.

[0031] The analysis unit analyzes the video footage collected by the collection unit. For example, the analysis unit uses image recognition algorithms to identify people and objects in the video. Specifically, it utilizes deep learning-based image recognition technology to accurately identify faces of people and license plates of vehicles in the video. The analysis unit can also detect suspicious activity using motion detection technology. For example, it tracks the movements of people in the video and detects unusual behavioral patterns. This includes changes in walking speed and prolonged stays in specific areas. Upon detecting these abnormal behaviors, the analysis unit prepares to immediately issue an alert. Furthermore, the analysis unit recognizes faces in the video and compares them to past criminal records. Using facial recognition technology, it can compare faces captured by surveillance cameras against a database to identify individuals with criminal records. The analysis unit can also analyze behavioral patterns in the video and detect abnormal activity. For example, it can detect suspicious movements in specific areas or unusual behavioral patterns. This allows the analysis unit to quickly and accurately analyze the collected data and understand the surrounding risk situation in real time. Furthermore, the analysis unit can utilize historical data and statistical information to conduct long-term risk assessments and trend analyses. For example, it can predict risk fluctuations in specific areas or time periods based on past crime data and formulate future countermeasures. The analysis unit can also use anomaly detection algorithms to detect unusual patterns and abnormal data, issuing early warnings. As a result, the analysis unit can not only grasp the situation in real time but also handle long-term risk management and anomaly detection, improving the reliability and security of the entire system.

[0032] The alert unit issues alerts based on the results analyzed by the analysis unit. For example, the alert unit can issue audio alerts to draw the attention of people in the vicinity. Specifically, it can place speakers at the locations where surveillance cameras are installed and issue audio warnings when an anomaly is detected. The alert unit can also send alerts to relevant parties via email or app notifications. For example, the alert unit can send alerts to the police or security personnel based on the analysis results. This includes real-time notifications to enable relevant parties to respond quickly. Furthermore, the alert unit provides detailed information in the alert to ensure that relevant parties accurately understand the situation. For example, the alert may include the location and time of the anomaly, and the type of anomaly detected. This allows the alert unit to provide relevant parties with quick and accurate information and encourage appropriate responses. In addition, the alert unit can reliably transmit information using multiple communication methods. For example, it can reliably deliver important information using not only smartphone notifications but also voice calls, SMS, and email. This allows the alert unit to provide users with quick and reliable instructions and minimize the risk of disaster.

[0033] The reporting unit notifies the police based on alerts issued by the alert unit. The reporting unit can, for example, automatically transmit video to the police in real time. Specifically, it transmits surveillance camera footage to the police monitoring center in real time, allowing for immediate understanding of the situation at the scene. The reporting unit can also transmit the content of the report to the police as text data. For example, the reporting unit sends detailed report content to the police based on analysis results. This includes the location and time the anomaly occurred, the type of anomaly detected, and contact information of those involved. Furthermore, the reporting unit automatically records the content of the report for later reference. This allows the reporting unit to provide the police with quick and accurate information and facilitate appropriate responses. In addition, the reporting unit manages the history of reports and can refer to past reports. This allows the reporting unit to analyze and improve based on past reports. For example, by analyzing past reports, it can understand risk trends in specific areas or time periods and plan future countermeasures. This allows the reporting unit to provide the police with quick and accurate information and minimize the risk of crime.

[0034] The inference unit uses generative AI to infer the suspect's next actions based on similar past cases. For example, the inference unit infers the suspect's escape route based on past crime data. It can also infer the most likely means of escape the suspect will choose. For example, it can infer whether the suspect will use a car or escape on foot based on past crime data. Furthermore, the inference unit can infer places the suspect might visit during their escape. For example, it can infer the likelihood of the suspect visiting a specific location based on past crime data. This helps the police respond quickly by inferring the suspect's next actions from similar past cases. Some or all of the above processing in the inference unit is performed using generative AI. For example, the inference unit inputs past crime data into the generative AI to infer the suspect's next actions. The generative AI performs inference using techniques such as deep learning and reinforcement learning. This allows the inference unit to perform highly accurate inferences using generative AI.

[0035] The action reasoning unit uses generative AI to infer the next actions the police should take in real time, derived from similar past cases. For example, the action reasoning unit infers the location the police should rush to based on past crime data. The action reasoning unit can also infer the response procedures the police should take. For example, the action reasoning unit infers what actions the police should take based on past crime data. The action reasoning unit can also infer the equipment and means the police should use. For example, the action reasoning unit infers what equipment the police should use based on past crime data. This supports an effective initial response by inferring the next actions the police should take in real time. Some or all of the above processing in the action reasoning unit is performed using generative AI. For example, the action reasoning unit inputs past crime data into the generative AI and infers the next actions the police should take. The generative AI performs inference using technologies such as deep learning and reinforcement learning. This allows the action reasoning unit to perform highly accurate inferences using generative AI.

[0036] The analysis unit can identify the physical characteristics and actions of people in the video and compare them with past criminal records. For example, the analysis unit can identify the faces of people in the video using facial recognition technology. The analysis unit can also identify the characteristics of clothing. For example, the analysis unit can identify the color and shape of the clothing of people in the video. The analysis unit can also identify actions such as walking patterns and hand movements. For example, the analysis unit can analyze the walking patterns of people in the video and compare them with past criminal records. By identifying the physical characteristics and actions of people in the video and comparing them with past criminal records, the risk of crime can be detected with high accuracy. Some or all of the above processing in the analysis unit may be performed using AI or not. For example, the analysis unit can input the physical characteristics of people in the video into AI and have the AI ​​perform the comparison with past criminal records.

[0037] The analysis unit can detect actions such as intruders attempting to break glass or suspicious individuals attempting to force their way through office gates. For example, the analysis unit can apply an algorithm to detect glass-breaking actions. The analysis unit can also detect actions such as forcing oneself through office gates. For example, the analysis unit can detect when a person in the video attempts to force their way through an office gate. The analysis unit can also detect suspicious actions such as looking around. For example, the analysis unit can detect when a person in the video looks around. By detecting actions such as intruders attempting to break glass or suspicious individuals attempting to force their way through office gates, the risk of crime can be detected early. Some or all of the above processing in the analysis unit may be performed using AI or not. For example, the analysis unit can input actions in the video into an AI and have the AI ​​perform the detection of suspicious actions.

[0038] The reporting unit can automatically transmit video to the police in real time. For example, the reporting unit can transmit surveillance camera footage to the police in real time. The reporting unit can also transmit text data along with the video. For example, the reporting unit can transmit detailed report information to the police based on analysis results. The reporting unit can also transmit audio data. For example, the reporting unit can transmit audio from the scene to the police in real time. This enables a rapid response by automatically transmitting video to the police in real time. Some or all of the above processing in the reporting unit may be performed using AI or not. For example, the reporting unit can input surveillance camera footage into AI and have the AI ​​perform the real-time transmission.

[0039] The data collection unit can dynamically adjust the placement and angle of surveillance cameras to collect optimal footage. For example, the data collection unit can use AI to dynamically change the placement of surveillance cameras to ensure an optimal field of view. Furthermore, the data collection unit can use AI to adjust the angle of surveillance cameras in real time, constantly monitoring important areas. For example, the data collection unit can use AI to utilize the zoom function of surveillance cameras to collect detailed footage. This allows for the collection of optimal footage by dynamically adjusting the placement and angle of surveillance cameras. Some or all of the above-described processes in the data collection unit may be performed using AI or not. For example, the data collection unit can have AI perform the adjustment of the placement and angle of surveillance cameras.

[0040] The collection unit can change its collection range based on specific time periods or events when collecting video footage. For example, the collection unit can expand its collection range during specific times, such as at night or on holidays, to reduce the risk of crime. It can also expand its collection range during events to monitor crowd movements. For example, the collection unit can change its collection range and collect detailed video when a specific alert occurs. This reduces the risk of crime by changing the collection range based on specific time periods or events. Some or all of the above processes in the collection unit may be performed using AI or not. For example, the collection unit can have AI perform the changes to the collection range based on specific time periods or events.

[0041] The data collection unit can collect ambient audio data simultaneously with video data and use it for analysis. For example, the data collection unit can collect ambient audio data along with the video and provide it to the analysis unit. The data collection unit can also collect video when it detects a specific audio event (for example, the sound of breaking glass). For example, the data collection unit can analyze the audio data and collect video if it detects an abnormal sound. This improves the accuracy of the analysis by simultaneously collecting ambient audio data. Some or all of the above processing in the data collection unit may be performed using AI or not. For example, the data collection unit can have AI perform the collection and analysis of audio data.

[0042] The data collection unit can collect video footage in conjunction with data from environmental sensors (temperature, humidity, light intensity, etc.). For example, the data collection unit can collect video footage based on temperature sensor data when an abnormal temperature change is detected. It can also collect video footage based on humidity sensor data when an abnormal humidity change is detected. For example, the data collection unit can collect video footage based on light intensity sensor data when an abnormal light intensity change is detected. By using data from environmental sensors in conjunction with this data, abnormal situations can be detected more quickly. Some or all of the above processing in the data collection unit may be performed using AI, or it may be performed without AI. For example, the data collection unit can input environmental sensor data into AI and have the AI ​​perform the detection of abnormal situations.

[0043] The analysis unit can improve the accuracy of video analysis by integrating multiple camera images. For example, the analysis unit can integrate multiple camera images to perform analysis with a wide field of view. The analysis unit can also integrate images from different angles to perform detailed analysis. For example, the analysis unit can integrate images from different time periods to perform long-term analysis. This improves the accuracy of analysis by integrating multiple camera images. Some or all of the above processing in the analysis unit may be performed using AI or not. For example, the analysis unit can have AI perform the integration of multiple camera images.

