JP2019152943A - Degree-of-risk detection device, degree-of-risk detection method, and degree-of-risk detection program - Google Patents

Abstract

To provide a crime prevention technique for reducing a workload of a security guard or the like.SOLUTION: An event detection section 23 detects an event that has occurred in an object area 6. An environment determination section 24 determines an environment of an object area 6. A state determination section 25 determines a state of an event that has occurred in an object area 6. A degree-of-risk estimation section 27 estimates a degree of risk with the use of an environment of an object area 6 determined by the environment determination section 24 and a state of an event that has occurred in an object area 6 determined by the state determination section 25 for each kind of risk elements corresponding to an event that has occurred in an object area 6 detected by the event detection section 23.SELECTED DRAWING: Figure 2

Description

  The present invention relates to crime prevention technology.

  Conventionally, a captured image of a target area captured by an installed security camera (hereinafter simply referred to as a camera) is processed to detect a person such as a suspicious person (hereinafter simply referred to as a suspicious person) who has taken a specific action. It is carried out. The target area is a place where an unspecified number of people gather, for example, a station, a shopping center, or a downtown area. In this system, the frame image captured by the camera is processed, the behavior of the person being captured is analyzed, and it is determined whether the behavior is a peculiar behavior. In addition, when a person who takes a peculiar behavior related to fraud is detected, the fact is reported to a security guard or the like.

  For example, Patent Literature 1 describes a configuration for determining the behavior of a subject based on the posture estimated for the subject being imaged and the environment at the location where the subject is located.

Japanese Patent No. 5559335

  However, the conventional techniques described in Patent Document 1 and the like are based on related parties (including parties) of events (such as snatching, fighting, wandering, etc.) caused by the behavior of the target person, and third parties in the vicinity. We did not judge the magnitude of the damage (physical damage, mental damage, financial damage, etc.). Therefore, security guards had to determine the magnitude of damage to the parties involved in the event that occurred and the third parties in the vicinity, and act accordingly. For this reason, the work burden of security guards was large.

  An object of the present invention is to provide a crime prevention technique capable of reducing the work burden of a guard or the like.

  The risk detection apparatus according to the present invention is configured as follows to achieve the above object.

  The event detection unit detects an event that has occurred in the target area. The event referred to here is an event in the target area, such as snatching, fighting, wandering, and the like. The event detection unit detects an event that has occurred in the target area by processing the image of the target area captured by the camera or the sound of the target area collected by the microphone.

  The environment determination unit determines the environment of the target area. The environment of the target area mentioned here is the situation of the target area, and includes not only the weather that is dark, bright, rainy, and sunny, but also that that is related to places such as stairs and underground.

  The status determination unit determines the status of an event that has occurred in the target area detected by the event detection unit. The situation of the event mentioned here is only the parties related to the event that occurred (including parties), and there are no third parties in the vicinity, many third parties are in the vicinity, etc. Shows the situation.

  The state determination unit determines the state of the subject related to the event that occurred in the target area detected by the event detection unit. The state of the subject here refers to a state in which the subject has a knife, the subject has fallen, or has fallen for a predetermined time or more. Target people include not only parties to the event that occurred, but also third parties in the vicinity.

  The risk detection apparatus according to the present invention includes at least two of the above-described three determination units: the environment determination unit, the situation detection unit, and the state determination unit. In other words, the risk detection device of the present invention determines at least two of the above-described target area environment, event status, and target person status.

  The risk level estimation unit estimates the risk level for each type of risk element corresponding to the event that has occurred in the target area, using the determined environment of the target area, the situation of the event, or the state of the target person. The risk factors mentioned here are physical damage, mental damage, financial damage, etc. caused by the event that occurred this time.

  As described above, in this configuration, the risk level is estimated for each risk element type using two or more of the environment of the target area, the situation of the event, and the state of the target person. Therefore, it is not necessary for the security guard or the like to determine the magnitude of damage to the parties involved in the event that occurred and the third parties in the vicinity. For this reason, the work burden of a security guard etc. can be reduced.

