Visual inspection system

Abstract

A visual inspection system (100) can be obtained that can improve the accuracy of visual inspections of fire prevention equipment (200) installed in a fire monitoring environment. [Solution] The visual inspection system (100) is a system for inspecting the appearance of disaster prevention equipment (200) installed at a high place, and comprises an unmanned aerial vehicle (1) equipped with an imaging unit (13) that approaches the disaster prevention equipment (200) and flies around it, a remote controller (2) for operating the unmanned aerial vehicle (1), and a display unit (24) for displaying the captured image taken by the imaging unit (13), wherein the imaging unit (13) is mounted on the unmanned aerial vehicle (1) so as to be able to capture the disaster prevention equipment (200) from the horizontal direction.

Description

Technical Field

[0001] This disclosure relates to an appearance inspection system for inspecting the appearance of disaster prevention equipment.

Background Art

[0002] Maintenance work such as operation confirmation and appearance inspection is regularly performed on disaster prevention equipment installed in buildings.

[0003] In addition, some disaster prevention equipment such as fire detectors and sprinklers is installed at high places such as the ceiling of a room in a fire monitoring environment.

[0004] As a conventional technique for confirming the operation of a fire detector installed at such a high place, the following is disclosed (for example, see Patent Document 1). The unmanned aircraft according to Patent Document 1 approaches the fire detector while confirming the position of the fire detector using an imaging unit, and generates a simulated fire using a simulated fire generation device after approaching. Thereby, it is possible to confirm whether the fire detector operates correctly.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] Appearance inspection, which is one of the maintenance work, is mainly performed by visual inspection of workers. Here, in a fire monitoring environment, when disaster prevention equipment is installed at a high place such as the ceiling, the object is located far away, so the accuracy of visual appearance inspection is reduced. In addition, when disaster prevention equipment is installed on the ceiling of a place with dim indoor lighting such as a concert hall, the accuracy of appearance inspection is further reduced.

[0007] Furthermore, depending on the shape of the disaster prevention equipment, there may be parts that cannot be seen from below. For this reason, when conducting visual inspections of disaster prevention equipment installed at high altitudes or other distant locations, it can be difficult to inspect every detail.

[0008] The unmanned aerial vehicle described in Patent Document 1 is used to verify whether a fire detector works correctly using a simulated fire generation device, and is not used for visual inspection.

[0009] This disclosure is made to solve the above-mentioned problems and aims to provide a visual inspection system that can improve the accuracy of visual inspections of fire prevention equipment installed in a fire monitoring environment. [Means for solving the problem]

[0010] The visual inspection system relating to this disclosure is a visual inspection system for inspecting the appearance of fire prevention equipment installed in a fire monitoring environment, and comprises an unmanned aerial vehicle equipped with an imaging unit that approaches the fire prevention equipment and flies around it, a remote controller for operating the unmanned aerial vehicle, and a display unit for displaying the images captured by the imaging unit, wherein the imaging unit is mounted on the unmanned aerial vehicle so as to be able to capture images of the fire prevention equipment from a horizontal direction. [Effects of the Invention]

[0011] According to this disclosure, it is possible to obtain a visual inspection system that can improve the accuracy of visual inspections of fire prevention equipment installed in a fire monitoring environment. [Brief explanation of the drawing]

[0012] [Figure 1] This figure shows the configuration of the visual inspection system in Embodiment 1 of the present disclosure. [Figure 2] Figure 1 is a block diagram showing an example configuration of the visual inspection system. [Figure 3] This diagram shows the appearance of a fire detector to be inspected, and illustrates an example of how it looks when installed on the ceiling of a room. [Figure 4] This diagram illustrates how the protective filter shown in Figure 3 looks when viewed from below and when viewed from the side. [Figure 5] This is a flowchart showing the operation of the visual inspection system in Embodiment 1 of this disclosure. [Modes for carrying out the invention]

[0013] Hereinafter, preferred embodiments of the visual inspection system of this disclosure will be described with reference to the drawings. The visual inspection system described herein performs a visual inspection of fire prevention equipment installed in a fire monitoring environment by approaching the equipment with an unmanned aerial vehicle and taking images with its imaging unit.

[0014] In particular, in fire monitoring environments, if there are recessed areas on the sides of fire prevention equipment installed at a distance, such as at a high altitude, it becomes difficult to visually inspect the inside of these recesses from below. The visual inspection system disclosed herein is characterized by its ability to perform visual inspections of such areas by imaging them from a horizontal direction. This makes it possible to improve the accuracy of visual inspections, especially for fire prevention equipment installed at a distance.