[0044] The analysis unit can apply algorithms to identify specific motion patterns or action sequences during video analysis. For example, the analysis unit can apply an algorithm to identify a specific motion pattern (e.g., the action of breaking glass). It can also apply an algorithm to identify a specific action sequence (e.g., an escape route after intrusion). For example, the analysis unit can identify the motion patterns of a specific person and compare them with past crime records. This allows for highly accurate detection of crime risk by identifying specific motion patterns and action sequences. Some or all of the above-described processes in the analysis unit may be performed using AI or not. For example, the analysis unit can have AI perform the identification of motion patterns and action sequences.

[0045] The analysis unit can detect suspicious sounds by combining video analysis with audio analysis. For example, the analysis unit can detect the sound of glass breaking by combining video and audio analysis. The analysis unit can also detect unusual sounds (e.g., screams) by combining video and audio analysis. For example, the analysis unit can detect specific audio events (e.g., gunshots) by combining video and audio analysis. By combining audio analysis, suspicious sounds can be detected, and the risk of crime can be identified early. Some or all of the above processing in the analysis unit may be performed using AI or not. For example, the analysis unit can have AI perform the analysis of audio data.

[0046] The analysis unit can correct the analysis results by referring to past weather data during video analysis. For example, the analysis unit can correct the video analysis results for rainy weather by referring to past weather data. The analysis unit can also correct the video analysis results for snowy days by referring to past weather data. For example, the analysis unit can correct the video analysis results for sunny weather by referring to past weather data. In this way, the accuracy of the analysis results is improved by referring to past weather data. Some or all of the above processing in the analysis unit may be performed using AI or not. For example, the analysis unit can have AI perform the referencing of weather data and the correction of analysis results.

[0047] The alert unit can use multiple communication methods (SMS, email, app notifications, etc.) when sending an alert. For example, the alert unit can send an alert using both SMS and email. Alternatively, the alert unit can send an alert using both app notifications and email. For example, the alert unit can send an alert using both SMS and app notifications. This ensures that alerts are reliably transmitted by using multiple communication methods. Some or all of the above-described processes in the alert unit may be performed using AI or not. For example, the alert unit can have AI select and send alerts using multiple communication methods.

[0048] The alert unit can select different notification methods depending on the priority of the alert when an alert is issued. For example, the alert unit may send high-priority alerts via SMS and low-priority alerts via email. Alternatively, the alert unit may send high-priority alerts via app notifications and low-priority alerts via email. For example, the alert unit may send high-priority alerts via both SMS and app notifications and low-priority alerts via email. This ensures that important alerts are reliably communicated by selecting the appropriate notification method according to the alert priority. Some or all of the above processing in the alert unit may be performed using AI or not. For example, the alert unit can have AI perform the determination of alert priority and the selection of notification methods.

[0049] The alert unit can notify the police and relevant parties by adding geographical information when an alert is issued. For example, the alert unit can notify the police by adding geographical information of the current location when an alert is issued. The alert unit can also notify the police by adding geographical information of the crime scene when an alert is issued. For example, the alert unit can notify the police by adding the location information of relevant parties when an alert is issued. This allows the police and relevant parties to respond quickly by adding geographical information. Some or all of the above processing in the alert unit may be performed using AI or not using AI. For example, the alert unit can have AI perform the acquisition and notification of geographical information.

[0050] The alert unit can improve the accuracy of alerts by referring to past alert history when issuing an alert. For example, the alert unit can refer to past alert history and issue similar alerts. The alert unit can also refer to past alert history and optimize the content of alerts. For example, the alert unit can refer to past alert history and adjust the timing of alert issuance. In this way, the accuracy of alerts is improved by referring to past alert history. Some or all of the above processing in the alert unit may be performed using AI or not. For example, the alert unit can have AI perform the referencing of alert history and the optimization of alerts.

[0051] The reporting unit can monitor the police response in real time upon receiving a report and provide additional information as needed. For example, the reporting unit can monitor the police response in real time and provide additional video footage as needed. The reporting unit can also monitor the police response in real time and provide additional audio data as needed. For example, the reporting unit can monitor the police response in real time and provide additional text information as needed. This allows for the provision of additional information as needed by monitoring the police response in real time. Some or all of the above processes in the reporting unit may be performed using AI or not. For example, the reporting unit can have AI perform the monitoring of the police response and the provision of additional information.

[0052] The reporting department can automatically route reports to different police departments based on their content. For example, it can automatically route reports to the criminal investigation department. It can also automatically route reports to the traffic department. For example, it can automatically route reports to the community police department. This enables a swift response by automatically routing reports to the appropriate police department based on their content. Some or all of the above processing in the reporting department may be performed using AI, or not. For example, the reporting department can have AI perform the classification and automatic routing of reports.

[0053] The reporting unit can transmit audio and text data along with video when a report is made. For example, the reporting unit can transmit audio data along with video when a report is made. The reporting unit can also transmit text data along with video when a report is made. For example, the reporting unit can transmit video, audio data, and text data simultaneously when a report is made. This allows for the provision of more detailed information by transmitting audio and text data along with video. Some or all of the above processing in the reporting unit may be performed using AI or not. For example, the reporting unit can have AI perform the transmission of video, audio data, and text data.

[0054] The reporting unit can optimize the content of a report by referring to past reporting history when a report is made. For example, the reporting unit can refer to past reporting history and optimize similar report content. The reporting unit can also refer to past reporting history and optimize the content of the report. For example, the reporting unit can refer to past reporting history and adjust the timing of the report's submission. This optimizes the report content and improves accuracy by referring to past reporting history. Some or all of the above processes in the reporting unit may be performed using AI or not. For example, the reporting unit can have AI perform the referencing of reporting history and the optimization of report content.

[0055] The inference unit can improve its inference accuracy by referring to past crime data during inference. For example, the inference unit can refer to past crime data and perform inferences based on similar crime patterns. The inference unit can also refer to past crime data and infer the probability of a crime occurring. For example, the inference unit can refer to past crime data and infer the location where a crime occurred. In this way, the inference unit improves its inference accuracy by referring to past crime data. Some or all of the above processing in the inference unit is performed using a generative AI. For example, the inference unit can input past crime data into the generative AI and have the generative AI perform the improvement of inference accuracy.

[0056] The inference unit can improve accuracy by combining multiple inference algorithms during inference. For example, the inference unit can combine different inference algorithms to infer the probability of a crime occurring. It can also combine different inference algorithms to infer the location where a crime occurred. For example, the inference unit can combine different inference algorithms to infer the time of a crime. This improves inference accuracy by combining multiple inference algorithms. Some or all of the above processing in the inference unit is performed using generative AI. For example, the inference unit can have the generative AI execute combinations of multiple inference algorithms.

[0057] The inference unit can correct its inference results by referring to geographic information during the inference process. For example, the inference unit can correct the location of a crime by referring to geographic information. It can also correct the probability of a crime occurring by referring to geographic information. For example, the inference unit can correct the time of a crime by referring to geographic information. This improves the accuracy of the inference results by referring to geographic information. Some or all of the above processing in the inference unit is performed using a generative AI. For example, the inference unit can have the generative AI perform the referencing of geographic information and the correction of inference results.

[0058] The inference unit can correct its inference results by referring to past weather data during the inference process. For example, the inference unit can correct the probability of crime occurring during rainy weather by referring to past weather data. It can also correct the probability of crime occurring on snowy days by referring to past weather data. For example, the inference unit can correct the probability of crime occurring during sunny weather by referring to past weather data. In this way, the accuracy of the inference results is improved by referring to past weather data. Some or all of the above processing in the inference unit is performed using a generative AI. For example, the inference unit can have the generative AI perform the referencing of weather data and the correction of inference results.

[0059] The action inference unit can improve its inference accuracy by referring to past police response data during action inference. For example, the action inference unit can refer to past police response data and perform inferences based on similar response patterns. The action inference unit can also refer to past police response data and infer the effectiveness of a response. For example, the action inference unit can refer to past police response data and infer the speed of a response. In this way, the inference accuracy is improved by referring to past police response data. Some or all of the above processing in the action inference unit is performed using generative AI. For example, the action inference unit can have the generative AI perform the referencing of police response data and the improvement of inference accuracy.

[0060] The action inference unit can improve accuracy by combining multiple inference algorithms during action inference. For example, the action inference unit can combine different inference algorithms to infer the effectiveness of a response. It can also combine different inference algorithms to infer the speed of a response. For example, the action inference unit can combine different inference algorithms to infer the appropriateness of a response. This improves inference accuracy by combining multiple inference algorithms. Some or all of the above processing in the action inference unit is performed using generative AI. For example, the action inference unit can have the generative AI execute combinations of multiple inference algorithms.

[0061] The action inference unit can correct its inference results by referring to geographic information during action inference. For example, the action inference unit can refer to geographic information to correct the effectiveness of a response. It can also refer to geographic information to correct the speed of a response. For example, the action inference unit can refer to geographic information to correct the appropriateness of a response. In this way, the accuracy of the action inference results is improved by referring to geographic information. Some or all of the above processing in the action inference unit is performed using a generative AI. For example, the action inference unit can have the generative AI perform the referencing of geographic information and the correction of the inference results.

[0062] The action inference unit can correct its inference results by referring to past weather data during action inference. For example, the action inference unit can refer to past weather data to correct the response effect during rainy weather. It can also refer to past weather data to correct the response effect on snowy days. For example, the action inference unit can refer to past weather data to correct the response effect during sunny weather. In this way, the accuracy of the action inference results is improved by referring to past weather data. Some or all of the above processing in the action inference unit is performed using a generative AI. For example, the action inference unit can have the generative AI perform the referencing of weather data and the correction of the inference results.