  Further, the risk detection device includes a notification destination determination unit that determines a notification destination for notifying the occurrence of an event in the target area according to the risk level for each risk element type by the risk estimation unit, and a notification destination determination unit And a notification unit that notifies the occurrence of an event in the target area to the notification destination determined. With this configuration, it is possible to promptly notify the appropriate organizations such as the police station and fire department about the events that have occurred. Expansion of damage (physical damage, mental damage, financial damage, etc.) can be suppressed.

  According to this invention, it is possible to reduce the work burden on security guards and the like.

It is the schematic which shows the crime prevention system to which the danger level detection apparatus concerning this example is applied. It is a block diagram which shows the structure of the principal part of the danger detection apparatus concerning this example. It is a flowchart which shows operation | movement of the danger level detection apparatus concerning this example.

  Embodiments of the present invention will be described below.

<1. Application example>
FIG. 1 is a schematic diagram showing a crime prevention system to which a risk detection device according to the present invention is applied. A camera 2 and a microphone 3 are connected to the risk detection device 1 of this example. The risk detection device 1 detects an event that has occurred in the target area 6 and estimates the risk level for each type of risk element corresponding to the event. The target area 6 is a space that suppresses the occurrence of crimes and the like assigned to the risk detection device 1. The camera 2 outputs an image obtained by capturing the target area 6 to the risk detection device 1. The camera 2 is a general video camera, and its frame rate is about 10 to 30 frames / sec. The microphone 3 collects environmental sounds (speaking voice, music, noise, etc.) in the target area 6 and outputs the collected sounds to the risk detection device 1.

  The danger level detection device 1 processes the frame image of the target area 6 input from the camera 2 and the environmental sound of the target area 6 input from the microphone 3 to detect an event occurring in the target area 6. The event referred to here is an event in the target area, such as snatching, fighting, wandering, and the like. Here, snatching, fighting, wandering, etc. are called event types.

  When the risk detection device 1 detects an event that has occurred in the target area 6, the environment of the target area 6, the status of the event that has occurred in the target area 6 that has been detected this time, and the event that has occurred in the target area 6 that has been detected this time Determine the state of the subject. The environment of the target area here refers to the situation of the target area, and includes not only those that are affected by weather such as dark, bright, rainy, sunny, but also those that are related to attributes of places such as stairs, up escalators, down escalators, etc. It is out. Here, dark, bright, rainy, sunny, stairs, ascending escalator, descending escalator, and the like are referred to as environment types of the target area 6. Further, the situation of an event referred to here is a situation in which only a party related to the event that occurred (including a party), a third party is not in the vicinity, and a large number of third parties are in the vicinity. Here, a situation where the third party is not in the vicinity, a large number of third parties are in the vicinity, and the like is referred to as an event situation type. Further, the state of the subject referred to here indicates a state in which the subject has a blade, the subject has fallen, or has fallen for a predetermined time or more. Target people include not only parties to the event that occurred, but also third parties in the vicinity. Here, the subject has a knife, the subject has fallen, or has fallen for a predetermined time or more, etc., is referred to as the type of the subject's state.

  The danger level detection device 1 estimates the danger level for each type of danger element corresponding to the event that has occurred in the target area 6. The risk factors referred to here are physical damage, mental damage, monetary damage, etc. caused to the target person by the event that occurred in the target area 6 this time. Here, physical damage, mental damage, financial damage, etc. are referred to as types of risk factors. The estimation of the degree of risk is comprehensively estimated using the environment of the target area 6 determined for the event detected this time, the situation of the event that has occurred this time, and the state of the target person related to the event that has occurred this time. Accordingly, it is not necessary for the security guard or the like to judge by himself the type and magnitude of the damage to the parties involved in the event that occurred and the third parties around it. For this reason, the work burden of a security guard etc. can be reduced.

<2. Configuration example>
FIG. 2 is a block diagram showing a configuration of a main part of the risk detection device according to this example. The danger level detection device 1 includes a control unit 11, a storage unit 12, an image input unit 13, a voice input unit 14, a communication unit 15, and an output unit 16.

  The control unit 11 includes an image processing unit 21, an audio processing unit 22, an event detection unit 23, an environment determination unit 24, a situation determination unit 25, a state determination unit 26, and a risk level estimation unit 27. Details of the image processing unit 21, the sound processing unit 22, the event detection unit 23, the environment determination unit 24, the situation determination unit 25, the state determination unit 26, and the risk estimation unit 27 will be described later.