[0015] Furthermore, in the following embodiment, the captured image captured by the imaging unit is displayed on the display unit, and the operator visually inspects the captured image displayed on the display unit to perform a visual inspection. Alternatively, a method may be used in which a pass / fail judgment is made by image analysis processing of the captured image, and the judgment result is obtained as the result of the visual inspection.

[0016] Furthermore, the following embodiments describe an example in which a fire detector, which is one of the fire prevention devices, is the object of inspection. Note that the embodiments described below can be applied to various fire prevention devices installed in fire monitoring environments, such as sprinklers, foam fire extinguishing heads, and red lights.

[0017] Embodiment 1. FIG. 1 is a diagram showing the configuration of the appearance inspection system according to Embodiment 1 of the present disclosure.

[0018] The appearance inspection system 100 includes a drone 1 and a remote controller 2, and inspects the appearance of a fire detector 200 installed on the ceiling C of a room.

[0019] The drone 1 is a flying object that autonomously maintains its attitude while flying. The drone 1 performs wireless communication with the remote controller 2, approaches the fire detector 200 according to an instruction from the remote controller 2, and flies around it.

[0020] The drone 1 is equipped with an imaging unit 13. The imaging unit 13 is provided above or in front of the main body of the drone 1, and is provided so that the fire detector 200 can be imaged at least from the horizontal direction when the drone 1 is stationary in the air.

[0021] The remote controller 2 is a control device for operating the drone 1 by manual operation of an operator. The remote controller 2 includes a display unit 24, and causes the display unit 24 to display the captured image captured by the imaging unit 13. Thereby, the operator can perform an appearance inspection of the fire detector 200 by visually observing the display unit 24 while operating the drone 1.

[0022] In Embodiment 1, it is assumed that the remote controller 2 is a dedicated device having a shape and function specialized for the operation of the drone 1. On the other hand, a device in which a predetermined operation program is introduced into a smartphone, a tablet PC, a notebook PC, etc. may be used as the remote controller 2.

[0023] Also, in Embodiment 1, the display unit 24 is configured to be integrated with the remote controller 2, but the display unit 24 may be prepared separately.

[0024] Alternatively, an external server capable of communicating with the unmanned aerial vehicle 1 or remote controller 2 may be provided within the visual inspection system 100, and various calculation processes such as image processing and analysis processing may be performed on the external server.

[0025] Figure 2 is a block diagram showing an example configuration of the visual inspection system 100 shown in Figure 1.

[0026] In Figure 2, the unmanned aerial vehicle 1 comprises a communication unit 11, a control unit 12, an imaging unit 13, an illumination unit 14, a storage unit 15, and an autonomous flight control unit 16.

[0027] The communication unit 11 communicates wirelessly with the remote controller 2. In this embodiment 1, the communication unit 11 transmits data in a specific frequency band such as the 2.4GHz band. The communication unit 11 also receives operation signals from the remote controller 2 using the FH (Frequency Hopping) method and transmits captured images to the remote controller 2 using the FM (Frequency Modulation) method.

[0028] The control unit 12 has a configuration that includes an arithmetic processing unit, a main memory, and an interface for controlling various hardware. The control unit 12 performs various arithmetic processing and controls various hardware within the unmanned aerial vehicle 1 according to commands from the remote controller 2.

[0029] The imaging unit 13 images fire detectors 200 and other disaster prevention equipment. As described above, the imaging unit 13 is installed on the unmanned aerial vehicle 1 so that it can image the fire detectors 200 from the horizontal direction while the unmanned aerial vehicle 1 is stationary in the air. This allows the imaging unit 13 to image the side of the fire detector 200 from a direction facing the side of the fire detector 200.

[0030] The lighting unit 14 is an LED light that illuminates the area around the unmanned aerial vehicle 1. The lighting unit 14 is used to inspect fire detectors 200 installed in areas with low light levels. The lighting unit 14 is also used to illuminate parts of the fire detectors 200 that are in shadow due to their structure.

[0031] If the illumination light from the lighting unit 14 reflects off the surface of the fire detector 200 (resulting in so-called "overexposure"), the imaging unit 13 will either take multiple images while adjusting its exposure, or the unmanned aerial vehicle 1 will be moved to a position where it does not reflect light.

[0032] The memory unit 15 is a secondary memory device that stores parameters for flying the unmanned aerial vehicle 1. The memory unit 15 also records images captured by the imaging unit 13 during flight and records operation logs.

