Equipment verification system, entry / exit management system, and equipment verification method

Abstract

To provide an equipment confirmation system, an entrance / exit management system and equipment confirmation method capable of automatically confirming whether or not a worker of a construction site appropriately wear equipment.SOLUTION: The present invention relates to an equipment confirmation system 1 for confirming a condition of safety equipment 8 required for a worker. The equipment confirmation system comprises: an imaging section 12 for imaging the worker; a determination section for determining the condition of the safety equipment 8 based on image data which are imaged by the imaging section 12; and a warning section 13 which issues a warning in a case where it is determined that the condition of the safety equipment 8 is not appropriate. In a case where there is deficiency of the safety equipment 8 or a violation in wearing, the determination section determines that the condition of the safety equipment 8 is not appropriate.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to an equipment confirmation system, an entrance / exit management system, and an equipment confirmation method.

Background Art

[0002] Currently, various entrance / exit management systems according to purposes have been proposed. For example, those for the purpose of suppressing and preventing the occurrence of fraud and crime (see Patent Document 1), those for the purpose of preventing the spread of infectious diseases into buildings (see Patent Documents [2, 3]), etc. are known.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Patent Document 3

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, a system for confirming the status of equipment required by construction site workers has not been developed. Conventionally, humans visually confirm the presence or absence of equipment, and automation of confirmation is desired in terms of convenience and reliability. From such a perspective, the present invention provides an equipment confirmation system, an entrance / exit management system, and an equipment confirmation method that can automatically confirm whether construction site workers are appropriately wearing equipment.

Means for Solving the Problems

[0005] The equipment confirmation system according to the present invention is an equipment confirmation system for confirming the status of equipment required by workers. This equipment confirmation system, Over time, multiple timesA camera unit that photographs the aforementioned worker, and the camera unit that photographs the worker multiple The system includes a determination unit that determines the status of the equipment based on image data, and an alarm unit that issues an alarm if it is determined that the status of the equipment is inappropriate. The system detects defects or violations of the equipment's wearing status in multiple image data sets and performs statistical processing on the detection results to determine whether the equipment's condition is appropriate. In the equipment verification system according to the present invention, the status of the equipment can be determined from image data of the worker, eliminating the need for a human to visually check whether the equipment is being worn. Furthermore, since it can detect violations of proper equipment wearing (for example, holding the helmet in the hand or wearing a dust mask pulled down to the chin), a highly effective verification is possible. Furthermore, this method can reduce false positives compared to using only a single image file to determine equipment defects or violations of wearing regulations. For example, the determination unit may acquire skeletal coordinates indicating the position of the worker's skeleton as seen in the image data, and equipment coordinates indicating the position of the equipment as seen in the image data, and determine the violation of wearing the equipment based on the skeletal coordinates and the equipment coordinates. Furthermore, the determination unit may determine whether a violation of the wearing requirement has occurred by detecting a specific part of the worker's body.

[0006] The equipment verification system may include a data storage unit that stores image data when the equipment is determined to be in an inappropriate condition, and a verification terminal that acquires worker information from the data storage unit and displays it on a screen. In this way, workers whose equipment is in an inappropriate condition can be identified, making it easier for site managers to provide guidance to workers. Alternatively, the entry and exit management system may include the equipment verification system and a gate installed at least at one of the entrance and exit points of the work site. In this case, the camera unit photographs workers entering the work site or workers leaving the work site. The gate unit may close the entrance or exit point if it is determined that the equipment is not in order. This makes it possible to prevent workers with inadequate equipment from entering the work site. In addition, since equipment is checked when workers leave the work site, it is possible to make workers more strongly aware of the importance of wearing their equipment properly.