The storage unit 12 stores estimation information 31, event history information 32, and notification destination information 33. The estimation information 31 is information used to estimate the degree of danger for each type of danger element for an event. For example, the estimation information 31 is information classified by event types such as snatching, fighting, and wandering. The estimation information 31 is information classified according to the types of risk factors such as physical damage, mental damage, and monetary damage for each event type. Further, the estimation information 31 is information classified by combination patterns of the type of environment of the target area 6, the type of situation of the event, and the type of state of the target person for each type of risk element. This combination pattern includes the type of environment in the target area 6 (dark, bright, rainy, clear, stairs, up escalator, down escalator, etc.) and the type of event situation (no third person is in the vicinity) For example, the person is in the vicinity) and the type of state of the subject (the subject has a knife, the subject has fallen, or has fallen for a predetermined time or more), for example,
(1) A pattern that the environment of the target area 6 is dark, rainy, staircase, the situation of the event is a situation where a third party is not in the vicinity, and the state of the target person has a knife,
(2) A pattern in which the environment of the target area 6 is dark, rainy, staircase, the situation of the event is a situation where many third parties are in the vicinity, and the state of the target person is collapsed,
(3) A pattern in which the environment of the target area 6 is bright and clear, the situation of the event is that there is no third person in the vicinity, and the state of the target person has a knife.

  The estimation information 31 is information in which the degree of risk is associated with each combination pattern of the type of environment of the target area 6, the type of event status, and the type of state of the target person.

  That is, the estimation information 31 is information in which the risk level is associated with each combination of the five items of the event, the risk factor, the environment of the target area 6, the status of the event, and the status of the target person.

  The event history information 32 is information in which, for each event that occurs in the target area 6, the environment of the target area related to the event, the status of the event, the state of the target person, the risk level of various risk factors, and the like are associated with each other. The risk level of each risk factor in the event history information 32 may be estimated or determined after the fact.

  The notification destination information 33 is information that associates the contact information (the police station, the fire department, the security room, etc.) that notifies the occurrence of the event and the degree of danger that is determined as requiring notification for each risk element type. is there.

  The storage unit 12 may be a non-volatile memory (EPROM, EEPROM, etc.) or an auxiliary storage device such as a hard disk drive or a solid state drive. The storage unit 12 may be configured with a plurality of storage media.

  The camera 2 is connected to the image input unit 13. The image input unit 13 receives a frame image of the target area 6 captured by the camera 2. Further, the image input unit 13 has an image memory (not shown) that temporarily stores the frame image of the target area 6 input from the camera 2.

  The microphone 3 is connected to the voice input unit 14. The sound input unit 14 receives the environmental sound of the target area 6 collected by the microphone 3.

  The communication unit 15 notifies the occurrence of the event to the police station, the fire station, the security room, and the like via the network 5.

  A display, a speaker, and the like are connected to the output unit 16. The output unit 16 displays a moving image of the target area 6 imaged by the camera 2 on a connected display, and outputs a sound for notifying a detected event from a connected speaker.

  Next, the image processing unit 21, the sound processing unit 22, the event detection unit 23, the environment determination unit 24, the situation determination unit 25, the state determination unit 26, and the risk estimation unit 27 included in the control unit 11 will be described.

  The image processing unit 21 processes the frame image of the target area 6 captured by the camera 2 and detects a moving body (a person, a vehicle, etc.) and an abandoned object in the target area 6. In addition, the image processing unit 21 detects the detected posture of the person and tracks the moving object in the target area 6. The image processing unit 21 detects the type and size of the detected left object. The above processing of the image processing unit 21 can be realized by using a technique related to object detection by a known background difference method or the like, a technique related to object attribute determination (type, posture, etc.) by pattern matching or the like.

  The voice processing unit 22 performs voice recognition processing or the like on the environmental sound collected by the microphone, and detects angry voices, moans, abnormal sounds (metal sounds, firing sounds, etc.) and the like.

