Collision risk notification system and collision risk notification method

Abstract

To provide a collision risk notification system and a collision risk notification method to notify a driver of a risk of collision with a person, a vehicle, or the like in a blind spot.SOLUTION: A collision risk notification system includes: a captured image acquisition unit that acquires captured images; an image coordinate estimation unit that estimates in-image coordinates indicating positions of movable bodies within the captured images; a time acquisition unit that acquires time; an area identification unit that identifies an area where the movable body is located based on the in-image coordinates of the movable body; an in-image destination estimation unit that estimates the in-image coordinates of the movable body at a third time point after a second time point based on a time change from a first time point to the second time point in the in-image coordinates of the movable body; an area destination identification unit that identifies the area where the movable body is located at the third time point based on the in-image coordinates of the movable body at the third time point; a risk determination unit that determines a collision risk based on a relationship between the area where a first movable body and a second movable body are located at the second time point and the area where the first movable body and the second movable body are located at the third time point; and a notification unit that notifies of the collision risk.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a collision risk notification system and a collision risk notification method. [Background technology]

[0002] In factories, warehouses, etc., a technology has been disclosed that uses a camera attached to a vehicle to detect the presence of a worker in order to prevent a driver from driving a vehicle without noticing the presence of a worker and colliding with the worker (see, for example, Patent Document 1).There is also a technology that has workers carry RFID (Radio Frequency Identification) tags, and by reading the RFID tags using a receiver attached to the vehicle, detects workers within the readable range of the RFID. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-069671 Summary of the Invention [Problem to be solved by the invention]

[0004] However, factories, warehouses, and other locations have many blind spots due to the presence of cargo and other items. A worker detection method using a camera mounted on a vehicle may not be able to detect a worker hiding in a blind spot due to cargo or around a corner, resulting in a collision with a worker appearing from a blind spot or around a corner. Furthermore, a worker detection method using RFID tags and a receiver may detect a worker who is within the readable range of the RFID tag even when there is a wall or other obstacle between the worker and the vehicle, eliminating the risk of a collision. Furthermore, because the receiver detects workers by reading the RFID tag, it is unable to detect visitors or other people without RFID tags.

[0005] Therefore, the present invention has been made in consideration of the above points, and aims to provide a collision risk notification system and collision risk notification method that notify a driver of the risk of collision with a person, vehicle, etc. in a blind spot. [Means for solving the problem]

[0006] [1] One aspect of the present invention is a collision risk notification system having an image acquisition unit that acquires, from each of the multiple imaging devices, images of areas into which detection areas are divided by the multiple imaging devices; an extraction unit that extracts a plurality of different moving objects from the image images obtained from each of the multiple imaging devices; a destination estimation unit that estimates to which area the future position of each of the extracted different moving objects will fall based on a position history identified from image images taken at different times; a risk determination unit that determines, in accordance with the estimated areas, that there is a risk of collision for moving objects among the multiple moving objects whose areas estimated based on the future positions are close to each other; and a notification unit that notifies the result determined by the risk determination unit.

[0007] [2] Also, one aspect of the present invention is a collision risk notification system that uses a plurality of imaging devices installed at different positions to notify that there is a risk of collision between a first moving body and a second moving body among moving bodies, the collision risk notification system including: an image acquisition unit that acquires a plurality of captured images of the moving body taken between a second time corresponding to a current time and a first time prior to the second time; an image coordinate estimation unit that estimates, based on the captured images, image coordinates that are coordinates indicating a position of the captured moving body shown in the captured image and whether the captured moving body is the first moving body or the second moving body; a time acquisition unit that acquires the time at which the captured image was taken; a first storage unit that stores in advance correspondence information for each of the plurality of imaging devices, in which coordinates that indicate a position in the captured image taken by the imaging device are associated with identifiers that identify areas into which an area captured by the plurality of imaging devices is divided based on the risk of collision between the first moving body and the second moving body; and a first storage unit that, by referring to the correspondence information, estimates the position of the moving body at the second time from the image coordinates of the moving body at the second time estimated by the image coordinate estimation unit. a second storage unit that stores the in-image coordinates of the moving object estimated from the captured image by the in-image coordinate estimation unit, moving object information indicating whether the moving object is the first moving object or the second moving object, and a time at which the captured image was captured in association with each other; an in-image destination estimation unit that estimates coordinates indicating a position at which the moving object will be displayed in the captured image at a third time after the second time based on a time change from the first time to the second time of the in-image coordinates of the moving object stored in the second storage unit; an area destination specification unit that specifies the area where the moving body is located at the third time from coordinates indicating the position of the moving body represented in the captured image at the third time estimated by the intra-image destination estimation unit by referring to the information; the area where the first moving body is located at the second time specified by the area specification unit; the area where the first moving body is located at the third time specified by the area destination specification unit; the area where the second moving body is located at the second time specified by the area specification unit;A collision risk notification system including a risk determination unit that determines the degree of risk of collision between the first moving body and the second moving body based on the relationship with the area where the second moving body is located at the third time, and a notification unit that notifies the degree of risk of collision based on the degree determined by the risk determination unit.

[0008] [3] Another aspect of the present invention is a collision risk notification system that uses a plurality of image capturing devices installed at different positions to notify a user of a risk of collision between a first moving body and a second moving body, the system including: an image capturing unit that captures a plurality of captured images of the moving body between a second time corresponding to a current time and a first time prior to the second time; and an image capturing unit that captures, based on the captured images, image coordinates that indicate the position of the captured moving body shown in the captured image; and an image capturing unit that estimates whether the captured moving body is the first moving body or the second moving body. an internal coordinate estimation unit; a first storage unit that stores in advance correspondence information in which, for each of the plurality of imaging devices, coordinates indicating a position in the captured image captured by the imaging device correspond to coordinates indicating a position in an area captured by the plurality of imaging devices; an internal area coordinate specification unit that specifies internal area coordinates that are coordinates indicating a position in the area where the moving object is located at the second time from the internal image coordinates of the moving object at the second time estimated by the internal image coordinate estimation unit by referring to the correspondence information; a second storage unit that stores the coordinates within the area where the moving body is located, moving body information indicating whether the moving body is the first moving body or the second moving body, and the time at which the captured image was captured, in association with each other; an intra-area destination estimation unit that estimates coordinates indicating a position within the area where the moving body is located at a third time after the second time based on a time change from the first time to the second time of the intra-area coordinates of the moving body stored in the second storage unit; and a coordinate indicating a position within the area where the first moving body is located at the second time, which is identified by the intra-area coordinate identification unit, and a time at which the captured image was captured. a coordinate specifying unit that indicates the position within the area where the first moving body is located at the third time, a coordinate specified by the intra-area coordinate specifying unit that indicates the position within the area where the second moving body is located at the second time, and a coordinate specified by the intra-area destination estimating unit that indicates the position within the area where the second moving body is located at the third time, and based on whether at least two of the coordinates are included in any of the areas that are one or more areas set within the area based on the risk of collision between the first moving body and the second moving body,The collision risk notification system includes a risk determination unit that determines the degree of risk of collision between the first moving body and the second moving body, and a notification unit that notifies the degree of risk of collision based on the degree determined by the risk determination unit.

[0009] [4] Also, one aspect of the present invention is a collision risk notification system that uses a plurality of imaging devices installed at different positions to notify that there is a risk of collision between a first moving body and a second moving body among moving bodies, the collision risk notification system including: an image acquisition step of acquiring a plurality of images of the moving body captured between a second time corresponding to the current time and a first time prior to the second time; an image coordinate estimation step of estimating, based on the captured images, in-image coordinates that are coordinates indicating the position of the captured moving body shown in the captured image and whether the captured moving body is the first moving body or the second moving body; a time acquisition step of acquiring the time at which the captured image was captured; a first storage step of storing in advance in a first storage unit, for each of the plurality of imaging devices, correspondence information that associates coordinates indicating the position in the captured image captured by the imaging device with an identifier that identifies an area into which an area captured by the plurality of imaging devices is divided based on the risk of collision between the first moving body and the second moving body; an area specifying step of specifying the area where the moving object is located at a certain time; a second storage step of storing in a second storage unit the in-image coordinates of the moving object estimated from the captured image by the in-image coordinate estimation step, moving object information indicating whether the moving object is the first moving object or the second moving object, and the time at which the captured image was captured, in association with each other; and estimating coordinates indicating the position at which the moving object will be represented in the captured image at a third time after the second time, based on the change over time from the first time to the second time of the in-image coordinates of the moving object stored in the second storage unit. an area destination specification step of specifying the area where the moving body is located at the third time from coordinates indicating the position of the moving body represented in the captured image at the third time estimated by the area destination specification step by referring to the correspondence information; the area where the first moving body is located at the second time specified by the area specification step; the area where the first moving body is located at the third time specified by the area destination specification step; and the area where the second moving body is located at the second time specified by the area specification step.The collision risk notification method includes a risk determination step of determining a degree of risk of collision between the first moving body and the second moving body based on the relationship with the area where the second moving body is located at the third time, which is specified by the area destination specification step, and a notification step of notifying the degree of risk of collision based on the degree determined by the risk determination step.

