Elevator monitoring system

Abstract

This invention provides a technology that allows for effective monitoring of elevator shafts. [Solution] Based on a shooting instruction that includes a first instruction to photograph the elevator shaft 8 at a predetermined interval, the recording processing device 500 causes the shooting device 19 to photograph the elevator shaft 8 and acquires the image taken by the shooting device 19 as the current image. Based on a comparison of past images and current images, the recording processing device 500 determines whether or not there is an abnormality in the elevator shaft 8. If the recording processing device 500 determines that there is an abnormality, it notifies the monitoring device 80 of the occurrence of the abnormality.

Description

Technical Field

[0001] The present disclosure relates to an elevator shaft monitoring system.

Background Art

[0002] The elevator shaft is a normally closed vertical shaft, and building users and owners cannot inspect the inside of the elevator shaft. On the other hand, problems may occur in the elevator shaft during disasters such as earthquakes. For example, cracks, peeling, falling objects, or water leakage may occur in the walls, columns, or beams of the elevator shaft, and repair work may be required depending on the degree.

[0003] Generally, since the building owner cannot directly check the situation inside the elevator shaft, it is usually done by visual inspection by the elevator maintenance staff and reports by taking photos. Therefore, it is difficult to grasp the situation of the elevator shaft in real time.

[0004] Japanese Unexamined Patent Application Publication No. 2008-1494 (Patent Document 1) discloses an elevator earthquake recovery device that allows a general person who has received earthquake recovery training to compare, on a terminal, the video of each device in its normal state corresponding to the car position with the video of the monitoring camera during inspection operation to determine the presence or absence of elevator abnormalities.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] The aforementioned earthquake recovery device is disclosed to be capable of detecting abnormalities through automatic video comparison. However, because it focuses on detecting abnormalities in each component of the elevator, it does not consider abnormalities in the elevator shaft itself, such as cracks, peeling, falling objects, or water leaks in walls, columns, or beams. Furthermore, it does not disclose how abnormalities in each component of the elevator are automatically detected.

[0007] This disclosure has been made to solve the above-mentioned problems, and its purpose is to provide a technology that can suitably monitor an elevator shaft. [Means for solving the problem]

[0008] The elevator shaft monitoring system according to this disclosure comprises a camera and a recording device. The camera photographs the elevator shaft through which the elevator car travels. The recording device processes the images captured by the camera. The recording device includes a storage device. The storage device stores past images of the elevator shaft previously captured by the camera. The recording device is configured to communicate with a monitoring device. The monitoring device monitors the elevator. Based on a photography instruction that includes a first instruction to photograph the elevator shaft at a predetermined interval, the recording device causes the camera to photograph the elevator shaft and acquires the images captured by the camera as current images. Based on a comparison of past images and current images, the recording device determines whether or not there is an abnormality in the elevator shaft. If an abnormality is determined to exist, the recording device notifies the monitoring device of the occurrence of the abnormality. [Effects of the Invention]

[0009] According to this disclosure, the elevator shaft can be suitably monitored. [Brief explanation of the drawing]

[0010] [Figure 1] This figure shows an example of the overall configuration of the elevator shaft monitoring system according to the first embodiment. [Figure 2] This figure shows an example of the hardware configuration of an elevator shaft monitoring system. [Figure 3] This is a diagram illustrating the process of photographing an elevator shaft. [Figure 4] This is a diagram illustrating an example of a judgment result. [Figure 5] This is a flowchart of the processes performed by the elevator shaft monitoring system. [Figure 6] This is a flowchart of the processes performed by the elevator shaft monitoring system. [Figure 7] This figure shows an example of training data according to the second embodiment. [Figure 8] This is a diagram showing an example of the configuration of a learning device. [Figure 9] This is a flowchart showing the learning process of the learning device. [Figure 10] This figure shows an example of the configuration of the inference processing unit. [Figure 11] This is a flowchart illustrating the inference procedure performed by the inference processing unit. [Modes for carrying out the invention]

[0011] The embodiments will be described below with reference to the drawings. In the following description, identical parts are denoted by the same reference numerals. Their names and functions are also the same. Therefore, detailed descriptions of them will not be repeated.

[0012] [First Embodiment] Figure 1 shows an example of the overall configuration of the elevator shaft monitoring system 100 according to the first embodiment. Figure 2 shows an example of the hardware configuration of the elevator shaft monitoring system 100.

[0013] The elevator shaft monitoring system 100 comprises a monitoring device 80, a communication device 50, an elevator 1, a recording and processing device 500, a plurality of imaging devices 19, a maintenance terminal 600, and a user terminal 700 (Figures 1 and 2). The elevator 1 includes an elevator control device 20 and an elevator device 30.

[0014] The recording processing device 500 and the elevator control device 20 are configured to be communicably connected to the monitoring device 80 via the communication device 50. The monitoring device 80 is provided in the information center 3 of the maintenance company that maintains the elevator 1. The monitoring device 80 managed by this maintenance company performs the management (monitoring and control) of the elevators 1 installed in a plurality of buildings (in this example, buildings 2, 2a, and 2b).

[0015] The monitoring device 80 is configured to be communicable with these plurality of elevators 1. The monitoring device 80 monitors each elevator 1 and performs management and the like of each elevator 1. When a disaster such as an earthquake or a fire occurs, the elevator 1 executes a control operation to be executed at the time of the disaster and reports the disaster information to the monitoring device 80.

[0016] For example, when an earthquake sensor provided in the elevator 1 senses an earthquake, the elevator control device 20 can execute an earthquake-time control operation in which the car 10 travels to the nearest floor and then the car 10 is opened. The elevator 1 reports the information at the time of the earthquake to the monitoring device 80. In addition, the elevator control device 20 can execute a fire-time control operation in which the car 10 is called back to the evacuation floor during a fire, a flooding-time control operation in which the car 10 is stopped during flooding of the pit, and a power failure-time control operation in which the car 10 is stopped at the nearest floor by a battery during a power failure.

