JP7574418B2 - Accident cause detection system, method, and program - Google Patents

Description

The present invention relates to an accident cause detection system, method and program that ensures the safety of workers by properly managing work tools stored in warehouses, etc.

Traditionally, the logistics industry has adopted a management method in which workers move and store luggage in a designated location, with the aim of properly managing luggage and ensuring the safety of workers. Here, the designated location could be, for example, a storage shelf, where luggage brought into a warehouse or the like is placed, and when it is time to ship it, it is taken out and taken out of the warehouse.

However, luggage stored in designated locations may leave those locations due to natural disasters or other circumstances. In such cases, luggage may be damaged by falling off shelves, or may block aisles, making it impossible to ensure the safety of workers.

To address this issue, Patent Document 1 provides technology that flies a drone equipped with a camera inside a warehouse to detect luggage that has fallen off a storage shelf.

JP 2020-55672 A

However, items stored in warehouses and the like also include work tools such as pallets, carts, hand trucks, forklifts, cardboard boxes, carts, and whiteboards, and these are generally not stored on storage shelves, so the technology in Patent Document 1 mentioned above cannot ensure the safety of workers.

Therefore, the present invention aims to provide an accident cause detection system, method, and program that can ensure the safety of workers, provided that work tools such as pallets are properly positioned and managed.

The present invention provides the following solutions:

The present invention provides an accident cause detection system that prevents a worker from causing an accident due to work tools, comprising an image acquisition unit that acquires image data within a predetermined range, a person detection unit that detects the worker based on the image data, an accident cause detection unit that detects a state in which at least a part of the work tool has come out of a predetermined location based on the image data, and a warning unit that issues a warning when it is predicted that the worker will cause an accident due to the accident cause.

The present invention also provides an accident cause detection system in which the accident cause detection unit determines that an accident has occurred when at least a portion of the work tool remains outside the location for a predetermined period of time.

The present invention also provides an accident cause detection system in which the person detection unit has an authentication function for identifying the worker, and the warning unit issues a warning to the identified worker if it is predicted that the worker identified by the authentication function will cause an accident.

The present invention also provides an accident cause detection system in which the person detection unit monitors the identified worker, and the warning unit issues a warning when the distance between the worker and the work tool becomes less than a predetermined distance.

The present invention also provides an accident cause detection system in which the person detection unit monitors the identified worker, and the warning unit issues a warning if contact is predicted within a predetermined time based on the relative speed between the work tool and the worker.

The present invention also provides an accident cause detection system, wherein the accident cause detection unit identifies the accident cause as being any one of a pallet, a dolly, a hand porter, a forklift, a cardboard box, a basket cart, and a whiteboard .

The present invention also provides an accident cause detection system that includes a mobile terminal carried by the worker and capable of receiving the warning.

The present invention also provides an accident cause detection system in which the warning unit transmits to the mobile terminal information that the cause of the accident is one of a pallet, a dolly, a hand porter, a forklift, a cardboard box, a basket cart, or a whiteboard.

The present invention also provides an accident cause detection system in which the accident cause detection unit acquires location information of the work tool, and the warning unit transmits the acquired location information of the accident cause to the mobile terminal.

Also, although the present invention is in the category of a system, it may also be a method and a program.

The present invention provides a system, method and program that can ensure the safety of workers, provided that work tools are properly positioned and managed.

1 is a block diagram showing an overall configuration and a functional configuration of an accident cause detection system according to an embodiment of the present invention. FIG. 2 is a diagram showing an accident cause detection processing flow according to the embodiment of the present invention. FIG. 2 is a diagram showing an example of a camera image according to an embodiment of the present invention. FIG. 11 is a diagram showing another example of a camera image according to the embodiment of the present invention. FIG. 11 is a diagram showing another example of a camera image according to the embodiment of the present invention. FIG. 13 is a diagram showing an example of a notification screen according to an embodiment of the present invention.

Hereinafter, a detailed description of the embodiment of the present invention will be given with reference to the accompanying drawings. In the following drawings, the same elements are designated by the same numbers or symbols throughout the description of the embodiment.