[0010] [5] Another aspect of the present invention is a collision risk notification system that uses a plurality of imaging devices installed at different positions to notify a user of a risk of collision between a first moving body and a second moving body among moving bodies, the collision risk notification system including: an image acquisition step of acquiring a plurality of images of the moving body captured between a second time corresponding to a current time and a first time prior to the second time; an image coordinate estimation step of estimating, based on the captured images, image coordinates that indicate the position of the captured moving body shown in the captured image and whether the captured moving body is the first moving body or the second moving body; a first storage step of storing in advance in a first storage unit, for each of the plurality of imaging devices, correspondence information that associates coordinates that indicate the position in the captured image captured by the imaging device with coordinates that indicate the position in the area captured by the plurality of imaging devices; an area coordinate identification step of identifying area coordinates that indicate the position in the area where the moving body is located at the second time from the image coordinates of the moving body at the second time estimated by the image coordinate estimation step by referring to the correspondence information; and a second storage step of storing in a second storage unit coordinates within the area where the moving object identified by the step is located, moving object information indicating whether the moving object is the first moving object or the second moving object, and a time at which the captured image was captured, in association with each other; an intra-area destination estimation step of estimating coordinates indicating a position within the area where the moving object is located at a third time after the second time, based on a time change from the first time to the second time of the intra-area coordinates of the moving object stored in the second storage unit; a relationship between coordinates indicating a position within the area where the first moving body is located, coordinates indicating a position within the area where the first moving body is located at the third time estimated by the intra-area destination estimation process, coordinates indicating a position within the area where the second moving body is located at the second time identified by the intra-area coordinate identification process, and coordinates indicating a position within the area where the second moving body is located at the third time estimated by the intra-area destination estimation process; and one or more areas set in the area based on the risk of collision between the first moving body and the second moving body,The collision risk notification method includes a risk determination step of determining the degree of risk of collision between the first moving body and the second moving body based on whether at least two of the above-mentioned coordinates are included, and a notification step of notifying the degree of risk of collision based on the degree determined by the risk determination step. [Effects of the Invention]

[0011] According to the present invention, it is possible to notify a driver of the risk of collision with a person, vehicle, etc. in a blind spot. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a schematic diagram for explaining a collision risk notification system. [Figure 2] 1 is a block diagram for explaining the functional configuration of a collision risk notification system according to a first embodiment. [Figure 3] FIG. 10 is a diagram illustrating an example of an area set by dividing a detection region. [Figure 4] FIG. 10 is an image diagram showing each area in an image captured by the first imaging device. [Figure 5] 10A and 10B are conceptual diagrams for explaining changes over time in the position of an image of an imaged worker displayed in a captured image. [Figure 6] 10A and 10B are diagrams illustrating an example of the imaging areas and detection results of each imaging device. [Figure 7] 4 is a flowchart illustrating an example of a processing flow of the collision risk notification system according to the first embodiment. [Figure 8] FIG. 10 is a sequence diagram illustrating an example of a process flow for registering a map. [Figure 9] FIG. 10 is a sequence diagram illustrating an example of a process flow for registering an area. [Figure 10] FIG. 10 is a sequence diagram illustrating an example of a process flow for registering a criterion. [Figure 11]3 is a sequence diagram for explaining an example of a processing flow of each device included in the collision risk notification system according to the first embodiment. FIG. [Figure 12] FIG. 10 is a block diagram illustrating the functional configuration of a collision risk notification system according to a second embodiment. [Figure 13] FIG. 10 is a diagram showing an example of an image displayed on a position display monitor. [Figure 14] 10 is a flowchart illustrating an example of a processing flow of a collision risk notification system according to the second embodiment. [Figure 15] FIG. 10 is a sequence diagram illustrating an example of a process flow for registering coordinates. [Figure 16] FIG. 10 is a sequence diagram illustrating an example of a process flow for displaying an image on a position display monitor. [Figure 17] FIG. 10 is a block diagram illustrating the functional configuration of a collision risk notification system according to a third embodiment. [Figure 18] 11 is a flowchart illustrating an example of a processing flow of a collision risk notification system according to a third embodiment. [Figure 19] FIG. 11 is a sequence diagram for explaining an example of the flow of processing by each device included in the collision risk notification system according to the third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] A collision risk determination system and a collision risk determination method according to preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Note that the present embodiments are not limited to these embodiments and include various modifications and improvements. In other words, the components described below include those that can be easily imagined by a person skilled in the art and those that are substantially the same, and the components described below can be combined as appropriate. Furthermore, various omissions, substitutions, or modifications of the components of the present embodiments can be made without departing from the spirit and scope of the present invention.

[0014] Factories, warehouses, and other locations have many blind spots due to obstacles such as luggage, shelves, and equipment installed within the building. Therefore, it is difficult for workers and vehicle drivers to stay aware of all blind spots, and they may collide with a worker or vehicle that appears from a blind spot they are not aware of. The collision risk notification system 1 according to this embodiment determines whether there is a risk of collision between moving objects within a detection area DA and issues a notification if there is a risk of collision. The detection area DA is an area captured by imaging devices 10 installed at different locations. For example, the detection area DA may be a part or entire area of a building such as a factory or warehouse, or may be an area that further includes the periphery of the building. Examples of moving objects include people such as workers and visitors to the factory or warehouse, and vehicles such as automated guided vehicles (AGVs), transport robots (AMRs), forklifts, and trucks. Of two moving objects for which the collision risk notification system 1 determines the collision risk, one is referred to as a first moving object, and the other is referred to as a second moving object. The first and second moving bodies may be the same type of moving body, such as a person and a person, or a vehicle and a vehicle, or may be different types of moving bodies, such as a person and a vehicle. Hereinafter, the description will be given assuming that the first moving body is a worker MO1 and the second moving body is a vehicle MO2. The collision risk notification system 1 may notify that there is a risk of collision between two of multiple moving bodies within the detection area DA. That is, the collision risk notification system 1 may determine whether there is a risk of collision between two or more moving bodies, and notify the moving bodies that are at risk of collision that there is a risk of collision. The collision risk notification system 1 may also determine and notify whether there is a risk of collision between three or more moving bodies (multi-body collision). Furthermore, when a set of two or more moving bodies is defined as a "set of moving bodies," the collision risk notification system 1 may perform the determination and notification for multiple sets of moving bodies.

[0015] FIG. 1 is a schematic diagram illustrating a collision risk notification system 1. FIG. 1 shows an example in which a worker MO1 and a vehicle MO2 move within a detection area DA in which a package B is placed. When the worker MO1 and the vehicle MO2 move to the center of the detection area DA, there is a risk that the worker MO1 and the vehicle MO2, who are in each other's blind spots of the package B, may collide without realizing that they are approaching each other. Therefore, the collision risk notification system 1 uses an alarm device 2 to notify the worker MO1 and the vehicle MO2 of the risk of a collision. This allows the worker MO1 and the vehicle MO2 to notice the presence of the worker MO1 and the vehicle MO2 in their blind spots and avoid a collision. [Embodiment 1]

[0016] FIG. 2 is a block diagram illustrating the functional configuration of a collision risk notification system 1 according to the first embodiment. The collision risk notification system 1 includes multiple image capture devices 10, a collision risk determination device 40, a warning device 2, a captured image display monitor 3, and a management device 4. The first image capture device 10-1 to the third image capture device 10-3 exemplify the multiple image capture devices 10, and when there is no need to distinguish between them, they may be simply referred to as the image capture devices 10. The image capture devices 10, the collision risk determination device 40, the warning device 2, the captured image display monitor 3, and the management device 4 are connected to one another via a network NW. The network NW is a predetermined communication network configured using lines. The network NW is, for example, the Internet. Each functional unit of the image capture device 10 and the collision risk determination device 40 may be realized using an electronic circuit, if necessary. Furthermore, each functional unit does not have to be included in a single device, and the image capture device 10 and the collision risk determination device 40 may be configured from multiple devices.

[0017] The imaging device 10 has a functional unit that processes captured images. The imaging device 10 is, for example, an AI camera. The imaging device 10 includes an imaging unit 11, a captured image acquisition unit 12, an in-image coordinate estimation unit 13, an area identification unit 14, a time acquisition unit 15, an in-image destination estimation unit 16, an area destination identification unit 17, a detection result output unit 18, a first storage unit 21, a second storage unit 22, an instruction information acquisition unit 31, and a captured image output unit 32.

[0018] The imaging unit 11 captures an image of an imaging area in front of the vehicle and outputs the captured image to the captured image acquisition unit 12.

[0019] The captured image acquisition unit 12 acquires the captured image from the imaging unit 11. The captured image acquisition unit 12 outputs the acquired captured image to the intra-image coordinate estimation unit 13. Furthermore, the captured image acquisition unit 12 acquires instruction information from the instruction information acquisition unit 31. The instruction information is information that instructs the captured image acquisition unit 12 to output the captured image acquired by the captured image acquisition unit 12 to the captured image display monitor 3. When acquiring the instruction information, the captured image acquisition unit 12 outputs the captured image to the captured image output unit 32.

[0020] The image coordinate estimation unit 13 acquires the captured image from the captured image acquisition unit 12. Based on the captured image, the image coordinate estimation unit 13 estimates whether the moving object depicted in the captured image is a worker MO1 or a vehicle MO2 when the moving object is captured. Based on the captured image, the image coordinate estimation unit 13 also estimates image coordinates, which are coordinates indicating the position in the captured image where the captured moving object is depicted. For example, the image coordinate estimation unit 13 may estimate an area in which the moving object is depicted in the captured image, and may use coordinates indicating the center position of the estimated area as the image coordinates, or may use coordinates indicating the center position of the bottom edge of the estimated area (e.g., the feet of the worker MO1) as the image coordinates. The image coordinate estimation unit 13 outputs the estimated image coordinates to the area identification unit 14. The image coordinate estimation unit 13 also outputs detection information indicating that the moving object has been captured to the time acquisition unit 15. Furthermore, the image coordinate estimation unit 13 outputs moving object information indicating the type of moving object (whether it is a worker MO1 or a vehicle MO2) to the detection result output unit 18. The image coordinate estimation unit 13 also outputs the moving object information and the image coordinates to the second storage unit 22. The image coordinate estimation unit 13 may also be referred to as an extraction unit.

[0021] Correspondence information is stored in advance in the first storage unit 21. The correspondence information is information in which coordinates indicating a position in the captured image are associated with identifiers identifying each of multiple areas SA. An area SA is an area set within the detection area DA based on the risk of a collision between the worker MO1 and the vehicle MO2. In the first embodiment, the area SA is set by dividing the detection area DA based on the risk of a collision between the worker MO1 and the vehicle MO2. For example, the area SA may be set so that each area that is empirically known to be prone to collision accidents, such as a corner or an intersection where passageways intersect, is set as a single area SA. Different correspondence information may be stored in the first storage unit 21 for each imaging device 10, or correspondence information further associated with an identifier identifying each imaging device 10 may be stored commonly for each imaging device 10.