[0017] The elevator control device 20 is a control board that controls the elevator device 30. The elevator device 30 includes a hoist 250, landing devices (not shown) installed at the landings on each floor from the first floor to the top floor, a car device (not shown) installed in the car 10, and various sensors and various switches used in the elevator 1.

[0018] The hoist 250 is a motor that drives to raise and lower the car 10 of the elevator 1. The car device is various devices installed in the car 10 and includes a destination floor button (car call button) (not shown) for registering the destination floor. The landing device is various devices installed at the landings on each floor and includes a landing call button (not shown) for registering a landing call.

[0019] The elevator car 10 is installed in an elevator shaft 8 located within the building 2. The elevator car 10 moves up and down within the elevator shaft 8 to travel between multiple floors. In this embodiment, the elevator car 10 can stop at each floor from the 1st floor to the top floor (for example, the 6th floor). A machine room 5 is located directly above the elevator shaft 8. The machine room 5 is equipped with a hoisting machine 250, an elevator control device 20, a recording device 500, and a communication device 50.

[0020] In this embodiment, elevator 1 is a rope-type elevator in which the car 10 is suspended by a main rope 11. Elevator 1 further comprises an elevator device 30 consisting of a car 10, a counterweight 12, a main rope 11, and a deflector wheel 13. The main rope 11 is hoisted over the hoisting machine 250 and the deflector wheel 13. The car 10 and the counterweight 12 are suspended from both ends of the main rope 11.

[0021] Elevator 1 can move the car 10 installed in the hoistway 8 upward (also referred to as the "UP direction") or downward (also referred to as the "DN direction") by driving the hoisting machine 250. A buffer 14 is installed in the pit 6, which is the bottom of the hoistway 8. The buffer 14 is a device that absorbs the impact of a fall if the car 10 falls due to an abnormality.

[0022] As shown in Figure 2, the elevator control device 20 comprises a processing unit 21, a storage device 22, and a communication interface (IF) (not shown). These are connected to each other via a bus so that they can communicate with one another.

[0023] The processing unit 21 is, for example, a processor such as a CPU (Central Processing Unit). The storage device 22 includes, for example, ROM (Read Only Memory), RAM (Random Access Memory), and a non-volatile storage device such as an HDD (Hard Disk Drive) or SSD (Solid State Drive).

[0024] The processing unit 21 loads programs stored in ROM into RAM and executes them to realize various functions of the elevator control device 20. The processing unit 21 executes the process of controlling the elevator 1. ROM stores programs that describe the processing procedures for the processes executed by the elevator control device 20. RAM is the workspace for the processor when executing programs, and temporarily stores programs and data used when executing programs. The communication interface has the function of communicating with various devices connected to the elevator control device 20.

[0025] The recording processing device 500 processes images captured by multiple imaging devices 19. The recording processing device 500 is connected to multiple imaging devices 19 installed in the elevator car 10 and is configured to acquire images of the elevator shaft 8 captured by the multiple imaging devices 19. The recording processing device 500 can communicate with the monitoring device 80 via the communication device 50. The recording processing device 500 can also communicate with the elevator control device 20.

[0026] The recording device 500, like the elevator control device 20, includes a processing device 521, a storage device 522, and a communication interface (not shown). These are connected to each other via a bus so as to be able to communicate with each other. The processing device 521 is, for example, a CPU. The storage device 522 includes, for example, ROM, RAM, and a non-volatile storage device (HDD, SSD). The storage device 522 stores past images of the elevator shaft 8 taken by the imaging device 19 in the past.

[0027] The monitoring device 80 includes a control unit 81. Similar to the elevator control device 20, the control unit 81 also includes a processor (CPU), memory (ROM, RAM, etc.), and a communication interface (not shown). These are interconnected via a bus to enable communication with each other.

[0028] The monitoring device 80 has the function of acquiring and storing information transmitted (issued) by the elevator control device 20 or the recording device 500 via the communication device 50. The monitoring device 80 acquires information from the elevator control device 20 or the recording device 500 installed in each elevator 1 in each building and monitors the elevator 1 in each building.

[0029] The communication device 50 transmits (issues) various information about the elevator 1, etc., acquired from the elevator control device 20 or the recording device 500 to the monitoring device 80. The communication device 50 also includes a processor (CPU), memory (ROM, RAM, etc.), and a communication interface (not shown). These are connected to each other via a bus so that they can communicate with one another.

[0030] The monitoring device 80 communicates with elevators 1 (elevator control devices 20) or recording devices 500 installed in each building such as buildings 2a and 2b managed by the information center 3, via the communication devices 50 of each elevator 1. The communication devices 50 transmit various elevator signals, including earthquake signals detected by earthquake sensors, to the monitoring device 80. The monitoring device 80 acquires this various information. As a result, the monitoring device 80 can acquire various elevator information and understand whether an earthquake or other disaster has occurred in the building where the elevator is installed. In addition, the monitoring device 80 is configured to store the status of the hoistway 8 and whether or not there is an abnormality in the hoistway 8, which are transmitted by the recording device 500.

[0031] The maintenance terminal 600 is a terminal used by maintenance personnel of the maintenance company (maintenance personnel of elevator 1), and includes a processor (CPU), memory (ROM, RAM, etc.), and a communication interface (IF). The maintenance terminal 600 can communicate with the monitoring device 80 or the recording device 500, and can acquire information managed by the monitoring device 80 or the recording device 500.

[0032] The user terminal 700 is a terminal used by the owner of elevator 1 (a user such as the building owner or manager), and includes a processor (CPU), memory (ROM, RAM, etc.), and a communication interface (not shown). The user terminal 700 can communicate with the monitoring device 80 or the recording device 500, and can acquire information managed by the monitoring device 80 or the recording device 500.

[0033] Figure 3 is a diagram illustrating the photography of the elevator shaft 8. As shown in Figure 3, the elevator car 10 is equipped with multiple photography devices 19 for photographing the elevator shaft 8 on which the car 10 travels (in this example, two photography devices 19 are installed on the top of the car 10 and two photography devices 19 are installed on the bottom of the car 10). There may be only one photography device 19. The elevator car 10 is capable of photography operation. During photography operation, the elevator car 10 moves from the lowest floor to the highest floor of the elevator shaft 8. At that time, the multiple photography devices 19 photograph the entire elevator shaft 8 from the lowest to the highest point.