[0033] The autonomous flight control unit 16 is a module that controls the unmanned aircraft 1 so that it can autonomously maintain its attitude during flight. The autonomous flight control unit 16 consists of an inertial measurement unit, a rotor motor, an autonomous control circuit, and the like. The autonomous control circuit is a circuit that includes a motor driver that controls the rotor motor based on the values ​​measured by the inertial measurement unit.

[0034] The remote controller 2 comprises a communication unit 21, a control unit 22, an operation unit 23, a display unit 24, and a storage unit 25.

[0035] The communication unit 21 transmits data to the unmanned aerial vehicle 1 using a frequency band such as the 2.4GHz band. The communication unit 21 then transmits control signals to the unmanned aerial vehicle 1 using the FH method and receives captured images from the unmanned aerial vehicle 1 using the FM method.

[0036] The control unit 22 includes a processing unit, main memory, etc., and controls various hardware within the remote controller 2. The control unit 22 also performs various calculations, such as predicting the replacement timing, which will be described later.

[0037] The control unit 23 includes an operating stick and accepts operator input. The display unit 24 includes a liquid crystal monitor and displays captured images transmitted from the unmanned aerial vehicle 1.

[0038] The memory unit 25 stores analysis programs, parameters, and other necessary data for performing visual inspections and predicting replacement timing (described later). The memory unit 25 also stores captured images transmitted from the unmanned aerial vehicle 1.

[0039] Figure 3 shows the external appearance of the fire detector 200 to be inspected, and illustrates an example of its condition when installed on the ceiling C inside a room.

[0040] The fire detector 200 shown in Figure 3 is a photoelectric spot-type detector equipped with a photoelectric smoke detection unit inside, and monitors whether or not fire-related smoke is generated in the room being monitored. When a specified amount of smoke or more enters the internal smoke detection unit, the fire detector 200 outputs a fire occurrence signal to a receiver (not shown), illuminates a red light 212, and issues an audible alarm notification.

[0041] Furthermore, in Figure 3, the cover 210 constitutes the exterior of the fire detector 200. The cover 210 is composed of a main body cover 210A, a top cover 210B, and a plurality of connecting parts 210C that connect the main body cover 210A and the top cover 210B.

[0042] The opening 230, enclosed by the main body cover 210A, the top cover 210B, and the connecting part 210C, is an inlet for the smoke detection unit to take in air, and is equipped with a mesh protective filter 211. By providing this protective filter 211, it is possible to prevent foreign matter such as dust and insects from entering the smoke detection unit. Here, an insect screen can be used as the protective filter 211.

[0043] The protective filter 211 is installed on the fire detector 200 such that its surface faces horizontally when the fire detector 200 is installed on the ceiling C of the room.

[0044] Over time, dirt 251 accumulates on the surface of the cover 210, and foreign matter 252 such as dust accumulates on the surface of the protective filter 211.

[0045] If a significant amount of foreign matter 252 adheres to the protective filter 211, it will obstruct the entry of smoke particles into the smoke detection section, reducing the smoke detection performance of the fire detector 200. Therefore, in the visual inspection of the fire detector 200, it is important to inspect the protective filter 211 for contamination, as well as the cover 210.

[0046] Figure 4 illustrates how the protective filter 211 shown in Figure 3 looks when viewed from below and when viewed from the side.

[0047] The cover 210, which is the outer casing of the fire detector 200, needs to cover and protect all the components that make up the fire detector 200. Therefore, the top cover 210B, which is a part of the cover 210, is configured to protrude outward more than the protective filter 211, and as a result, the protective filter 211 is installed in a relatively recessed position inward.

[0048] Even if one attempts to view the protective filter 211, which is installed in such a recessed position, from a downward position L1, the protective filter 211 is obscured by the top cover 210B, making it difficult to see the entire filter.

[0049] On the other hand, when viewing the fire detector 200 horizontally from position L2, there is nothing obstructing the protective filter 211, so the entire protective filter 211 can be seen.

[0050] Therefore, in Embodiment 1, the unmanned aerial vehicle 1 is used to image the fire detector 200 from the horizontal direction, and the imaged protective filter 211 is visually inspected by a worker through the display unit 24. This allows a worker on the ground to visually inspect the protective filter 211 of the fire detector 200, which is installed at a distance, down to its smallest detail.

[0051] Figure 5 is a flowchart showing the operation of the visual inspection system 100 in Embodiment 1 of this disclosure.