[0022] Fig. 3 is a diagram illustrating an example of areas SA set by dividing the detection area DA. In Fig. 3, the detection area DA is divided into a first area SA1, a second area SA2, a third area SA3, a fourth area SA4, and a fifth area SA5. The upper area within the detection area DA is further divided into two areas SA: a right area and a left area. When worker MO1 moves from the fourth area SA4 to the second area SA2 and vehicle MO2 moves from the fifth area SA5 to the second area SA2, worker MO1 and vehicle MO2 cannot see each other because they are in each other's blind spots of the package B, which poses a risk of collision. In contrast, when worker MO1 moves from the third area SA3 to the first area SA1 and vehicle MO2 moves from the fifth area SA5 to the second area SA2, there is no package B between them when worker MO1 reaches the first area SA1 and vehicle MO2 reaches the second area SA2. Therefore, the drivers of worker MO1 and vehicle MO2 can see each other and may be able to avoid a collision. Therefore, if the area combining the first area SA1 and the second area SA2 is defined as a single area SA, the risk of collision varies depending on the positions of worker MO1 and vehicle MO2 in that area SA. Therefore, the first area SA1 and the second area SA2 are set separately.

[0023] Additionally, the lower area within the detection area DA is set as three separate areas SA, each separated by baggage B. If worker MO1 remains in the fourth area SA4 and vehicle MO2 moves from the second area SA2 to the fourth area SA4, the driver of vehicle MO2 may not be able to recognize worker MO1 in the blind spot created by baggage B, resulting in a potential collision. On the other hand, if worker MO1 remains in the fourth area SA4 and vehicle MO2 moves from the second area SA2 to the fifth area SA5, there is baggage B between the fourth area SA4 and the fifth area SA5, so worker MO1 and vehicle MO2 will not collide. Therefore, if the combined area of the fourth area SA4 and the fifth area SA5 is set as a single area SA, the risk of collision differs depending on the positions of worker MO1 and vehicle MO2 in that area SA. Therefore, the fourth area SA4 and the fifth area SA5 are set separately. Similarly, the third area SA3 and the fourth area SA4 are set separately.

[0024] The area identification unit 14 acquires the image coordinates of the moving object from the image coordinate estimation unit 13. The area identification unit 14 also acquires correspondence information by referring to the first storage unit 21. The area identification unit 14 identifies the area SA in which the moving object is located from the image coordinates of the moving object based on the correspondence information. The area identification unit 14 outputs an identifier of the area SA in which the moving object is located to the detection result output unit 18.

[0025] An example of how the area identification unit 14 identifies the area SA will be described in detail with reference to Fig. 4. Fig. 4 is an image diagram showing each area SA in a captured image captured by the first imaging device 10-1. In Fig. 4, the captured worker MO1 is shown in the center of the captured image. If the coordinates indicating the center position in the captured image are associated with an identifier indicating the fourth area SA4, the area identification unit 14 can identify that the worker MO1 is in the fourth area SA4 based on the coordinates indicating the center position in the captured image.

[0026] The time acquisition unit 15 acquires detection information from the intra-image coordinate estimation unit 13. When acquiring the detection information, the time acquisition unit 15 acquires time information indicating the time when the captured image was captured. The time acquisition unit 15 may acquire, for example, the time indicated by a clock built into the imaging device 10 or the standard time in the area where the imaging device 10 is installed, acquired from an NTP (Network Time Protocol) server, as the time when the captured image was captured. The time acquisition unit 15 outputs the acquired time information to the second storage unit 22.

[0027] The second storage unit 22 stores the moving object information estimated by the in-image coordinate estimation unit 13, the in-image coordinates estimated by the in-image coordinate estimation unit 13, and the time information acquired by the time acquisition unit 15 in association with each other. That is, the second storage unit 22 stores the history of the positions of the moving objects. Here, "storing the time information in association with each other" may mean storing, for each captured image, the time of capture, the in-image coordinates of the moving object depicted in the captured image, and the type of the moving object. Furthermore, "storing the time information in association with each other" may mean sequentially storing, in chronological order, the in-image coordinates of the moving object depicted in the captured image and the type of the moving object using a ring buffer or the like.

[0028] The in-image destination estimation unit 16 acquires corresponding information by referring to the second storage unit 22. The in-image destination estimation unit 16 estimates an in-image destination, which is a coordinate indicating a position in the captured image where the moving object is represented at a third time, based on the temporal change in the in-image coordinates of the moving object from the first time to the second time. The second time is a time corresponding to the current time. The second time may be, for example, a time closest to the current time among the times associated with the information stored in the second storage unit 22, or a time some distance before the time closest to the current time. The first time is a time before the second time. The first time and the second time are any of the times associated with the information stored in the second storage unit 22. The third time is a time after the second time. In other words, the in-image destination estimation unit 16 estimates future in-image coordinates based on the temporal change in the in-image coordinates from the past to the present. For example, the in-image destination estimation unit 16 may estimate a predicted path of movement between the second time and the third time based on a path (trajectory) along which the in-image coordinates of the moving object moved between the first time and the second time in the captured image. Specifically, the in-image destination estimation unit 16 may estimate a predicted path (predicted trajectory) by extending the trajectory in the movement direction. The in-image destination estimation unit 16 does not need to use all of the in-image coordinates from the first time to the second time stored in the second storage unit 22, but may use at least two or more in-image coordinates from the in-image coordinates from the first time to the second time. The in-image destination estimation unit 16 outputs the estimated in-image destination to the area destination identification unit 17. Furthermore, the in-image destination estimation unit 16 may estimate whether the moving object is moving based on the change in the in-image coordinates over time from the first time to the second time. For example, the in-image destination estimation unit 16 may estimate that the moving object is moving when the sum of the distances over which the in-image coordinates have changed in the captured image between the first time and the second time is equal to or greater than a threshold set by an administrator. The in-image destination estimation unit 16 outputs information indicating whether the moving object is moving to the detection result output unit 18.

[0029] An example of estimating an in-image destination by the in-image destination estimating unit 16 will be specifically described with reference to FIG. 5. FIG. 5 is an image diagram for explaining the change over time in the position of worker MO1 displayed in the captured image. In FIG. 5, the position of worker MO1 displayed in the captured image moves in a straight line downward and to the left from the position indicated by the in-image coordinates at the first time, from the first time, "11:55:28," to the second time, "11:55:30," every second. Therefore, the in-image destination estimating unit 16 estimates that the position of worker MO1 displayed in the captured image at the third time, "11:55:33," is a position obtained by moving in a straight line downward and to the left from the position indicated by the in-image coordinates at the second time. Furthermore, if the path taken by the position of worker MO1 shown in the captured image between the first time and the second time is curved, the in-image destination estimation unit 16 may estimate that the position of worker MO1 shown in the captured image at the third time will be a position to which worker MO1 has continued to curve between the second time and the third time.

[0030] The area destination identification unit 17 acquires an in-image destination, which is a coordinate indicating the position in the captured image where the moving object is represented at the third time, from the in-image destination estimation unit 16. The area destination identification unit 17 also acquires correspondence information by referring to the first storage unit 21. Similar to the area identification unit 14, the area destination identification unit 17 identifies the area SA where the moving object is located at the third time from the in-image destination of the moving object based on the correspondence information. The area destination identification unit 17 outputs an identifier of the area SA where the moving object is located at the third time to the detection result output unit 18. The area destination identification unit 17 is also referred to as a destination estimation unit.

[0031] The detection result output unit 18 acquires moving object information indicating whether the moving object is a worker MO1 or a vehicle MO2 from the intra-image coordinate estimation unit 13. Furthermore, the detection result output unit 18 acquires an identifier of the area SA in which the moving object is located at a second time from the area identification unit 14. Furthermore, the detection result output unit 18 acquires information indicating whether the moving object is moving or not from the intra-image destination estimation unit 16. Furthermore, the detection result output unit 18 acquires an identifier of the area SA in which the moving object is located at a third time from the area destination identification unit 17. The detection result output unit 18 associates the acquired information and outputs it to the collision risk determination device 40 as a detection result.

[0032] The instruction information acquisition unit 31 acquires instruction information from the collision risk determination device 40. The instruction information acquisition unit 31 outputs the acquired instruction information to the captured image acquisition unit 12.

[0033] The captured image output unit 32 acquires the captured image from the captured image acquisition unit 12. The captured image output unit 32 outputs the acquired captured image to the captured image display monitor 3.

[0034] The collision risk determination device 40 includes a detection result acquisition unit 41, a risk determination unit 42, an alarm information output unit 43, an instruction information output unit 44, a detection result memory unit 51, a determination information memory unit 52, a map information memory unit 53, and an area information memory unit 54.

[0035] The detection result acquisition unit 41 acquires detection results from each of the imaging devices 10. The detection result acquisition unit 41 also acquires identifiers that identify the imaging devices 10 that have output the detection results. The detection result acquisition unit 41 outputs the acquired detection results to the detection result storage unit 51 in association with the identifiers that identify the imaging devices 10 that have output the detection results.