[0034] Multiple imaging devices 19 are connected to a recording device 500 via communication, and transmit images captured by the imaging devices 19 to the recording device 500.

[0035] In the event of a disaster (earthquake, fire, flooding due to heavy rain, etc.) in Building 2, damage, cracks, water leaks, etc. may occur in the walls of the elevator shaft 8.

[0036] For example, in this case, due to the earthquake, a portion of beam 18a in the elevator shaft 8 peeled off as shown in section A0 and fell onto beam 18b as falling object A1. In addition, a portion of beam 18b cracked, creating crack A2. Water leakage occurred from crack A2, as shown in water leakage A3. Repair marks A4 indicate areas where cracks were repaired in the past.

[0037] In this embodiment, when the elevator 1 is installed, multiple imaging devices 19 capture images of the entire hoistway 8. The recording and processing device 500 stores the images taken by the multiple imaging devices 19 at the time of installation (also referred to as "past images") in the storage device 522.

[0038] In this embodiment, the recording processing device 500 requests the elevator control device 20 to perform a photography operation based on a photography instruction, and also causes the photography device 19 to photograph the hoistway 8. The photography instruction includes a first instruction, a second instruction, a third instruction, and a fourth instruction.

[0039] The first instruction is to have the imaging device 19 photograph the elevator shaft 8 at a predetermined interval (once a month). Multiple imaging devices 19 perform periodic imaging once a month in response to the first instruction, capturing images of the entire inside of the elevator shaft 8. The recording processing device 500 stores these images as past images in the storage device 522.

[0040] The second instruction is to have the camera 19 photograph the elevator shaft 8 at a shorter interval than the prescribed period (once a day) after controlled operation such as earthquake-resistant operation has been performed. This photography is called "controlled operation linked photography." The captured images are stored in the memory device 522 as past images. This allows, for example, if cracks appear in the walls inside the elevator shaft 8 due to an earthquake, the subsequent progress can be observed at short intervals.

[0041] The third instruction is to have the camera 19 photograph the elevator shaft 8 at a shorter interval than the prescribed period (once a day), based on a photography reservation made by a maintenance worker from the maintenance terminal 600. The fourth instruction is to have the camera 19 photograph the elevator shaft 8 at a shorter interval than the prescribed period (once a day), based on a photography reservation made by a user (building owner, etc.) from the user terminal 700.

[0042] In this way, maintenance personnel or users (such as building owners) can reserve the recording of the elevator shaft 8 using multiple recording devices 19. The recording reservation can be for the entire elevator shaft 8, or it can be for specific locations within the elevator shaft 8, such as areas where damage or cracks have occurred, to be specified for recording. When recording is reserved, the recording devices 19 record the elevator shaft 8 at short intervals (once a day). This recording is referred to as "scheduled recording". The recorded images are stored in the storage device 522 as past images. This allows users to observe the progress of areas of concern at short intervals, for example, after heavy rainfall, to check whether water leakage has occurred since repairs were made.

[0043] Maintenance personnel can visit the site during periodic inspections (for example, once every three months) to check the condition of the elevator shaft 8. With the configuration as in this embodiment, maintenance personnel can check the condition of the elevator shaft 8 at least once a month through periodic photography. Furthermore, by scheduling photography, it becomes possible to check a specified location once a day.

[0044] As described above, the multiple imaging devices 19 can take images of the inside of the elevator shaft 8 by periodic imaging, scheduled imaging, and imaging linked to controlled operation. The recording processing device 500 performs a comparison process between the image taken this time ("current image") and past images (for example, past images taken at the time of installation, or a series of time-series images taken up to the previous time).

[0045] In the comparison process described above, the recording device 500 uses known technology to take the difference between the current image and the past image, and records the result of comparing the current image and the past image as a determination result 90. Figure 4 is a diagram illustrating an example of the determination result 90. The determination result 90 records the past image, the current image, the position in the elevator shaft 8, the state of the elevator shaft 8, and whether or not there is an abnormality in the elevator shaft 8. The state of the elevator shaft 8 includes the occurrence of cracks in the wall surface of the elevator shaft 8, the occurrence of peeling of the wall surface, and the occurrence of water leakage from the wall surface at the above position. An elevator shaft abnormality is the presence or absence of an abnormality that arises based on the state of the elevator shaft 8. Here, in this embodiment, "wall surface of the elevator shaft 8" also includes the beams of the elevator shaft 8, the columns of the elevator shaft 8, etc.

[0046] In this example, past images are assumed to be images taken at the time of installation. Current images are current images of elevator shaft 8 taken during scheduled photography, scheduled photography, and control operation-linked photography, as described above.

[0047] The recording processing device 500 uses known techniques to calculate the difference between past images and present images and compares the past and present images. By comparing the images, it determines the location and condition of any changes, such as damage or cracks on walls, fallen objects, or water leaks. Furthermore, it determines whether or not there are any abnormalities based on these conditions.

[0048] For example, Figure 4 shows that, as a result of comparing past image P1 and current image Q1, it is determined that "delamination" has occurred at "position X1" (the delamination point A0 of beam 18a is at position X1, and delamination occurred at delamination point A0 (Figure 3)). When such delamination occurs, repair is necessary, so the recording device 500 determines that there is an "abnormality". Here, past image P1 is an image taken of the floor including beam 18a at the time of installation, and current image Q1 is an image taken of the floor including beam 18a after the earthquake occurred.

[0049] Furthermore, Figure 4 shows that, as a result of comparing past image P2 and current image Q2, a "crack" and "water leakage" were detected at "location X2" (in Figure 3, crack A2 in beam 18b is located at location X2, and crack A2 is judged to be a "crack". In addition, "water leakage A3" was detected, and it was judged that "water leakage" was occurring from the "crack"). Since such water leakage requires repair, the recording device 500 judges that there is an "abnormality". Here, past image P2 is an image taken of the floor including beam 18b at the time of installation, and current image Q2 is an image taken of the floor including beam 18a after heavy rain.

