KR102813879B1 - System for preventing suicide using artificial intelligence - Google Patents

Abstract

A suicide prevention system using AI is disclosed. A CCTV camera generating CCTV camera footage; an AI suicide prevention server analyzing CCTV camera footage generated from the CCTV camera for suicide signs or suicidal behavior using an AI engine and remotely controlling a suicide prevention/prevention process; and a suicide prevention/prevention notification device outputting a suicide prevention/prevention notification by remote control of the AI suicide prevention server are configured. According to the suicide prevention system using AI described above, the suicide prevention system is configured to quickly and accurately identify suicide signs and suicide attempts by comprehensively analyzing suicide behavior patterns or suicide tools using CCTV cameras or IP cameras, thereby having the effect of preventing suicide attempts and responding quickly in places where others are not watching, such as outdoors or at home. In addition, in such cases, there is the effect of increasing the accuracy of suicide attempts by comprehensively identifying suicide signs through messengers or SNS.

Description

AI-based suicide prevention system {SYSTEM FOR PREVENTING SUICIDE USING ARTIFICIAL INTELLIGENCE}

The present invention relates to a suicide prevention system, and more particularly, to a suicide prevention system using AI.

Recently, the number of single-person households, including elderly people living alone, is increasing.

The suicide rate is also increasing for various reasons, and single-person households are virtually defenseless against suicide.

Even outdoors, because the nature of suicide often involves avoiding the attention of those around you, it is very important to quickly identify and respond to signs of suicide or suicide attempts.

However, since people cannot monitor all suicidal behaviors, methods should be sought to monitor and respond to suicidal behaviors in places where suicides occur frequently or in indoor settings.

In other words, a means must be developed to prevent suicide by capturing suicidal behavior in real time in places where no one else is watching and reporting it to the outside world.

Patent registration publication 10-1874630 Patent registration publication 10-2013692

The purpose of the present invention is to provide a suicide prevention system using AI.

The suicide prevention system using AI according to the purpose of the present invention described above may be configured to include a CCTV camera that generates CCTV camera images; an AI suicide prevention server that analyzes the CCTV camera images generated from the CCTV camera for signs of suicide or suicidal behavior using an AI engine and remotely controls the suicide prevention/prevention process.

Here, the AI suicide prevention server may be configured to further include a suicide prevention/prevention notification device that outputs a suicide prevention/prevention notification by remote control.

And the above suicide prevention/prevention notification may consist of a message that discourages or persuades suicide.

According to the suicide prevention system using AI described above, it is configured to quickly and accurately identify signs of suicide and suicide attempts by comprehensively analyzing suicidal behavior patterns and suicide tools using CCTV cameras or IP cameras, thereby having the effect of preventing suicide attempts and responding quickly in places where others are not watching, such as outdoors or at home.

In addition, in these cases, it is possible to increase the accuracy of suicide attempts by comprehensively identifying signs of suicide through messengers or SNS.

Figure 1 is a block diagram of a suicide prevention system using AI according to one embodiment of the present invention.

The present invention can have various modifications and various embodiments, and specific embodiments are illustrated in the drawings and the specific contents for carrying out the invention are described in detail. However, this is not intended to limit the present invention to specific embodiments, but should be understood to include all modifications, equivalents, or substitutes included in the spirit and technical scope of the present invention. In describing each drawing, similar reference numerals are used for similar components.

The terms first, second, A, B, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are only used to distinguish one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as the second component, and similarly, the second component may also be referred to as the first component. The term and/or includes any combination of a plurality of related described items or any item among a plurality of related described items.

When it is said that a component is "connected" or "connected" to another component, it should be understood that it may be directly connected or connected to that other component, but that there may be other components in between. On the other hand, when it is said that a component is "directly connected" or "directly connected" to another component, it should be understood that there are no other components in between.