[0036] The risk determination unit 42 acquires the detection results of each image capture device 10 by referring to the detection result storage unit 51. The risk determination unit 42 also acquires the determination information by referring to the determination information storage unit 52. The determination information is information related to criteria for determining the degree of risk. The risk determination unit 42 determines the degree of risk of a collision between the worker MO1 and the vehicle MO2 based on the relationship between the area SA where the worker MO1 is located, the area SA to which the worker MO1 is moving, the area SA where the vehicle MO2 is located, and the area SA to which the vehicle MO2 is moving, as well as the determination information. Furthermore, if the vehicle MO2 is not moving, the worker MO1 is unlikely to be injured even if he collides with the stationary vehicle MO2. Therefore, the risk determination unit 42 may further determine the degree of risk of a collision based on whether the vehicle MO2 is moving. The degree of risk of a collision may be expressed as a binary value indicating whether there is a risk of a collision, or may be an index indicating each range of the risk of a collision, which is divided into upper and lower limits based on arbitrary conditions. The risk determination unit 42 outputs warning information indicating the degree of risk of collision to the warning information output unit 43. Furthermore, when there is a risk of collision between the worker MO1 and the vehicle MO2, the risk determination unit 42 outputs to the instruction information output unit 44 an identifier indicating the imaging device 10 that has output the detection result that is the basis for determining that there is a risk of collision.

[0037] Furthermore, the risk determination unit 42 can determine the degree of collision risk based on the proximity of the worker MO1 and the vehicle MO2, even without information on the distance between the worker MO1 and the vehicle MO2, based on the relationship between the area SA where the worker MO1 is located and the area SA where the vehicle MO2 is located. If the distance between the worker MO1 and the vehicle MO2 is short, even if one of them notices the approach, there is a risk of a collision. On the other hand, if the distance between the worker MO1 and the vehicle MO2 is long, there is a high probability that they will be able to see each other and avoid the approach before it occurs. Therefore, the closer the worker MO1 and the vehicle MO2 are to each other, the higher the degree of collision risk is likely to be, and the farther the vehicles MO2 are from each other, the lower the degree of collision risk is likely to be. For example, when the worker MO1 and the vehicle MO2 are in the same area SA or adjacent areas SA, the risk determination unit 42 may determine that the distance between the worker MO1 and the vehicle MO2 is short and the risk of a collision is high. For example, if there is another area SA between the area SA where worker MO1 is located and the area SA where vehicle MO2 is located, the risk determination unit 42 may determine that the distance between worker MO1 and vehicle MO2 is far and the risk of collision is low.

[0038] In this embodiment, the degree of risk of collision is expressed using indices of risk level 0, risk level 1, and risk level 2. The risk level index indicates that the higher the number, the higher the risk of collision. Risk level 2 indicates that there is an imminent risk of a collision between worker MO1 and vehicle MO2. Situations that are determined to be risk level 2 include, for example, when worker MO1 and vehicle MO2 are in the same area SA and vehicle MO2 is moving. A risk level of 1 indicates that there is a risk of collision after moving through area SA. Situations determined to be risk level 1 may include, for example, when vehicle MO2 moves into area SA where worker MO1 is located, or when worker MO1 and vehicle MO2, who are in different areas SA, move into the same area SA, resulting in the workers being in the same area SA after moving. Furthermore, situations determined to be risk level 1 may also include, for example, when the area SA to which worker MO1 is moving and the area SA to which vehicle MO2 is moving are close to each other, resulting in the workers being in the same area SA, resulting in the workers being in the same area SA, resulting in the workers being in the same area SA, resulting in the workers being in the same area SA, resulting in the workers being in the same area SA. A risk level of 0 indicates that there is no risk of a collision between worker MO1 and vehicle MO2. Information that is determined to be a risk level of 0 may include, for example, a case where the area SA where worker MO1 is located, the area SA where vehicle MO2 is located, and the area SA to which worker MO1 and vehicle MO2 are moving are different, or a case where vehicle MO2 is not moving.

[0039] An example of how the risk determination unit 42 determines the degree of collision risk will be described in detail with reference to FIG. 6. FIG. 6 is a diagram showing an example of the imaging areas and detection results of each imaging device 10. FIG. 6(A) shows the imaging area of the first imaging device 10-1. FIG. 6(B) shows the imaging area of the second imaging device 10-2. FIG. 6(C) shows the imaging area of the third imaging device 10-3. FIG. 6(D) shows the detection results of each imaging device 10. The first imaging device 10-1 and the third imaging device 10-3 estimate that the worker MO1 will move from the fourth area SA4 to the second area SA2. The second imaging device 10-2 estimates that the vehicle MO2 will move from the fifth area SA5 to the second area SA2. The risk determination unit 42 determines that the risk level is 1 because the area SA to which the worker MO1 and the vehicle MO2 are moving are the same. Furthermore, an identifier indicating at least one of the first imaging device 10-1 and the third imaging device 10-3 that has output the detection result regarding the worker MO1, and an identifier indicating the second imaging device 10-2 that has output the detection result regarding the vehicle MO2 are output to the instruction information output unit 44. When the detection result regarding the worker MO1 and the detection result regarding the vehicle MO2 are output from one imaging device 10, the warning information output unit 43 outputs an identifier indicating that imaging device 10 to the instruction information output unit 44.

[0040] In FIG. 6, the first imaging device 10-1 and the third imaging device 10-3 detect only the worker MO1 out of the worker MO1 and the vehicle MO2. The second imaging device 10-2 detects only the vehicle MO2 out of the worker MO1 and the vehicle MO2. The imaging devices 10 cannot detect moving objects in blind spots caused by obstacles such as luggage B. By using multiple imaging devices 10 installed to complement each other's blind spots to monitor for a risk of collision, a moving object that cannot be detected by one imaging device 10 can be detected by any of the other imaging devices 10. Therefore, the collision risk notification system 1, which includes multiple imaging devices 10 installed to complement each other's blind spots, can detect all moving objects within the detection area DA and determine and notify the risk of collision between the moving objects.

[0041] In this embodiment, the risk determination unit 42 can easily determine the degree of risk of a collision between the worker MO1 and the vehicle MO2 by using the area SA. For example, if the vehicle MO2 was traveling straight toward a corner between the first time and the second time, it would be difficult to estimate that the vehicle MO2 will turn based on the temporal change in the position of the vehicle MO2 displayed in the captured image from the first time to the second time. Therefore, even if there is a risk of a collision between the worker MO1 and the vehicle MO2 because the worker MO1 is in the passage ahead of the turn, it may not be possible to determine that there is a risk of a collision. Based on the area SA including the corner, the passage leading to the corner, and the passage ahead of the turn, the risk determination unit 42 can determine that there is a risk of a collision because the worker MO1 is in the area SA and the vehicle MO2 is heading toward the area SA.

[0042] The alarm information output unit 43 acquires the alarm information from the danger determination unit 42. The alarm information output unit 43 outputs the alarm information to the alarm device 2.

[0043] The instruction information output unit 44 acquires an identifier indicating the imaging device 10 that has output the detection result that is the basis for determining that there is a risk of collision from the risk determination unit 42. The instruction information output unit 44 outputs instruction information to the imaging device 10 indicated by the identifier.

[0044] The detection result storage unit 51 stores the detection results in association with the identifiers that identify the imaging devices 10 that have transmitted the detection results.

[0045] Determination information is stored in the determination information storage unit 52. The determination information is, for example, information that is determined by an administrator who manages the collision risk notification system 1, and includes determination criteria that indicate situations that may occur within the detection area DA and the degree of risk of collision between the worker MO1 and the vehicle MO2 in each situation.

[0046] The map information storage unit 53 stores information about the detection area DA. The information about the detection area DA may include, for example, information about the configuration within the detection area DA, such as the shape of the detection area DA, the position and shape of the passages, and the work space, as well as information about obstacles, such as luggage B, equipment, machinery, shelves, the exterior walls of the building, and the interior walls that separate the interior spaces of the building. Hereinafter, the information about the detection area DA may be referred to as map information. The map information storage unit 53 may also be referred to as a third storage unit.

[0047] Information about the area SA is stored in the area information storage unit 54. The information about the area SA may include, for example, an identifier for identifying each area SA set by dividing the detection area DA, the shape of the area SA, the position of each area SA within the detection area DA, etc. Hereinafter, the information about the area SA may be referred to as area information.

[0048] The warning device 2 acquires warning information from the collision risk determination device 40. Based on the warning information, the warning device 2 notifies the worker MO1 or the driver of the vehicle MO2 of the degree of risk of collision. The warning device 2 may be, for example, a speaker that notifies the worker MO1 or the driver of the vehicle MO2 of the risk of collision by sound, or a warning light that notifies the worker MO1 or the driver of the vehicle MO2 visually. The warning device 2 is also referred to as a notification unit. For example, when there is a risk of collision, the warning device 2 may sound an alarm indicating the risk of collision or may turn on a warning light. Furthermore, when there is no risk of collision, the warning device 2 may not turn on a warning light or may turn on a warning light of a color indicating there is no risk of collision. By the warning device 2 notifying the worker MO1 or the driver of the vehicle MO2 of the risk of collision, the worker MO1 or the driver of the vehicle MO2 can reduce the risk of collision by checking the surrounding situation and being aware of blind spots.

[0049] Furthermore, the warning device 2 may change the notification method, such as the warning sound or the color of the warning light, depending on the level of collision risk. For example, if the warning device 2 emits light in three colors, red, yellow, and green, red may indicate a risk level of 2, indicating an imminent risk of collision; yellow may indicate a risk level of 1, indicating a risk of collision after moving through area SA; and green may indicate a risk level of 0, indicating no risk of collision. By changing the notification method depending on the level of collision risk, the warning device 2 can enable the worker MO1 and the driver of the vehicle MO2 to take measures according to the notified level of collision risk. For example, if the driver of the vehicle MO2 is notified that the risk level is 2, he or she may be aware that the worker MO1 is nearby and may stop temporarily to check for the worker MO1 around or in the direction of travel of the vehicle MO2 before resuming movement. For example, if the driver of the vehicle MO2 is notified that the risk level is 1, he or she may be aware that the worker MO1 may appear at the destination and may reduce his or her speed to be able to respond if the worker MO1 appears from a blind spot.