[0050] Furthermore, Figure 4 shows that, as a result of comparing past image P3 and current image Q3, "post-repair" was detected at "location X3" (in Figure 3, repair mark A4 is located at location X3, and repair mark A4 is judged to be "post-repair"). Since no "water leakage" or other issues have occurred from repair mark A4, the recording device 500 has determined that there is "no abnormality". Past image P3 is an image taken of the floor including repair mark A4 at the time of installation, and current image Q3 is an image taken of the floor including repair mark A4 after the repair.

[0051] The above judgment result 90 is merely an example, and the results do not have to be as described above. For example, the presence or absence of an abnormality may be determined based on the size of the peeled area. Even if there is no water leakage, if the size of the crack is greater than a predetermined value, it may be determined to be abnormal. Furthermore, the presence or absence of water leakage may be determined by the difference in the color of the wall surface.

[0052] The recording and processing device 500 extracts elevator equipment and hoistway equipment from past images taken at the time of installation. For example, it extracts "rails," "beams," "walls," etc., from the images. Note that the method for detecting position, state, and presence or absence of abnormalities may be different from the method described above, and may be performed using the method described later in Figures 7 to 11.

[0053] The following explanation will use flowcharts. Figures 5 and 6 are flowcharts of the processes performed by the elevator monitoring system 100. Hereafter, "step" will also be simply referred to as "S".

[0054] Based on the shooting instruction, the recording processing device 500 requests the elevator control device 20 to perform a shooting operation, causes the hoistway 8 to be photographed by the shooting device 19, and acquires the image taken by the shooting device 19 (an overall image of the hoistway 8) as the current image. The shooting instruction includes, as described above, a first instruction to perform a regular shooting once a month, a second instruction to perform a shooting once a day after the execution of controlled operation, a third instruction to perform a shooting once a day based on a shooting reservation by maintenance personnel, and a fourth instruction to perform a shooting once a day based on a shooting reservation by a user (building owner, etc.).

[0055] First, as shown in Figure 5, the recording device 500 sets the shooting time according to the first instruction in S99. The first instruction is pre-recorded in the storage device 522. For example, if the first instruction specifies an shooting operation once a month, the shooting date and time is set to a predetermined time, for example, late at night when the operating rate of the cage 10 is low, which occurs once a month.

[0056] In S100, the recording device 500 sets the date and time of the photo shoot according to the second instruction after the controlled operation. For example, if elevator 1 is in controlled operation during an earthquake, the date and time of the photo shoot is set once a day for a predetermined period (for example, one month), for example, at a predetermined time in the middle of the night when the utilization rate of car 10 is low.

[0057] In S101, the recording processing device 500 determines whether or not the shooting date and time have arrived. This includes not only the shooting date and time set by the first or second instruction in S99 and S100, but also the shooting date and time set by the third or fourth instruction in S11, which will be described later using Figure 6.

[0058] In S102, the recording processing device 500 requests the elevator control device 20 to have the car 10 perform a photography operation. After the car 10 performs the photography operation, the device acquires an overall image of the hoistway 8 taken by multiple photography devices 19 as the current image. The acquired current image is recorded in the storage device 522.

[0059] In S103, the recording processing device 500 acquires past images. In this example, the past images are images taken at the time of installation, but they are not limited to these; they may also be images taken during the previous imaging operation, or time-series images taken during past imaging operations.

[0060] In S104, the recording processing device 500 performs a comparison process between past images and the captured current images. As explained using Figures 3 and 4, this allows for a comparison based on the difference between past and current images.

[0061] In the examples in Figures 3 and 4, for example, based on a comparison between past image P2 and current image Q2, it is detected that cracks and water leakage have occurred in the elevator shaft 8 at position X2. Furthermore, based on the occurrence of cracks and water leakage at position X2, it is determined that an abnormality has occurred in the elevator shaft 8.

[0062] In S105, the recording device 500 determines whether it has detected any condition of the elevator shaft 8, such as peeling, cracking, water leakage, or repair. If the recording device 500 determines YES in S105, it proceeds to S106; if it determines NO in S105, it returns to S100. In the above example, at position X2, the occurrence of cracks and water leakage is detected as conditions of the elevator shaft 8.

[0063] In S106, the recording device 500 determines whether or not an abnormality has been detected based on the above difference. If it determines YES in S106, the recording device 500 sets abnormality information in S107 to issue an alert that an abnormality has occurred, and proceeds to S108. On the other hand, if the recording device 500 determines NO in S105, it proceeds to S108. In the above example, an abnormality based on a crack and water leakage that occurred at position X2 is detected, and abnormality information is set to issue an alert to that effect.

[0064] In S108, the recording processing device 500 stores the determination result 90 (see Figure 4) obtained by comparing past images with current images in the storage device 522. In S109, the recording processing device 500 transmits the determination result 90 and abnormal information to the monitoring device 80 and returns the process to S100. As a result, each time the shooting date and time arrives, the shooting operation is performed and the determination result 90 and abnormal information are transmitted.

[0065] In step S110, the monitoring device 80 stores the received judgment result 90 and abnormal information in its storage device and displays the abnormal information on its display device. The maintenance terminal 600 or user terminal 700 can acquire the judgment result 90 and abnormal information and display them on its terminal screen. Furthermore, if an abnormality is detected, the monitoring device 80 is configured to immediately notify the maintenance terminal 600.

[0066] As explained above, the recording device 500 determines the state of the elevator shaft 8 and whether or not there is an abnormality in the elevator shaft 8 based on the state of the elevator shaft 8, based on a comparison of past images and present images. If the recording device 500 determines that there is an abnormality, it notifies the monitoring device 80 of the occurrence of the abnormality as abnormality information. The recording device 500 also notifies the monitoring device 80 of the state of the elevator shaft 8 and whether or not there is an abnormality in the elevator shaft 8.

[0067] Next, using Figure 6, we will explain the third instruction from the maintenance terminal 600 and the fourth instruction from the user terminal 700, and the corresponding processing by the recording device 500. The processing shown in Figure 6 can be initiated periodically, independently of the processing shown in Figure 5. In Figure 6, the processing executed by the recording device 500 may be configured to be executed by the monitoring device 80.