The terminology used in this application is only used to describe specific embodiments and is not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, it should be understood that the terms "comprises" or "has" and the like are intended to specify the presence of a feature, number, step, operation, component, part or combination thereof described in the specification, but do not exclude in advance the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts or combinations thereof.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms defined in commonly used dictionaries, such as those defined in common usage, should be interpreted as having a meaning consistent with the meaning they have in the context of the relevant art, and will not be interpreted in an idealized or overly formal sense unless expressly defined in this application.

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the attached drawings.

Figure 1 is a block diagram of a suicide prevention system using AI according to one embodiment of the present invention.

Referring to FIG. 1, a suicide prevention system using AI according to one embodiment of the present invention may be configured to include a CCTV camera (100), an AI suicide prevention server (200), a suicide prevention/prevention notification device (300), an IP camera (400), a Wi-Fi AP (500), and a user terminal (600).

Below, the detailed configuration is explained.

CCTV cameras (100) are configured to be installed anywhere outdoors, such as on bridges, cliffs, railings, and around parks or buildings.

The CCTV camera (100) can be configured to generate CCTV camera footage.

The AI suicide prevention server (200) can be configured to analyze CCTV camera footage generated from a CCTV camera (100) for signs of suicide or suicidal behavior using an AI engine and remotely control the suicide prevention/prevention process. Here, whether there are signs of suicide or suicidal behavior can be determined by the behavioral pattern of the subject of surveillance in the CCTV camera footage or suicide-related tools found in the CCTV camera footage.

The AI suicide prevention server (200) may be configured to include a server-side CCTV video reception module (201), a server-side suicide behavior pattern AI analysis module (202), a server-side suicide tool AI analysis module (203), a server-side suicide sign/behavior AI judgment module (204), and a server-side suicide prevention/prevention notification control module (205).

Below, the detailed configuration is explained.

The server-side CCTV video reception module (201) can be configured to receive CCTV camera video in real time from a CCTV camera (100).

The server-side suicide behavior pattern AI analysis module (202) can be configured to analyze the CCTV camera video received in real time from the server-side CCTV video reception module (201) using an AI engine to analyze the suicidal behavior pattern of the surveillance subject.

Suicidal behavior patterns may include, for example, attempting to jump off a dangerous area of a cliff, attempting to jump over a railing on a bridge, or attempting to climb onto the railing of a bridge.

The server-side suicide tool AI analysis module (203) can be configured to analyze CCTV camera images received in real time from the server-side CCTV image reception module (201) using an AI engine to analyze the suicide tool of the surveillance target.

Typical suicide tools include ropes, strings, belts, etc. used for tying a neck, and ropes or strings that can be tied to a tree or structure and knotted to form a ring.

The server-side suicide sign/behavior AI judgment module (204) can be configured to use an AI engine to determine the suicide sign or suicidal behavior of a surveillance subject based on the analysis results of the server-side suicide behavior pattern AI analysis module (202) and the server-side suicide tool AI analysis module (203).

The server-side suicide prevention/prevention notification control module (205) can be configured to remotely control the suicide prevention/prevention notification device (300) to output a suicide prevention/prevention notification based on the judgment of the server-side suicide sign/behavior AI judgment module (204).

The suicide prevention/prevention notification device (300) can be configured to output a suicide prevention/prevention notification by remote control of the AI suicide prevention server (200).

A suicide prevention/prevention notification device (300) may be configured to be attached to a CCTV camera (100). It may be attached to a CCTV camera (100) in an area where suicides frequently occur.

The suicide prevention/prevention notification device (300) may be configured to output, through a speaker, messages such as “Please step back from the cliff,” “Please step back from the bridge railing,” or other persuasive messages to prevent suicide.

In addition, the suicide prevention/prevention notification device (300) can be configured to transmit a suicide prevention/prevention notification to a nearby terminal (10) connected to a nearby public Wi-Fi AP by remote control of the AI suicide prevention server (200).

Here, suicide prevention/prevention alerts are intended to notify people around them of a person attempting suicide or a person under surveillance with signs of suicide.