[0050] The warning device 2 is installed in a position where at least one of the worker MO1 and the driver of the vehicle MO2, upon receiving a notification indicating a risk of collision, can recognize that an area SA where there is a risk of collision is nearby. The warning device 2 may be installed, for example, in each area SA, or in an area SA that must be passed through when traveling to another area SA, such as an area SA that includes an intersection of passageways. When the worker MO1 or the vehicle MO2 travels to the third area SA3, they must pass through the first area SA1. Therefore, if there is a risk of collision between moving objects in the third area SA3, the warning device 2 installed in the first area SA1 notifies the worker MO1 or the driver of the vehicle MO2 that there is a risk of collision. Therefore, it is sufficient that the warning device 2 is installed in at least the first area SA1 among the first area SA1, the third area SA3, and the fourth area SA4. Furthermore, for the same reason as in the first area SA1, the alarm device 2 may be installed in at least the second area SA2 out of the second area SA2, the fourth area SA4, and the fifth area SA5.

[0051] The captured image display monitor 3 is installed in the same position as the warning device 2. The captured image display monitor 3 displays the captured image output by the imaging device 10 in accordance with instruction information. The captured image display monitor 3 is sometimes referred to as a captured image display unit. The imaging device 10, which has output the detection result that serves as the basis for determining that there is a risk of collision, is likely to have captured images of the current location of the worker MO1 or the vehicle MO2 and their destination location. After recognizing the risk of collision through the warning device 2, the driver of the vehicle MO2 can determine the location of the worker MO1, who is at risk of collision, by looking at the captured image display monitor 3, which displays the captured image of the worker MO1. By determining the location of the worker MO1, the driver of the vehicle MO2 can narrow down the target to which he or she should pay attention in order to avoid a collision. The location of the worker MO1 can also be determined for the vehicle MO2, just as can the driver of the vehicle MO2.

[0052] The management device 4 is operated by an administrator of the collision risk notification system 1. The management device 4 outputs to the collision risk determination device 40 determination information including a determination criterion, a binary expression indicating whether or not there is a risk, a method of expressing the degree of collision risk such as the risk level, information about the detection area DA, information about the area SA, etc. Furthermore, the management device 4 outputs information to the collision risk determination device 40 instructing the collision risk determination device 40 to generate correspondence information for each image capture device 10 based on the information about the detection area DA and the information about the area SA and to output the information to the image capture device 10. The management device 4 also outputs to the image capture device 10 information used by the in-image destination estimation unit 16, such as the interval between the first time and the second time, the interval between the second time and the third time, and a threshold value used to estimate whether or not a moving object is moving.

[0053] 7 is a flowchart illustrating an example of the processing flow of the collision risk notification system 1 according to the first embodiment. The collision risk determination device 40 stores the map information output from the management device 4 in the map information storage unit 53 (step S101). Furthermore, the collision risk determination device 40 stores the area information output from the collision risk determination device 40 in the area information storage unit 54 (step S102). Furthermore, the collision risk determination device 40 stores the determination information output from the management device 4 in the determination information storage unit 52 (step S103).

[0054] The collision risk notification system 1 monitors whether a worker MO1 and a vehicle MO2 are present within the detection area DA using multiple image capture devices 10. If the worker MO1 and the vehicle MO2 are not detected by the multiple image capture devices 10 (step S104; No), the collision risk notification system 1 continues to monitor whether a worker MO1 and a vehicle MO2 are present.

[0055] If the worker MO1 and the vehicle MO2 are detected by multiple image capture devices 10 (step S104; Yes), the image capture device 10 that detected the moving object estimates the area SA where the detected moving object is located and the area SA to which the moving object will move (step S105). The collision risk determination device 40 determines the degree of risk of collision between the worker MO1 and the vehicle MO2 based on the detection results including information estimated by each image capture device 10 (step S106). If it is determined that there is no risk of collision (step S106; No), the collision risk notification system 1 continues to monitor whether there is a risk of collision between the worker MO1 and the vehicle MO2 by performing the processes of steps S104 to S106 again.

[0056] If it is determined that there is a risk of collision (step S106; Yes), the warning device 2 notifies the worker MO1 and the vehicle MO2 of the risk of collision by lighting up or sounding an alarm (step S107). Also, if it is determined that there is a risk of collision (step S106; Yes), the captured image display monitor 3 displays the captured image captured by the imaging device 10 that detected the worker MO1 or the vehicle MO2 (step S108).

[0057] An example of the flow of the process of registering a map and an area will be described with reference to Figures 8 to 10. The process shown in Figures 8 to 10 is one continuous process.

[0058] FIG. 8 is a sequence diagram illustrating an example of the flow of a process for registering a map. The management device 4 acquires authentication information for logging in through an operation by an administrator (step S201). The management device 4 outputs the acquired authentication information to the collision risk determination device 40 (step S202). The collision risk determination device 40 performs authentication processing based on the authentication information acquired from the management device 4 (step S203). When authentication is complete, the collision risk determination device 40 outputs display information including information on a menu screen to be displayed on the management device 4 to the management device 4 (step S204). The management device 4 acquires the display information from the collision risk determination device 40 and displays the menu screen (step S205). The management device 4 outputs map information indicating the layout, etc., of the detection area DA to the collision risk determination device 40 (step S206). The collision risk determination device 40 outputs the acquired map information to the map information storage unit 53 (step S207). The map information storage unit 53 stores the output map information (step S208). Hereinafter, storing the map information uploaded by the management device 4 in the map information storage unit 53 may be referred to as "registering a map."

[0059] FIG. 9 is a sequence diagram illustrating an example of the process flow for registering an area. The management device 4 instructs the collision risk determination device 40 to call up a map (step S209). In response to the instruction from the management device 4, the collision risk determination device 40 instructs the map information storage unit 53 to output map information (step S210). In response to the instruction from the collision risk determination device 40, the map information storage unit 53 outputs the map information to the collision risk determination device 40 (step S211). Based on the acquired map information, the collision risk determination device 40 outputs display information including information such as the location of obstacles within the detection area DA to the management device 4 (step S212). The management device 4 acquires the display information from the collision risk determination device 40 and displays the map (step S213). The administrator operates the management device 4 to set the area SA while viewing the displayed map. The management device 4 outputs area information regarding the area SA set by the administrator to the collision risk determination device 40 (step S214). The collision risk determination device 40 outputs area information regarding the area SA set within the detection region DA to the area information storage unit 54 (step S215). The area information storage unit 54 stores the output area information (step S216). Hereinafter, storing the area information uploaded by the management device 4 in the area information storage unit 54 may be referred to as "registering an area." The area information storage unit 54 outputs completion information indicating that the area SA has been registered to the collision risk determination device 40. When the collision risk determination device 40 acquires the completion information, it outputs display information to the management device 4 for displaying a result screen indicating that the registration of the area SA has been completed (step S217). The management device 4 displays a result screen indicating that the registration of the area SA has been completed, based on the display information output from the collision risk determination device 40 (step S218).

[0060] FIG. 10 is a sequence diagram illustrating an example of the flow of a process for registering a judgment criterion. The management device 4 outputs judgment information indicating the judgment criterion for the risk of collision determined by the administrator to the collision risk judgment device 40 (step S219). The collision risk judgment device 40 outputs the judgment information to the judgment information storage unit 52 (step S220). The judgment information storage unit 52 stores the output judgment information (step S221). Hereinafter, storing the judgment information uploaded by the management device 4 in the judgment information storage unit 52 may be referred to as "registering the judgment criterion." The judgment information storage unit 52 outputs completion information indicating that the judgment criterion has been stored to the collision risk judgment device 40. When the collision risk judgment device 40 acquires the completion information, it outputs display information to the management device 4 for displaying a result screen indicating that the registration of the area SA has been completed (step S222). The management device 4 displays a result screen indicating that the registration of the judgment criterion has been completed based on the display information output from the result screen (step S223).

[0061] 11 is a sequence diagram for explaining an example of the flow of processing by each device included in the collision risk notification system 1 according to the first embodiment. The risk determination unit 42 instructs the determination information storage unit 52 to output determination information in order to refer to the determination criterion (step S301). The determination information storage unit 52 outputs determination information indicating the determination criterion in accordance with the instruction from the risk determination unit 42 (step S302). The risk determination unit 42 acquires the determination information output from the determination information storage unit 52 (step S303).

[0062] The imaging device 10 detects a moving object in the detection area DA based on the captured image (step S304). The imaging device 10 also estimates the area SA where the detected moving object is currently located and the area SA to which the moving object will move (step S305). The imaging device 10 outputs the type of the detected moving object, the area SA where the moving object is currently located, and the area SA to which the moving object will move as detection results to the collision risk determination device 40. The detection result storage unit 51 included in the collision risk determination device 40 stores the detection results estimated by the imaging device 10 (step S306). The risk determination unit 42 acquires the detection results of each imaging device 10 by referring to the detection result storage unit 51. The risk determination unit 42 determines the degree of collision risk by comparing the relationship between the area SA where the worker MO1 is currently located, the area SA to which the worker MO1 will move, the area SA where the vehicle MO2 is located, and the area SA to which the vehicle MO2 will move, with a determination criterion (step S307).

[0063] When there is a risk of collision, the collision risk determination device 40 outputs warning information indicating the degree of risk of collision to the warning device 2. The warning device 2 notifies the worker MO1 and the vehicle MO2 that there is a risk of collision by sounding an alarm or turning on a light in response to the warning information (step S308).

[0064] Furthermore, when there is a risk of collision, the collision risk determination device 40 instructs the imaging device 10, which has captured an image of the worker MO1 and the vehicle MO2, to output the captured image to the captured image display monitor 3. In accordance with the instruction from the collision risk determination device 40, the imaging device 10 outputs the captured image that is currently being captured to the captured image display monitor 3 (step S309). The captured image display monitor 3 displays the captured image output from the imaging device 10 (step S310). [Embodiment 2]

[0065] The collision risk notification system 1A according to the second embodiment is similar to the collision risk notification system 1 according to the first embodiment in that it determines the risk of collision for multiple moving objects in a detection area DA and issues a notification if there is a risk of collision. The collision risk notification system 1A further differs from the collision risk notification system 1 in that it determines the degree of risk of collision based on the relationship between coordinates indicating the position within the detection area DA where the moving object is located at a second time and coordinates indicating the position within the detection area DA where the moving object is located at a third time.