  Based on the detection result of the image processing unit 21 and the detection result of the sound processing unit 22, the event detection unit 23 determines whether an event such as snatching, fighting, wandering, and the like has occurred in the target area 6. If so, the type of the event is determined.

  The environment determination unit 24 determines the status of the target area 6 based on the detection result of the image processing unit 21 and the detection result of the sound processing unit 22. The environment determination unit 24 is identified from the detection results of the image processing unit 21 and the sound processing unit 22, and the brightness of the place where the event in the target area 6 occurs and the attribute of the place (stair, up escalator, Etc.).

  The status determination unit 25 determines the status of the event detected this time based on the detection result of the image processing unit 21 and the detection result of the audio processing unit 22. The situation determination unit 25 is based on an object such as a person detected around the place where the event in the target area 6 is identified, which is specified from the detection results of the image processing unit 21 and the audio processing unit 22. It is determined that the three parties are not in the vicinity or that many third parties are in the vicinity.

  The state determination unit 26 determines the state of the subject based on the detection result of the image processing unit 21 and the detection result of the sound processing unit 22. The state determination unit 26 is identified from the detection results of the image processing unit 21 and the audio processing unit 22, and the behavior, posture, etc. of an object such as a person detected around the place where the event in the target area 6 occurs. Based on the above, it is determined whether the subject has a knife, the subject has fallen, or has fallen for a predetermined time or more.

  For the event types determined by the event detection unit 23, the risk level estimation unit 27 determines, for each risk element type, the environment of the target area 6 determined by the environment determination unit 24, the event status determined by the status determination unit 25, And the information 31 for estimation memorize | stored in the memory | storage part 12 is searched with the state of the subject determined by the state determination unit 26, and the risk level is estimated (acquired).

  The control unit 11 of the risk detection device 1 is constituted by a hardware CPU, a memory, and other electronic circuits. When the hardware CPU executes the risk detection program according to the present invention, the image processing unit 21, the sound processing unit 22, the event detection unit 23, the environment determination unit 24, the situation determination unit 25, the state determination unit 26, and It operates as the risk estimation unit 27. Further, the memory has an area for developing the risk detection program according to the present invention and an area for temporarily storing data generated when the risk detection program is executed. The control unit 11 may be an LSI in which a hardware CPU, a memory, and the like are integrated. The hardware CPU is a computer that executes the risk detection method according to the present invention.

<3. Example of operation>
FIG. 3 is a flowchart showing the operation of the risk detection device according to this example. In the risk detection device 1 according to this example, the event detection unit 23 repeatedly detects whether an event such as snatching, fighting, wandering, etc. has occurred in the target area 6 (s1). In the risk detection device 1, the image processing unit 21 processes a frame image of the target area 6 captured by the camera 2, and detects a moving object (a person, a vehicle, etc.) and an abandoned object in the target area 6. Yes. In addition, the image processing unit 21 detects the detected posture of the person and tracks the moving object in the target area 6. Further, the image processing unit 21 detects the type and size of the detected abandoned object. In addition, the voice processing unit 22 performs voice recognition processing or the like on the environmental sound collected by the microphone, and detects angry voices, moans, abnormal sounds (metal sounds, firing sounds, etc.) and the like. The event detection unit 23 detects whether an event such as snatching, fighting, or wandering has occurred in the target area 6 based on the detection result of the image processing unit 21 and the detection result of the sound processing unit 22. When detecting that an event has occurred in the target area 6 at s1, the event detection unit 23 determines the type of the event (s2).

  In addition, when the event detection unit 23 detects that an event has occurred in the target area 6 at s1 based on the detection result of the image processing unit 21 and the detection result of the audio processing unit 22, the environment determination unit 24 The status of area 6 is determined (s3). The environment determination unit 24 is identified from the detection results of the image processing unit 21 and the sound processing unit 22, and the brightness of the place where the event in the target area 6 occurs and the attribute of the place (stair, up escalator, Etc.).

  In addition, when the event detection unit 23 detects that an event has occurred in the target area 6 at s1 based on the detection result of the image processing unit 21 and the detection result of the audio processing unit 22, the situation determination unit 25 The status of the detected event is determined (s4). The situation determination unit 25 is based on an object such as a person detected around the place where the event in the target area 6 is identified, which is specified from the detection results of the image processing unit 21 and the audio processing unit 22. It is determined that the three parties are not in the vicinity or that many third parties are in the vicinity.