This is to prevent suicide by transmitting a suicide prevention/prevention notification to a surrounding terminal (10) connected to a public Wi-Fi AP surrounding a CCTV camera (100). At this time, the suicide prevention/prevention notification may include the location information of the CCTV camera (100) so that people around can easily identify its location.

The above descriptions relate to a configuration for capturing outdoor suicide attempts or signs of suicide using a CCTV camera (100), and the following description relates to a configuration for capturing indoor suicide attempts or signs of suicide using an indoor IP camera (400).

The IP camera (400) is configured to be installed indoors, such as in a home or office, and can be configured to generate indoor IP camera images.

The WiFi AP (500) can be configured to transmit IP camera images generated from the IP camera (400) to an indoor user terminal (600) in real time.

And the Wi-Fi AP (500) can be configured to receive IP camera video clips generated by the user terminal (600) in real time from the user terminal (600) and transmit them in real time to a designated terminal (20) through the network. Here, the IP camera video clips can be composed of videos showing the user's suicidal behavior pattern or signs of suicide. At this time, the designated terminal (20) is a terminal pre-designated in the application of the user terminal (600), and can be a terminal of a family member, guardian, or close acquaintance.

The user terminal (600) can be configured to receive IP camera images in real time from a Wi-Fi AP (500), analyze signs of suicide or suicidal behavior using an AI engine, and perform a suicide prevention/prevention process.

The user terminal (600) may be configured to include an IP camera video real-time collection module (601), an IP camera video storage module (602), a suicide behavior pattern AI analysis module (603), a suicide tool AI analysis module (604), a suicide-related information automatic collection module (605), a suicide sign/action AI judgment module (606), an IP camera video clip generation module (607), an IP camera video clip automatic transmission module (608), a real-time location information calculation module (609), a suicide prevention/prevention control module (610), and an emergency reporting module (611).

Below, the detailed configuration is explained.

The IP camera video real-time collection module (601) can be configured to collect IP camera video in real time from the IP camera (400).

The IP camera image storage module (602) can be configured to store IP camera images collected in real time from the IP camera image real-time collection module (601).

The suicide behavior pattern AI analysis module (603) can be configured to analyze a user's suicidal behavior pattern in real time from IP camera images stored in the IP camera image storage module (602) using an AI engine.

Here, suicidal behavior patterns can appear in various ways, such as jumping off an apartment balcony, tying a leash, cutting one's wrists with a knife, overdosing on sleeping pills and collapsing, cutting one's wrists in a bathtub and staying in it for a long time until the water turns blood red, or drinking excessively along with these behaviors.

Suicidal behavior patterns are determined through image pattern analysis, and suicidal behavior patterns can be configured to be inferred by analyzing how the user's behavior patterns appear as a foreground compared to the background.

The suicide tool AI analysis module (604) can be configured to analyze the suicide tool in real time using an AI engine based on IP camera images stored in the IP camera image storage module (602).

Here, suicide tools may include a rope or string for tying a leash, a belt, a tie, and also a chair, sleeping pills, alcohol, a knife, etc. that are used together with these.

The AI analysis of the suicide behavior pattern AI analysis module (603) and the suicide tool AI analysis module (604) can be performed in a complex manner, and the suicide tool and the user's behavior pattern can be analyzed in a complex manner as a foreground. The user's behavior pattern related to the suicide tool must be judged to be for suicide.

The oral medication monitoring module (not shown) may be configured to monitor in real time whether oral medication, particularly sleeping pills, have been taken. When sleeping pills with an RFID tag attached thereto are taken, the RFID tag protected by the protective film may be configured to output an RF signal as it is removed by stomach acid. At this time, the oral medication monitoring module (not shown) may be configured to monitor in real time whether an overdose of sleeping pills has occurred and how many pills have been taken. Each RFID tag is configured to emit an RF signal composed of a unique frequency for each pill, and the oral medication monitoring module (not shown) may identify each RF signal to identify the number of oral medications taken. Based on the monitoring results, the suicide tool AI analysis module (604) may analyze that the sleeping pill intake is an overdose as a suicide tool, not a general intake.