[0066] 12 is a block diagram for explaining the functional configuration of a collision risk notification system 1A according to embodiment 2. The collision risk notification system 1A differs from the collision risk notification system 1 in that it further includes an imaging device 10A, a collision risk determination device 40A, and a position display monitor 5. Descriptions of matters already explained in embodiment 1 may be omitted.

[0067] The imaging device 10A differs from the imaging device 10 in that it further includes an intra-area coordinate specifying unit 61 and an intra-area destination specifying unit 62. The first storage unit 21 stores correspondence information in which coordinates indicating a position in the captured image, identifiers for identifying each of the plurality of areas SA, and coordinates indicating a position in the detection region DA are further associated with each other.

[0068] The image coordinate estimation unit 13 outputs the image coordinates of the moving object at the second time to the area coordinate specification unit 61. The intra-region coordinate specifying unit 61 acquires the intra-image coordinates of the moving object at the second time from the intra-image coordinate estimating unit 13. The intra-region coordinate specifying unit 61 also acquires correspondence information by referring to the first storage unit 21. The intra-region coordinate specifying unit 61 estimates coordinates indicating the position within the detection region DA where the moving object is located (hereinafter, may be referred to as intra-region coordinates) from the intra-image coordinates based on the correspondence information. The intra-region coordinate specifying unit 61 outputs information indicating the intra-region coordinates of the moving object at the second time to the detection result output unit 18.

[0069] The in-image destination estimation unit 16 outputs to the in-area destination specification unit 62 the in-image destination, which is the position in the captured image where the moving object is represented at the third time. The intra-area destination identification unit 62 acquires the intra-image destination from the intra-image destination estimation unit 16. The intra-area destination identification unit 62 also acquires correspondence information by referring to the first storage unit 21. The intra-area destination identification unit 62 estimates coordinates indicating the position within the detection area DA where the moving object is located at the third time (hereinafter, may be referred to as the intra-area destination) from the intra-image destination based on the correspondence information. The intra-area destination identification unit 62 outputs the intra-area destination of the moving object to the detection result output unit 18.

[0070] The detection result output unit 18 acquires the intra-area coordinates of the moving object at the second time from the intra-area coordinate specification unit 61. The detection result output unit 18 also acquires the intra-area destination of the moving object from the intra-area destination specification unit 62. The detection result output unit 18 outputs the type of moving object, the intra-area coordinates of the moving object at the second time, whether the moving object is moving or not, and the intra-area destination of the moving object as detection results to the collision risk determination device 40A.

[0071] The collision risk determination device 40A differs from the collision risk determination device 40 in that it further includes a generation unit 63, a generated information output unit 64, and a coordinate information storage unit 65.

[0072] The risk determination unit 42 further determines the degree of risk of a collision between worker MO1 and vehicle MO2 based on the absolute distance between the coordinates where worker MO1 is located and the coordinates where vehicle MO2 is located, and the absolute distance between the coordinates where worker MO1 is to move and the coordinates where vehicle MO2 is to move. Even if the worker MO1 and the vehicle MO2 are in the same area SA, the risk of collision is low if the location of the worker MO1 is far from the location of the vehicle MO2. Also, even if the worker MO1 and the vehicle MO2 are in adjacent areas SA, the risk of collision is high if the location of the worker MO1 and the location of the vehicle MO2 are close. Therefore, by using the coordinates within the area of the moving object at the second time and the destination of the moving object within the area, the risk determination unit 42 can more accurately determine the degree of risk of collision between the worker MO1 and the vehicle MO2.

[0073] The generation unit 63 refers to the detection result storage unit 51 and acquires the intra-area coordinates of each moving object present in the detection area DA at the second time. The generation unit 63 also acquires map information. The generation unit 63 generates information relating to an image showing the position of the moving object on a plan view showing the detection area DA, based on the intra-area coordinates of each moving object and the map information. The generation unit 63 outputs the generated information to the generation information output unit 64.

[0074] The generated information output unit 64 acquires the generated information from the generation unit 63. The generated information output unit 64 outputs the generated information to the position display monitor 5.

[0075] Coordinate information relating to coordinates indicating a position within the detection area DA is stored in the coordinate information storage unit 65. The coordinate information is output from the management device 4 in response to an operation by the administrator.

[0076] The position display monitor 5 may be provided in each moving object, or only in some moving objects. The position display monitor 5 may be provided in the vehicle MO2 so that the driver of the vehicle MO2 can view it. The position display monitor 5 may also be referred to as a position display unit. The position display monitor 5 displays an image showing the location of the moving object on a plan view showing the detection area DA based on the generated information output from the collision risk determination device 40A. FIG. 13 is a diagram showing an example of an image displayed on the position display monitor. In FIG. 13, an image showing the locations of three workers MO1 and one vehicle MO2 is displayed on a map showing the detection area DA. After the driver of the vehicle MO2 recognizes the risk of collision through the warning device 2, he or she can check the location of the worker MO1, who is at risk of collision, by looking at the position display monitor 5. By knowing the location of the worker MO1, the driver of the vehicle MO2 can narrow down the target to focus his or her attention on in order to avoid a collision.

[0077] FIG. 14 is a flowchart illustrating an example of the processing flow of the collision risk notification system 1A according to the second embodiment. The collision risk determination device 40A performs the processes from step S101 to step S103 and further stores the coordinate information output from the management device 4 in the coordinate information storage unit 65 (step S401). If the worker MO1 and the vehicle MO2 are detected by the multiple image capture devices 10A (step S104; Yes), the image capture device 10A estimates the area SA where the detected moving object is located, its coordinates within the area, and the area SA and its destination within the area to which the moving object will move (step S402). The collision risk determination device 40A outputs generation information generated based on the current coordinates within the area of the moving object identified by the image capture device 10A and map information to the position display monitor 5. The position display monitor 5 displays an image showing the location of the moving object on a plan view showing the detection area DA based on the generation information (step S403). The processing from step S106 onward is performed in the same manner as in the collision risk notification system 1 according to the first embodiment.

[0078] Fig. 15 is a sequence diagram for explaining an example of the flow of a process for registering coordinates. The process for registering coordinates is performed after calling up a map by the same processes as steps S201 to S213 shown in Fig. 8 and Fig. 9. After the process for registering coordinates, a determination criterion using the coordinates is uploaded by the same processes as steps S219 to S223 shown in Fig. 10. The administrator operates the management device 4 while viewing the displayed map to associate a position within the detection area DA with a coordinate. The management device 4 outputs coordinate information regarding the coordinates associated by the administrator to the collision risk determination device 40A (step S501). The collision risk determination device 40A outputs the coordinate information to the coordinate information storage unit 65 (step S502). The coordinate information storage unit 65 stores the output coordinate information (step S503). Hereinafter, storing the coordinate information uploaded by the management device 4 in the coordinate information storage unit 65 may be referred to as "registering coordinates." The coordinate information storage unit 65 outputs completion information indicating that the coordinates have been registered to the collision risk determination device 40A. When the collision risk determination device 40A acquires the completion information, it outputs display information to the management device 4 for displaying a result screen indicating that the registration of the coordinates has been completed (step S504). The management device 4 displays a result screen indicating that the registration of the coordinates has been completed, based on the display information output from the collision risk determination device 40A (step S505).

[0079] 16 is a sequence diagram illustrating an example of a process flow for displaying an image on a position display monitor. In order to refer to the map information, the generation unit 63 instructs the map information storage unit 53 to output the map information (step S601). The map information storage unit 53 outputs the map information in accordance with the instruction from the generation unit 63 (step S602). The generation unit 63 acquires the map information output from the map information storage unit 53 (step S603).

[0080] The imaging device 10A detects a moving object in the imaging area based on the captured image (step S604). Furthermore, the imaging device 10A estimates coordinates indicating the position of the moving object within the detection area DA based on the coordinates indicating the position where the moving object is located and the correspondence information (step S605). Furthermore, the imaging device 10A acquires time information indicating the time when the captured image was captured (step S606). The imaging device 10A outputs a detection result including the type of the detected moving object, coordinates indicating the position of the moving object within the detection area DA, and time information to the collision risk determination device 40A. The detection result storage unit 51 included in the collision risk determination device 40A stores the detection result output from the imaging device 10A (step S607). The generation unit 63 refers to the detection result storage unit 51 to acquire the type of moving object detected at the second time, coordinates indicating the position of the moving object within the detection area DA at the second time, and the type of the moving object. The generation unit 63 generates generated information based on the acquired information (step S608). The collision risk determination device 40A outputs the generated information generated by the generation unit 63 to the position display monitor 5. The position display monitor 5 displays an image showing the position of the moving object on a plan view showing the detection area DA (step S609). [Embodiment 3]

[0081] The collision risk notification system 1B according to the third embodiment is similar to the collision risk notification system 1A according to the second embodiment in that it determines whether there is a risk of collision between the worker MO1 and the vehicle MO2 based on the intra-area coordinates and the area SA. The collision risk notification system 1B differs from the collision risk notification system 1A in that it estimates the location of the moving body at a third time based on the temporal change in the intra-area coordinates of the moving body from a first time to a second time. The collision risk notification system 1B includes a plurality of image capture devices 10B, a collision risk determination device 40B, an alarm device 2, an captured image display monitor 3, a management device 4, and a position display monitor 5. Descriptions of matters described in the first and second embodiments may be omitted.

[0082] 17 is a block diagram for explaining the functional configuration of a collision risk notification system 1B according to embodiment 3. The imaging device 10B according to embodiment 3 differs from the imaging device 10A according to embodiment 2 in that it does not include the area identification unit 14, the intra-image destination estimation unit 16, the area destination identification unit 17, the intra-region destination identification unit 62, and the second storage unit 22. In other words, the imaging device 10B does not estimate the destination of the moving object at the third time.

