Utility Pole Monitoring System

Abstract

To provide a power pole monitoring system which can reliably recognize an abnormality of a power pole in an early stage.SOLUTION: A power pole monitoring system comprises: a camera 2 which is arranged in a power pole P1 to photograph a peripheral power pole P2 except the power pole P1; and a monitoring device 3 which is arranged in the power pole P1 to determine presence / absence of an abnormality of the peripheral power pole P2 on the basis of an image photographed with the camera 2 and notifies a monitoring computer 4 when it is determined that there is the abnormality.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a utility pole monitoring system that monitors the condition of a utility pole. [Background technology]

[0002] In order to ensure the soundness of utility poles and prevent any disruption to the power supply, patrol inspections have been conducted to monitor whether the poles are in contact with trees or tilted. If any abnormality is found in the poles, necessary measures such as cutting down trees or reinforcing the poles are taken.

[0003] Meanwhile, a road condition measuring device is known that can measure the inclination of utility poles and the like by image processing of photographs (for example, see Patent Document 1). This road condition measuring device takes photographs of the road surface and roadside while moving a dolly carrying a road surface camera and a roadside camera during on-site surveys, and generates point cloud data indicating the three-dimensional positions of utility poles and the like from these photographs by matching, relative orientation, and absolute orientation. Then, it generates ortho-images of utility poles and the like from the point cloud data, and measures the inclination of the utility poles and the like. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-39390 Summary of the Invention [Problem to be solved by the invention]

[0005] By the way, there are cases where residents or pedestrians discover an abnormality in a utility pole, and in such cases, the discoverer contacts the power company, which then takes the necessary measures. However, there is a risk that the person will not contact the power company because it is too much of a hassle to contact them, and if the abnormality is left unattended for a long period of time without being contacted, it could cause a disruption to the power supply.

[0006] In addition, although electric power companies periodically (for example, once a year) conduct visual inspections, abnormalities can only be discovered during the periodic inspections, and there is a risk of detection being delayed.Similarly, with the road condition measuring device described in Patent Document 1, photographs of the road surface and roadside must be taken during on-site investigations, and abnormalities in utility poles cannot be discovered until the on-site investigations are conducted.

[0007] Therefore, an object of the present invention is to provide a utility pole monitoring system that can detect abnormalities in utility poles early and reliably. [Means for solving the problem]

[0008] In order to solve the above problem, the invention of claim 1 is a utility pole monitoring system characterized by comprising: an imaging means that is disposed on a utility pole and captures images of surrounding utility poles other than the utility pole in question; and a monitoring means that is disposed on the utility pole and determines whether or not there is an abnormality in the surrounding utility pole based on an image captured by the imaging means, and reports the presence of an abnormality to a specified reporting destination if it is determined that there is an abnormality.

[0009] The invention of claim 2 is the utility pole monitoring system according to claim 1, characterized in that the monitoring means reports recommended measures for the surrounding utility pole to the predetermined reporting destination in response to the abnormality.

[0010] The invention of claim 3 is characterized in that, in the utility pole monitoring system described in claim 1 or 2, the photographing means is disposed on a utility pole capable of photographing a plurality of the surrounding utility poles, and the monitoring means determines the presence or absence of the abnormality for each of the photographed plurality of the surrounding utility poles.

[0011] The invention of claim 4 is characterized in that, in the utility pole monitoring system described in claims 1 to 3, the monitoring means uses a monitoring learning model that has been machine-learned based on past performance data so that when an image captured by the photographing means is input, the presence or absence of an abnormality in the surrounding utility pole is output. Effect of the Invention

[0012] According to the invention described in claim 1, the surrounding utility poles are photographed by the photographing means installed on the utility poles, and the presence or absence of abnormalities in the surrounding utility poles is judged based on the images, and if there is an abnormality, a report is made to the specified reporting destination, so that it is possible to know the abnormality of the utility poles early and reliably. In other words, even if an investigator or the like does not go to the site, the surrounding utility poles are photographed, the presence or absence of abnormalities is judged, and if there is an abnormality, a report is made to the reporting destination, so that it is possible to know the abnormality of the utility poles early and reliably. In addition, since the photographing means and the monitoring means are installed on the utility poles, there is no need to secure a separate installation location, and the system can be easily and reliably constructed. Moreover, if it is judged that there is an abnormality, a report is made to the specified reporting destination, and if it is judged that there is no abnormality, no report is made, so not only can the amount of communication be reduced, but also the complexity of processing due to frequent reports can be reduced.

[0013] According to the invention described in claim 2, since the recommended measures according to the abnormality of the surrounding utility poles are notified to the specified reporting destination, it becomes possible to take appropriate measures according to the abnormality of the surrounding utility poles early and easily. In other words, since the recommended measures are automatically determined and notified to the reporting destination, not only can the time and effort required for a person to consider measures be reduced, but also it becomes possible to take appropriate measures reliably (without variation).