The suicide-related information automatic collection module (605) can be configured to automatically collect suicide-related information in real time from the user's SNS application (30) and messenger application (31).

Here, suicide-related information refers to information about whether there are posts suggesting suicide or posts expressing depressive feelings by analyzing recent SNS posts or messenger conversations.

Meanwhile, the suicide-related information automatic collection module (605) can be configured to additionally automatically collect current physical condition information and past physical condition information in real time from the user's health application (32).

Here, the current physical condition information can be the current pulse, blood pressure, etc., and the past physical condition information can be the past average of the pulse, blood pressure, etc. The current physical condition information can be configured to be measured in real time through a smart watch (40), and the suicide-related information automatic collection module (605) can control the user's smart watch (40) to cause the smart watch (40) to measure the pulse and blood pressure in real time, and the suicide-related information automatic collection module (605) can be configured to automatically collect the current pulse and blood pressure obtained from the health application (32).

The suicide sign/behavior AI judgment module (606) can be configured to capture and judge the user's suicide signs and suicidal behavior in real time based on the suicidal behavior pattern analyzed in real time by the suicide behavior pattern AI analysis module (603), the suicidal behavior pattern analyzed in real time by the suicide tool AI analysis module (604), the suicide-related information automatically collected in real time by the suicide-related information automatic collection module (605), and the current physical condition information using the AI engine.

In other words, it is possible to comprehensively analyze whether there is a correlation between suicidal behavior patterns and related suicide tools, and if such a correlation is acknowledged, it is possible to further analyze suicidal hints or signs through the user's SNS posts or messenger posts, and infer the current psychological state by analyzing past and current pulse and blood pressure, etc., and further increase the accuracy of AI analysis through the current physical state during the suicidal act.

On the other hand, if the current pulse and blood pressure are different from the past pulse and blood pressure, the suicidal behavior pattern and suicide tool can be combined and used for analysis. For example, the pulse and blood pressure can be much higher than usual when jumping off a railing or tying a leash, and the pulse and blood pressure can be significantly higher in the case of an overdose of sleeping pills. The suicidal behavior pattern, suicide tool, and current physical condition information are all data that can be combined and judged.

The IP camera video clip generation module (607) can be configured to automatically generate an IP camera video clip from the IP camera video stored in the IP camera video storage module (602) when a suicidal behavior pattern is analyzed in real time by the suicide behavior pattern AI analysis module (603) or a suicide tool is analyzed in real time by the suicide tool AI analysis module (604). That is, a video portion in which a suicidal behavior pattern appears or an object recognized as a suicide tool appears can be generated as an IP camera video clip.

The IP camera video clip automatic transmission module (608) can be configured to automatically transmit the IP camera video clip automatically generated by the IP camera video clip generation module (607) to a designated terminal (20) in real time via a WiFi AP (500).

Here, the designated terminal (20) can be a terminal of a family member, close friend, guardian, etc. A family member, guardian, etc. can see this and dissuade the suicide or take prompt follow-up measures.

The real-time location information generation module (609) can be configured to generate real-time location information of a user terminal (600).

The suicide prevention/prevention control module (610) controls the SNS application (30) and messenger application (31) to automatically send a suicide sign/behavior notification message to the user's acquaintances when the suicide sign/behavior AI judgment module (606) determines that the user has signs of suicide or suicidal behavior, and can control the voice call application (33) to automatically connect a call to the user's acquaintances.

That is, the user's suicidal potential can be notified to acquaintances in real time, allowing acquaintances to dissuade or assist in suicide.

The emergency reporting module (611) may be configured to perform an emergency report to the police station or 119 rescue team if the suicide sign/behavior AI judgment module (606) determines that the user has signs of suicide or suicidal behavior, and to transmit IP camera video automatically generated by the IP camera video clip generation module (607) to the police station or 119 rescue team in real time.