[0007] Furthermore, the equipment verification method according to the present invention is an equipment verification method for verifying the status of equipment necessary for a worker. This equipment verification method is Over time, multiple times A photographic process for photographing the aforementioned worker, and the photograph taken in the aforementioned photographic process multiple The system includes a determination step that determines the status of the equipment based on image data, and an alarm generation step that issues an alarm if it is determined that the status of the equipment is not appropriate. In the determination step, The system detects defects or violations of the equipment's wearing status in multiple image data sets and performs statistical processing on the detection results to determine whether the equipment's condition is appropriate. In the equipment verification method according to the present invention, the status of the equipment can be determined from image data of the worker, eliminating the need for a human to visually check whether the equipment is being worn. Furthermore, since it can detect violations of proper equipment wearing (for example, holding the helmet in the hand or wearing a dust mask pulled down to the chin), a highly effective verification is possible. Furthermore, this method can reduce false positives compared to determining equipment defects or violations based on a single image file. [Effects of the Invention]

[0008] According to the present invention, it is possible to automatically verify whether workers at a construction site are properly wearing their equipment. [Brief explanation of the drawing]

[0009] [Figure 1] This is a configuration diagram of an equipment verification system according to an embodiment of the present invention. [Figure 2] This is an illustrative diagram of an equipment verification system according to an embodiment of the present invention. [Figure 3] These are illustrative diagrams showing the results of the equipment assessment. (a) indicates a state where all safety equipment is determined to be properly worn, (b) indicates a state where a reflective vest and gloves are not worn, (c) indicates a state where a helmet is not properly worn, and (d) indicates a state where a dust mask is not properly worn. [Figure 4] This is an example flowchart illustrating the operation of the equipment verification system according to an embodiment of the present invention. [Modes for carrying out the invention]

[0010] Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the drawings as appropriate. Each figure is only a schematic representation to the extent that the present invention can be fully understood. Therefore, the present invention is not limited to the illustrated examples. In each figure, common or similar components are denoted by the same reference numerals, and their redundant descriptions are omitted. <Configuration of the equipment verification system according to the embodiment> The equipment verification system 1 according to the embodiment will be described with reference to Figures 1 and 2. Figure 1 is a configuration diagram of the equipment verification system 1 according to the embodiment. Figure 2 is an image diagram of the equipment verification system according to the embodiment. The equipment verification system 1 shown in Figure 1 is a system for verifying the status of equipment worn by workers at a construction site. Equipment includes various tools used at the construction site and items worn on parts of the body. Verifying the status of equipment means confirming that the equipment is worn correctly in the designated places according to its purpose and use. If the equipment status is not appropriate, it includes equipment defects and violations of wearing rules. Equipment defects are when the equipment itself is not worn, and violations of wearing rules are when the equipment is worn but not in the correct way or when the equipment is held in the hand. For example, if a worker is holding a helmet that should be worn on the head in their hand, it will be judged that the wearing method is incorrect (i.e., the equipment status is not appropriate).

[0011] The equipment verification system 1 can be used in various situations where it is necessary to verify the equipment of workers, and the equipment to be verified is not limited in terms of type, number, etc. In this embodiment, we assume that the equipment verification system 1 is used when entering a work site, and we will illustrate the case in which the status of safety equipment (referred to as "safety equipment") is checked. The equipment confirmation system 1 in this embodiment mainly includes a construction site system 10, a cloud system 20, and a management system 30. The construction site system 10 is a system constructed within the construction site, and the cloud system 20 and the management system 30 are systems constructed outside the construction site. The construction site system 10 and other systems are communicably connected via, for example, the Internet 2.

[0012] The construction site system 10 mainly includes a gate unit 11, a photographing unit 12, an alarm unit 13, a control unit 14, and a confirmation terminal 15. As shown in FIG. 2, the gate unit 11 is provided at the entrance of the work site, and workers entering the work site pass through the gate unit 11. Only workers wearing safety equipment 8 are permitted to enter the work site, and entry of other persons is not permitted. The safety equipment 8 in this embodiment is a helmet 8A, a dust mask 8B, a reflective vest 8C, and gloves 8D. The configuration and shape of the gate unit 11 are not particularly limited. The gate unit 11 in this embodiment is a sliding open / close door in the left-right direction. The gate unit 11 automatically opens when a person whose entry is permitted passes through, and remains closed when a person whose entry is not permitted attempts to pass through. That is, the gate unit 11 closes the entrance when the situation of the safety equipment 8 of the worker entering the site is inappropriate.