<Basic concept/basic configuration>
1 is a diagram showing the overall configuration and functional configuration of an accident cause detection system according to an embodiment of the present invention. The accident cause detection system 1 includes a camera 30 fixed in a warehouse or the like and a an accident factor detection device 100 for detecting accident factors; an administrator terminal 50 connected to the accident factor detection device 100 via a network 3 and receiving a warning issued by the accident factor detection device 100 when an accident is predicted; and an operator terminal 60. Although the camera 30 is described as a fixed camera in the embodiment, the camera 30 may be a mobile camera mounted on a drone or the like that moves within a warehouse or the like.

In the accident cause detection system 1, the camera 30 is fixedly installed at a predetermined location in a warehouse (not shown in the figure) in this embodiment. Work tools 10 such as pallets, carts, hand porters, forklifts, cardboard boxes, carts, and whiteboards that could be the cause of an accident involving a worker 20 are placed at predetermined locations in the warehouse.

In this embodiment, the administrator terminal 50 is a smartphone, but may be a tablet terminal or other network-connectable terminal. Similarly, in this embodiment, the worker terminal 60 is a smartphone, but may be a smart watch or other network-connectable terminal.

In this embodiment, the accident cause detection device 100 is a cloud server and is managed, for example, by a warehouse manager (a logistics company, etc.).

In this embodiment, the present invention is applied to the detection of accident factors in a warehouse. In a warehouse, predetermined work tools (pallets, dollies, hand porters, forklifts, cardboard boxes, basket carts, whiteboards, etc.) 10 are placed in predetermined positions and used for transporting luggage, etc., but they may move from their predetermined positions due to reasons such as insufficient organization, and the safety of the worker 20 may not be ensured. In view of this, the present invention ensures the safety of the worker 20 by properly managing the work tools 10. In this embodiment, the present invention is applied to the inside of a warehouse as an example, but the present invention is not limited to this, and can naturally be applied to a predetermined location outdoors.

The accident cause detection system 1 captures an image of a specified range and acquires image data from the camera 30. The image captured of the specified range may be a video or a still image. In addition, multiple cameras 30 may be fixedly installed in the warehouse so as to cover necessary locations in the warehouse.

Next, the accident cause detection system 1 detects the worker 20 based on the acquired image data. The detection of the worker 20 is performed by image analysis of the acquired image data. In addition, a facial image of the detected worker 20 is acquired and compared with a registration database 105 described below to identify the person (e.g., the worker 20) and monitor their movements.

The accident cause detection system 1 also detects, based on the acquired image data, a state in which at least a part of the work tool 10 has come out of a predetermined location. The work tool 10 is detected by image analysis of the acquired image data. At this time, the work tool is compared with preregistered work tool data of size and shape to identify the type of tool, such as a pallet, dolly, hand truck, forklift, cardboard box, basket cart, whiteboard, etc.

Next, the accident cause detection system 1 issues a warning when it is predicted that the worker 20 will cause an accident due to the accident cause. The warning may be transmitted to the worker terminal 60 carried by the worker 10, and may be conveyed by outputting a warning sound, displaying the warning content on a screen, or vibrating the worker terminal 60. A similar warning may also be transmitted to the manager terminal 50. Furthermore, the warning is not limited to these forms, and may be issued from a speaker installed near the accident cause.

<Functional configuration>
Next, a functional configuration of an accident factor detection system 1 according to an embodiment of the present invention will be described with reference to Fig. 1. The accident factor detection system 1 includes a camera 30 fixedly installed in a warehouse, an accident factor detection device 100 connected to the camera 30, a manager terminal 50 connected to the accident factor detection device 100 via a network 3, and a worker terminal 60 connected to the accident factor detection device 100 via the network 3.

In this embodiment, the accident factor detection device 100, the manager terminal 50, and the worker terminal 60 are described as separate devices, but the manager terminal 50 may be omitted and the accident factor detection device 100 may have the functions of the manager terminal 50 described below. Conversely, the accident factor detection device 100 may be omitted and the functions of the accident factor detection device 100 described below may be provided by the manager terminal 50, in which case the camera 30 is connected to the manager terminal 50.

(Functional configuration of the accident cause detection device)
The accident cause detection device 100 includes an image acquisition unit 101 , a person detection unit 102 , an accident cause detection unit 103 , a warning unit 104 , and a registration database 105 .

The image acquisition unit 101 acquires an image of a predetermined position in the warehouse. In detail, the image acquisition unit 101 acquires an image taken by a camera 30 fixed at a predetermined position in the warehouse. The image may be a video or a still image. Furthermore, the camera 30 and the image acquisition unit 101 may be directly connected or may be remotely connected.

