Safety monitoring system

Abstract

To efficiently determine safety equipment, a safety action, or a safety state.SOLUTION: A safety monitoring system includes: a photographing section 20 for photographing a worker or a working area; and a determination section 30 for determining at least one of safety equipment and a safety action of the worker or determining a safety state of a working area based on an image photographed by the photographing section 20. The determination section 30 starts the determination when the photographing section 20 photographs a trigger image.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present invention relates to a safety monitoring system. [Background technology]

[0002] Patent Document 1 describes a safety monitoring system that determines the safety equipment, etc., of workers. 12, the safety monitoring system 100 judges the safety equipment and safety behavior of the worker 400 from an image captured by a monitoring camera 200 as an imaging device. If at least one of the following is found to be inadequate in the safety equipment and / or failure to perform safe behavior by the worker 400, the system issues a warning by turning on a patrol lamp 300. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2023-48521 A Summary of the Invention [Problem to be solved by the invention]

[0004] At a work site, appropriate safety equipment and safety behavior are required depending on the content of the work. In the safety monitoring system 100 of Patent Document 1 and the like, the safety equipment and safety behavior of the worker 400 are constantly judged while the monitoring camera 200 is capturing images. For example, even when the worker 400 is performing work that does not require wearing protective equipment, a warning may be issued that the protective equipment is not being worn. For this reason, there is a demand for more efficient judgment. In addition, if the worker 400 becomes accustomed to such unnecessary warnings, the warnings may become meaningless. In addition, at a work site, there is a demand for efficient judgment of the safety state of the work area.

[0005] The present invention has been made in consideration of the above circumstances, and an object of the present invention is to provide a safety monitoring system that makes it possible to more efficiently determine safety equipment, safe actions, or safe conditions. [Means for solving the problem]

[0006] The safety monitoring system of aspect 1 includes an imaging unit that images a worker or a work area, and a judgment unit that judges at least one of the safety equipment and safety behavior of the worker from the image captured by the imaging unit, or judges the safety state of the work area, and the judgment unit starts the judgment when a trigger image is captured by the imaging unit.

[0007] According to this configuration, the judgment can be started when the trigger image is captured by the imaging unit. In the case of work that does not necessarily require safety monitoring, unnecessary judgment can be prevented by not capturing the trigger image. Therefore, the safety equipment, the safe behavior, or the safe state can be judged more efficiently.

[0008] A second aspect of the safety monitoring system of the first aspect further includes a notification unit that issues a warning when the judgment result by the judgment unit corresponds to at least one of a deficiency in the safety equipment and a failure to take the safe action, or when the judgment result indicates that the work area is not in a safe state. With this configuration, the issuance of the warning can suppress deficiencies in the safety equipment and failure to take the safe action by the worker. In addition, the worker can be prompted to take action to avoid danger.

[0009] In a third aspect, in the safety monitoring system of the first or second aspect, the determination unit continues to perform the determination while the trigger image is being captured by the imaging unit. With this configuration, at least one of the safety equipment and the safety behavior of the worker can be determined while the trigger image is being captured by the imaging unit. Also, the safety state of the work area can be determined while the trigger image is being captured by the imaging unit.

[0010] A fourth aspect of the present invention relates to a safety monitoring system according to any one of the first to third aspects, wherein the determination unit uses a plurality of AI models according to the work content. According to this configuration, it is possible to use different AI models according to the work content, thereby improving the accuracy of the determination result by the determination unit.

[0011] A fifth aspect of the present invention relates to a safety monitoring system according to any one of the first to fourth aspects, and when the imaging unit captures the trigger image, the determination unit sets a determination area and performs the determination within the determination area. With this configuration, by performing the determination within the determination area, it is possible to prevent unnecessary determinations from being made for workers or work areas outside the determination area.

[0012] A sixth aspect is a safety monitoring system according to any one of the first to fifth aspects, wherein the trigger image is a QR code (registered trademark). With this configuration, when the QR code is imaged by the imaging unit, the judgment is started. Therefore, for example, it becomes possible to start the judgment by having an operator present the QR code to the imaging unit. Effect of the Invention

[0013] According to the safety monitoring system of the present invention, safety equipment, safety actions, or safety conditions can be determined more efficiently. [Brief description of the drawings]