[0068] As described above, the judgment result 90 and abnormal information can be displayed on the maintenance terminal 600. Maintenance personnel using the maintenance terminal 600 can check the judgment result 90 and abnormal information, and if they wish to further observe the progress, they can make a shooting reservation on the screen of the maintenance terminal 600.

[0069] For example, in the event of heavy rainfall, it is possible to observe whether water leakage occurs from the repair site at location X3 shown in Figure 4 and to monitor the subsequent developments. When a maintenance worker performs a shooting reservation operation on the screen of the maintenance terminal 600, as shown in Figure 6, the maintenance terminal 600 transmits the shooting reservation according to the third instruction to the recording processing device 500 in S20. For example, if the maintenance worker specifies a period (for example, two weeks), the shooting operation will be performed once a day during that period.

[0070] In S10, the recording processing device 500 receives a recording reservation based on the third instruction transmitted by the maintenance terminal 600.

[0071] Based on the third or fourth instruction, the recording device 500 selects a time period during the day when the utilization rate of the elevator car 10 is low, and has the camera 19 photograph the elevator shaft 8.

[0072] Specifically, in S11, the recording processing device 500 sets the shooting date and time for the scheduled shooting based on the third instruction. In the above example, for example, the shooting date and time is set to 2:00 AM, when the operating rate of the cage 10 is low, once a day for a specified two weeks. As a result, in the flowchart shown in Figure 5, the shooting operation according to the fourth instruction is executed at the specified shooting date and time.

[0073] In S12, if the recording device 500 performs an imaging operation based on the third instruction, it transmits the determination result to the maintenance terminal 600. The maintenance terminal 600 obtains the determination result based on the third instruction (S21) and displays it on the screen (S22).

[0074] In the example above, the assessment results can be checked daily for two weeks after heavy rainfall. If water leakage occurs from the repair site, "water leakage" will be detected as the status of elevator shaft 8. If the result is determined to be "abnormal," the fact that an abnormality has occurred can also be confirmed.

[0075] Furthermore, for example, if a "crack" is detected in the condition of the elevator shaft 8, its progress can be observed. For instance, it is possible to monitor whether water is leaking from the crack, whether the crack is getting bigger, or whether delamination is occurring from the crack, and to observe the progress of any areas of concern.

[0076] In this way, the maintenance terminal 600 acquires the status of the elevator shaft 8 and whether or not there is an abnormality in the elevator shaft 8, which is determined based on a comparison of past images with current images taken based on the third instruction. The maintenance terminal 600 displays the acquired status of the elevator shaft 8 and whether or not there is an abnormality in the elevator shaft 8.

[0077] Furthermore, as described above, the judgment result 90 and abnormal information can be displayed on the user terminal 700. Users using the user terminal 700 (building owners, administrators, etc.) can check the judgment result 90 and abnormal information, and if they wish to observe the progress further, they can make a shooting reservation on the screen of the user terminal 700.

[0078] For example, similar to the above example, in the event of heavy rain, it is possible to observe whether water leakage occurs from the repair site at location X3 shown in Figure 4 and to monitor the subsequent developments. When a user performs a shooting reservation operation on the screen of the user terminal 700, as shown in Figure 6, the user terminal 700 transmits the shooting reservation according to the fourth instruction to the recording processing device 500 in S30. For example, if the user specifies a period (for example, two weeks), the shooting operation will be performed once a day during that period.

[0079] In S10, the recording processing device 500 receives a shooting reservation based on the fourth instruction transmitted by the user terminal 700. Specifically, in S11, the recording processing device 500 sets the shooting date and time of the scheduled shooting based on the fourth instruction. As a result, in the flowchart shown in Figure 5, the shooting operation based on the fourth instruction is executed at the specified shooting date and time.

[0080] In S12, if the recording device 500 performs the shooting operation based on the fourth instruction, it transmits the determination result to the user terminal 700. The user terminal 700 obtains the determination result based on the fourth instruction (S31) and displays it on the screen (S32). In the above example, the determination result can be checked every day for two weeks after the heavy rain.

[0081] Furthermore, the recording device 500 may have a function to output the judgment results to a recording medium (CD / DVD / SD card, etc.). The user terminal 700 may also be configured to access the recording device 500 or the monitoring device 80 via a browser and view the judgment results on a web screen. Alternatively, the user terminal 700 may be configured to access various SNS (Social Networking Services) to view the judgment results. Users (such as building owners) can use these services to check the judgment results, develop building maintenance plans, and monitor the progress after repairs to defective areas.

[0082] In this way, the user terminal 700 acquires the status of the elevator shaft 8 and whether or not there is an abnormality in the elevator shaft 8, which is determined based on a comparison of past images with the current image taken based on the fourth instruction. The user terminal 700 displays the acquired status of the elevator shaft 8 and whether or not there is an abnormality in the elevator shaft 8.

[0083] As described above, the elevator shaft monitoring system 100 comprises a camera 19 and a recording device 500. The camera 19 photographs the elevator shaft 8 through which the elevator car 10 of the elevator 1 travels. The recording device 500 processes the images captured by the camera 19. The recording device 500 includes a storage device 522. The storage device 522 stores past images of the elevator shaft 8 that the camera 19 has previously photographed. The recording device 500 is configured to communicate with the monitoring device 80. The monitoring device 80 monitors the elevator 1. Based on a photography instruction that includes a first instruction to photograph the elevator shaft 8 at a predetermined interval (for example, once a month), the recording device 500 causes the camera 19 to photograph the elevator shaft 8 and acquires the image captured by the camera 19 as the current image. Based on a comparison of the past image and the current image, the recording device 500 determines whether or not there is an abnormality in the elevator shaft 8. If the recording device 500 determines that an abnormality has occurred, it notifies the monitoring device 80 of the occurrence of the abnormality.