The person detection unit 102 detects the worker 20 based on the acquired image data. Specifically, the worker 20 is detected by image analysis of the acquired image data. In addition, a facial image of the detected worker is acquired and compared with a registration database 105 (described later) to identify the person and monitor their movements. Note that the people registered in the registration database 105 include not only workers, but also visitors who come for the purpose of transactions or sightseeing.

The accident cause detection unit 103 detects, based on the acquired image data, a state in which at least a part of the work tool 10 has left a predetermined location. The work tool 10 is detected by image analysis of the acquired image data. At this time, the work tool is compared with data on the size and shape of the work tool registered in advance in the registration database 105, and the type of tool, such as a pallet, dolly, hand truck, forklift, cardboard box, basket cart, whiteboard, etc., is identified.

The warning unit 104 issues a warning when it is predicted that the worker 20 will cause an accident due to the accident cause. The warning is issued by transmitting the warning to the worker terminal 60 carried by the worker 20. In response to the transmitted warning, the worker terminal 60 outputs a warning sound or displays the contents of the warning on a screen. Alternatively, the warning may be conveyed by vibrating the worker terminal 60, or a warning may be similarly transmitted to the manager terminal 50. Of course, this does not preclude issuing warnings in other forms.

The registration database 105 stores images of a plurality of people that have been registered in advance, and is used by the person detection unit 102 when matching registered faces with images.
Furthermore, the registration database 105 stores image data in which the size and shape of the work tool 10 are registered in advance, and is collated when the accident cause detection unit 103 identifies the work tool.

(Functional configuration of administrator terminal)
The administrator terminal 50 is composed of, for example, a smartphone, tablet terminal, personal computer, etc., operated by each administrator, and is connected to the accident cause detection device 100 via the network 3 so as to be able to send and receive information, and is equipped with a terminal control unit 51, an input/output unit 52, and a communication unit 53.

The terminal control unit 51 receives, via the communication unit 53, a warning or the like when the warning unit 104 of the accident cause detection device 100 predicts that the worker 20 will have an accident due to an accident cause. The input/output unit 52 is composed of, for example, a touch panel, and accepts operations by the administrator, and displays or outputs a warning by the warning unit 104 received from the accident cause detection device 100 via the network 3 under the control of the terminal control unit 51. It may also display the type and location of the work tool, or the need for correction of the current situation.

(Functional configuration of the operator terminal)
The worker terminal 60 is carried and worn by each worker 20, and may be implemented, for example, as a smartphone or a smart watch. Each worker terminal 60 is connected to the accident cause detection device 100 via the network 3 so as to be able to send and receive information, and is equipped with a terminal control unit 61, an input/output unit 62, a communication unit 63, and a vibration unit 64.

The terminal control unit 61 receives, via the communication unit 63, a warning or the like when the warning unit 104 of the accident cause detection device 100 predicts that the worker 20 will have an accident due to an accident cause. The input/output unit 62 is, for example, composed of a touch panel, and accepts operations by the worker 20, and displays or outputs, for example, a warning by the warning unit 104 received from the accident cause detection device 100 via the network 3 under the control of the terminal control unit 61. For example, the type or location of the work tool 10 may be displayed, or the vibration unit 64 may vibrate the worker terminal 60 to convey the warning by vibration.

The above-mentioned functional configuration of the present system is merely an example, and one functional block may be divided, or multiple functional blocks may be combined into one functional block. Each functional processing unit is realized by a computer program executed by a CPU (Central Processing Unit) built into the device or terminal, which reads out a computer program (e.g., a core software or an application that causes the CPU to execute the above-mentioned various processes) stored in a storage device (storage unit) such as a ROM (Read Only Memory), a flash memory, an SSD (Solid State Drive), or a hard disk. In other words, each functional processing unit is realized by the computer program reading and writing necessary data such as tables from a database (DB; Data Base) stored in the storage device or a storage area on the memory, and in some cases controlling related hardware (e.g., an input/output device, a display device, a communication interface device).

<Processing flow>
2 is a diagram showing a flow of an accident cause detection process executed by the accident cause detection system according to the embodiment of the present invention. In this embodiment, the accident cause detection process is executed by the accident cause detection device 100, the manager terminal 50, and the worker terminal 60.