In addition, in the case of the suicide prevention/prevention notification device (300), a speaker that outputs a siren sound while outputting a suicide prevention/prevention notification and a smoke generating unit (not shown) that generates colored smoke that can identify a location can be further included. The location can be easily identified and found by those around the device to prevent a suicide attempt.

Meanwhile, if clear signs of a suicide attempt, such as a collar with a circular ring formed indoors, are detected, the suicide prevention/sedation control module (610) can remotely control a sleeping/sedating gas spraying device (not shown) installed in a place where the suicide attempt is to be prevented using the IP camera (400) to spray sleeping/sedating gas toward the user. In this case, the suicide attempt can be prevented by quickly causing the user to enter a state of sleep or calming down before the act of tying the collar, etc. In the case of a user whose suicide attempt has been prevented by the system of the present invention or whose suicide signs are frequently detected, accidental suicide or additional suicide attempts can be prevented by installing such a sleeping/sedating gas spraying device (not shown).

On the other hand, if there is a history of suicide signs or suicidal behavior determined by the suicide signs/behavior AI judgment module (606), the history may be stored in the suicide signs/behavior history database (not shown) of the user terminal (600). In addition, the suicide risk area confirmation module (not shown) of the user terminal (600) may be configured to check whether the real-time location of the real-time location information calculation module (609) is in an area where suicides frequently occur, and if it is confirmed that the user is in an area where suicides frequently occur, the suicide risk area appearance notification module (not shown) of the user terminal (600) may be configured to transmit a suicide risk area appearance notification to a designated terminal (20), a police station, or a 119 rescue team.

In addition, if the above history is recorded in the suicide symptom/action history database (not shown), the deactivation/deletion prohibition module (not shown) of the user terminal (600) can be set to deactivate each component of the user terminal (600) or prevent the deletion of an application in which the corresponding function is implemented.

Although the present invention has been described with reference to the above embodiments, it will be understood by those skilled in the art that various modifications and changes can be made to the present invention without departing from the spirit and scope of the present invention as set forth in the claims below.

100: CCTV Camera
200: AI Suicide Prevention Server
201: Server-side CCTV video reception module
202: Server-side suicide behavior pattern AI analysis module
203: Server-side suicide tool AI analysis module
204: Server-side suicide signs/behavior AI judgment module
205: Server-side suicide prevention/prevention notification control module
300: Suicide Prevention/Prevention Alert Device
400: IP Camera
500: WiFi AP
600: User Terminal
601: IP camera video real-time collection module
602: IP camera video storage module
603: Suicidal Behavior Pattern AI Analysis Module
604: Suicide Tool AI Analysis Module
605: Automatic Suicide-Related Information Collection Module
606: Suicide Signs/Behavior AI Decision Module
607: IP Camera Video Clip Generation Module
608: IP camera video clip automatic transmission module
609: Real-time location information generation module
610: Suicide Prevention/Prevention Control Module
611: Emergency Report Module

Claims (3)