[0084] The condition of the elevator shaft 8 can normally be checked by maintenance personnel when they visit the site (for example, once every three months). The camera 19 photographs the elevator shaft 8 at a predetermined interval (for example, once a month) that is more frequent than the frequency of maintenance personnel visits to the site. If an abnormality is determined based on a comparison of past images with the current image taken by the camera 19, the monitoring device 80 is notified of the occurrence of the abnormality. This allows for accurate detection of abnormalities in the elevator shaft 8 without maintenance personnel having to visit the site, and enables early detection of abnormalities. This allows for effective monitoring of the elevator shaft 8 of the elevator 1.

[0085] The recording device 500 determines the state of the elevator shaft 8 and whether or not there are any abnormalities based on that state, based on a comparison of past images and current images. The states include the occurrence of cracks in the wall surface of the elevator shaft 8, the occurrence of peeling of the wall surface, and the occurrence of water leakage from the wall surface. The recording device 500 notifies the monitoring device 80 of the state and whether or not there are any abnormalities. This makes it possible to accurately grasp the state of the elevator shaft 8, such as the occurrence of cracks in the wall surface, the occurrence of peeling of the wall surface, and the occurrence of water leakage from the wall surface, and to observe their progress. Furthermore, it is possible to detect abnormalities caused by these states at an early stage.

[0086] The photography instruction further includes a second instruction. The second instruction is to photograph the elevator shaft 8 at a shorter interval (e.g., once a day) than the prescribed interval (e.g., once a month) after the controlled operation of elevator 1, which is performed when a disaster occurs, has been carried out. When controlled operation is performed due to an earthquake or other disaster, there is a high possibility that the elevator shaft 8 will be affected by cracks or other adverse effects, so photography is automatically carried out at a shorter interval. This allows for monitoring the progress after a disaster occurs and enables earlier detection of abnormalities. In addition, it is possible to make judgments, including observation of the elevator shaft condition, during temporary restoration by the automatic diagnosis and recovery system in the event of an earthquake, and it is possible to construct a system that shuts down the elevator in the event of elevator shaft malfunction.

[0087] The elevator shaft monitoring system 100 further includes a maintenance terminal 600. The maintenance terminal 600 is a terminal used by maintenance personnel of elevator 1. The shooting instruction further includes a third instruction. The third instruction is an instruction to photograph the elevator shaft 8 at a shorter interval than the prescribed interval (for example, once a day) based on a shooting reservation from the maintenance terminal 600. The maintenance terminal 600 acquires the status and presence or absence of abnormalities determined by comparing past images with the current image taken based on the third instruction. The maintenance terminal 600 displays the acquired status and presence or absence of abnormalities. This allows maintenance personnel to manually set the shooting period and shooting date when they find a faulty area (such as water leakage marks) during a periodic inspection of elevator 1, enabling them to identify the location of the fault and perform continuous observation. It can also be used as material for maintenance personnel to explain the need for repairs.

[0088] The elevator shaft monitoring system 100 further includes a user terminal 700. The user terminal 700 is a terminal used by the owner of elevator 1. The shooting instruction further includes a fourth instruction to photograph the elevator shaft 8 at a shorter interval than the prescribed interval (for example, once a day) based on a shooting reservation from the user terminal 700. The user terminal 700 acquires the status and presence or absence of abnormalities determined by comparing past images with the current image taken based on the fourth instruction. The user terminal 700 displays the acquired status and presence or absence of abnormalities. This allows the user (building owner, etc.) to monitor the elevator shaft 8 over time and to create repair plans. This allows maintenance personnel to check for water leaks that can only be confirmed after rain or heavy rain, and to check the elevator shaft 8 during and after disasters such as earthquakes, without having to go to the site directly, and users can also inspect the inside of the elevator shaft 8 at the time they wish to check.

[0089] Based on the third or fourth instruction, the recording device 500 selects a time period during the day when the elevator car 10 is in low operation and has the camera 19 photograph the elevator shaft 8 (photography operation). This allows the camera operation to be performed without disrupting the operation of the elevator car 10. In addition to the above time period, the camera operation may be performed on an elevator car 10 that is not currently in operation. Furthermore, instead of performing a camera operation to photograph from the lowest floor to the highest floor, it may be set to photograph only a specified location. For example, if one wants to observe the repair marks at location X3, one may specify "location X3" and have only location X3 photographed.

[0090] [Second Embodiment] In the second embodiment, the elevator shaft monitoring system 100 is configured to estimate the position, state, and presence or absence of anomalies from past and present images using AI (Artificial Intelligence). Specifically, the recording processing device 500 infers the position, state, and presence or absence of anomalies from past and present images using a trained model. The trained model is a model for inferring the position, state, and presence or absence of anomalies from past and present images. The following describes the differences from the elevator shaft monitoring system 100 according to the first embodiment, and omits the description of the parts that are the same as the elevator shaft monitoring system 100 according to the first embodiment.

[0091] Figure 7 shows an example of training data 91 according to the second embodiment. Training data 91 is data that represents the training dataset used for training.

[0092] The input data for training data 91 includes past and present images. The output data for training data 91 is the output result given the input data. The output result includes the position in the elevator shaft 8, the state of the elevator shaft 8 at that position, and whether or not there is an abnormality in the elevator shaft 8 based on that state. In training data 91, the output result is used as the correct answer data (training data). The input and output data for training data 91 are the same as those for judgment result 90.

[0093] For example, suppose the input data consists of a past image P1 taken during installation and a current image Q1 taken during an earthquake, and the current image Q1 includes the image shown in Figure 3 where delamination occurred (delamination location A0). In this case, the output data (ground truth data) for the input data of past image P1 and current image Q1 is prepared as a training dataset by a worker (e.g., a maintenance worker) who sets the location as "Location X1", the state as "Delamination", and the presence or absence of an abnormality as "Abnormality present".

[0094] For example, suppose the input data includes a past image P2 taken during installation and a current image Q2 taken during the earthquake, and the current image Q2 is an image that includes the image in Figure 3 (crack A2, leak A3) where cracks and water leakage occurred. In this case, the worker sets the output data (ground truth data) for the input data of past image P2 and current image Q2 as "location X2", "crack" and "water leakage" as the state, and "abnormality present" as the presence or absence of abnormality, and prepares these as a training dataset.