  Further, when the event detection unit 23 detects that an event has occurred in the target area 6 at s1 based on the detection result of the image processing unit 21 and the detection result of the audio processing unit 22, the state determination unit 26 The state of the person is determined (s5). The state determination unit 26 is identified from the detection results of the image processing unit 21 and the audio processing unit 22, and the behavior, posture, etc. of an object such as a person detected around the place where the event in the target area 6 occurs. Based on the above, it is determined whether the subject has a knife, the subject has fallen, or has fallen for a predetermined time or more.

  In addition, the process concerning s2-s5 may be performed in what order, and two or more processes may be performed simultaneously in parallel.

  For the event types determined by the event detection unit 23, the risk level estimation unit 27 determines, for each risk element type, the environment of the target area 6 determined by the environment determination unit 24, the event status determined by the status determination unit 25, And the information 31 for estimation memorize | stored in the memory | storage part 12 is searched with the state of the subject determined by the state determination unit 26, and the risk level is estimated (acquired) (s6). The risk level estimation unit 27 determines the necessity of notification for the event detected this time for each type of risk element (s7). The risk level estimation unit 27 compares, for each risk element type, the risk level estimated in s6 and the level of risk level determined to be notification required in the notification destination information 33 stored in the storage unit 12, If the degree of risk estimated in s6 is larger, it is determined that notification is necessary. If there is no risk factor determined to be required for notification in s7, the risk level estimation unit 27 returns to s1.

  On the other hand, if there is even one risk factor determined to be required for notification in s7, the risk level estimation unit 27 indicates the notification destination of the event detected this time for each type of risk element determined to be required for notification. (S8). Based on the environment of the target area 6 determined by the environment determination unit 24, the situation of the event determined by the situation determination unit 25, the state of the target person determined by the state determination unit 26, etc. Is created (s9). For example, “The subject is rampaging with a knife at the point of ○. There are many ordinary people around.” And a black jumper is worn. "

  The risk detection device 1 notifies the notification text created in s9 to the notification destination determined in s8 (s10), and returns to s1. The notification in s10 is transmitted via the network 5 to the police station, the fire station, the security room, etc., which are the notification destinations determined in s8.

  Accordingly, it is not necessary for the security guard or the like to judge by himself the type and magnitude of the damage to the parties involved in the event that occurred and the third parties around it. For this reason, the work burden of a security guard etc. can be reduced.

  In addition, the occurrence of the event can be notified to the police station, fire department, security room, etc. according to the type and magnitude of damage to the parties involved and the third parties in the vicinity. In addition to being able to deal with victims quickly and appropriately, it is possible to appropriately prevent the damage caused by the events that have occurred.

  Here, the risk level estimation in the risk level estimation unit 27 will be described with a specific example.

(Specific example A)
Event detected by the event detection unit: snatching Situation of the target area 6 determined by the environment determination unit 24: dark The event status determined by the status determination unit 25: There is no third party around The target determined by the state determination unit 26 In this case, the risk estimator 27 determines that notification of financial damage, which is a risk factor, is necessary, and requests a police station to investigate the financial damage. Etc. are notified.

(Specific example B)
Event detected by the event detection unit: snatching Situation of the target area 6 determined by the environment determination unit 24: dark The event status determined by the status determination unit 25: There is no third party around The target determined by the state determination unit 26 Person's condition: falls and does not stand up for more than 3 minutes In this case, the risk estimation unit 27 determines that notification of financial damage and physical damage, which are risk factors, is necessary, and sends money to the police station. In addition to notifications of investigation requests related to physical damage, it also notifies the fire department of requests to dispatch ambulances related to physical damage.

(Specific example C)
Event detected by the event detection unit: fighting The situation of the target area 6 determined by the environment determination unit 24: dark The status of the event determined by the status determination unit 25: There is no third party around The target determined by the state determination unit 26 In this case, the risk estimation unit 27 determines that notification of physical damage, which is a risk factor, is necessary, and notifies the police station of a dispatch request for physical damage, etc. At the same time, the fire department will be notified of ambulance requests for physical damage.