[0083] The detection result output unit 18 acquires moving object information indicating the type of moving object from the intra-image coordinate estimation unit 13. The detection result output unit 18 also acquires intra-area coordinates indicating the position within the detection area DA where the moving object is located from the intra-area coordinate identification unit 61. The detection result output unit 18 also acquires time information indicating the time when the captured image was captured from the time acquisition unit 15. The detection result output unit 18 outputs the moving object information, intra-area coordinates, and time information as detection results to the collision risk determination device 40B.

[0084] The imaging device 10B does not estimate the area SA where the moving object is located at the second time or the area SA where the moving object is located at the third time. Therefore, the first storage unit 21 stores correspondence information in which coordinates indicating a position in the captured image are associated with coordinates indicating a position in the detection area DA, and does not need to store an identifier that identifies each area SA.

[0085] Collision risk determination device 40B differs from collision risk determination device 40A in that it includes a second storage unit 71 instead of detection result storage unit 51. Collision risk determination device 40B also differs from collision risk determination device 40A in that it further includes an intra-area destination estimation unit 72.

[0086] The detection result acquisition unit 41 outputs the acquired detection results to the second storage unit 71. The second storage unit 71 stores the detection results in association with identifiers that identify the imaging devices 10B that have transmitted the detection results.

[0087] The intra-area destination estimation unit 72 refers to the second storage unit 71 and acquires information relating to the type of moving object, the intra-area coordinates of the moving object, and time information, which are associated with each other for the moving object between the first time and the second time. The intra-area destination estimation unit 72 estimates coordinates indicating the position within the detection area DA where the moving object is located at the third time, based on the temporal change in the intra-area coordinates of the moving object between the first time and the second time acquired by referring to the second storage unit 71. The intra-area destination estimation unit 72 outputs the coordinates indicating the position within the detection area DA where the moving object is located at the third time to the risk determination unit 42.

[0088] The risk determination unit 42 refers to the second storage unit 71 to acquire the type of moving object and the in-area coordinates of the moving object at the second time. The risk determination unit 42 also acquires coordinates indicating the position within the detection area DA where the moving object is located at the third time from the in-area destination estimation unit 72. The risk determination unit 42 also acquires determination information related to the determination criteria by referring to the determination information storage unit 52. The risk determination unit 42 also acquires information indicating the area SA set as an area with a high risk of collision by referring to the area information storage unit 54 (for example, if the area SA is a rectangle, the coordinates of each of its four vertices). The risk determination unit 42 determines the degree of collision risk based on the relationship between the coordinates indicating the location of the worker MO1 and the coordinates indicating the location of the destination of the worker MO1, the coordinates of the location of the vehicle MO2 and the coordinates indicating the location of the destination of the worker MO2, whether these locations are included in the area SA, and the determination information. For example, if the absolute distance between the worker MO1 and the vehicle MO2 is shorter than the distance determined by the manager and the vehicle MO2 is moving, the risk determination unit 42 may determine the risk level to be 2, indicating an imminent risk of collision. Alternatively, if the vehicle MO2 is moving near the location of the worker MO1, the risk determination unit 42 may determine the risk level to be 1, indicating a risk of collision at the destination of the vehicle MO2. Furthermore, if the destination locations of the worker MO1 and the vehicle MO2 are included in a single area SA, the risk determination unit 42 may determine the risk level to be 1, indicating a risk of collision between the worker MO1 and the vehicle MO2 at the destination. By using the coordinates within the area and the destination location within the area, the risk determination unit 42 can determine the degree of risk of collision based on the exact positions of the worker MO1 and the vehicle MO2.

[0089] The risk determination unit 42 may also acquire information indicating the predicted routes that the worker MO1 and the vehicle MO2 will take between the second time and the third time from the intra-area destination estimation unit 72. In this case, the risk determination unit 42 may determine that there is a risk of collision when the predicted route of the worker MO1 intersects with the predicted route of the vehicle MO2, or when the predicted routes of the worker MO1 and the vehicle MO2 pass through area SA. When the predicted route of the worker MO1 intersects with the predicted route of the vehicle MO2, there is a risk of collision between the worker MO1 and the vehicle MO2 even if the location within the area where the worker MO1 is located and the destination location are far apart, and the location within the area where the vehicle MO2 is located and the destination location are far apart. By making a determination based on the locations of the moving objects at the second time and the third time, as well as their locations at times between the second time and the third time, it is possible to notify the moving objects of the risk of collision if there is a risk of collision between the second time and the third time.

[0090] FIG. 18 is a flowchart illustrating an example of the processing flow of the collision risk notification system 1B according to the third embodiment. The collision risk determination device 40B performs the same processes as in steps S101 to S104 and S401 to monitor whether the worker MO1 and the vehicle MO2 are present within the image capture areas of the multiple image capture devices 10B. If the worker MO1 and the vehicle MO2 are detected by the multiple image capture devices 10B (step S104; Yes), the image capture device 10B that detected the moving object identifies the intra-area coordinates based on the intra-image coordinates of the detected moving object (step S701). The image capture device 10B that detected the moving object also acquires the time at which the moving object was detected. The second storage unit 71 included in the collision risk determination device 40B stores the intra-area coordinates and the time at which the moving object was detected in association with each other (step S702). The intra-area destination estimation unit 72 estimates the intra-area destination where the moving object will be located at a third time based on the information stored in the second storage unit 71 (step S703). The position display monitor 5 displays an image generated based on the information stored in the second storage unit 71 and the map information (step S403). The processes from step S106 onwards are performed in the same manner as in the collision risk notification system 1 according to the first embodiment.

[0091] Fig. 19 is a sequence diagram for explaining an example of the processing flow of each device included in a collision risk notification system 1B according to embodiment 3. In the sequence diagram shown in Fig. 19, processing from step S801 to step S803 is performed instead of the processing of step S305 and step S306 shown in Fig. 11.

[0092] The risk determination unit 42 performs the same processes as in steps S301 to S303 to obtain determination information. The imaging device 10B detects a moving object in the detection area DA based on the captured image (step S304). The imaging device 10B also identifies the coordinates within the area based on the coordinates within the image of the detected moving object (step S801).

[0093] The imaging device 10B outputs the type of the detected moving object, the current coordinates of the moving object within the area, and time information to the collision risk determination device 40B. The second storage unit 71 included in the collision risk determination device 40B stores the type of the detected moving object, the coordinates of the moving object within the area, and the time information output from the imaging device 10B in association with each other (step S802).

[0094] The intra-area destination estimation unit 72 acquires the type of moving object and the intra-area coordinates of the moving object detected between the first time and the second time by referring to the second storage unit 71. The intra-area destination estimation unit 72 estimates coordinates indicating the position within the detection area DA where the moving object is located at the third time, based on the time change of the intra-area coordinates of the moving object between the first time and the second time (step S803).

[0095] The risk determination unit 42 acquires coordinates indicating the position within the detection area DA where the moving object is located at the third time and information indicating the type of the moving object from the in-area destination estimation unit 72. The risk determination unit 42 references the second storage unit 71 to acquire coordinates indicating the position within the detection area DA where the moving object is located at the second time and information indicating the type of the moving object. Based on the acquired information, the risk determination unit 42 determines whether there is a risk of collision between the worker MO1 and the vehicle MO2 (step S307). Thereafter, the collision risk notification system 1B performs the same processes as steps S308 to S310.

[0096] In the above description, the imaging devices 10, 10A, and 10B are shown as an example of AI cameras having functional units that process captured images. However, the present embodiment is not limited to this example, and the imaging devices 10, 10A, and 10B may be general cameras that only include the imaging unit 11 and an output unit that outputs captured images. In this case, the functional units other than the imaging unit 11 that are included in the imaging devices 10, 10A, and 10B may be included in the collision risk determination device 40.

[0097] Note that all or part of the functions of each unit of the image capture device 10, image capture device 10A, image capture device 10B, collision risk determination device 40, collision risk determination device 40A, and collision risk determination device 40B in the above-described embodiments may be realized by recording a program for realizing these functions on a computer-readable recording medium, and reading and executing the program recorded on the recording medium into a computer system. Note that the term "computer system" here includes hardware such as an OS and peripheral devices.

[0098] Although one embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to the above, and various design modifications can be made without departing from the spirit of the present invention. Furthermore, the configurations described in the above-described embodiments and examples can be combined. [Explanation of symbols]

[0099] 1...Collision risk notification system, 10...imaging device, 11...imaging unit, 12...captured image acquisition unit, 13...image coordinate estimation unit, 14...area identification unit, 15...time acquisition unit, 16...image destination estimation unit, 17...area destination identification unit, 18...detection result output unit, 21...first memory unit, 22...second memory unit, 31...instruction information acquisition unit, 32...captured image output unit, 40...collision risk determination device, 41...detection result acquisition unit, 4 2...Danger determination unit, 43...Warning information output unit, 44...Instruction information output unit, 51...Detection result storage unit, 52...Determination information storage unit, 53...Map information storage unit, 54...Area information storage unit, 61...Intra-area coordinate identification unit, 62...Intra-area destination identification unit, 63...Generation unit, 64...Generated information output unit, 65...Coordinate information storage unit, 72...Intra-area destination estimation unit, DA...Detection area, SA...Area, MO1...Worker, MO2...Vehicle

Claims

1. an image acquisition unit that acquires, from each of the plurality of imaging devices, an image of an area into which a detection region defined by the plurality of imaging devices is divided; an extraction unit that extracts a plurality of different moving objects from the captured images obtained from the plurality of image capture devices; a destination estimation unit that estimates to which area a future location of each of the extracted different moving bodies will fall based on a history of locations identified from captured images taken at different times; and a risk determination unit that determines, according to the estimated areas, that there is a risk of collision for moving bodies among the plurality of moving bodies whose areas estimated based on the future positions are close to each other; a notification unit that notifies the result of the determination by the risk determination unit; A collision risk notification system having:

2. The destination estimation unit The future location is estimated by extending the trajectory according to the location history. The collision risk notification system of claim 1 .