[0095] In this way, the operator sets the position, state, and presence or absence of anomalies for the set of past and present images. Note that this training data 91 may belong to the target building, but is not limited to that; data from other buildings may also be used.

[0096] Figure 8 shows an example configuration of the learning device 201. In the second embodiment, the elevatorshaft monitoring system 100 includes the learning device 201. The functions realized by the learning device 201 may also be realized by the recording device 500.

[0097] The learning device 201 includes a data acquisition unit 202 and a model generation unit 203. The functions of the data acquisition unit 202 and the model generation unit 203 are realized by the processor executing a program stored in memory.

[0098] The processor acquires training data including past images, current images, location, state, and anomaly status, and uses this training data to generate a trained model that infers location, state, and anomaly status (output data) from past and current images (input data).

[0099] The data acquisition unit 202 acquires input data and output data (ground truth data (training data)) as training data. The model generation unit 203 generates a trained model that infers output data from input data based on the training data consisting of the combination of input data and output data output from the data acquisition unit 202. The model generation unit 203 can perform training using, for example, supervised learning using a neural network.

[0100] The model generation unit 203 can also use deep learning, which learns to extract features themselves, as its learning algorithm. The model generation unit 203 stores the trained model generated by performing the learning process in the trained model memory device 300.

[0101] Figure 9 is a flowchart showing the learning process of the learning device 201. In S501, the data acquisition unit 202 acquires input data and output data (training data).

[0102] In S502, the model generation unit 203 generates a trained model that infers output data from input data using so-called supervised learning, according to training data consisting of a combination of input data and output data (training data) acquired by the data acquisition unit 202.

[0103] In other words, as shown in Figure 8, the trained model, when given past and present images as input data, infers location, state, and presence or absence of anomalies as output data.

[0104] In S503, the trained model storage device 300 stores the trained model generated by the model generation unit 203.

[0105] Figure 10 shows an example of the configuration of the inference processing unit 401. In the elevator shaft monitoring system 100 according to the second embodiment, the recording processing unit 500 includes the inference processing unit 401.

[0106] The inference processing unit 401 includes a data acquisition unit 402 and an inference unit 403. The functions of the data acquisition unit 402 and the inference unit 403 are realized, for example, by the processing unit 521 of the recording processing unit 500 executing a program stored in the storage device 522.

[0107] The recording processing device 500 infers location, state, and presence or absence of anomalies from past and present images using a trained model for inferring location, state, and presence or absence of anomalies from past and present images. The data acquisition unit 402 acquires past and present images as input data.

[0108] The inference unit 403 uses the trained model to infer location, state, and presence or absence of anomalies as output data. In other words, the inference unit 403 can output output data inferred from the input data by inputting the input data acquired by the data acquisition unit 402 into the trained model.

[0109] Figure 11 is a flowchart showing the inference procedure performed by the inference processing unit 401. In S601, the data acquisition unit 402 acquires input data. In S602, the inference unit 403 inputs the input data to the trained model stored in the trained model memory device 300 and obtains output data. In S603, the inference unit 403 outputs the output data obtained by the trained model, namely the position, state, and presence or absence of anomalies.

[0110] In the first embodiment, in S104, a comparison process is performed between past images taken during installation and currently captured images. In the second embodiment, the comparison process is performed using the method described above.

[0111] As explained above, the trained model is a model for inferring the state and presence or absence of anomalies from past and present images. The recording processing device 500 uses the trained model to infer the state and presence or absence of anomalies from past and present images. This allows maintenance personnel and building owners to more accurately understand the state and presence or absence of anomalies of the elevator shaft 8 using the trained model.

[0112] In the second embodiment, the inference processing by the inference unit 403 may be performed on an external server. The external server is, for example, a web server equipped with an artificial intelligence inference processing unit. For example, the inference unit 403 in Figure 10 transmits past images and current images (input information) acquired from the data acquisition unit 402 to the external server. The past images and current images are input to the artificial intelligence inference processing unit on the external server, and location, state, and presence or absence of anomalies are obtained as output information. For example, when the inference unit 403 receives past images and current images as input, it sends a request to the external server via an API (Application Programming Interface) to obtain location, state, and presence or absence of anomalies as output information. The inference unit 403 acquires the location, state, and presence or absence of anomalies from the external server and outputs the location, state, and presence or absence of anomalies as estimation results.

[0113] The artificial intelligence described above may be a generative AI (Generative Artificial Intelligence) that includes large language models (LLMs) and diffusion models. The generative AI may also be a multimodal AI that handles input and output of text, images, audio, and video, such as GPT, Gemini, Claude, Llama, and Grok. Furthermore, the generative AI may be configured to undergo additional training through fine-tuning to improve the accuracy of the output (location, state, and presence or absence of anomalies) in relation to the input (past and present images). Alternatively, Retrieval-Augmented Generation (RAG) may be applied to improve output accuracy by searching a database of past maintenance information accumulated by the maintenance company and then inputting the search results, along with other input information, into the generative AI.

[0114] [Note] The embodiments described above are specific examples of the following appendix.

[0115] (Note 1) A camera that photographs the elevator shaft in which the elevator car travels, The camera device comprises a recording and processing device that processes images captured by the camera, The recording processing device includes a storage device that stores past images of the elevator shaft taken by the imaging device in the past. The recording processing device is configured to communicate with a monitoring device that monitors the elevator. The aforementioned recording processing device is Based on a shooting instruction that includes a first instruction to photograph the elevator shaft at a predetermined interval, the shooting device is made to photograph the elevator shaft, and the image captured by the shooting device is acquired as the current image. Based on the comparison of the past image and the current image, it is determined whether or not there is an abnormality in the elevator shaft. A hoistway monitoring system that, when it is determined that the aforementioned abnormality exists, notifies the monitoring device of the occurrence of the abnormality.

[0116] (Note 2) The recording device determines the state of the elevator shaft and whether or not there is an abnormality that occurs based on the state, based on a comparison between the past image and the current image. The aforementioned conditions include the occurrence of cracks in the wall surface of the elevator shaft, the occurrence of peeling of the wall surface, and the occurrence of water leakage from the wall surface. The elevator shaft monitoring system as described in Appendix 1, wherein the recording processing device notifies the monitoring device of the state and the presence or absence of the abnormality.