(Specific example D)
Event detected by the event detection unit: fighting The situation of the target area 6 determined by the environment determination unit 24: dark The status of the event determined by the status determination unit 25: There is a third party around The target determined by the state determination unit 26 In this case, the risk estimation unit 27 determines that notification of physical damage, which is a risk factor, is necessary, and notifies the security room or the like of an arbitration request or the like.

(Specific example E)
Event detected by the event detection unit: wandering The situation of the target area 6 determined by the environment determination unit 24: dark, the state of the event determined by the bicycle parking state determination unit 25: There is no third party around The state determination unit 26 determines In this case, the risk estimation unit 27 determines that notification about the damage to the article, which is a risk factor, is necessary, and notifies the security room or the like of the site confirmation.

  In this case, the notification device provided in the target area 6 may be notified to instruct the output of the alert sound.

<4. Modification>
The risk detection device 1 may be configured not to determine any one of the environment of the target area 6, the status of the event that has occurred this time, and the state of the target person related to the event that has occurred this time. That is, the risk detection device 1 may be configured not to include any one of the environment determination unit 24, the situation determination unit 25, and the state determination unit 26. With this configuration, the risk level detection device 1 estimates the risk level by any one of the environment of the target area 6 that is not determined, the situation of the event that has occurred this time, or the state of the subject person who has experienced the event that has occurred this time. Do not use items, but comprehensively use the determination results of the remaining two items.

  In addition, the configuration relating to the event detection unit 23, the environment determination unit 24, the situation determination unit 25, the state determination unit 26, the risk level estimation unit 27, and the like may be a configuration using artificial intelligence (AI). Good.

  In addition, for each event occurring in the target area 6, the person in charge of the risk detection device 1 determines, for each event, the environment of the target area, the situation of the event, the state of the target person, and the risk level of various risk factors. A configuration in which the result determined after the fact is stored in the storage unit 12 as the event history information 32 may be adopted. In this case, the risk detection apparatus 1 may be configured to update the estimation information 31 by using the event history information 32 as learning data.

  Further, when the target area 6 is relatively wide, the target area 6 is divided into a plurality of areas, and for each of the divided areas, the type of environment of the area, the type of event status, and the type of state of the target person For each combination pattern, information associated with the degree of risk may be stored in the storage unit 12 as the estimation information 31.

  Note that the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. Further, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, you may combine suitably the component covering different embodiment.

Furthermore, the correspondence between the configuration according to the present invention and the configuration according to the above-described embodiment can be described as in the following supplementary notes.
<Appendix>
An event detector (23) for detecting an event that has occurred in the target area (6);
An environment determination unit (24) for determining the environment of the target area (6);
A situation determination section (25) for determining the situation of an event that has occurred in the target area (6) detected by the event detection section (23);
The environment of the target area (6) determined by the environment determination unit (24) for each type of risk factor corresponding to the event that occurred in the target area (6) detected by the event detection unit (23), And a risk level estimation unit (1) that estimates the risk level using the status of the event that occurred in the target area (6) determined by the status determination unit (25). .

DESCRIPTION OF SYMBOLS 1 ... Risk detection apparatus 2 ... Camera 3 ... Microphone 5 ... Network 6 ... Target area 11 ... Control unit 12 ... Memory | storage part 13 ... Image input part 14 ... Audio | voice input part 15 ... Communication part 16 ... Output part 21 ... Image processing part 22 ... Audio processing unit 23 ... Event detection unit 24 ... Environment determination unit 25 ... Situation determination unit 26 ... State determination unit 27 ... Risk estimation unit 31 ... Information for estimation 32 ... Event history information 33 ... Notification destination information

Claims (12)