[0014] [Figure 1] FIG. 1 is a schematic diagram showing the configuration of a safety monitoring system according to the first embodiment. [Diagram 2] FIG. 2 is a schematic diagram showing the configuration of the determination unit of the first embodiment. [Diagram 3] FIG. 3 is a schematic diagram of a determination mechanism in the determination unit of the first embodiment. [Figure 4] FIG. 4 is a flowchart showing a flow of judgment using the safety monitoring system of the first embodiment. [Diagram 5]FIG. 5 is a schematic diagram showing a state in which a trigger image is captured by an imaging section in the safety monitoring system of the first embodiment. [Figure 6] FIG. 6 is a schematic diagram showing a determination result of a worker within a determination area in the safety monitoring system of the first embodiment. [Figure 7] FIG. 7 is an image showing a state in which it has been determined that a worker performing work is wearing safety equipment in the safety monitoring system of the first embodiment. [Figure 8] FIG. 8 is an image showing a state in which it has been determined that a worker working does not have safety equipment in the safety monitoring system of the first embodiment. [Figure 9] FIG. 9 is a schematic diagram showing a state in which a trigger image is captured by an imaging section in the safety monitoring system of the second embodiment. [Figure 10] FIG. 10 is a schematic diagram showing a state in which a trigger image is captured by an imaging section in the safety monitoring system of the modified example. [Figure 11] FIG. 11 is a schematic diagram showing a state in which a trigger image is captured by an imaging section in a safety monitoring system according to another modified example. [Figure 12] FIG. 12 is a schematic diagram showing the configuration of a safety monitoring system according to the prior art. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] First Embodiment A first embodiment of the safety monitoring system 10 will be described. In the first embodiment, the safety equipment and safety actions of the worker are determined from the image captured by the imaging unit.

[0016] 1 and 6, the safety monitoring system 10 includes an imaging unit 20 that captures an image of a work area A, and a judgment unit 30 that judges at least one of the safety equipment and the safety behavior of a worker 60 in the work area A from an image 21 captured by the imaging unit 20. The judgment unit 30 starts judgment when a trigger image is captured by the imaging unit 20. The safety monitoring system 10 also includes a notification unit 40 that issues a warning when the judgment result by the judgment unit 30 matches at least one of a deficiency in the safety equipment and a failure to perform a safe behavior. The judgment unit 30 also includes a memory unit 34 (see FIG. 2) that stores the judgment result.

[0017] The safety monitoring system 10 will be described in detail below. (Imaging unit) 1 and 5, the imaging unit 20 captures an image of the work area A. The imaging unit 20 outputs the captured image 21 to the determination unit 30. The imaging unit 20 is, for example, a camera installed in the work area A. The imaging unit 20 is not particularly limited, but an optical or electronic camera can be used. The image 21 captured by the imaging unit 20 may be a moving image or a still image.

[0018] (Judgment Department) As shown in FIG. 2, the determination unit 30 includes a CPU 31 that performs various arithmetic processing, a communication interface (also referred to as a communication I / F) 32 that transmits and receives signals to and from other components of the safety monitoring system 10, a storage 33, a memory unit 34 that stores image data 34a and the like, an input interface (also referred to as an input I / F) 35, and an output interface (also referred to as an output I / F) 36. The memory unit 34 stores, for example, image data 34a, thumbnail data 34b, and log data 34c. The determination unit 30 uses a plurality of AI models according to the work content of the worker to determine at least one of the safety equipment and the safety behavior of the worker from the captured image. The determination unit 30 outputs the result of the determination to the notification unit 40. The determination unit 30 is not particularly limited, but an edge device or a computer can be used.

[0019] (Notification Department) 1, the notification unit 40 notifies the result of the determination. The notification unit 40 is not particularly limited, but may be, for example, a speaker or a light.

[0020] (Determination mechanism of the determination unit) The determination mechanism by the determination unit 30 will be described. As shown in FIG. 3, the judgment unit 30 judges the state by detecting an "OK" state in which the safety equipment and the safety behavior have been confirmed, an "NG" state in which the non-compliance of at least one of the safety equipment and the safety behavior has been confirmed, and a "NO" state in which the trigger image has not been captured.

[0021] As shown in FIG. 6, for example, when a worker 60 transfers liquid from a drum 50 to a plastic container 61, the worker may be required to wear goggles 62 and gloves 63 as safety equipment.

[0022] In FIG. 3, the "OK" state indicates, for example, a state in which a drum 50 is captured as the trigger image and an operator 60 is wearing goggles 62 and gloves 63.