[0063] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0064] The surveillance system may also include an audio analysis unit. The audio analysis unit can analyze audio data collected along with the video from the surveillance camera and detect abnormal sounds. For example, the audio analysis unit can detect abnormal sounds such as the sound of breaking glass or screaming. The audio analysis unit can also identify specific audio patterns and detect the risk of crime with high accuracy. For example, the audio analysis unit can analyze specific audio patterns and compare them with past crime records. This allows for the early detection of anomalies that cannot be detected by video data alone by analyzing audio data. Some or all of the above processing in the audio analysis unit may be performed using AI or not. For example, the audio analysis unit can input audio data into an AI and have the AI ​​perform the detection of abnormal sounds.

[0065] The monitoring system may also include an environmental sensor unit. The environmental sensor unit can collect environmental data such as temperature, humidity, and light intensity and provide it to the analysis unit. For example, the environmental sensor unit can detect abnormal temperature and humidity changes and detect crime risks with high accuracy. The environmental sensor unit can also detect changes in light intensity and detect nighttime crime risks early. For example, the environmental sensor unit can analyze sudden changes in light intensity and compare them with past crime records. By analyzing environmental data in this way, anomalies that cannot be detected by video data alone can be detected early. Some or all of the above processing in the environmental sensor unit may be performed using AI or not. For example, the environmental sensor unit can input environmental data into the AI ​​and have the AI ​​perform the detection of abnormal situations.

[0066] The surveillance system may further include a behavior prediction unit. The behavior prediction unit can predict a person's next action from surveillance camera footage and provide it to the analysis unit. For example, the behavior prediction unit can analyze a person's movements in the footage and predict the most likely next action. The behavior prediction unit can also detect the risk of crime with high accuracy based on past crime records. For example, the behavior prediction unit can analyze past crime data and identify specific behavior patterns. By analyzing behavioral data, it is possible to detect anomalies that cannot be detected by video data alone at an early stage. Some or all of the above processing in the behavior prediction unit may be performed using AI or not. For example, the behavior prediction unit can input behavioral data into AI and have the AI ​​perform the prediction of the next action.

[0067] The surveillance system may also include an anomaly detection unit. The anomaly detection unit can detect abnormal movements or situations from surveillance camera footage and provide this information to the analysis unit. For example, the anomaly detection unit can analyze the movements of people in the footage and detect abnormal movements. The anomaly detection unit can also detect specific situations (e.g., abnormal movements of a crowd) and detect the risk of crime with high accuracy. For example, the anomaly detection unit can analyze the movements of a crowd in the footage and compare them with past crime records. By analyzing the anomaly data, it is possible to detect anomalies that cannot be detected by video data alone at an early stage. Some or all of the above-described processes in the anomaly detection unit may be performed using AI or not. For example, the anomaly detection unit can input the anomaly data into the AI ​​and have the AI ​​perform the detection of abnormal movements.

[0068] The surveillance system may also include an anomaly detection unit. The anomaly detection unit can detect abnormal movements or situations from surveillance camera footage and provide this information to the analysis unit. For example, the anomaly detection unit can analyze the movements of people in the footage and detect abnormal movements. The anomaly detection unit can also detect specific situations (e.g., abnormal movements of a crowd) and detect the risk of crime with high accuracy. For example, the anomaly detection unit can analyze the movements of a crowd in the footage and compare them with past crime records. By analyzing the anomaly data, it is possible to detect anomalies that cannot be detected by video data alone at an early stage. Some or all of the above-described processes in the anomaly detection unit may be performed using AI or not. For example, the anomaly detection unit can input the anomaly data into the AI ​​and have the AI ​​perform the detection of abnormal movements.

[0069] The following briefly describes the processing flow for example form 1.

[0070] Step 1: The collection unit collects video footage from surveillance cameras. For example, the collection unit acquires video data from surveillance cameras in real time. The collection unit can also store past video data and use it for analysis as needed. For example, the collection unit can store the surveillance camera footage in cloud storage and make it accessible to the analysis unit. Furthermore, the collection unit can perform noise reduction and image correction to improve the quality of the video data. Step 2: The analysis unit analyzes the video collected by the collection unit. The analysis unit can, for example, use image recognition algorithms to identify people and objects in the video. The analysis unit can also detect suspicious movements using motion detection technology. For example, the analysis unit can recognize the faces of people in the video and compare them with past criminal records. The analysis unit can also analyze movement patterns in the video and detect abnormal movements. Step 3: The alert unit issues an alert based on the results analyzed by the analysis unit. The alert unit can, for example, issue an audio alert to draw the attention of people in the vicinity. The alert unit can also send alerts to relevant parties via email or app notifications. For example, the alert unit can send an alert to the police or security personnel based on the analysis results. Step 4: The reporting unit notifies the police based on the alert issued by the alert unit. The reporting unit can, for example, automatically send video to the police in real time. The reporting unit can also send the report content to the police as text data. For example, the reporting unit can send detailed report content to the police based on the analysis results.

[0071] (Example of form 2) The surveillance system according to an embodiment of the present invention is a system that uses AI to analyze crime records and pre-crime footage that may lead to future crimes recorded by surveillance cameras in real time, and issues alerts and notifies the police. The surveillance system detects risk from "footage of the few minutes before and after an incident or accident" captured by surveillance cameras, and automatically notifies the police in near real time only when a risk is actually detected. Specifically, it identifies "the physical characteristics of the suspect" and "actions at a scene where an incident or accident is likely" from the image, performs the following matching / inference using a generating AI that has been fed past crime records, and automatically generates the following information and notifies / recommends the police at the same time as transmitting the above image to the police in real time. First, the AI ​​analyzes the footage recorded by the surveillance camera in real time. At this time, the AI ​​identifies the physical characteristics and actions of people in the footage and compares them with past crime records. For example, it detects actions such as an intruder trying to break glass or a suspicious person trying to force their way through an office gate. If it is determined that there is a high risk of crime, the AI ​​automatically transmits the footage to the police in real time. Next, the generating AI infers the suspect's next actions (such as specific escape methods) based on similar past cases. For example, it presents a ranking of escape routes and destinations chosen by suspects in similar past cases. This allows the police to respond quickly. Furthermore, the generating AI infers the next actions the police should take in real time based on similar past cases. For example, it suggests locations where the suspect is likely to escape or where the police should rush. This enables the police to make an effective initial response. This system enables a higher level of incident and accident resolution and crime deterrence than existing security providers, leading to a reduction in crime rates. In addition, automated analysis and reporting by AI reduces labor costs, making it possible to provide high-performance security services at a low price. As a result, the monitoring system can detect crime risks in real time and quickly report them to the police.

[0072] The surveillance system according to this embodiment comprises a collection unit, an analysis unit, an alert unit, and a notification unit. The collection unit collects video from surveillance cameras. For example, the collection unit acquires video data from surveillance cameras in real time. The collection unit can also store past video data and use it for analysis as needed. For example, the collection unit stores video from surveillance cameras in cloud storage so that the analysis unit can access it. Furthermore, the collection unit can perform noise reduction and image correction to improve the quality of the video data. The analysis unit analyzes the video collected by the collection unit. For example, the analysis unit identifies people and objects in the video using an image recognition algorithm. The analysis unit can also detect suspicious movements using motion detection technology. For example, the analysis unit recognizes the faces of people in the video and compares them with past crime records. The analysis unit can also analyze movement patterns in the video and detect abnormal movements. The alert unit issues alerts based on the results of the analysis performed by the analysis unit. For example, the alert unit issues an audio alert to draw the attention of people in the vicinity. Furthermore, the alert unit can send alerts to relevant parties via email or app notifications. For example, the alert unit sends alerts to the police or security personnel based on the analysis results. The reporting unit reports to the police based on the alerts issued by the alert unit. The reporting unit can, for example, automatically transmit video to the police in real time. The reporting unit can also send the report content to the police as text data. For example, the reporting unit sends detailed report content to the police based on the analysis results. As a result, the surveillance system according to this embodiment can analyze surveillance camera footage in real time, detect the risk of crime, and report it to the police.

[0073] The data collection unit collects video footage from surveillance cameras. For example, the data collection unit acquires video data from surveillance cameras in real time. Specifically, surveillance cameras capture high-resolution video, and this video data is transmitted to a central server via a network. The data collection unit receives this video data and prepares it for real-time processing. The data collection unit can also store past video data and use it for analysis as needed. For example, the data collection unit can store surveillance camera footage in cloud storage, making it accessible to the analysis unit. Cloud storage can efficiently store large amounts of data, enabling long-term storage of video data. Furthermore, the data collection unit can perform noise reduction and image correction to improve the quality of the video data. Noise reduction removes unwanted noise from the video, while image correction adjusts brightness and contrast to improve the visibility of the video. This allows the data collection unit to collect high-quality video data from surveillance cameras, providing a foundation for accurate analysis by the analysis unit. Additionally, the data collection unit can integrate video data from multiple surveillance cameras to achieve wide-area surveillance. For example, video from multiple cameras can be displayed on a single screen to grasp the overall situation. This allows the data collection unit to collect data efficiently and effectively, improving the overall performance of the system.