[0013] The photographing unit 12 is a video camera capable of acquiring digital images. The photographing unit 12 is installed near the entrance and photographs workers (sometimes referred to as "entrants") entering the work site. The photographing unit 12 in this embodiment is a network camera and is connected to the control unit 14 and the router 17 via a PoE (Power over Ethernet) hub 16. The photographed data taken by the photographing unit 12 is used to determine the situation of the safety equipment 8. The photographing unit 12 transmits the photographed video data to the control unit 14. The warning unit 13 notifies those around the gate unit 11 (including those entering the site) that a worker with inappropriate safety equipment 8 is attempting to enter. The warning unit 13 is preferably installed near the gate unit 11. The warning unit 13 preferably notifies those around the gate unit 11 of the inappropriate situation of the safety equipment 8 through vision or hearing. The warning unit 13 is, for example, a rotating light or a buzzer.

[0014] The control unit 14 controls the overall operation of the equipment confirmation system 1 (mainly the construction site system 10). The control unit 14 is, for example, a computer for IoT (Internet of Things) sensing provided inside the construction site (also called an "edge computer" etc.), and is connected to the gate unit 11, the imaging unit 12, the warning unit 13, the router 17, etc. via the PoE hub 16. The control unit 14 includes a determination unit 14a that determines the status of the safety equipment 8 based on the imaging data captured by the imaging unit 12. The determination unit 14a is realized, for example, by a hardware engine for AI (Artificial Intelligence) processing. The determination unit 14a has, for example, a posture detection AI and an object detection AI. The posture detection AI can obtain coordinates (skeleton coordinates) indicating the positions of the skeletons of the workers shown in an image (including video frames). On the other hand, the object detection AI can determine whether a specific object (here, mainly the safety equipment 8) is shown in an image (including video frames), and can also obtain coordinates (equipment coordinates) indicating the position of the identified object. The determination unit 14a determines the deficiencies and violations of wearing of the safety equipment 8 based on the two detection results of the posture detection AI and the object detection AI.

[0015] For example, the determination unit 14a uses object detection AI to determine the presence or absence of helmet 8A, and if helmet 8A is not visible along with the worker, it determines that helmet 8A is not being worn. The determination unit 14a also uses posture detection AI to identify the position (coordinates) of the worker's head, and object detection AI to identify the position (coordinates) of helmet 8A, and determines that helmet 8A is not being worn properly if the position of the head and the position of helmet 8A do not match. An example of a situation where the position of the head and the position of helmet 8A do not match is when the worker is holding helmet 8A in their hand. Furthermore, the determination unit 14a uses object detection AI to determine the presence or absence of the dust mask 8B, and determines that the dust mask 8B is not being worn if the dust mask 8B is not visible along with the worker. The determination unit 14a also uses object detection AI to determine whether the worker's nose is exposed, and determines that the dust mask 8B is not being worn properly if the nose is exposed even if the dust mask 8B is visible. Examples of not wearing the dust mask 8B properly include cases where the dust mask 8B is worn in a position such as being pulled down to the chin. Alternatively, the position (coordinates) of the nose and mouth may be identified using posture detection AI, and the position (coordinates) of the dust mask 8B may be identified using object detection AI, and if the position of the nose and mouth does not match the position of the dust mask 8B, it may be determined that the dust mask 8B is not being worn properly.

[0016] Furthermore, the determination unit 14a uses object detection AI to determine the presence or absence of the reflective vest 8C, and determines that the worker is not wearing the reflective vest 8C if the reflective vest 8C is not visible in the image along with the worker. Additionally, the determination unit 14a uses posture detection AI to identify the position (coordinates) of the worker's torso (e.g., chest or abdomen), and uses object detection AI to identify the position (coordinates) of the reflective vest 8C, and determines that the worker is not properly wearing the reflective vest 8C if the position of the torso and the position of the reflective vest 8C do not match. Furthermore, the determination unit 14a uses the object detection AI to determine the presence or absence of the glove 8D, and if the glove 8D is not visible along with the worker, it determines that the glove 8D is not being worn. Regarding the detection of the glove 8D, for example, the object detection AI could be trained to recognize human skin color, the posture detection AI could be used to identify the position (coordinates) of the worker's hand, and the object detection AI could be used to determine the color of the hand position to determine whether the glove 8D is being worn. This method allows for compatibility with various types of gloves. Additionally, the determination unit 14a uses the posture detection AI to identify the position (coordinates) of the worker's hand, and the object detection AI to identify the position (coordinates) of the glove 8D, and determines that the glove 8D is not being worn properly if the hand position and the glove 8D position do not match.