CCTV camera that produces CCTV camera footage;
An AI suicide prevention server that analyzes CCTV camera footage generated from the above CCTV cameras using an AI engine for signs of suicide or suicidal behavior and remotely controls the suicide prevention/prevention process;
A suicide prevention/prevention notification device that outputs a suicide prevention/prevention notification by remote control of the above AI suicide prevention server;
IP camera that produces indoor IP camera footage;
A WiFi AP that transmits IP camera images generated from the above IP camera in real time, receives IP camera image clips in real time, and transmits them in real time to a designated terminal through a network;
A user terminal that receives IP camera images in real time from the above Wi-Fi AP, analyzes signs of suicide or suicidal behavior using an AI engine, and performs a suicide prevention/prevention process;
A suicide prevention/prevention notification device that transmits a suicide prevention/prevention notification including the location information of the IP camera to a nearby terminal connected to a public Wi-Fi AP by remote control of the AI suicide prevention server,
The above AI suicide prevention server is,
Automatically collects suicide-related information in real time from the user's SNS application and messenger application, and automatically collects current and past physical condition information in real time from the user's health application.
The above AI suicide prevention server is,
It is configured to analyze the user's suicide suggestions or signs using the real-time automatically collected suicide-related information,
The above AI suicide prevention server is,
It is configured to determine the possibility of suicide by comparing and analyzing the current physical condition information and past physical condition information automatically collected in real time using an AI algorithm.
The above AI suicide prevention server is,
A suicide prevention system using AI, characterized in that it is configured to measure pulse and blood pressure in real time as the current physical condition information through a smart watch, and is configured to compare and analyze the pulse and blood pressure measured in real time with the past average values of pulse and blood pressure to determine the possibility of suicide.
In the first paragraph, the AI suicide prevention server,
A server-side CCTV video reception module that receives CCTV camera video in real time from the above CCTV camera;
A server-side suicide behavior pattern AI analysis module that analyzes CCTV camera images received in real time from the server-side CCTV image reception module using an AI engine to analyze the suicidal behavior pattern of the subject of surveillance;
A server-side suicide tool AI analysis module that analyzes CCTV camera images received in real time from the server-side CCTV image reception module using an AI engine to analyze the suicide tool of the surveillance subject;
A server-side suicide sign/behavior AI judgment module that uses an AI engine to determine signs of suicide or suicidal behavior of the subject of surveillance based on the analysis results of the server-side suicide behavior pattern AI analysis module and the server-side suicide tool AI analysis module;
A suicide prevention system using AI, characterized in that it comprises a server-side suicide prevention/prevention notification control module that remotely controls suicide prevention/prevention notification based on the judgment of the server-side suicide sign/behavior AI judgment module.
In the second paragraph, the user terminal,
An IP camera video real-time collection module that collects IP camera video in real time from the above IP camera;
An IP camera video storage module in which IP camera video collected in real time from the above IP camera video real-time collection module is stored;
A suicide behavior pattern AI analysis module that analyzes a user's suicidal behavior pattern in real time from IP camera images stored in the IP camera image storage module using an AI engine;
A suicide tool AI analysis module that analyzes a suicide tool in real time using an AI engine based on IP camera images stored in the above IP camera image storage module;
A suicide-related information automatic collection module that automatically collects suicide-related information in real time from the user's SNS application and messenger application, and automatically collects current and past physical condition information in real time from the user's health application;
A suicide sign/behavior AI judgment module that captures and judges in real time the user's suicide signs and suicidal behaviors based on the suicidal behavior patterns analyzed in real time by the AI engine in the suicide behavior pattern AI analysis module, the suicide tools analyzed in real time by the suicide tool AI analysis module, the suicide-related information automatically collected in real time by the suicide-related information automatic collection module, and the current physical condition information;
An IP camera video clip generation module that automatically generates an IP camera video clip from an IP camera video stored in the IP camera video storage module when a suicidal behavior pattern is analyzed in real time in the above-described suicide behavior pattern AI analysis module or a suicide tool is analyzed in real time in the above-described suicide tool AI analysis module;
An IP camera video clip automatic transmission module that automatically transmits an IP camera video clip automatically generated by the above IP camera video clip generation module to a designated terminal in real time through the above Wi-Fi AP;
A real-time location information calculation module that calculates real-time location information of the user terminal;
A suicide prevention/prevention control module that controls the SNS application and the messenger application to automatically send a suicide sign/behavior notification message to the user's acquaintances when the suicide sign/behavior AI judgment module determines that the user has suicide signs or suicidal behavior, and controls the voice call application to automatically connect a call to the user's acquaintances;
A suicide prevention system using AI, characterized in that it comprises an emergency reporting module that performs an emergency report to the police station or 119 rescue team when the suicide sign/behavior AI judgment module determines that the user has signs of suicide or suicidal behavior, and transmits the IP camera video automatically generated by the IP camera video clip generation module to the police station or 119 rescue team in real time.