[0074] The analysis unit analyzes the video footage collected by the collection unit. For example, the analysis unit uses image recognition algorithms to identify people and objects in the video. Specifically, it utilizes deep learning-based image recognition technology to accurately identify faces of people and license plates of vehicles in the video. The analysis unit can also detect suspicious activity using motion detection technology. For example, it tracks the movements of people in the video and detects unusual behavioral patterns. This includes changes in walking speed and prolonged stays in specific areas. Upon detecting these abnormal behaviors, the analysis unit prepares to immediately issue an alert. Furthermore, the analysis unit recognizes faces in the video and compares them to past criminal records. Using facial recognition technology, it can compare faces captured by surveillance cameras against a database to identify individuals with criminal records. The analysis unit can also analyze behavioral patterns in the video and detect abnormal activity. For example, it can detect suspicious movements in specific areas or unusual behavioral patterns. This allows the analysis unit to quickly and accurately analyze the collected data and understand the surrounding risk situation in real time. Furthermore, the analysis unit can utilize historical data and statistical information to conduct long-term risk assessments and trend analyses. For example, it can predict risk fluctuations in specific areas or time periods based on past crime data and formulate future countermeasures. The analysis unit can also use anomaly detection algorithms to detect unusual patterns and abnormal data, issuing early warnings. As a result, the analysis unit can not only grasp the situation in real time but also handle long-term risk management and anomaly detection, improving the reliability and security of the entire system.

[0075] The alert unit issues alerts based on the results analyzed by the analysis unit. For example, the alert unit can issue audio alerts to draw the attention of people in the vicinity. Specifically, it can place speakers at the locations where surveillance cameras are installed and issue audio warnings when an anomaly is detected. The alert unit can also send alerts to relevant parties via email or app notifications. For example, the alert unit can send alerts to the police or security personnel based on the analysis results. This includes real-time notifications to enable relevant parties to respond quickly. Furthermore, the alert unit provides detailed information in the alert to ensure that relevant parties accurately understand the situation. For example, the alert may include the location and time of the anomaly, and the type of anomaly detected. This allows the alert unit to provide relevant parties with quick and accurate information and encourage appropriate responses. In addition, the alert unit can reliably transmit information using multiple communication methods. For example, it can reliably deliver important information using not only smartphone notifications but also voice calls, SMS, and email. This allows the alert unit to provide users with quick and reliable instructions and minimize the risk of disaster.

[0076] The reporting unit notifies the police based on alerts issued by the alert unit. The reporting unit can, for example, automatically transmit video to the police in real time. Specifically, it transmits surveillance camera footage to the police monitoring center in real time, allowing for immediate understanding of the situation at the scene. The reporting unit can also transmit the content of the report to the police as text data. For example, the reporting unit sends detailed report content to the police based on analysis results. This includes the location and time the anomaly occurred, the type of anomaly detected, and contact information of those involved. Furthermore, the reporting unit automatically records the content of the report for later reference. This allows the reporting unit to provide the police with quick and accurate information and facilitate appropriate responses. In addition, the reporting unit manages the history of reports and can refer to past reports. This allows the reporting unit to analyze and improve based on past reports. For example, by analyzing past reports, it can understand risk trends in specific areas or time periods and plan future countermeasures. This allows the reporting unit to provide the police with quick and accurate information and minimize the risk of crime.

[0077] The inference unit uses generative AI to infer the suspect's next actions based on similar past cases. For example, the inference unit infers the suspect's escape route based on past crime data. It can also infer the most likely means of escape the suspect will choose. For example, it can infer whether the suspect will use a car or escape on foot based on past crime data. Furthermore, the inference unit can infer places the suspect might visit during their escape. For example, it can infer the likelihood of the suspect visiting a specific location based on past crime data. This helps the police respond quickly by inferring the suspect's next actions from similar past cases. Some or all of the above processing in the inference unit is performed using generative AI. For example, the inference unit inputs past crime data into the generative AI to infer the suspect's next actions. The generative AI performs inference using techniques such as deep learning and reinforcement learning. This allows the inference unit to perform highly accurate inferences using generative AI.

[0078] The action reasoning unit uses generative AI to infer the next actions the police should take in real time, derived from similar past cases. For example, the action reasoning unit infers the location the police should rush to based on past crime data. The action reasoning unit can also infer the response procedures the police should take. For example, the action reasoning unit infers what actions the police should take based on past crime data. The action reasoning unit can also infer the equipment and means the police should use. For example, the action reasoning unit infers what equipment the police should use based on past crime data. This supports an effective initial response by inferring the next actions the police should take in real time. Some or all of the above processing in the action reasoning unit is performed using generative AI. For example, the action reasoning unit inputs past crime data into the generative AI and infers the next actions the police should take. The generative AI performs inference using technologies such as deep learning and reinforcement learning. This allows the action reasoning unit to perform highly accurate inferences using generative AI.

[0079] The analysis unit can identify the physical characteristics and actions of people in the video and compare them with past criminal records. For example, the analysis unit can identify the faces of people in the video using facial recognition technology. The analysis unit can also identify the characteristics of clothing. For example, the analysis unit can identify the color and shape of the clothing of people in the video. The analysis unit can also identify actions such as walking patterns and hand movements. For example, the analysis unit can analyze the walking patterns of people in the video and compare them with past criminal records. By identifying the physical characteristics and actions of people in the video and comparing them with past criminal records, the risk of crime can be detected with high accuracy. Some or all of the above processing in the analysis unit may be performed using AI or not. For example, the analysis unit can input the physical characteristics of people in the video into AI and have the AI ​​perform the comparison with past criminal records.

[0080] The analysis unit can detect actions such as intruders attempting to break glass or suspicious individuals attempting to force their way through office gates. For example, the analysis unit can apply an algorithm to detect glass-breaking actions. The analysis unit can also detect actions such as forcing oneself through office gates. For example, the analysis unit can detect when a person in the video attempts to force their way through an office gate. The analysis unit can also detect suspicious actions such as looking around. For example, the analysis unit can detect when a person in the video looks around. By detecting actions such as intruders attempting to break glass or suspicious individuals attempting to force their way through office gates, the risk of crime can be detected early. Some or all of the above processing in the analysis unit may be performed using AI or not. For example, the analysis unit can input actions in the video into an AI and have the AI ​​perform the detection of suspicious actions.

[0081] The reporting unit can automatically transmit video to the police in real time. For example, the reporting unit can transmit surveillance camera footage to the police in real time. The reporting unit can also transmit text data along with the video. For example, the reporting unit can transmit detailed report information to the police based on analysis results. The reporting unit can also transmit audio data. For example, the reporting unit can transmit audio from the scene to the police in real time. This enables a rapid response by automatically transmitting video to the police in real time. Some or all of the above processing in the reporting unit may be performed using AI or not. For example, the reporting unit can input surveillance camera footage into AI and have the AI ​​perform the real-time transmission.

[0082] The collection unit can estimate the user's emotions and adjust the timing of video collection based on the estimated emotions. For example, if the user is tense, the collection unit can immediately collect video and begin real-time analysis. Alternatively, if the user is relaxed, the collection unit can periodically collect video and perform analysis as needed. For example, if the user is relaxed, the collection unit can collect video of the normal monitoring area. If the user is excited, the collection unit can frequently collect video and perform detailed analysis. For example, if the user is excited, the collection unit can prioritize collecting video of specific events or activities. This allows for more appropriate timing of video collection by adjusting the timing of video collection based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI includes, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the collection unit may be performed using AI or not. For example, the data collection unit can input user emotion data into a generating AI and have the generating AI adjust the timing of video collection.

[0083] The data collection unit can dynamically adjust the placement and angle of surveillance cameras to collect optimal footage. For example, the data collection unit can use AI to dynamically change the placement of surveillance cameras to ensure an optimal field of view. Furthermore, the data collection unit can use AI to adjust the angle of surveillance cameras in real time, constantly monitoring important areas. For example, the data collection unit can use AI to utilize the zoom function of surveillance cameras to collect detailed footage. This allows for the collection of optimal footage by dynamically adjusting the placement and angle of surveillance cameras. Some or all of the above-described processes in the data collection unit may be performed using AI or not. For example, the data collection unit can have AI perform the adjustment of the placement and angle of surveillance cameras.

[0084] The collection unit can change its collection range based on specific time periods or events when collecting video footage. For example, the collection unit can expand its collection range during specific times, such as at night or on holidays, to reduce the risk of crime. It can also expand its collection range during events to monitor crowd movements. For example, the collection unit can change its collection range and collect detailed video when a specific alert occurs. This reduces the risk of crime by changing the collection range based on specific time periods or events. Some or all of the above processes in the collection unit may be performed using AI or not. For example, the collection unit can have AI perform the changes to the collection range based on specific time periods or events.

[0085] The data collection unit can estimate the user's emotions and determine the priority of the video to collect based on the estimated emotions. For example, if the user is tense, the data collection unit will prioritize collecting video of important areas. Conversely, if the user is relaxed, the data collection unit can also collect video of normal monitoring areas. For example, if the user is excited, the data collection unit will prioritize collecting video of specific events or activities. This allows for the priority collection of important video by determining the priority of the video to collect based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the data collection unit may be performed using AI or not. For example, the data collection unit can input user emotion data into a generative AI and have the generative AI determine the priority of the video.

[0086] The data collection unit can collect ambient audio data simultaneously with video data and use it for analysis. For example, the data collection unit can collect ambient audio data along with the video and provide it to the analysis unit. The data collection unit can also collect video when it detects a specific audio event (for example, the sound of breaking glass). For example, the data collection unit can analyze the audio data and collect video if it detects an abnormal sound. This improves the accuracy of the analysis by simultaneously collecting ambient audio data. Some or all of the above processing in the data collection unit may be performed using AI or not. For example, the data collection unit can have AI perform the collection and analysis of audio data.

[0087] The data collection unit can collect video footage in conjunction with data from environmental sensors (temperature, humidity, light intensity, etc.). For example, the data collection unit can collect video footage based on temperature sensor data when an abnormal temperature change is detected. It can also collect video footage based on humidity sensor data when an abnormal humidity change is detected. For example, the data collection unit can collect video footage based on light intensity sensor data when an abnormal light intensity change is detected. By using data from environmental sensors in conjunction with this data, abnormal situations can be detected more quickly. Some or all of the above processing in the data collection unit may be performed using AI, or it may be performed without AI. For example, the data collection unit can input environmental sensor data into AI and have the AI ​​perform the detection of abnormal situations.