[0117] (Note 3) The elevator shaft monitoring system according to Appendix 1 or Appendix 2, wherein the aforementioned photographic instruction further includes a second instruction to photograph the elevator shaft at a shorter interval than the prescribed period after the elevator control operation, which is performed in the event of a disaster, has been executed.

[0118] (Note 4) The elevator is further equipped with a maintenance terminal for use by maintenance personnel, The aforementioned shooting instruction further includes a third instruction to photograph the elevator shaft at a shorter interval than the prescribed interval, based on the shooting reservation from the maintenance terminal. The aforementioned maintenance terminal is The state and the presence or absence of the abnormality are determined based on a comparison between the past image and the current image taken based on the third instruction. A hoistway monitoring system as described in Appendix 2 or Appendix 3, which displays the acquired status and whether or not there is an abnormality.

[0119] (Note 5) The elevator shaft monitoring system according to Appendix 4, wherein the recording processing device, based on the third instruction, selects a time period during the day when the elevator car's operating rate is low and causes the imaging device to photograph the elevator shaft.

[0120] (Note 6) The elevator further includes a user terminal for use by the owner of the elevator, The aforementioned shooting instruction further includes a fourth instruction to photograph the elevator shaft at a shorter interval than the prescribed interval, based on the shooting reservation from the user terminal. The aforementioned user terminal is The state and the presence or absence of the abnormality are determined based on a comparison between the past image and the current image taken based on the fourth instruction. A hoistway monitoring system as described in any of Appendix 2 to Appendix 5, which displays the acquired state and whether or not there is an abnormality.

[0121] (Note 7) The elevator shaft monitoring system according to any one of the appendices 2 to 6, wherein the recording processing device infers the state and the presence or absence of abnormalities from the past images and the current images using a trained model for inferring the state and the presence or absence of abnormalities from the past images and the current images.

[0122] The embodiments disclosed herein are illustrative and not limited to those described herein. The scope of the present invention is defined by the claims, and all modifications within the meaning and scope equivalent to the claims are intended to be included. [Explanation of Symbols]

[0123] 1 Elevator, 2,2a,2b Building, 3 Information Center, 5 Machine Room, 6 Pit, 8 Hoistway, 10 Car, 11 Main Rope, 12 Counterweight, 13 Deflection Vehicle, 14 Shock Absorber, 18a,18b Beam, 19 Imaging Device, 20 Elevator Control Device, 21 Processing Device, 22 Storage Device, 30 Elevator Equipment, 50 Communication Device, 80 Monitoring Device, 81 Control Unit, 90 Judgment Result, 91 Training Data, 100 Hoistway Monitoring System, 201 Learning Device, 202 Data Acquisition Unit, 203 Model Generation Unit, 300 Trained Model Storage Device, 401 Inference Processing Unit, 402 Data Acquisition Unit, 403 Inference Unit, 500 Recording Device, 521 Processing Device, 522 Storage Device, 600 Maintenance Terminal, 700 User Terminal, A0 Peeling Location, A1 Falling Object, A2 Cracks, A3 water leakage, A4 repair marks.

Claims

1. A camera that photographs the elevator shaft in which the elevator car travels, The camera device comprises a recording and processing device that processes images captured by the camera, The recording processing device includes a storage device that stores past images of the elevator shaft taken by the imaging device in the past. The recording processing device is configured to communicate with a monitoring device that monitors the elevator. The aforementioned recording processing device is Based on a shooting instruction that includes a first instruction to photograph the elevator shaft at a predetermined interval, the shooting device is made to photograph the elevator shaft, and the image captured by the shooting device is acquired as the current image. Based on the comparison of the past image and the current image, it is determined whether or not there is an abnormality in the elevator shaft. A hoistway monitoring system that, when it is determined that the aforementioned abnormality exists, notifies the monitoring device of the occurrence of the abnormality.

2. The recording device determines the state of the elevator shaft and whether or not there is an abnormality that occurs based on the state, based on a comparison between the past image and the current image. The aforementioned conditions include the occurrence of cracks in the wall surface of the elevator shaft, the occurrence of peeling of the wall surface, and the occurrence of water leakage from the wall surface. The elevator shaft monitoring system according to claim 1, wherein the recording processing device notifies the monitoring device of the state and the presence or absence of the abnormality.

3. The elevator shaft monitoring system according to claim 2, wherein the aforementioned photographic instruction further includes a second instruction to photograph the elevator shaft at a shorter interval than the prescribed period after the elevator control operation, which is performed in the event of a disaster, has been executed.

4. The elevator is further equipped with a maintenance terminal for use by maintenance personnel, The aforementioned shooting instruction further includes a third instruction to photograph the elevator shaft at a shorter interval than the prescribed interval, based on the shooting reservation from the maintenance terminal. The aforementioned maintenance terminal is The state and the presence or absence of the abnormality are determined based on a comparison between the past image and the current image taken based on the third instruction. The elevator shaft monitoring system according to claim 2, which displays the acquired state and whether or not there is an abnormality.

5. The elevator shaft monitoring system according to claim 4, wherein the recording processing device, based on the third instruction, selects a time period during the day when the elevator car's operating rate is low and causes the imaging device to photograph the elevator shaft.

6. The elevator further includes a user terminal for use by the owner of the elevator, The aforementioned shooting instruction further includes a fourth instruction to photograph the elevator shaft at a shorter interval than the prescribed interval, based on the shooting reservation from the user terminal. The aforementioned user terminal is The state and the presence or absence of abnormalities are determined based on a comparison between the past image and the current image taken based on the fourth instruction. The elevator shaft monitoring system according to claim 2, which displays the acquired state and whether or not there is an abnormality.

7. The elevator shaft monitoring system according to any one of claims 2 to 6, wherein the recording processing device infers the state and the presence or absence of abnormalities from the past images and the current images using a trained model for inferring the state and the presence or absence of abnormalities from the past images and the current images.

Images