[0017] As shown in Figure 3(a), the determination unit 14a opens the gate unit 11 when it determines that all safety equipment 8 is being worn properly. Figure 3(a) shows the state in which it is determined that all safety equipment 8 is being worn properly. In Figure 3(a), the location of area 9A where the helmet 8A is detected corresponds to the head, the location of area 9B where the dust mask 8B is detected corresponds to the nose and mouth, the location of area 9C where the reflective vest 8C is detected corresponds to the torso, and the location of area 9D where the gloves 8D are detected corresponds to the hands. Furthermore, as shown in Figures 3(b) to (d), the determination unit 14a closes the gate unit 11 and activates the alarm unit 13 to notify that the safety equipment 8 is not being worn properly if it is determined that at least one piece of safety equipment 8 is not being worn properly. Figures 3(b) to (d) show the state in which it is determined that at least one piece of safety equipment 8 is not being worn properly. In Figure 3(b), the reflective vest 8C and gloves 8D are not detected, so it is determined that the safety equipment 8 is not being worn properly. In Figure 3(c), the helmet 8A is detected, and the location of the area 9A does not correspond to the head, so it is determined that the helmet is not being worn properly. In Figure 3(d), the dust mask 8B is detected, and the location of the area 9B does not correspond to the nose (only the mouth is covered, not the nose), so it is determined that the mask is not being worn properly.

[0018] The verification terminal 15 shown in Figure 1 has a web application that allows users to check the results of the safety equipment 8 in various formats. Only authenticated users can access terminal 15, and it is operated by, for example, a site manager. The web application has functions such as displaying images of violations, downloading images of violations, downloading monitoring results for a specified period, and drawing graphs by specifying deadlines and violation items (if there are multiple locations, comparison graphs between locations can be drawn). The cloud system 20 shown in Figure 1 stores image data when the status of the safety equipment 8 is determined to be inappropriate. The cloud system 20 is an example of a "data storage unit". The management system 30 is where the equipment verification system 1 is managed, and for example, a system administrator uses the management system 30 to perform system maintenance. The system administrator can, for example, retrain the posture detection AI and object detection AI, or change settings to improve detection accuracy. The management system 30 also collects information used by the web application on terminal 15 from the cloud system 20 and provides it to terminal 15.

[0019] ≪Operation of the equipment verification system according to this embodiment≫ Referring to Figure 4 (and Figures 1 to 3 as appropriate), an example of the operation of the equipment verification system 1 will be explained. Figure 4 is an example of a flowchart showing the operation of the equipment verification system 1. The process shown in Figure 4 begins when the equipment verification system 1 is started. After the system is started, the input of the video captured by the shooting unit 12 to the control unit 14 begins (step S1). The processes in steps S2 to S7 shown in Figure 4 are performed for each frame (image data) and are executed sequentially and repeatedly. The determination unit 14a of the control unit 14 processes the input frame (image data) using posture detection AI (step S2) to obtain, for example, the coordinates of the worker's skeleton (skeletal coordinates). The determination unit 14a also processes using object detection AI (step S3) to obtain, for example, the coordinates of the safety equipment 8 (equipment coordinates). Then, the determination unit 14a determines whether the worker is properly wearing the safety equipment 8 based on the detection results of both the posture detection AI and the object detection AI.