[0088] The analysis unit can estimate the user's emotions and adjust the accuracy of the analysis based on the estimated emotions. For example, if the user is tense, the analysis unit can perform a highly accurate analysis and provide detailed results. Conversely, if the user is relaxed, the analysis unit can perform an analysis with normal accuracy. For example, if the user is excited, the analysis unit can perform a rapid analysis and provide results immediately. This allows for more appropriate analysis results by adjusting the accuracy of the analysis based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generative AI. The generative AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processes in the analysis unit may be performed using AI or not. For example, the analysis unit can input user emotion data into the generative AI and have the generative AI adjust the accuracy of the analysis.

[0089] The analysis unit can improve the accuracy of video analysis by integrating multiple camera images. For example, the analysis unit can integrate multiple camera images to perform analysis with a wide field of view. The analysis unit can also integrate images from different angles to perform detailed analysis. For example, the analysis unit can integrate images from different time periods to perform long-term analysis. This improves the accuracy of analysis by integrating multiple camera images. Some or all of the above processing in the analysis unit may be performed using AI or not. For example, the analysis unit can have AI perform the integration of multiple camera images.

[0090] The analysis unit can apply algorithms to identify specific motion patterns or action sequences during video analysis. For example, the analysis unit can apply an algorithm to identify a specific motion pattern (e.g., the action of breaking glass). It can also apply an algorithm to identify a specific action sequence (e.g., an escape route after intrusion). For example, the analysis unit can identify the motion patterns of a specific person and compare them with past crime records. This allows for highly accurate detection of crime risk by identifying specific motion patterns and action sequences. Some or all of the above-described processes in the analysis unit may be performed using AI or not. For example, the analysis unit can have AI perform the identification of motion patterns and action sequences.

[0091] The analysis unit can estimate the user's emotions and adjust the display method of the analysis results based on the estimated emotions. For example, if the user is tense, the analysis unit can provide a simple and highly visible display method. It can also provide a display method that includes detailed information if the user is relaxed. For example, if the user is in a hurry, the analysis unit can provide a concise display method. This allows for the provision of more appropriate information by adjusting the display method of the analysis results based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generative AI. The generative AI may be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processes in the analysis unit may be performed using AI or not. For example, the analysis unit can input user emotion data into a generative AI and have the generative AI adjust the display method of the analysis results.

[0092] The analysis unit can detect suspicious sounds by combining video analysis with audio analysis. For example, the analysis unit can detect the sound of glass breaking by combining video and audio analysis. The analysis unit can also detect unusual sounds (e.g., screams) by combining video and audio analysis. For example, the analysis unit can detect specific audio events (e.g., gunshots) by combining video and audio analysis. By combining audio analysis, suspicious sounds can be detected, and the risk of crime can be identified early. Some or all of the above processing in the analysis unit may be performed using AI or not. For example, the analysis unit can have AI perform the analysis of audio data.

[0093] The analysis unit can correct the analysis results by referring to past weather data during video analysis. For example, the analysis unit can correct the video analysis results for rainy weather by referring to past weather data. The analysis unit can also correct the video analysis results for snowy days by referring to past weather data. For example, the analysis unit can correct the video analysis results for sunny weather by referring to past weather data. In this way, the accuracy of the analysis results is improved by referring to past weather data. Some or all of the above processing in the analysis unit may be performed using AI or not. For example, the analysis unit can have AI perform the referencing of weather data and the correction of analysis results.

[0094] The alert unit can estimate the user's emotions and adjust how alerts are issued based on the estimated emotions. For example, if the user is tense, the alert unit can issue an alert immediately. It can also issue alerts periodically if the user is relaxed. For example, if the user is excited, the alert unit can issue alerts frequently. This allows for more appropriate alerts to be issued by adjusting the alert method based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above-described processes in the alert unit may be performed using AI or not. For example, the alert unit can input user emotion data into a generative AI and have the generative AI adjust the alert method.

[0095] The alert unit can use multiple communication methods (SMS, email, app notifications, etc.) when sending an alert. For example, the alert unit can send an alert using both SMS and email. Alternatively, the alert unit can send an alert using both app notifications and email. For example, the alert unit can send an alert using both SMS and app notifications. This ensures that alerts are reliably transmitted by using multiple communication methods. Some or all of the above-described processes in the alert unit may be performed using AI or not. For example, the alert unit can have AI select and send alerts using multiple communication methods.

[0096] The alert unit can select different notification methods depending on the priority of the alert when an alert is issued. For example, the alert unit may send high-priority alerts via SMS and low-priority alerts via email. Alternatively, the alert unit may send high-priority alerts via app notifications and low-priority alerts via email. For example, the alert unit may send high-priority alerts via both SMS and app notifications and low-priority alerts via email. This ensures that important alerts are reliably communicated by selecting the appropriate notification method according to the alert priority. Some or all of the above processing in the alert unit may be performed using AI or not. For example, the alert unit can have AI perform the determination of alert priority and the selection of notification methods.

[0097] The alert unit can estimate the user's emotions and customize the content of the alert based on the estimated emotions. For example, if the user is tense, the alert unit can issue an alert containing detailed information. If the user is relaxed, the alert unit can also issue an alert containing concise information. For example, if the user is excited, the alert unit can issue an alert containing information for quick action. This allows for the provision of more appropriate information by customizing the content of the alert based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the alert unit may be performed using AI or not. For example, the alert unit can input user emotion data into a generative AI and have the generative AI customize the content of the alert.

[0098] The alert unit can notify the police and relevant parties by adding geographical information when an alert is issued. For example, the alert unit can notify the police by adding geographical information of the current location when an alert is issued. The alert unit can also notify the police by adding geographical information of the crime scene when an alert is issued. For example, the alert unit can notify the police by adding the location information of relevant parties when an alert is issued. This allows the police and relevant parties to respond quickly by adding geographical information. Some or all of the above processing in the alert unit may be performed using AI or not using AI. For example, the alert unit can have AI perform the acquisition and notification of geographical information.

[0099] The alert unit can improve the accuracy of alerts by referring to past alert history when issuing an alert. For example, the alert unit can refer to past alert history and issue similar alerts. The alert unit can also refer to past alert history and optimize the content of alerts. For example, the alert unit can refer to past alert history and adjust the timing of alert issuance. In this way, the accuracy of alerts is improved by referring to past alert history. Some or all of the above processing in the alert unit may be performed using AI or not. For example, the alert unit can have AI perform the referencing of alert history and the optimization of alerts.

[0100] The notification unit can estimate the user's emotions and adjust the content of the notification based on the estimated emotions. For example, if the user is tense, the notification unit will provide a notification with detailed information. If the user is relaxed, the notification unit can also provide a notification with concise information. For example, if the user is excited, the notification unit will provide a notification with information for a quick response. This allows for the provision of more appropriate information by adjusting the content of the notification based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the notification unit may be performed using AI or not. For example, the notification unit can input user emotion data into a generative AI and have the generative AI adjust the content of the notification.

[0101] The reporting unit can monitor the police response in real time upon receiving a report and provide additional information as needed. For example, the reporting unit can monitor the police response in real time and provide additional video footage as needed. The reporting unit can also monitor the police response in real time and provide additional audio data as needed. For example, the reporting unit can monitor the police response in real time and provide additional text information as needed. This allows for the provision of additional information as needed by monitoring the police response in real time. Some or all of the above processes in the reporting unit may be performed using AI or not. For example, the reporting unit can have AI perform the monitoring of the police response and the provision of additional information.

[0102] The reporting department can automatically route reports to different police departments based on their content. For example, it can automatically route reports to the criminal investigation department. It can also automatically route reports to the traffic department. For example, it can automatically route reports to the community police department. This enables a swift response by automatically routing reports to the appropriate police department based on their content. Some or all of the above processing in the reporting department may be performed using AI, or not. For example, the reporting department can have AI perform the classification and automatic routing of reports.

[0103] The notification unit can estimate the user's emotions and determine the priority of notifications based on the estimated emotions. For example, if the user is stressed, the notification unit will issue a high-priority notification. If the user is relaxed, the notification unit can issue a normal-priority notification. For example, if the user is agitated, the notification unit will issue a high-priority notification requiring a quick response. This allows important notifications to be processed preferentially by determining the priority of notifications based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the notification unit may be performed using AI or not. For example, the notification unit can input user emotion data into a generative AI and have the generative AI determine the priority of notifications.

[0104] The reporting unit can transmit audio and text data along with video when a report is made. For example, the reporting unit can transmit audio data along with video when a report is made. The reporting unit can also transmit text data along with video when a report is made. For example, the reporting unit can transmit video, audio data, and text data simultaneously when a report is made. This allows for the provision of more detailed information by transmitting audio and text data along with video. Some or all of the above processing in the reporting unit may be performed using AI or not. For example, the reporting unit can have AI perform the transmission of video, audio data, and text data.

[0105] The reporting unit can optimize the content of a report by referring to past reporting history when a report is made. For example, the reporting unit can refer to past reporting history and optimize similar report content. The reporting unit can also refer to past reporting history and optimize the content of the report. For example, the reporting unit can refer to past reporting history and adjust the timing of the report's submission. This optimizes the report content and improves accuracy by referring to past reporting history. Some or all of the above processes in the reporting unit may be performed using AI or not. For example, the reporting unit can have AI perform the referencing of reporting history and the optimization of report content.

[0106] The inference unit can estimate the user's emotions and adjust how the inference results are displayed based on the estimated emotions. For example, if the user is nervous, the inference unit can provide a simple and highly visible display method. It can also provide a display method that includes detailed information if the user is relaxed. For example, if the user is in a hurry, the inference unit can provide a concise display method. This allows for the provision of more appropriate information by adjusting the display method of the inference results based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may include, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above-described processing in the inference unit is performed using generative AI. For example, the inference unit can input user emotion data into the generative AI and have the generative AI adjust how the inference results are displayed.