[0052] In step S101, the remote controller 2 of the visual inspection system 100 moves the unmanned aerial vehicle 1 closer to the fire detector 200. This approach operation is performed manually by an operator, and is carried out by the remote controller 2 outputting a control signal to the unmanned aerial vehicle 1.

[0053] In step S102, the remote controller 2 activates the imaging unit 13 based on manual operation by the operator, and begins imaging the fire detector 200. As a result, the image of the fire detector 200 captured by the imaging unit 13 is displayed on the display unit 24 of the remote controller 2.

[0054] Next, in step S103, the visual inspection system 100 performs a dirt inspection of the cover 210 within the range that can be confirmed from the current position of the unmanned aircraft 1. This dirt inspection of the cover 210 may be performed by an operator visually inspecting the captured image via the display unit 24, or a qualified person may remotely check the captured image via a server or the like. Alternatively, the dirt inspection of the cover 210 may be performed by image analysis processing for dirt inspection. When the visual inspection is performed by image analysis processing, the display unit 24 displays the judgment result.

[0055] Next, in step S104, if the visual inspection system 100 can confirm the red light 212 from the current position of the unmanned aircraft 1, it performs a lighting inspection of the red light 212 by visual inspection or image analysis processing. This lighting inspection is performed, for example, by transmitting a simulated fire signal from the receiver to the fire detector 200. When the lighting inspection is performed by image analysis processing, the display unit 24 displays the determination result as described above.

[0056] Next, in step S105, the visual inspection system 100 inspects the protective filter 211 for contamination by visual inspection or image analysis, within the range that can be confirmed from the current position of the unmanned aerial vehicle 1. At this time, the unmanned aerial vehicle 1 is stationary in a position where the fire detector 200 can be imaged from the horizontal direction, while imaging the protective filter 211. When the inspection is performed by image analysis, the display unit 24 displays the judgment result as described above.

[0057] In the process described in steps S103 to S105 above, if the inspection is to be performed in an area with insufficient light, the illumination unit 14 is used to illuminate the area around the unmanned aircraft 1 and the object to be inspected.

[0058] Then, in step S106, if the inspection of the fire detector 200 from all directions has not been completed, the remote controller 2 of the visual inspection system 100 flies the unmanned aircraft 1 to a position where it can image the uninspected direction of the fire detector 200 in step S107. After that, the visual inspection system 100 returns to step S102 and continues to perform a visual inspection of the fire detector 200 at the uninspected position.

[0059] In other words, after imaging of the fire detector 200, if there are any directions that have not been imaged, the unmanned aerial vehicle 1 flies to a location where it can image those unimaged directions. Then, the imaging unit 13 images the fire detector 200 from those unimaged directions.

[0060] In step S106, once the inspection of the fire detector 200 from all directions is complete, the remote controller 2 predicts in step S110 when it is time to replace the fire detector 200 based on the current level of dirt accumulation on the cover 210 and protective filter 211.

[0061] The remote controller 2 determines the degree of contamination by comparing images of the fire detector 200 taken in the past with images taken in the current image, and predicts when the fire detector 200 should be replaced. Alternatively, the replacement time for the fire detector 200 may be predicted using AI with a model that has been trained on the degree of contamination, or it may be predicted using conventional statistical methods.

[0062] After step S110, the flowchart shown in Figure 5 is completed. If there is another fire detector 200 in the facility, the worker performs the process shown in Figure 5 for that other fire detector 200.

[0063] The data captured by the imaging unit 13 is to be transmitted to the remote controller 2 in video format, but in this case, the amount of data to be transmitted will be large. Also, saving it as video data will result in a large file size. Therefore, the unmanned aerial vehicle 1 may combine multiple still images or individual frames of the video into a 360-degree panoramic image before transmitting it to the remote controller 2. Alternatively, the remote controller 2 may receive the unprocessed imaging data and process it into a 360-degree panoramic image.

[0064] Furthermore, it is conceivable that the fire detector 200 may be installed on a sloped ceiling rather than a horizontal ceiling. In order to perform a high-precision visual inspection in such cases as well, a PTZ camera may be used as the imaging unit 13, and imaging may be performed by tilting it up and down along the horizontal or sloped ceiling surface. In other words, the imaging unit 13 may be mounted on the unmanned aerial vehicle 1 so that imaging can be performed from a direction along the ceiling surface on which the fire detector 200 is installed.

[0065] The features of this visual inspection system 100 can be summarized as follows, and it will be able to achieve its intended effects.