In step S10, the image acquisition unit 101 of the accident cause detection device 100 acquires the image captured by the camera 30.

In step S12, the person detection unit 102 of the accident cause detection device 100 detects a person from the camera image acquired by the image acquisition unit 101 through image analysis.

The accident cause detection unit 103 also detects work tools by image analysis from the camera image acquired by the image acquisition unit 101. The work tools 10 are detected by comparing them with work tool data of size and shape registered in advance in the registration database 105, and the type of tool, such as a pallet, dolly, hand truck, forklift, cardboard box, basket cart, whiteboard, etc., is identified.

In step S14, the person detection unit 102 acquires a facial image of the detected person and identifies the person by comparing it with the registration database 105. Then, the person detection unit 102 monitors the movement of the identified worker 20.

Next, in step S16, the accident cause detection unit 103 determines, based on the acquired image data, whether a predetermined time has elapsed with at least a part of the work tool 10 outside a predetermined location.

Figure 3 shows a camera image 32 in which the work tool 10 (a pallet in the illustrated example) is in a predetermined location 40 and is not likely to be a cause of an accident. Note that the predetermined location 40 is specified in advance by pixels from the image of the camera 30, but this does not prevent it from being specified by other methods.

If the work tool 10 is inside the predetermined location 40 as shown in Figure 3, step S16 is NO and the process flow proceeds to the end. Also, if at least a portion of the work tool 10 has left the predetermined location 40 and the predetermined time has not elapsed, step S16 is NO and the process flow proceeds to the end.

On the other hand, if a predetermined time has elapsed with at least a part of the work tool 10 having left the predetermined location as shown in FIG. 4, step S16 becomes YES, and the accident cause detection unit 103 detects that at least a part of the work tool 10 has left the predetermined location 40, and proceeds to step S18.

In addition, the determination of whether at least a part of the work tool 10 has left the predetermined place 40 may be performed, for example, as shown in FIG. 4, by image-recognizing the work tool 10, measuring the shortest distance D1 from the center position C of the recognized frame to the predetermined place 40, and if the shortest distance D1 is equal to or greater than a threshold, determining that at least a part of the work tool 10 has left the predetermined place 40. In addition, the determination of whether at least a part of the work tool 10 has left the predetermined place 40 may be performed by image-recognizing the work tool 10 and recognizing the shape of its bottom surface, and if the shape does not fit within the predetermined place 40, determining that at least a part of the work tool 10 has left. The shape of the bottom surface of the work tool 10 may be a shape registered in advance in the registration database 105 when the work tool is detected in S12. The shape may also be defined based on the pixels of the image.

Next, in step S18, the warning unit 104 determines whether the distance between the work tool 10, at least a portion of which has left the predetermined location 40, and the worker 20 remains below a predetermined distance for a predetermined time. Specifically, as shown in Fig. 5, the person detection unit 102 monitors the identified worker 20, and when the distance D2 between the worker 20 and the work tool 10 becomes equal to or less than a predetermined value, it determines that there is a danger and measures the time.

If it is determined in step S18 that the distance between the work tool 10, at least a portion of which has left the specified location 40, and the worker 20 is equal to or less than a specified value and that a specified time has elapsed, step S18 becomes YES and the process proceeds to step 20. In step 20, the warning unit 104 notifies the worker terminal 60 of a warning via the network 3.

In response to the warning notified by the warning unit 104, the worker terminal 60 may output a warning sound or display the contents of the warning on the screen. The warning may be transmitted by displaying the type and location of the work tool 10 on the input/output unit 62, or by vibrating the worker terminal 60 via the vibration unit 64. The upper part of Figure 6 shows an example of a display displayed on the input/output unit 62 of the worker terminal 60. As shown in Figure 6, the words "Approach detected" may be displayed, or a warning may be issued by vibration. Furthermore, a warning may be issued by combining these displays with vibration.

In step 22, the warning unit 104 may also notify the administrator terminal of a warning as necessary. The warning to the administrator terminal 50 may be sent to the administrator terminal 50 by outputting a warning sound, displaying the warning content on a screen, or vibrating the administrator terminal 50, or a combination of multiple types of warnings.

In addition to the warning, it may also be possible to transmit information indicating that the detected cause of the accident was a pallet, cart, hand truck, forklift, cardboard box, basket cart, whiteboard, etc., and it may also be possible to transmit the location information of the detected work tool 10.