[0107] The inference unit can improve its inference accuracy by referring to past crime data during inference. For example, the inference unit can refer to past crime data and perform inferences based on similar crime patterns. The inference unit can also refer to past crime data and infer the probability of a crime occurring. For example, the inference unit can refer to past crime data and infer the location where a crime occurred. In this way, the inference unit improves its inference accuracy by referring to past crime data. Some or all of the above processing in the inference unit is performed using a generative AI. For example, the inference unit can input past crime data into the generative AI and have the generative AI perform the improvement of inference accuracy.

[0108] The inference unit can improve accuracy by combining multiple inference algorithms during inference. For example, the inference unit can combine different inference algorithms to infer the probability of a crime occurring. It can also combine different inference algorithms to infer the location where a crime occurred. For example, the inference unit can combine different inference algorithms to infer the time of a crime. This improves inference accuracy by combining multiple inference algorithms. Some or all of the above processing in the inference unit is performed using generative AI. For example, the inference unit can have the generative AI execute combinations of multiple inference algorithms.

[0109] The inference unit can estimate the user's emotions and determine the priority of inference results based on the estimated emotions. For example, if the user is tense, the inference unit will provide high-priority inference results. It can also provide normal-priority inference results if the user is relaxed. For example, if the user is excited, the inference unit will provide high-priority inference results for a quick response. This allows for the priority of important inference results by determining the priority of inference results based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may include, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the inference unit is performed using generative AI. For example, the inference unit can input user emotion data into the generative AI and have the generative AI determine the priority of inference results.

[0110] The inference unit can correct its inference results by referring to geographic information during the inference process. For example, the inference unit can correct the location of a crime by referring to geographic information. It can also correct the probability of a crime occurring by referring to geographic information. For example, the inference unit can correct the time of a crime by referring to geographic information. This improves the accuracy of the inference results by referring to geographic information. Some or all of the above processing in the inference unit is performed using a generative AI. For example, the inference unit can have the generative AI perform the referencing of geographic information and the correction of inference results.

[0111] The inference unit can correct its inference results by referring to past weather data during the inference process. For example, the inference unit can correct the probability of crime occurring during rainy weather by referring to past weather data. It can also correct the probability of crime occurring on snowy days by referring to past weather data. For example, the inference unit can correct the probability of crime occurring during sunny weather by referring to past weather data. In this way, the accuracy of the inference results is improved by referring to past weather data. Some or all of the above processing in the inference unit is performed using a generative AI. For example, the inference unit can have the generative AI perform the referencing of weather data and the correction of inference results.

[0112] The action inference unit can estimate the user's emotions and adjust the display method of the action inference results based on the estimated user emotions. For example, if the user is nervous, the action inference unit can provide a simple and highly visible display method. It can also provide a display method that includes detailed information if the user is relaxed. For example, if the user is in a hurry, the action inference unit can provide a concise display method. This allows for the provision of more appropriate information by adjusting the display method of the action inference results based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may include, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above-described processing in the action inference unit is performed using generative AI. For example, the action inference unit can input user emotion data into the generative AI and have the generative AI adjust the display method of the action inference results.

[0113] The action inference unit can improve its inference accuracy by referring to past police response data during action inference. For example, the action inference unit can refer to past police response data and perform inferences based on similar response patterns. The action inference unit can also refer to past police response data and infer the effectiveness of a response. For example, the action inference unit can refer to past police response data and infer the speed of a response. In this way, the inference accuracy is improved by referring to past police response data. Some or all of the above processing in the action inference unit is performed using generative AI. For example, the action inference unit can have the generative AI perform the referencing of police response data and the improvement of inference accuracy.

[0114] The action inference unit can improve accuracy by combining multiple inference algorithms during action inference. For example, the action inference unit can combine different inference algorithms to infer the effectiveness of a response. It can also combine different inference algorithms to infer the speed of a response. For example, the action inference unit can combine different inference algorithms to infer the appropriateness of a response. This improves inference accuracy by combining multiple inference algorithms. Some or all of the above processing in the action inference unit is performed using generative AI. For example, the action inference unit can have the generative AI execute combinations of multiple inference algorithms.

[0115] The action reasoning unit can estimate the user's emotions and determine the priority of action reasoning results based on the estimated emotions. For example, if the user is tense, the action reasoning unit can provide high-priority action reasoning results. It can also provide normal-priority action reasoning results if the user is relaxed. For example, if the user is excited, the action reasoning unit can provide high-priority action reasoning results for quick response. This allows for the priority of important action reasoning results by determining the priority of action reasoning results based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may include, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above-described processing in the action reasoning unit is performed using generative AI. For example, the action reasoning unit can input user emotion data into the generative AI and have the generative AI determine the priority of action reasoning results.

[0116] The action inference unit can correct its inference results by referring to geographic information during action inference. For example, the action inference unit can refer to geographic information to correct the effectiveness of a response. It can also refer to geographic information to correct the speed of a response. For example, the action inference unit can refer to geographic information to correct the appropriateness of a response. In this way, the accuracy of the action inference results is improved by referring to geographic information. Some or all of the above processing in the action inference unit is performed using a generative AI. For example, the action inference unit can have the generative AI perform the referencing of geographic information and the correction of the inference results.

[0117] The action inference unit can correct its inference results by referring to past weather data during action inference. For example, the action inference unit can refer to past weather data to correct the response effect during rainy weather. It can also refer to past weather data to correct the response effect on snowy days. For example, the action inference unit can refer to past weather data to correct the response effect during sunny weather. In this way, the accuracy of the action inference results is improved by referring to past weather data. Some or all of the above processing in the action inference unit is performed using a generative AI. For example, the action inference unit can have the generative AI perform the referencing of weather data and the correction of the inference results.

[0118] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0119] The surveillance system may also include an audio analysis unit. The audio analysis unit can analyze audio data collected along with the video from the surveillance camera and detect abnormal sounds. For example, the audio analysis unit can detect abnormal sounds such as the sound of breaking glass or screaming. The audio analysis unit can also identify specific audio patterns and detect the risk of crime with high accuracy. For example, the audio analysis unit can analyze specific audio patterns and compare them with past crime records. This allows for the early detection of anomalies that cannot be detected by video data alone by analyzing audio data. Some or all of the above processing in the audio analysis unit may be performed using AI or not. For example, the audio analysis unit can input audio data into an AI and have the AI ​​perform the detection of abnormal sounds.

[0120] The monitoring system may also include an environmental sensor unit. The environmental sensor unit can collect environmental data such as temperature, humidity, and light intensity and provide it to the analysis unit. For example, the environmental sensor unit can detect abnormal temperature and humidity changes and detect crime risks with high accuracy. The environmental sensor unit can also detect changes in light intensity and detect nighttime crime risks early. For example, the environmental sensor unit can analyze sudden changes in light intensity and compare them with past crime records. By analyzing environmental data in this way, anomalies that cannot be detected by video data alone can be detected early. Some or all of the above processing in the environmental sensor unit may be performed using AI or not. For example, the environmental sensor unit can input environmental data into the AI ​​and have the AI ​​perform the detection of abnormal situations.

[0121] The surveillance system may also include an emotion estimation unit. The emotion estimation unit can estimate a person's emotions from surveillance camera footage and provide this information to the analysis unit. For example, the emotion estimation unit can analyze a person's facial expressions and movements in the footage to estimate emotions such as tension or excitement. The emotion estimation unit can also detect the risk of crime with high accuracy based on the estimated emotions. For example, the emotion estimation unit can analyze the movements of a tense person and compare them with past crime records. By analyzing the emotion data, it is possible to detect anomalies that cannot be detected by video data alone at an early stage. Some or all of the above processing in the emotion estimation unit may be performed using AI or not. For example, the emotion estimation unit can input emotion data into AI and have the AI ​​perform the detection of abnormal emotions.

[0122] The surveillance system may further include a behavior prediction unit. The behavior prediction unit can predict a person's next action from surveillance camera footage and provide it to the analysis unit. For example, the behavior prediction unit can analyze a person's movements in the footage and predict the most likely next action. The behavior prediction unit can also detect the risk of crime with high accuracy based on past crime records. For example, the behavior prediction unit can analyze past crime data and identify specific behavior patterns. By analyzing behavioral data, it is possible to detect anomalies that cannot be detected by video data alone at an early stage. Some or all of the above processing in the behavior prediction unit may be performed using AI or not. For example, the behavior prediction unit can input behavioral data into AI and have the AI ​​perform the prediction of the next action.

[0123] The surveillance system may also include an anomaly detection unit. The anomaly detection unit can detect abnormal movements or situations from surveillance camera footage and provide this information to the analysis unit. For example, the anomaly detection unit can analyze the movements of people in the footage and detect abnormal movements. The anomaly detection unit can also detect specific situations (e.g., abnormal movements of a crowd) and detect the risk of crime with high accuracy. For example, the anomaly detection unit can analyze the movements of a crowd in the footage and compare them with past crime records. By analyzing the anomaly data, it is possible to detect anomalies that cannot be detected by video data alone at an early stage. Some or all of the above-described processes in the anomaly detection unit may be performed using AI or not. For example, the anomaly detection unit can input the anomaly data into the AI ​​and have the AI ​​perform the detection of abnormal movements.