[0014] According to the invention described in claim 3, a photographing means is provided on a utility pole capable of photographing multiple surrounding utility poles, and the presence or absence of an abnormality is judged for each photographed surrounding utility pole by the monitoring means. This makes it possible to reliably judge the presence or absence of an abnormality for multiple surrounding utility poles while reducing the number of photographing means provided. In other words, it becomes possible to monitor more utility poles while keeping costs low.

[0015] According to the invention described in claim 4, the presence or absence of an abnormality in the surrounding utility poles is output using a machine-learned monitoring learning model, so it is possible to more accurately determine the presence or absence of an abnormality in the surrounding utility poles. As a result, it is possible to more accurately and reliably know the abnormality of the utility poles. [Brief description of the drawings]

[0016] [Figure 1] 1 is a schematic configuration diagram showing a utility pole monitoring system according to an embodiment of the present invention. [Diagram 2] 2A and 2B are side views showing the arrangement of the camera and monitoring device in the utility pole monitoring system of FIG. 1, where (a) shows the camera facing forward, and (b) shows the camera facing to the side. [Diagram 3] FIG. 2 is a diagram showing an image of a utility pole in a normal state in the utility pole monitoring system of FIG. 1. [Figure 4] FIG. 2 is a diagram showing a single image including a plurality of utility poles in the utility pole monitoring system of FIG. 1. [Diagram 5] 2 is a diagram showing the positional relationship between a utility pole on which a camera and a monitoring device are mounted and surrounding utility poles in the utility pole monitoring system of FIG. 1. FIG. [Figure 6] 2 is a schematic configuration block diagram of a monitoring device in the utility pole monitoring system of FIG. 1. [Figure 7] FIG. 2 is a diagram showing an image of a utility pole in a slightly tilted state in the utility pole monitoring system of FIG. 1. [Figure 8] FIG. 2 is a diagram showing an image of a utility pole in a state where the utility pole is greatly tilted, in the utility pole monitoring system of FIG. [Figure 9] 7 is a functional block diagram showing a schematic configuration of a monitoring learning model of the monitoring device of FIG. 6. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] Hereinafter, the present invention will be described based on the illustrated embodiment.

[0018] 1 to 9 show an embodiment of the present invention, and Fig. 1 is a schematic diagram showing a utility pole monitoring system 1 according to this embodiment. This utility pole monitoring system 1 is a system that monitors the state of a utility pole P, and in this embodiment, a case will be described in which the utility pole P is a power pole, but it is of course applicable to a communication pole, etc. Also, utility poles in general are indicated by the symbol "P", utility poles on which a camera 2, which will be described later, is disposed are indicated by the symbol "P1", and utility poles on which a camera 2, which will be described later, is not disposed, i.e., surrounding utility poles, are indicated by "P2".

[0019] The utility pole monitoring system 1 mainly comprises a camera (imaging means) 2, a monitoring device (monitoring means) 3, and a monitoring computer (predetermined reporting destination) 4.

[0020] The camera 2 is disposed on the utility pole P1, and is an imaging device that captures a neighboring utility pole P2 other than the utility pole P1 and located in the vicinity to generate an image as shown in Fig. 3. In this embodiment, as described later, the camera 2 is disposed on the utility pole P1 via the monitoring device 3, and is disposed on the utility pole P1 that can capture a plurality of neighboring utility poles P2. In other words, the camera 2 is disposed on the utility pole P1 that overlooks as many neighboring utility poles P2 as possible and can capture as many neighboring utility poles P2 as possible. For example, as shown in Fig. 4, the camera 2 is disposed on the utility pole P1 that can capture two neighboring utility poles P2 in one capture (can generate an image including two neighboring utility poles P2).

[0021] Furthermore, as shown in FIG. 2, it is movably connected to the monitoring device 3, and can change its orientation to photograph multiple surrounding utility poles P2. Specifically, it is connected to the monitoring device 3 via a vertical shaft 21 provided at the bottom, and can change its orientation to photograph multiple surrounding utility poles P2 by rotating the vertical shaft 21. In this embodiment, it rotates at preset angles (for example, front, 45 degrees left and right, 90 degrees left and right), stops, and photographs the surrounding utility poles P2 facing each other. However, it may be possible to generate a panoramic image in which multiple surrounding utility poles P2 are included in one image by photographing while rotating.

[0022] Since the camera 2 is thus rotatable, it becomes possible to photograph more surrounding utility poles P2 with fewer cameras 2. For example, as shown in FIG. 5, when a plurality of utility poles P are installed along a road R that extends in a substantially straight line on both sides, a camera 2 is disposed on the utility pole P1 indicated by a black circle. This allows five surrounding utility poles P2 to be photographed with one camera 2, and for example, it becomes possible to photograph 16 surrounding utility poles P2 with four cameras 2. Here, the utility pole P1 on which the camera 2 is disposed must also be photographed as a surrounding utility pole P2 by the other cameras 2, and the cameras 2 are disposed so that all utility poles P, including the utility pole P1 on which the camera 2 is disposed, can be photographed.