[0023] The "NG" state refers to a state in which, for example, drum 50 is captured as the trigger image and worker 60 is not wearing either goggles 62 or gloves 63. The "NO" state refers to a state in which, for example, the drum can 50 is not imaged.

[0024] The judgment unit 30 detects the "OK", "NG", and "NO" states every second, which is a unit time, and adopts the judgment result according to the number of times each state is detected. In the first embodiment, the detection time is set to 10 seconds, and detection is performed every second to make a judgment.

[0025] In FIG. 3, case 1 is shown in the upper part, case 2 is shown in the middle part below that, and case 3 is shown in the lower part below that. In Case 1, the "OK" state is detected for the first second of the 10-second detection time, and the "NO" state is detected for the rest. The number of times the "NO" state is detected is greater than the number of times the "OK" state is detected, and the "NO" state is detected last, so the result of the judgment at this time is "NO".

[0026] In Case 2, the "OK" state is detected for the first 7 seconds of the 10-second detection time, and the "NO" state is detected for the remaining 3 seconds. Although the number of times the "OK" state is detected is greater than the number of times the "NO" state is detected, the "NO" state is detected last, so the judgment result in this case is "NO".

[0027] From Case 1 and Case 2, even if an "OK" state is detected first during the 10-second detection time, if a "NO" state is detected at the end, the result of the judgment will be "NO". In case 3, the "OK" state is detected 5 times and the "NG" state is detected 5 times during the 10-second detection time. Since the number of times the "OK" state is detected is the same as the number of times the "NG" state is detected, and the "OK" state is detected last, the judgment result in this case is "OK."

[0028] From Case 3, if the number of times that the "OK" state and the "NG" state are detected within the detection time are the same, and the "OK" state is detected last, the judgment result will be "OK". In addition, even if an "OK" state is detected last, if the number of times that the "OK" state is detected is less than the number of times that the "NG" state is detected within the 10-second detection time, the judgment result will be "NG".

[0029] The determination by the determination section 30 starts when the trigger image is captured, and continues while the trigger image is being captured. Furthermore, the safety equipment is not limited to goggles 62 and gloves 63, and known safety equipment can be used. Examples of known safety equipment include dust masks, hats, and safety boots. In addition to the safety equipment, safe behavior can also be determined. For example, in the case of a task of transferring liquid from a drum can 50 to a plastic container 61, a specific example of safe behavior can be the use of a dedicated jig 64 for transferring liquid (see FIG. 8).

[0030] (Judgment process using the safety monitoring system) The flow of judgment using the safety monitoring system 10 will be described. A flowchart is shown in Fig. 4. Also, schematic diagrams of an image 21 captured by the imaging unit 20 are shown in Figs.

[0031] (Step S101) Step S101 is a step in which the imaging unit 20 captures an image of a work area A such as a manufacturing site. The imaging unit 20 outputs the captured image 21 to the determination unit 30. If a worker 60 is already present in the work area A, the imaging unit 20 captures an image of the worker 60 in the work area A at the same time as capturing an image of the work area A. That is, the worker 60 may be captured in the image of the work area A.

[0032] (Step S102) Step S102 is a classification step in which the determination unit 30 classifies the image 21 captured by the imaging unit 20 as to whether or not a trigger image has been captured. The trigger image is not particularly limited, but is, for example, a drum 50 containing liquid. If a trigger image has been captured, the process proceeds to the next step S103. If a trigger image has not been captured, the process returns to the previous step S101. It is not necessary for the entire trigger image to be captured, and the image may be classified as having a trigger image when only a portion of the trigger image has been captured.

[0033] (Step S103) Step S103 is a step of setting a determination area B. In step S103, a certain range around the analyzed trigger image is set as a determination area B.

[0034] 5, for example, when the trigger image is a drum 50, the drum 50 is recognized as the trigger image and is displayed so as to be surrounded by a frame 22. In addition, a certain range around the frame 22 surrounding the drum 50, i.e., the trigger image, is set as a judgment area B and is displayed so as to be surrounded by a frame 23.

[0035] The range of the judgment area B is not particularly limited, but can be set appropriately, for example, within a radius of 5 m, within a radius of 3 m, within a radius of 1 m, etc., centered on the drum 50 that is the trigger image.

[0036] (Step S104) Step S104 is a step of analyzing the image 21 in which the trigger image is captured. The judgment unit 30 detects the safety equipment and safety behavior of the worker 60 in the judgment area B using multiple AI models according to the work content. Using multiple AI models according to the work content means using AI models that correspond individually to multiple types of work content. For example, if the work content is to transfer liquid from a drum can 50 to a plastic container 61, an AI model that corresponds to this work content is used. That is, it means using multiple AI models according to the work content. By using multiple AI models according to the work content, the accuracy of the judgment result in the judgment unit 30 can be improved.