An event detector for detecting events that occur in the target area;
An environment determination unit for determining the environment of the target area;
A status determination unit that determines the status of an event that has occurred in the target area detected by the event detection unit;
The environment of the target area determined by the environment determination unit and the target area determined by the situation determination unit for each type of risk element corresponding to the event that occurred in the target area detected by the event detection unit A risk level detection device comprising: a risk level estimation unit that estimates a risk level using a situation of an event that has occurred. An event detector for detecting events that occur in the target area;
An environment determination unit for determining the environment of the target area;
A state determination unit that determines the state of the subject related to the event that occurred in the target area detected by the event detection unit;
The environment of the target area determined by the environment determination unit and the target person determined by the state determination unit for each type of risk element corresponding to the event that occurred in the target area detected by the event detection unit And a risk level estimation unit that estimates the level of risk using the state of the risk. An event detector for detecting events that occur in the target area;
A status determination unit that determines the status of an event that has occurred in the target area detected by the event detection unit;
A state determination unit that determines the state of the subject related to the event that occurred in the target area detected by the event detection unit;
For each type of risk element corresponding to the event that occurred in the target area detected by the event detection unit, the status of the event that occurred in the target area determined by the status determination unit, and the status determination unit A danger level detection device comprising: a risk level estimation unit that estimates a risk level using the state of the subject person. A state determination unit that determines the state of the subject related to the event that occurred in the target area detected by the event detection unit;
The risk level detection device according to claim 1, wherein the risk level estimation unit estimates the risk level by using the state of the subject determined by the state determination unit for each type of the risk factor. A notification destination determination unit for determining a notification destination for notifying the occurrence of an event in the target area, according to the risk level for each type of the risk factor by the risk level estimation unit;
The risk detection device according to claim 1, further comprising: a notification unit that notifies the notification destination determined by the notification destination determination unit that an event has occurred in the target area.   The risk detection device according to claim 1, wherein the event detection unit processes a captured image obtained by imaging the target area and detects an event that has occurred in the target area. An event detection step for detecting an event occurring in the target area; and
An environment determination step for determining an environment of the target area;
A situation determination step for determining a situation of an event that has occurred in the target area detected in the event detection step;
The environment of the target area determined in the environment determination step and the target determined in the situation determination step for each type of risk element corresponding to the event that occurred in the target area detected in the event detection step A risk detection method in which a computer executes a risk estimation step of estimating a risk using the state of an event occurring in an area. An event detection step for detecting an event occurring in the target area; and
An environment determination step for determining an environment of the target area;
A state determination step for determining the state of the subject related to the event that occurred in the target area detected in the event detection step;
For each type of risk factor corresponding to the event that occurred in the target area detected in the event detection step, the environment of the target area determined in the environment determination step and the target person determined in the state determination step A risk detection method in which a computer executes a risk estimation step of estimating a risk using a state. An event detection step for detecting an event occurring in the target area; and
A situation determination step for determining a situation of an event that has occurred in the target area detected in the event detection step;
A state determination step for determining the state of the subject related to the event that occurred in the target area detected in the event detection step;
For each type of risk factor corresponding to the event that occurred in the target area detected in the event detection step, the status of the event that occurred in the target area determined in the status determination step, and the status determination step A risk level detection method in which a computer executes a risk level estimation step of estimating a risk level using the state of the subject. An event detection step for detecting an event occurring in the target area; and
An environment determination step for determining an environment of the target area;
A situation determination step for determining a situation of an event that has occurred in the target area detected in the event detection step;
The environment of the target area determined in the environment determination step and the target determined in the situation determination step for each type of risk element corresponding to the event that occurred in the target area detected in the event detection step A risk detection program for causing a computer to execute a risk estimation step for estimating a risk using the status of an event occurring in an area. An event detection step for detecting an event occurring in the target area; and
An environment determination step for determining an environment of the target area;
A state determination step for determining the state of the subject related to the event that occurred in the target area detected in the event detection step;
For each type of risk factor corresponding to the event that occurred in the target area detected in the event detection step, the environment of the target area determined in the environment determination step and the target person determined in the state determination step A risk level detection program for causing a computer to execute a risk level estimation step for estimating a risk level using a state. An event detection step for detecting an event occurring in the target area; and
A situation determination step for determining a situation of an event that has occurred in the target area detected in the event detection step;
A state determination step for determining the state of the subject related to the event that occurred in the target area detected in the event detection step;
For each type of risk factor corresponding to the event that occurred in the target area detected in the event detection step, the status of the event that occurred in the target area determined in the status determination step, and the status determination step A risk detection program for causing a computer to execute a risk estimation step of estimating a risk using the state of the subject.

Images