The middle part of Fig. 6 shows an example of a display displayed on the input/output unit 52 of the administrator terminal 50. For example, in the case of a warning that requires correction, text such as "Empty pallet left at location: xx" may be displayed on the input/output unit 52, and a notification may be given by vibration upon detection.

In addition, the lower part of Fig. 6 shows examples of displays displayed on the input/output unit 52 of the administrator terminal 50, in which no correction is required. Characters such as "Approach detection location: xx" may be displayed on the input/output unit 52, and a notification may be given by vibration upon detection.

Returning to step S18, if the distance between the work tool 10, at least a portion of which has left the specified location 40, and the worker 20 is less than the specified value and the specified time has not elapsed, step S18 becomes NO and the processing ends.

In addition, in the above-mentioned step S18, the person detection unit 102 monitors the identified worker 20, and the warning unit 104 may issue a warning to the identified worker 20 if contact is predicted within a predetermined time based on the relative speed of the worker 20 to the work tool 10.

＜Effects＞
According to such an accident cause detection system 1, it is possible to determine whether or not at least a part of the work tool 10 has left a designated location 40 in a warehouse or the like, and to appropriately manage that the work tool 10 is positioned in the designated location 40, thereby ensuring the safety of the worker 20.

In addition, according to the accident cause detection system 1, the accident cause detection unit 103 determines that an accident has occurred when at least a part of the work tool 10 has been left outside the specified location for a specified period of time. This makes it possible to eliminate unnecessary notifications.

According to the accident cause detection system 1, the person detection unit 102 has an authentication function for identifying the worker 20, and the warning unit 104 issues a warning to the worker 20 identified by the authentication function when the accident is predicted. This calls the attention of the worker 20 and makes it possible to prevent accidents.

In addition, according to the accident cause detection system 1, the person detection unit 102 monitors the identified worker 20, and the warning unit 104 warns the identified worker 20 when the distance between the worker 20 and the work tool 10 becomes equal to or less than a predetermined distance. This makes it possible to alert the worker 20 and prevent accidents.

According to the accident cause detection system 1, the person detection unit 102 monitors the identified worker 20, and the warning unit 104 warns the identified worker 20 when contact is predicted within a predetermined time based on the relative speed between the work tool 10 and the worker 20. This makes it possible to alert the worker 20 and prevent accidents.

Furthermore, according to the accident cause detection system 1, the accident cause detection unit 103 identifies the accident cause as being one of a pallet, a cart, a hand porter, a forklift, a cardboard box, a basket cart, a whiteboard, etc.

In addition, the accident cause detection system 1 includes a worker terminal 60 carried by the worker that can receive the warning. This makes it possible to directly warn the worker 20 and encourage him or her to be careful.

Furthermore, according to the accident cause detection system 1, the warning unit 104 transmits to the worker terminal 60 that the accident cause is one of a pallet, a dolly, a hand porter, a forklift, a cardboard box, a basket cart, a whiteboard, etc.

According to the accident cause detection system 1, the accident cause detection unit 103 acquires the position information of the work tool 10, and the warning unit 104 transmits the acquired position information of the accident cause to the worker terminal 60. This alerts the worker 20 to prevent accidents such as falls.

Although the present invention has been described above using the embodiments, it goes without saying that the technical scope of the present invention is not limited to the scope described in the above embodiments. It is clear to those skilled in the art that various modifications or improvements can be made to the above embodiments. Furthermore, it is clear from the claims that such modifications or improvements can also be included in the technical scope of the present invention. In the above embodiments, the present invention has been described as an accident factor detection system as an invention of a product, but it can also be considered as an invention of a method executed by the accident factor detection system in the present invention, or a program that causes the accident factor detection system to function as various means. Furthermore, the detection to which the present invention can be applied includes not only inside a warehouse, but also all places where proper management of work tools is required, such as designated outdoor locations.

The accident cause detection system of the present invention may be configured as follows.
(1) An accident cause detection system that prevents a worker from causing an accident due to work tools, comprising: an image acquisition unit that acquires image data of a predetermined range; a person detection unit that detects the worker based on the image data; an accident cause detection unit that detects the work tool when at least a portion of it has come out of a predetermined location based on the image data; and a warning unit that issues a warning when it is predicted that the worker will cause an accident due to the accident cause.