[0037] As shown in Figs. 6 and 7, when a worker 60 is inside the frame 23 set as the judgment area B, the wearing state of the worker 60, for example, goggles 62 and gloves 63 are analyzed. An image of the worker 60 is analyzed, and the areas where the goggles 62 and gloves 63 should be checked are displayed surrounded by frames 24 and 25, respectively. Note that Fig. 6 shows a case where the analysis of the wearing state of the goggles 62 has been completed, but the analysis of the wearing state of the gloves 63 has not been completed. In Fig. 6, the worker 60 is wearing the gloves 63, so the circumference of the arms of the worker 60 is also surrounded by the frame 25 (see Fig. 7).

[0038] As shown in Fig. 7, when a worker 60 in the judgment area B has appropriate safety equipment, "Set", which means that the equipment is being worn, is displayed in the frames 24 and 25. When this display is displayed, the judgment result for the detection in the unit time becomes "OK".

[0039] As shown in Fig. 8, if a worker 60 in judgment area B does not have safety equipment, "Out", meaning that the equipment is not worn, is displayed in frames 24 and 25. When this display is displayed, the judgment result for the detection in the unit time is "NG." Also, if a worker 60 in judgment area B does not use a dedicated tool 64 for transferring liquid from drum can 50 to plastic container 61, that is, if the dedicated tool 64 is not detected, the worker is not performing safe behavior and the result is "NG."

[0040] It should be noted that detection of safety equipment and safe behavior is not performed for the worker 60 outside the judgment area B. Also, when the worker 60 who was in the judgment area B leaves the judgment area B, detection of the safety equipment and safe behavior is terminated.

[0041] (Step S105) Step S105 is a step for judging the safety equipment and safe behavior of the worker 60 in the judgment area B. Judgment is made by the judgment mechanism described in FIG. 3, and if the judgment result is "NG", that is, at least one of a deficiency in the safety equipment and a failure to perform a safe behavior is confirmed, the "NG" result is output to the notification unit 40 and the process proceeds to the next step S106. If the judgment result is "OK", that is, the safety equipment and the safe behavior are confirmed, the judgment ends. At that time, the "OK" result may be output to the notification unit 40. If the judgment result is "NO", the judgment also ends.

[0042] (Step S106) Step S106 is a step in which a warning is issued by the notification unit 40. If the result of the determination is "NG", the notification unit 40 issues a warning to the operator 60.

[0043] As shown in Fig. 1, the warning is not particularly limited, but may be, for example, turning on a red lamp 40a. Also, an alarm may be issued. When the judgment result is "OK" in the notification unit 40 of Fig. 1, a green lamp 40b may be turned on to notify that both the safety equipment and the safe behavior have been confirmed. In other words, when the judgment result of the judgment unit 30 is "OK", it may be notified that the judgment is good.

[0044] Through the above process, the safety equipment and the safe behavior of the worker 60 can be judged. Since the image 21 captured by the imaging unit 20 is output to the judgment unit 30 in real time, the judgment by the judgment unit 30 is continuously performed while the trigger image is being captured by the imaging unit 20.

[0045] <Action and Effects> The operation and effects of the first embodiment will be described. (1-1) The safety monitoring system 10 includes an imaging unit 20 that captures an image of a work area A, and a judgment unit 30 that judges at least one of the safety equipment and the safety behavior of a worker 60 from an image 21 captured by the imaging unit 20. When a trigger image is captured by the imaging unit 20, the judgment unit 30 sets a judgment area B and starts judgment within the judgment area B. According to this configuration, the judgment can be started when the trigger image is captured by the imaging unit 20. In work that does not necessarily require safety monitoring, unnecessary judgment can be prevented by not capturing a trigger image. In addition, unnecessary judgment can be suppressed for the worker 60 outside the judgment area B. Therefore, the safety equipment and the safe behavior can be judged more efficiently.

[0046] (1-2) The system includes a notification unit 40 that issues a warning when the judgment result by the judgment unit 30 matches at least one of a deficiency in safety equipment and a failure to take safe actions. With this configuration, a warning is issued to the worker 60, so that the worker 60 can be prevented from having a deficiency in safety equipment or from taking safe actions.