[0020] Next, the determination unit 14a performs statistical processing on the detection results of a predetermined number of frames, including the latest frame (image data), to obtain a value indicating the effectiveness of the detection results (step S4). For example, it determines what percentage of the predetermined number of frames were found to have a defect or violation of the safety equipment 8, and how many consecutive frames contained a defect or violation of the safety equipment 8. Subsequently, the determination unit 14a determines whether the value obtained in step S4 exceeds a threshold (step S5), and if it does not exceed the threshold ("No" in step S5), it proceeds to processing the next frame (image data). On the other hand, if the threshold is exceeded ("Yes" in step S5), the determination unit 14a activates the alarm unit 13 to generate an alert, indicating a high probability of a defect or violation of the safety equipment 8 (step S6), and also uploads the image data to the cloud system 20 (step S7). In this way, by performing the processing in steps S4 and S5, false detections can be suppressed compared to determining a defect or violation of the safety equipment 8 based on only one frame (image data).

[0021] As described above, in the equipment verification system 1 according to this embodiment, the status of the safety equipment 8 can be determined from image data of the worker, eliminating the need for a human to visually confirm the presence or absence of the safety equipment 8. Furthermore, since it can detect violations of wearing the equipment (for example, holding the helmet 8A in the hand or wearing the dust mask 8B pulled down to the chin), a highly effective verification is possible. While embodiments of the present invention have been described so far, the present invention is not limited thereto and can be implemented without changing the spirit of the claims. In this embodiment, the equipment verification system 1 was described assuming its use upon entry to the work site. However, it is also possible to use the equipment verification system 1 upon exiting the work site. In that case, the equipment verification system 1 functions as an "entry and exit management system." [Explanation of symbols]

[0022] 1. Equipment Verification System (Entry / Exit Management System) 8. Safety Equipment (Equipment) 8A Helmet (equipment) 8B Dust mask (equipped) 8C Reflective Vest (Equipment) 8D gloves 10 Construction Site Systems 11 Gate section 12 Photography Department 13 Alarm section 14 Control Unit 14a Judgment Unit 15 devices 16 PoE Hubs 17 Routers 20. Cloud System (Data Storage Section) 30 Management Systems

Claims

1. An equipment verification system that checks the status of the equipment required by workers, A camera unit that photographs the worker multiple times over time, A determination unit that determines the status of the equipment based on a plurality of image data captured by the aforementioned imaging unit, It includes an alarm unit that issues an alarm if it is determined that the condition of the aforementioned equipment is not appropriate, The determination unit detects defects or violations of the wearing of the equipment in a plurality of image data, and performs statistical processing on the detection results to determine whether the condition of the equipment is appropriate. A system for verifying equipment, characterized by the following features.

2. The determination unit acquires skeletal coordinates indicating the position of the worker's skeleton as seen in the image data, and equipment coordinates indicating the position of the equipment as seen in the image data, and determines a violation of the wearing requirement based on the skeletal coordinates and the equipment coordinates. The equipment verification system according to feature 1.

3. The determination unit determines the violation of the wearing requirement by detecting a specific part of the worker. The equipment verification system according to feature 1.

4. A data storage unit that stores image data when the condition of the aforementioned equipment is determined to be inappropriate, The system includes a confirmation terminal that retrieves the worker's information from the data storage unit and displays it on a screen. The equipment verification system according to feature 1.

5. The equipment verification system according to any one of claims 1 to 4, It comprises a gate section installed at least at one of the entrance and exit points of the work site, The aforementioned camera unit photographs workers entering the work site or workers leaving the work site. The gate section closes the entrance or exit if it is determined that the equipment is not in an appropriate condition. An access control system characterized by the following features.

6. A method for checking the status of equipment necessary for workers, A filming process in which the worker is filmed multiple times over time, A determination step in which the status of the equipment is determined based on a plurality of image data captured in the aforementioned shooting step, The system includes an alarm generation step that issues an alarm if it is determined that the condition of the aforementioned equipment is not appropriate, In the determination step, defects or violations of wearing the equipment are detected in a plurality of image data, and statistical processing is performed on the detection results to determine whether the condition of the equipment is appropriate. A method for verifying equipment, characterized by the following features.

Images