[0066] The visual inspection system 100 is a system for inspecting the appearance of fire prevention equipment installed in a fire monitoring area, and includes an unmanned aerial vehicle 1 equipped with an imaging unit 13 that approaches the fire prevention equipment and flies around it. The visual inspection system 100 also includes a remote controller 2 for operating the unmanned aerial vehicle 1 and a display unit for displaying the images captured by the imaging unit 13. The imaging unit 13 is mounted on the unmanned aerial vehicle so that it can capture images of the fire prevention equipment from a horizontal direction.

[0067] Therefore, it is possible to obtain a visual inspection system 100 that can improve the accuracy of visual inspections of disaster prevention equipment installed at high places or other distant locations. In particular, even if there are parts of the disaster prevention equipment that cannot be seen from below, such as recessed parts on the sides of the equipment, the visual inspection of the disaster prevention equipment can be performed with high accuracy, including these parts.

[0068] Furthermore, the fire prevention equipment is a fire detector 200 equipped with a protective filter 211 to prevent foreign objects from entering the smoke detection section located inside. The surface of the protective filter 211 is oriented horizontally when the fire detector 200 is installed. The imaging unit 13 then images the protective filter 211.

[0069] Therefore, the protective filter 211 installed in the fire detector 200 can also be visually inspected down to its smallest detail.

[0070] Furthermore, the remote controller 2 predicts the replacement time for disaster prevention equipment based on the captured images taken by the imaging unit 13.

[0071] Therefore, it becomes possible to know when disaster prevention equipment needs to be replaced and to take early action such as making preparations.

[0072] Furthermore, after imaging the fire detector 200, if there are any directions that have not been imaged, the unmanned aerial vehicle 1 will fly to a location where it can image those unimaged directions. Then, the imaging unit 13 will image the fire detector 200 from those unimaged directions.

[0073] Therefore, the fire detector 200 can be visually inspected from multiple directions, thereby improving the accuracy of the visual inspection.

[0074] Furthermore, the imaging unit 13 is mounted on the unmanned aerial vehicle 1 so that it can capture images from a direction parallel to the ceiling surface where the fire detector 200 is installed. Therefore, even if the ceiling surface is not horizontal but inclined, the imaging unit 13 can capture images from an oblique upward or oblique downward direction that is parallel to the inclination. [Explanation of Symbols]

[0075] 1 Unmanned aircraft, 2 Remote controller, 11, 21 Communication unit, 12, 22 Control unit, 13 Imaging unit, 14 Lighting unit, 15, 25 Memory unit, 16 Autonomous flight control unit, 23 Operation unit, 24 Display unit, 100 Visual inspection system, 200 Fire detector (fire prevention equipment), 210 Cover, 210A Main unit cover, 210B Top cover, 210C Connecting unit, 211 Protective filter, 212 Red light, 230 Opening, 251 Dirt, 252 Foreign object, C Ceiling.

Claims

1. A visual inspection system for inspecting the appearance of fire prevention equipment installed in a fire monitoring environment, An unmanned aerial vehicle equipped with an imaging unit, which approaches the disaster prevention equipment and flies around it, A remote controller for operating the aforementioned unmanned aircraft, A display unit that displays the captured image captured by the aforementioned imaging unit, It has, The imaging unit is mounted on the unmanned aerial vehicle so as to enable imaging of the disaster prevention equipment from the horizontal direction. Visual inspection system.

2. The aforementioned fire prevention device is a fire detector equipped with a protective filter to prevent foreign matter from entering the smoke detection section located inside, The surface of the protective filter is oriented in the horizontal direction when the disaster prevention equipment is installed. The imaging unit images the protective filter. The visual inspection system according to claim 1.

3. The remote controller predicts the replacement time of the disaster prevention equipment based on the image captured by the imaging unit. The visual inspection system according to claim 1 or 2.

4. After imaging has been taken of the disaster prevention equipment, if there are any directions that have not been imaged, the unmanned aerial vehicle will fly to a location where it can image those unimaged directions. The imaging unit images the disaster prevention equipment from the direction that has not been imaged. The visual inspection system according to claim 1 or 2.

5. A visual inspection system for inspecting the appearance of fire prevention equipment installed in a fire monitoring environment, An unmanned aerial vehicle equipped with an imaging unit, which approaches the disaster prevention equipment and flies around it, A remote controller for operating the aforementioned unmanned aircraft, A display unit that displays the captured image captured by the aforementioned imaging unit, It has, The imaging unit is mounted on the unmanned aerial vehicle so as to be able to take images from a direction along the ceiling surface on which the disaster prevention equipment is installed. Visual inspection system.

Images