[0047] (1-3) The determination unit 30 continues to perform the determination while the trigger image is being captured by the imaging unit 20. According to this configuration, while the trigger image is being captured by the imaging unit 20, at least one of the safety equipment and the safety behavior of the worker 60 can be determined.

[0048] (1-4) The judgment unit 30 uses a plurality of AI models according to the work content of the worker 60. According to this configuration, it is possible to use different AI models depending on the work content of the worker 60, so that the accuracy of the judgment result in the judgment unit 30 can be improved.

[0049] <Second embodiment> A second embodiment of the safety monitoring system 10 will now be described. 9, the second embodiment assumes a case where, for example, a worker 60 is required to wear a helmet 65, gloves 66, a full harness 67, etc. as safety equipment when working at a height. The imaging unit 20 is, for example, a camera of a tablet terminal.

[0050] As shown in Fig. 9, in the second embodiment, the imaging unit 20 captures an image of a worker 60. In addition, in the second embodiment, a QR code 70 is adopted as a trigger image. The determination by the determination unit 30 is started when the QR code 70 is captured. When the QR code 70 is captured, a determination is made regarding at least one of safety equipment and safety behavior for the worker 60 in the image 21 captured by the imaging unit 20.

[0051] 9, a QR code 70 is displayed on a plate 71 such as an employee ID card or an admission card. When a worker 60 holds the plate 71 over the imaging unit 20, the QR code 70 can be captured as an image 21a including the plate 71. Alternatively, the image 21a including the plate 71 may be captured by the imaging unit 20 while the worker 60 is wearing the plate 71.

[0052] It is preferable that the QR code 70 is linked to information about the worker 60. If the QR code 70 is linked to information about the worker 60, it becomes easier to identify the worker 60 when checking the determination results later. Alternatively, the imaging unit 20 may be configured to acquire an image 21a including the plate 71, so that the worker 60 can be identified from the plate 71.

[0053] Before moving to the high altitude of the work area A, the worker 60 may make a judgment using the safety monitoring system 10 installed on the ground, and move to the high altitude after obtaining an "OK" judgment. Alternatively, the worker 60 may make a judgment using the safety monitoring system 10 installed in the work area A after moving to the high altitude of the work area A, and start work after obtaining an "OK" judgment. If the judgment result is "NG", a warning may be issued by the notification unit 40. The tablet terminal used as the imaging unit 20 may be used as the notification unit 40, or the notification unit 40 may be installed separately.

[0054] In the second embodiment, a specific example of the safety behavior is, for example, a hook (not shown) of the full harness 67 being locked at a specific locking position. <Action and Effects> The operation and effects of the second embodiment will be described.

[0055] (2-1) The trigger image is a QR code 70. According to this configuration, when the QR code 70 is imaged by the imaging unit 20, the judgment is started. Therefore, for example, it is possible to start the judgment by having the worker 60 present the QR code 70 to the imaging unit 20. Therefore, it is possible to more efficiently judge safety equipment and safety actions.

[0056] <Example of change> The first embodiment and the second embodiment (hereinafter, collectively referred to as the present embodiment) can be modified as follows. The present embodiment and the following modified examples can be combined with each other to the extent that there is no technical contradiction.

[0057] The judgment area B may be set to an area that is a certain distance away from the trigger image. For example, when the worker 60 transports a heavy object, the worker 60 may be required to use a dolly as a safety action.

[0058] As shown in FIG. 10, in this modification, an area away from the dolly 51 is set as a judgment area B, and is displayed so as to be surrounded by a frame 23. The range of the judgment area B is not particularly limited, and can be set appropriately, for example, to a radius of 1 m or more and 3 m or less, or a radius of 1 m or more and 5 m or less, with the dolly 51 serving as the trigger image at the center. When the worker 60 is within the frame 23 set as the judgment area B, the judgment unit 30 notifies the result of the judgment of "NG". In other words, since the worker 60 is transporting a heavy object without using the dolly 51, the judgment result of "NG" is notified.

[0059] In the first embodiment, the safety equipment and safety behavior of the worker 60 are determined from the image of the work area A captured by the imaging unit 20, but the present invention is not limited to this. The safety state of the work area A may be determined from the image of the work area A captured by the imaging unit 20.