3. A collision risk notification system that notifies a user of a risk of collision between a first moving body and a second moving body using a plurality of imaging devices installed at different positions, a captured image acquisition unit that acquires a plurality of captured images captured by the moving object between a second time corresponding to a current time and a first time prior to the second time; an image coordinate estimation unit that estimates, based on the captured image, image coordinates that indicate the position of the captured moving object in the captured image, and whether the captured moving object is the first moving object or the second moving object; a time acquisition unit that acquires the time when the captured image was captured; a first storage unit that stores in advance, for each of the plurality of imaging devices, correspondence information in which coordinates indicating a position in the captured image captured by the imaging device are associated with identifiers that identify areas into which the areas captured by the plurality of imaging devices are divided based on the risk of collision between the first moving body and the second moving body; an area identification unit that identifies the area where the moving object is located at the second time from the image coordinates of the moving object at the second time estimated by the image coordinate estimation unit by referring to the correspondence information; a second storage unit that stores the in-image coordinates of the moving object estimated from the captured image by the in-image coordinate estimation unit, moving object information indicating whether the moving object is the first moving object or the second moving object, and a time when the captured image was captured, in association with each other; an in-image destination estimation unit that estimates coordinates indicating a position where the moving object will be displayed in the captured image at a third time point after the second time point, based on a time change from the first time point to the second time point of the in-image coordinates of the moving object stored in the second storage unit; an area destination identification unit that identifies the area where the moving object is located at the third time from coordinates indicating a position where the moving object is displayed in the captured image at the third time, which coordinates are estimated by the in-image destination estimation unit by referring to the correspondence information; a risk determination unit that determines the degree of risk of collision between the first moving body and the second moving body based on the relationship between the area identified by the area identification unit in which the first moving body is located at the second time, the area identified by the area destination identification unit in which the first moving body is located at the third time, the area identified by the area identification unit in which the second moving body is located at the second time, and the area identified by the area destination identification unit in which the second moving body is located at the third time; a notification unit that notifies a degree of collision risk based on the degree determined by the risk determination unit; A collision risk notification system.

4. The correspondence information is further associated with coordinates indicating a position within the area, an intra-region coordinate specifying unit that specifies coordinates indicating a position within the region where the moving object is located at the second time from the intra-image coordinates of the moving object at the second time estimated by the intra-image coordinate estimating unit by referring to the correspondence information; an intra-area destination identification unit that estimates coordinates indicating a position within the area where the moving object is located at the third time from coordinates indicating a position where the moving object is represented in the captured image at the third time, which coordinates are estimated by the intra-image destination estimation unit by referring to the correspondence information; Furthermore, The risk determination unit further determines the degree based on a relationship between coordinates indicating a position in the area where the first moving body is located at the second time, as specified by the intra-area coordinate determination unit, coordinates indicating a position in the area where the first moving body is located at the third time, as specified by the intra-area destination determination unit, coordinates indicating a position in the area where the second moving body is located at the second time, as specified by the intra-area coordinate determination unit, and coordinates indicating a position in the area where the second moving body is located at the third time, as specified by the intra-area destination determination unit. The collision risk notification system according to claim 3 .

5. A collision risk notification system that notifies a user of a risk of collision between a first moving body and a second moving body using a plurality of imaging devices installed at different positions, a captured image acquisition unit that acquires a plurality of captured images captured by the moving object between a second time corresponding to a current time and a first time prior to the second time; an image coordinate estimation unit that estimates, based on the captured image, image coordinates that indicate the position of the captured moving object in the captured image, and whether the captured moving object is the first moving object or the second moving object; a first storage unit configured to store in advance, for each of the plurality of imaging devices, correspondence information in which coordinates indicating a position in the captured image captured by the imaging device are associated with coordinates indicating a position in an area captured by the plurality of imaging devices; an intra-region coordinate specifying unit that specifies intra-region coordinates, which are coordinates indicating a position within the region where the moving object is located at the second time, from the intra-image coordinates of the moving object at the second time estimated by the intra-image coordinate estimating unit by referring to the correspondence information; a second storage unit that stores the coordinates within the area where the moving object identified by the coordinates within the area identification unit is located, moving object information indicating whether the moving object is the first moving object or the second moving object, and the time when the captured image was captured, in association with each other; an intra-area destination estimation unit that estimates coordinates indicating a position within the area where the moving object is located at a third time after the second time, based on a time change from the first time to the second time of the intra-area coordinates of the moving object stored in the second storage unit; a risk determination unit that determines the degree of risk of collision between the first moving body and the second moving body based on a relationship between coordinates indicating a position within the area where the first moving body is located at the second time, as identified by the intra-area coordinate determination unit, coordinates indicating a position within the area where the first moving body is located at the third time, as estimated by the intra-area destination estimation unit, coordinates indicating a position within the area where the second moving body is located at the second time, as identified by the intra-area coordinate determination unit, and coordinates indicating a position within the area where the second moving body is located at the third time, as estimated by the intra-area destination estimation unit, and whether at least two of the above-mentioned coordinates are included in any of areas, which are one or more areas set in the area based on the risk of collision between the first moving body and the second moving body; a notification unit that notifies a degree of collision risk based on the degree determined by the risk determination unit; A collision risk notification system.

6. The imaging device further includes a captured image display unit that displays captured images captured after the second time by the imaging device that captured the image of the first moving object at the second time and the imaging device that captured the image of the second moving object at the second time. The collision risk notification system according to any one of claims 3 to 5.

7. a third storage unit that stores information about the area; a generation unit that generates an image showing the position of each of the moving bodies within the area at the second time on a plan view showing a part or all of the area based on the information stored in the second storage unit and the information about the area stored in the third storage unit; a position display unit that displays the image generated by the generation unit.

6. A collision risk notification system according to claim 4 or 5.

8. A collision risk notification system that notifies a user of a risk of collision between a first moving body and a second moving body using a plurality of imaging devices installed at different positions, an image acquisition step of acquiring a plurality of images captured by the moving body between a second time corresponding to a current time and a first time preceding the second time; an image coordinate estimation step of estimating, based on the captured image, image coordinates that indicate the position of the captured moving object in the captured image, and estimating whether the captured moving object is the first moving object or the second moving object; a time acquisition step of acquiring a time when the captured image was captured; a first storage step of storing in advance in a first storage unit, for each of the plurality of imaging devices, correspondence information in which coordinates indicating a position in the captured image captured by the imaging device are associated with identifiers that identify areas into which the areas captured by the plurality of imaging devices are divided based on the risk of collision between the first moving body and the second moving body; an area specifying step of specifying the area where the moving object is located at the second time from the image coordinates of the moving object at the second time estimated by the image coordinate estimation step by referring to the correspondence information; a second storage step of storing the in-image coordinates of the moving object estimated from the captured image by the in-image coordinate estimation step, moving object information indicating whether the moving object is the first moving object or the second moving object, and the time at which the captured image was captured in a second storage unit in association with each other; an in-image destination estimation step of estimating coordinates indicating a position where the moving object will be displayed in the captured image at a third time point after the second time point, based on a time change from the first time point to the second time point of the in-image coordinates of the moving object stored in the second storage unit; an area destination identification step of identifying the area where the moving object is located at the third time from coordinates indicating the position where the moving object is represented in the captured image at the third time, which coordinates are estimated by the in-image destination estimation step by referring to the correspondence information; a risk determination step of determining the degree of risk of collision between the first moving body and the second moving body based on the relationship between the area identified by the area identification step in which the first moving body is located at the second time, the area identified by the area destination identification step in which the first moving body is located at the third time, the area identified by the area identification step in which the second moving body is located at the second time, and the area identified by the area destination identification step in which the second moving body is located at the third time; a notification step of notifying a degree of risk of collision based on the degree determined by the risk determination step; A collision risk notification method comprising:

9. A collision risk notification system that notifies a user of a risk of collision between a first moving body and a second moving body using a plurality of imaging devices installed at different positions, an image acquisition step of acquiring a plurality of images captured by the moving body between a second time corresponding to a current time and a first time preceding the second time; an image coordinate estimation step of estimating, based on the captured image, image coordinates that indicate the position of the captured moving object in the captured image, and estimating whether the captured moving object is the first moving object or the second moving object; a first storage step of storing in advance in a first storage unit, for each of the plurality of imaging devices, correspondence information in which coordinates indicating a position in the captured image captured by the imaging device are associated with coordinates indicating a position in an area captured by the plurality of imaging devices; an intra-region coordinate specifying step of specifying intra-region coordinates, which are coordinates indicating a position within the region where the moving object is located at the second time, from the intra-image coordinates of the moving object at the second time estimated by the intra-image coordinate estimating step by referring to the correspondence information; a second storage step of storing the coordinates within the area where the moving object identified by the coordinates within the area identification step is located, moving object information indicating whether the moving object is the first moving object or the second moving object, and the time at which the captured image was captured in a second storage unit in association with each other; an intra-area destination estimation step of estimating coordinates indicating a position within the area where the moving object is located at a third time after the second time, based on a time change from the first time to the second time of the intra-area coordinates of the moving object stored in the second storage unit; a risk determination process for determining the degree of risk of collision between the first moving body and the second moving body based on a relationship between coordinates indicating the position within the area where the first moving body is located at the second time, as identified by the intra-area coordinate determination process, coordinates indicating the position within the area where the first moving body is located at the third time, as estimated by the intra-area destination estimation process, coordinates indicating the position within the area where the second moving body is located at the second time, as identified by the intra-area coordinate determination process, and coordinates indicating the position within the area where the second moving body is located at the third time, as estimated by the intra-area destination estimation process, and whether at least two of the above-mentioned coordinates are included in any of the areas, which are one or more areas set in the area based on the risk of collision between the first moving body and the second moving body; a notification step of notifying a degree of risk of collision based on the degree determined by the risk determination step; A collision risk notification method comprising:

Images