[0124] The surveillance system may further include an emotion estimation unit. The emotion estimation unit can estimate a person's emotions from surveillance camera footage and adjust the alert delivery method based on the estimated emotions. For example, the emotion estimation unit analyzes the facial expressions and movements of a person in the footage to estimate emotions such as tension or excitement. The emotion estimation unit can also adjust the timing and content of alerts based on the estimated emotions. For example, the emotion estimation unit can immediately issue an alert to a person who is tense and periodically issue alerts to a person who is relaxed. By analyzing the emotion data, alerts can be issued at a more appropriate time. Some or all of the above processing in the emotion estimation unit may be performed using AI or not. For example, the emotion estimation unit can input emotion data into AI and have the AI ​​perform the adjustment of the alert delivery method.

[0125] The surveillance system may further include an emotion estimation unit. The emotion estimation unit can estimate a person's emotions from surveillance camera footage and adjust the content of the alert based on the estimated emotions. For example, the emotion estimation unit can analyze a person's facial expressions and movements in the footage to estimate emotions such as tension or excitement. The emotion estimation unit can also adjust the level of detail and urgency of the alert based on the estimated emotions. For example, the emotion estimation unit can provide a detailed alert to a tense person and a concise alert to a relaxed person. This allows for more appropriate alerts to be provided by analyzing the emotion data. Some or all of the above processing in the emotion estimation unit may be performed using AI or not. For example, the emotion estimation unit can input emotion data into AI and have the AI ​​adjust the content of the alert.

[0126] The surveillance system may further include an emotion estimation unit. The emotion estimation unit can estimate a person's emotions from surveillance camera footage and adjust the display method of the inference results based on the estimated emotions. For example, the emotion estimation unit analyzes the facial expressions and movements of a person in the footage to estimate emotions such as tension or excitement. The emotion estimation unit can also adjust the display format and content of the inference results based on the estimated emotions. For example, the emotion estimation unit may provide a simple and highly visible display method for a tense person and a display method that includes detailed information for a relaxed person. This allows for the provision of inference results in a more appropriate format by analyzing emotion data. Some or all of the above processing in the emotion estimation unit may be performed using AI or not. For example, the emotion estimation unit can input emotion data into AI and have the AI ​​adjust the display method of the inference results.

[0127] The surveillance system may further include an emotion estimation unit. The emotion estimation unit can estimate a person's emotions from surveillance camera footage and determine the priority of action inference results based on the estimated emotions. For example, the emotion estimation unit analyzes the facial expressions and movements of a person in the footage to estimate emotions such as tension or excitement. The emotion estimation unit can also adjust the priority of action inference results based on the estimated emotions. For example, the emotion estimation unit may provide high-priority action inference results for a tense person and normal-priority action inference results for a relaxed person. This allows for the provision of action inference results with more appropriate priorities by analyzing emotion data. Some or all of the above processing in the emotion estimation unit may be performed using AI or not. For example, the emotion estimation unit can input emotion data into AI and have the AI ​​determine the priority of action inference results.

[0128] The surveillance system may also include an anomaly detection unit. The anomaly detection unit can detect abnormal movements or situations from surveillance camera footage and provide this information to the analysis unit. For example, the anomaly detection unit can analyze the movements of people in the footage and detect abnormal movements. The anomaly detection unit can also detect specific situations (e.g., abnormal movements of a crowd) and detect the risk of crime with high accuracy. For example, the anomaly detection unit can analyze the movements of a crowd in the footage and compare them with past crime records. By analyzing the anomaly data, it is possible to detect anomalies that cannot be detected by video data alone at an early stage. Some or all of the above-described processes in the anomaly detection unit may be performed using AI or not. For example, the anomaly detection unit can input the anomaly data into the AI ​​and have the AI ​​perform the detection of abnormal movements.

[0129] The following briefly describes the processing flow for example form 2.

[0130] Step 1: The collection unit collects video footage from surveillance cameras. For example, the collection unit acquires video data from surveillance cameras in real time. The collection unit can also store past video data and use it for analysis as needed. For example, the collection unit can store the surveillance camera footage in cloud storage and make it accessible to the analysis unit. Furthermore, the collection unit can perform noise reduction and image correction to improve the quality of the video data. Step 2: The analysis unit analyzes the video collected by the collection unit. The analysis unit can, for example, use image recognition algorithms to identify people and objects in the video. The analysis unit can also detect suspicious movements using motion detection technology. For example, the analysis unit can recognize the faces of people in the video and compare them with past criminal records. The analysis unit can also analyze movement patterns in the video and detect abnormal movements. Step 3: The alert unit issues an alert based on the results analyzed by the analysis unit. The alert unit can, for example, issue an audio alert to draw the attention of people in the vicinity. The alert unit can also send alerts to relevant parties via email or app notifications. For example, the alert unit can send an alert to the police or security personnel based on the analysis results. Step 4: The reporting unit notifies the police based on the alert issued by the alert unit. The reporting unit can, for example, automatically send video to the police in real time. The reporting unit can also send the report content to the police as text data. For example, the reporting unit can send detailed report content to the police based on the analysis results.

[0131] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0132] Data generation model 58 is a form of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include text generation AI, image generation AI, and multimodal generation AI. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats from audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each of the above parts is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example.Furthermore, processing performed by AI, including generative AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by AI, including generative AI.

[0133] Furthermore, the processing performed by the data processing system 10 described above is carried out by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but it may also be carried out by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0134] Each of the multiple elements described above, including the collection unit, analysis unit, alert unit, notification unit, inference unit, and action inference unit, is implemented, for example, in at least one of the smart device 14 and the data processing unit 12. For example, the collection unit collects surveillance camera footage using the camera 42 and communication I / F 44 of the smart device 14 and analyzes it using the specific processing unit 290 of the data processing unit 12. The analysis unit is implemented, for example, in the specific processing unit 290 of the data processing unit 12 and analyzes the collected footage. The alert unit is implemented, for example, in the control unit 46A of the smart device 14 and issues an alert based on the analysis results. The notification unit is implemented, for example, in the specific processing unit 290 of the data processing unit 12 and notifies the police. The inference unit is implemented, for example, in the specific processing unit 290 of the data processing unit 12 and uses generated AI to infer the suspect's next action. The action inference unit is implemented, for example, in the specific processing unit 290 of the data processing unit 12 and infers the next action that the police should take. The correspondence between each part and the device or control unit is not limited to the examples described above, and various modifications are possible.

[0135] [Second Embodiment] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0136] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0137] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0138] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0139] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0140] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0141] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0142] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing by the processor 28. The storage 32 stores the specific processing program 56.

[0143] The processor 28 reads a specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0144] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0145] In the smart glasses 214, specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 acting as a control unit 46A according to the specific processing program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0146] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0147] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0148] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0149] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but it may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart glasses 214 or an external device, and the smart glasses 214 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0150] Each of the multiple elements described above, including the collection unit, analysis unit, alert unit, notification unit, inference unit, and action inference unit, is implemented, for example, in at least one of the smart glasses 214 and the data processing unit 12. For example, the collection unit collects images from surveillance cameras using the camera 42 and communication I / F 44 of the smart glasses 214 and analyzes them using the specific processing unit 290 of the data processing unit 12. The analysis unit is implemented, for example, by the specific processing unit 290 of the data processing unit 12 and analyzes the collected images. The alert unit is implemented, for example, by the control unit 46A of the smart glasses 214 and issues an alert based on the analysis results. The notification unit is implemented, for example, by the specific processing unit 290 of the data processing unit 12 and notifies the police. The inference unit is implemented, for example, by the specific processing unit 290 of the data processing unit 12 and uses generated AI to infer the suspect's next action. The action inference unit is implemented, for example, by the specific processing unit 290 of the data processing unit 12 and infers the next action that the police should take. The correspondence between each part and the device or control unit is not limited to the examples described above, and various modifications are possible.

[0151] [Third Embodiment] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0152] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0153] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0154] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0155] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0156] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0157] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0158] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0159] The processor 28 reads a specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0160] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0161] In the headset terminal 314, specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads the specific program 60 from the storage 50 and executes the read specific program 60 on the RAM 48. The specific processing is realized by the processor 46 acting as a control unit 46A according to the specific program 60 executed on the RAM 48. The headset terminal 314 also has a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0162] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0163] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0164] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0165] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset terminal 314, but may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset terminal 314. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the headset terminal 314 or an external device, and the headset terminal 314 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0166] Each of the multiple elements described above, including the collection unit, analysis unit, alert unit, notification unit, inference unit, and action inference unit, is implemented in at least one of the headset terminal 314 and the data processing unit 12. For example, the collection unit collects surveillance camera footage using the camera 42 and communication I / F 44 of the headset terminal 314 and analyzes it using the specific processing unit 290 of the data processing unit 12. The analysis unit is implemented in the specific processing unit 290 of the data processing unit 12 and analyzes the collected footage. The alert unit is implemented in the control unit 46A of the headset terminal 314 and issues an alert based on the analysis results. The notification unit is implemented in the specific processing unit 290 of the data processing unit 12 and notifies the police. The inference unit is implemented in the specific processing unit 290 of the data processing unit 12 and uses generated AI to infer the suspect's next action. The action inference unit is implemented in the specific processing unit 290 of the data processing unit 12 and infers the next action the police should take. The correspondence between each part and the device or control unit is not limited to the examples described above, and various modifications are possible.

[0167] [Fourth Embodiment] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0168] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0169] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0170] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0171] The microphone 238 receives voice signals from the user and accepts instructions from the user. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0172] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS image sensor or CCD image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0173] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0174] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. The robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0175] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0176] The processor 28 reads a specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0177] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0178] In robot 414, specific processing is performed by processor 46. A specific program 60 is stored in storage 50. Processor 46 reads the specific program 60 from storage 50 and executes it on RAM 48. The specific processing is achieved by processor 46 acting as a control unit 46A according to the specific program 60 executed on RAM 48. Robot 414 also has data generation model 58 and emotion identification model 59, similar to those of the robot, and can perform processing similar to that of the specific processing unit 290 using these models.