[0060] As shown in FIG. 11, for example, when an image is captured showing the presence of risk factor C within determination area B, it may be determined that the state is unsafe. Risk factor C is not particularly limited, but examples include industrial vehicles such as forklifts and turret trucks. That is, if an industrial vehicle is present near drum 50, there is a risk of it coming into contact with drum 50. Therefore, if an industrial vehicle is present within determination area B, it is determined that the state is unsafe. Determination unit 30 may make a determination using multiple AI models according to the work content in work area A.

[0061] Also, a warning may be issued that the state is not safe, and the worker 60 may be prompted to take action to avoid the danger. One example of action to avoid the danger is to move the industrial vehicle out of the judgment area B so as not to come into contact with the drum 50. According to this modification, the safety state of the work area A can be efficiently judged. Both the safety equipment and safe actions of the worker 60 and the safety state of the work area A may be judged.

[0062] In the present embodiment, when a trigger image is captured by the imaging unit 20, the determination unit 30 sets the determination area B and performs a determination on the worker 60 within the determination area B, but this is not limited to the embodiment. The determination unit 30 does not need to set the determination area B. The determination unit 30 may perform a determination on the worker 60 within the image 21 and a determination on the safety state for the entire image 21 captured by the imaging unit 20.

[0063] The trigger image is not limited to the drum 50 or the QR code 70. Other than the drum 50, an object essential for the work, such as a pallet or a flexible container bag, may be set as the trigger image. Also, other than the QR code 70, a barcode or a data matrix code may be set as the trigger image. A QR code may be displayed on the drum 50 as the trigger image.

[0064] In the safety monitoring system 10 of the present embodiment, the judgment unit 30 is capable of judging both the safety equipment and the safe behavior of the worker 60, but is not limited to this aspect. The judgment unit 30 may be configured to judge only one of the safety equipment and the safe behavior of the worker 60.

[0065] In the safety monitoring system 10 of the present embodiment, the judgment unit 30 includes the storage unit 34 that stores the judgment result, but the present invention is not limited to this. For example, the safety monitoring system 10 may include a storage unit that stores the judgment result in addition to the judgment unit 30.

[0066] The safety monitoring system 10 of this embodiment may have a display unit that displays the judgment result. For example, the display unit may display the judgment result or the image 21 on which the judgment was made on the web. Furthermore, the judgment results may be displayed in a list by date, etc. The display unit may be an edge device, a computer display, or the like. By having the display unit, it becomes easier to check the judgment result.

[0067] In the present embodiment, the notification unit 40 notifies the worker 60 of the determination result, but this is not limited to the embodiment. The notification unit 40 may also notify a supervisor of the determination result. The notification unit 40 may be omitted. That is, it is not necessary to notify a warning.

[0068] In the present embodiment, the determination unit 30 performs detection for the determination mechanism every second, which is a unit time, with a detection time of 10 seconds, but is not limited to this. The detection time and unit time can be set to any appropriate time.

[0069] The determination section 30 may continue making a determination for a certain period of time even after the imaging section 20 has stopped capturing a trigger image. The determination unit 30 does not have to continue the determination even if a trigger image is captured by the imaging unit 20. For example, the determination unit 30 may end the determination for an operator 60 that has already been determined.

[0070] The determination unit 30 does not have to use an AI model. For example, the determination unit 30 may be configured to perform the determination by an algorithm possessed by a computer. In this embodiment, multiple AI models are used depending on the work content, but a single AI model may be used. [Explanation of symbols]

[0071] 10...safety monitoring system, 20...imaging unit, 21...image, 30...judgment unit, 60...worker.

Claims

1. A safety monitoring system, comprising: An imaging unit that images a worker or a work area, and a determination unit that determines at least one of the safety equipment and safety behavior of the worker from the image captured by the imaging unit, or determines the safety state of the work area, The safety monitoring system according to claim 1, wherein the determination unit starts the determination when a trigger image is captured by the imaging unit.

2. The safety monitoring system of claim 1, further comprising a notification unit that issues a warning when the judgment result by the judgment unit matches at least one of a deficiency in the safety equipment and a failure to observe safe behavior, or when the judgment result indicates that the work area is not in a safe state.

3. The safety monitoring system according to claim 1 , wherein the determination section continues to perform the determination while the trigger image is being captured by the imaging section.

4. The safety monitoring system according to claim 1 , wherein the judgment unit uses a plurality of AI models according to the work content.

5. The safety monitoring system according to claim 1 , wherein the determination section sets a determination area when the trigger image is captured by the imaging section, and performs the determination within the determination area.

6. The security monitoring system according to claim 1 , wherein the trigger image is a QR code.

Images