(2) The accident cause detection unit is configured to determine that a state in which at least a portion of the work tool is outside the location continues for a predetermined period of time as a cause of an accident.

(3) The person detection unit has an authentication function for identifying the worker, and the warning unit is configured to warn the identified worker if the worker identified by the authentication function is predicted to cause an accident.

(4) The person detection unit monitors the identified worker, and the warning unit issues a warning when the distance between the worker and the work tool becomes less than a predetermined distance.

(5) The person detection unit monitors the identified worker, and the warning unit issues a warning when contact is predicted within a specified time based on the relative speed between the work tool and the worker.

(6) The accident cause detection unit is configured to identify the cause of the accident as being one of a pallet, a cart, a hand truck, a forklift, a cardboard box, a trolley, a whiteboard, etc.

(7) A configuration including a worker terminal carried by the worker that is capable of receiving the warning.

(8) The warning unit is configured to transmit to the mobile terminal information that the cause of the accident was one of a pallet, a cart, a hand truck, a forklift, a cardboard box, a basket cart, a whiteboard, etc.

(9) The accident cause detection unit acquires location information of the work tool, and the warning unit transmits the acquired location information of the accident cause to the mobile terminal.

Reference Signs List 1 Accident factor detection system 3 Network 10 Work tool 20 Worker 30 Camera 32 Camera image 40 Predetermined location 50 Manager terminal 51 Terminal control unit 52 Input/output unit 53 Communication unit 60 Worker terminal 61 Terminal control unit 62 Input/output unit 63 Communication unit 64 Vibration unit 100 Accident factor detection device 101 Image acquisition unit 102 Person detection unit 103 Accident factor detection unit 104 Warning unit 105 Registration database

Claims (11)

An accident cause detection system for preventing a worker from causing an accident due to a work tool, comprising:
an image acquisition unit that acquires image data of a predetermined range;
a person detection unit that detects the worker based on the image data;
an accident cause detection unit that detects the work tool at least partially protruding from a predetermined location based on the image data;
a warning unit that issues a warning when the worker is predicted to cause an accident due to the accident cause;
An accident cause detection system comprising: The accident cause detection unit is
2. The accident cause detection system according to claim 1, wherein when the state where at least a part of the work tool is out of the location continues for a predetermined period of time, it is determined that this is a cause of an accident. The person detection unit has an authentication function for identifying the worker,
3. The accident cause detection system according to claim 1, wherein the warning unit issues a warning to a worker identified by the authentication function when the worker is predicted to cause an accident. The person detection unit monitors the identified worker,
4. The accident cause detection system according to claim 3, wherein the warning unit issues a warning when a distance between the worker and the work tool becomes equal to or smaller than a predetermined distance. The person detection unit monitors the identified worker,
4. The accident cause detection system according to claim 3, wherein the warning unit issues a warning when a contact is predicted within a predetermined time from a relative speed between the work implement and the worker. The accident cause detection system according to any one of claims 1 to 5, characterized in that the accident cause detection unit identifies the accident cause as being one of a pallet, a cart, a hand porter, a forklift, a cardboard box, a basket cart, and a whiteboard . a mobile terminal carried by the worker and capable of receiving the warning;
7. The accident cause detection system according to claim 1, further comprising: 8. The accident cause detection system according to claim 7, wherein the warning unit transmits to the mobile terminal information that the cause of the accident is one of a pallet, a dolly, a hand porter, a forklift, a cardboard box, a basket cart, and a whiteboard . The accident cause detection unit acquires position information of the work tool,
9. The accident cause detection system according to claim 7, wherein the warning unit transmits the acquired location information of the accident cause to the mobile terminal. A method executed by an accident cause detection system for preventing a worker from causing an accident due to a work tool, comprising:
acquiring image data of a predetermined range;
detecting the worker based on the image data;
detecting the work implement at least partially exiting a predetermined location based on the image data;
a step of issuing a warning when the worker is predicted to cause an accident due to the accident cause;
An accident cause detection method comprising: An accident cause detection system that prevents workers from causing accidents due to work tools,
an image acquisition unit that acquires image data within a predetermined range;
a person detection unit that detects the worker based on the image data;
an accident cause detection unit that detects the work tool at least partially coming out of a predetermined location based on the image data;
a warning unit that issues a warning when it is predicted that the worker will cause an accident due to the accident cause;
An accident cause detection program that functions as a

Images