[0179] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0180] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0181] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0182] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but it may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the robot 414 or an external device, and the robot 414 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0183] Each of the multiple elements described above, including the collection unit, analysis unit, alert unit, notification unit, inference unit, and action inference unit, is implemented, for example, in at least one of the robot 414 and the data processing unit 12. For example, the collection unit collects surveillance camera footage using the camera 42 and communication I / F 44 of the robot 414 and analyzes it using the specific processing unit 290 of the data processing unit 12. The analysis unit is implemented, for example, by the specific processing unit 290 of the data processing unit 12 and analyzes the collected footage. The alert unit is implemented, for example, by the control unit 46A of the robot 414 and issues an alert based on the analysis results. The notification unit is implemented, for example, by the specific processing unit 290 of the data processing unit 12 and notifies the police. The inference unit is implemented, for example, by the specific processing unit 290 of the data processing unit 12 and uses generated AI to infer the suspect's next action. The action inference unit is implemented, for example, by the specific processing unit 290 of the data processing unit 12 and infers the next action that the police should take. The correspondence between each part and the device or control unit is not limited to the examples described above, and various modifications are possible.

[0184] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0185] Figure 9 shows the emotion map 400, in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0186] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0187] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0188] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, and motorcycles, emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated based, for example, on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0189] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0190] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0191] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing method for the specific process may be used, which includes computer 22 and multiple other computers.

[0192] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0193] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[0194] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0195] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0196] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0197] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0198] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0199] Furthermore, although the above-described examples were divided into four embodiments, some or all of these embodiments may be combined. Also, the smart device 14, smart glasses 214, headset terminal 314, and robot 414 are just examples, and they may be combined, or other devices may be used. Also, although the above-described examples were divided into two embodiments, Embodiment 1 and Embodiment 2, these may be combined.

[0200] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and other things that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0201] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0202] (Note 1) The collection unit collects footage from surveillance cameras, An analysis unit analyzes the video collected by the aforementioned collection unit, An alert unit that issues an alert based on the results analyzed by the aforementioned analysis unit, A reporting unit that notifies the police based on the alert issued by the aforementioned alert unit, Equipped with A system characterized by the following features. (Note 2) The generating AI includes an inference unit that infers the suspect's next action based on similar past cases. The system described in Appendix 1, characterized by the features described herein. (Note 3) The generating AI is equipped with an action inference unit that infers the next action the police should take in real time, derived from similar past cases. The system described in Appendix 1, characterized by the features described herein. (Note 4) The aforementioned analysis unit, Identify the physical characteristics and actions of individuals in the video and compare them with past criminal records. The system described in Appendix 1, characterized by the features described herein. (Note 5) The aforementioned analysis unit, It detects actions such as intruders attempting to break glass or suspicious individuals trying to force their way through office gates. The system described in Appendix 1, characterized by the features described herein. (Note 6) The aforementioned reporting unit, Automatically transmits video to the police in real time. The system described in Appendix 1, characterized by the features described herein. (Note 7) The aforementioned collection unit is The system estimates the user's emotions and adjusts the timing of video collection based on those emotions. The system described in Appendix 1, characterized by the features described herein. (Note 8) The aforementioned collection unit is Dynamically adjust the placement and angle of surveillance cameras to collect optimal footage. The system described in Appendix 1, characterized by the features described herein. (Note 9) The aforementioned collection unit is When collecting video footage, the collection range can be changed based on specific time periods or events. The system described in Appendix 1, characterized by the features described herein. (Note 10) The aforementioned collection unit is It estimates the user's emotions and determines the priority of the videos to collect based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 11) The aforementioned collection unit is When collecting video footage, ambient audio data is also collected simultaneously and used for analysis. The system described in Appendix 1, characterized by the features described herein. (Note 12) The aforementioned collection unit is When collecting video footage, data from environmental sensors will be collected in conjunction with it. The system described in Appendix 1, characterized by the features described herein. (Note 13) The aforementioned analysis unit, It estimates the user's emotions and adjusts the accuracy of the analysis based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 14) The aforementioned analysis unit, When analyzing video, integrating footage from multiple cameras improves the accuracy of the analysis. The system described in Appendix 1, characterized by the features described herein. (Note 15) The aforementioned analysis unit, When analyzing video footage, an algorithm is applied to identify specific motion patterns or behavioral sequences. The system described in Appendix 1, characterized by the features described herein. (Note 16) The aforementioned analysis unit, It estimates the user's emotions and adjusts how the analysis results are displayed based on the estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 17) The aforementioned analysis unit, Suspicious sounds are detected by combining video analysis with audio analysis. The system described in Appendix 1, characterized by the features described herein. (Note 18) The aforementioned analysis unit, During video analysis, past weather data is referenced to correct the analysis results. The system described in Appendix 1, characterized by the features described herein. (Note 19) The alert unit is, It estimates the user's emotions and adjusts how alerts are sent based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 20) The alert unit is, When issuing an alert, use multiple communication methods in combination. The system described in Appendix 1, characterized by the features described herein. (Note 21) The alert unit is, When an alert is issued, different notification methods are selected depending on the alert's priority. The system described in Appendix 1, characterized by the features described herein. (Note 22) The alert unit is, It estimates the user's emotions and customizes the content of alerts based on those emotions. The system described in Appendix 1, characterized by the features described herein. (Note 23) The alert unit is, When an alert is issued, geographical information will be added to notify the police and relevant parties. The system described in Appendix 1, characterized by the features described herein. (Note 24) The alert unit is, When an alert is issued, past alert history is referenced to improve the accuracy of the alert. The system described in Appendix 1, characterized by the features described herein. (Note 25) The aforementioned reporting unit, The system estimates the user's emotions and adjusts the content of the report based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 26) The aforementioned reporting unit, When a report is made, the police response is monitored in real time, and additional information is provided as needed. The system described in Appendix 1, characterized by the features described herein. (Note 27) The aforementioned reporting unit, When a report is made, it is automatically routed to the appropriate police department based on the content of the report. The system described in Appendix 1, characterized by the features described herein. (Note 28) The aforementioned reporting unit, The system estimates the user's emotions and prioritizes reports based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 29) The aforementioned reporting unit, When making a report, audio and text data are sent simultaneously along with the video. The system described in Appendix 1, characterized by the features described herein. (Note 30) The aforementioned reporting unit, When making a report, the report content is optimized by referring to past report history. The system described in Appendix 1, characterized by the features described herein. (Note 31) The inference unit, It estimates the user's emotions and adjusts how the inference results are displayed based on the estimated user emotions. The system described in Appendix 2, characterized by the features described herein. (Note 32) The inference unit, During inference, past crime data is referenced to improve inference accuracy. The system described in Appendix 2, characterized by the features described herein. (Note 33) The inference unit, During inference, multiple inference algorithms are combined to improve accuracy. The system described in Appendix 2, characterized by the features described herein. (Note 34) The inference unit, It estimates the user's emotions and determines the priority of the inference results based on the estimated user emotions. The system described in Appendix 2, characterized by the features described herein. (Note 35) The inference unit, During inference, the inference results are corrected by referring to geographic information. The system described in Appendix 2, characterized by the features described herein. (Note 36) The inference unit, During inference, the inference results are corrected by referring to past weather data. The system described in Appendix 2, characterized by the features described herein. (Note 37) The aforementioned action inference unit, It estimates the user's emotions and adjusts how action inference results are displayed based on the estimated user emotions. The system described in Appendix 3, characterized by the features described herein. (Note 38) The aforementioned action inference unit, During action inference, past police response data is referenced to improve inference accuracy. The system described in Appendix 3, characterized by the features described herein. (Note 39) The aforementioned action inference unit, When performing action inference, combine multiple inference algorithms to improve accuracy. The system described in Appendix 3, characterized by the features described herein. (Note 40) The aforementioned action inference unit, It estimates the user's emotions and prioritizes action inference results based on the estimated user emotions. The system described in Appendix 3, characterized by the features described herein. (Note 41) The aforementioned action inference unit, During action inference, refer to geographic information to correct the inference results. The system described in Appendix 3, characterized by the features described herein. (Note 42) The aforementioned action inference unit, During action inference, the inference results are corrected by referring to past weather data. The system described in Appendix 3, characterized by the features described herein. [Explanation of Symbols]

[0203] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots

Claims

1. The collection unit collects footage from surveillance cameras, An analysis unit analyzes the video collected by the aforementioned collection unit, An alert unit that issues an alert based on the results analyzed by the aforementioned analysis unit, A reporting unit that notifies the police based on the alert issued by the aforementioned alert unit, Equipped with A system characterized by the following features.

2. The generating AI includes an inference unit that infers the suspect's next action based on similar past cases. The system according to feature 1.

3. The generating AI is equipped with an action inference unit that infers the next action the police should take in real time, derived from similar past cases. The system according to feature 1.

4. The aforementioned analysis unit, Identify the physical characteristics and actions of individuals in the video and compare them with past criminal records. The system according to feature 1.

5. The aforementioned analysis unit, It detects actions such as intruders attempting to break glass or suspicious individuals trying to force their way through office gates. The system according to feature 1.

6. The aforementioned reporting unit, Automatically transmits video to the police in real time. The system according to feature 1.

7. The aforementioned collection unit is The system estimates the user's emotions and adjusts the timing of video collection based on those emotions. The system according to feature 1.

8. The aforementioned collection unit is Dynamically adjust the placement and angle of surveillance cameras to collect optimal footage. The system according to feature 1.

9. The aforementioned collection unit is When collecting video footage, the collection range can be changed based on specific time periods or events. The system according to feature 1.

10. The aforementioned collection unit is It estimates the user's emotions and determines the priority of the videos to collect based on the estimated user emotions. The system according to feature 1.

Images