[0023] Such a camera 2 is controlled by the monitoring device 3 to continuously capture images at a predetermined frequency. Here, the predetermined frequency is a frequency at which the change in the state of the surrounding utility pole P2 can be reliably and quickly determined and monitored by the image capture, and at which the image capture is not excessive and wasteful (for example, once a month). This frequency may be changed according to the state of the surrounding utility pole P2. For example, as described later, when it is determined that the surrounding utility pole P2 is slightly tilted (tilted at a first tilt angle or more), the frequency may be increased slightly (for example, once a week), and when it is determined that the surrounding utility pole P2 is significantly tilted (tilted at a second tilt angle or more that is larger than the first tilt angle), the frequency may be further increased (for example, once a day).

[0024] The monitoring device 3 is mounted on a utility pole P and is a device that judges the presence or absence of abnormalities in a nearby utility pole P2 based on images captured by the camera 2. That is, holes and hooks for passing the band 20 through are provided on the side of the case / housing, and the monitoring device 3 and camera 2 are attached to the utility pole P by passing the band 20 through these holes and fastening the band 20 to the utility pole P. The monitoring device 3 and camera 2 are supplied with power from an electric wire installed on the utility pole P. Alternatively, a solar panel or a battery may be provided.

[0025] As shown in FIG. 6, the monitoring device 3 mainly includes a communication unit 31, a control unit 32, a memory unit 33, a monitoring task (monitoring means) 34, a learning task 35, and a central processing unit 36 ​​that controls these units.

[0026] The communication unit 31 is an interface for communicating with the outside via the Internet network, a telephone communication network, etc., and in this embodiment, it is connected to a monitoring computer 4 installed in a monitoring center C of an electric power company via a communication line L installed between utility poles P for free communication. The control unit 32 is a control device that controls the camera 2, and rotates the vertical shaft 21 to change the orientation of the camera 2 and starts the camera 2 to take pictures. Here, any mechanism may be used for rotating the vertical shaft 21, and for example, the vertical shaft 21 is rotated by turning a gear that meshes with a gear attached to the vertical shaft 21.

[0027] The storage unit 33 mainly includes an image database (image storage means) 331, a monitoring learning model 332, and a monitoring performance database 333. The image database 331 is a database that stores images captured by the camera 2 connected to this monitoring device 3, and stores images captured when the surrounding utility pole P2 is in a normal state at the beginning of monitoring, images captured thereafter, and judgment results (to be described later) in chronological order for each surrounding utility pole P2 (for each identification information of utility pole P). The monitoring learning model 332 and the monitoring performance database 333 will be described later.

[0028] The monitoring task 34 is a program task that judges whether or not there is an abnormality in the surrounding utility pole P2 based on the image captured by the camera 2, and reports the abnormality to the monitoring computer 4 via the communication unit 31 when it is judged that there is an abnormality, and is started every time an image is captured by the camera 2. First, based on the image captured by the camera 2, it is judged whether or not there is an abnormality in the surrounding utility pole P2 to be captured, based on the image in normal state, etc. That is, based on the image of the surrounding utility pole P2 stored in the image database 331 in a normal state and the image of the judgment target captured this time, it is judged by image analysis whether or not there is an abnormality in the surrounding utility pole P2 compared to normal state. Alternatively, it is judged by image analysis whether or not the state of the surrounding utility pole P2 is abnormal, based on the general state of the utility pole P in normal state (standing vertically on the ground, etc.)

[0029] Here, abnormalities include the tilting of the surrounding utility pole P2, contact of trees with the surrounding utility pole P2, deformation of the cross arm of the surrounding utility pole P2, etc., but in this embodiment, the tilting of the surrounding utility pole P2 will be mainly described. For example, when the image captured this time is as shown in Fig. 7 compared to the image of the surrounding utility pole P2 in a normal state as shown in Fig. 3, it is determined that the surrounding utility pole P2 is slightly tilted and abnormal, and when the image captured this time is as shown in Fig. 8, it is determined that the surrounding utility pole P2 is greatly tilted and abnormal.

[0030] Such a determination is made for each photographed surrounding utility pole P2. That is, as described above, when one camera 2 changes direction and photographs different surrounding utility poles P2 in sequence, the presence or absence of an abnormality is determined for each surrounding utility pole P (for each identification information of utility pole P). Also, as shown in FIG. 4, when one image contains multiple surrounding utility poles P2, the presence or absence of an abnormality is determined for each surrounding utility pole P2. In this case, multiple surrounding utility poles P2 are also included in a normal image, and a determination is made for each surrounding utility pole P2 by comparing and analyzing the images based on this image and the image photographed this time.

[0031] If it is determined that there is an abnormality, the abnormality information is reported to the monitoring computer 4. This abnormality information includes identification information of this monitoring device 3, i.e., the camera 2, identification information of the surrounding utility pole P2 determined to be abnormal, the contents of the determined abnormality, the image when the abnormality was determined, etc. Here, in the identification information of the surrounding utility pole P2, NULL (zero) is input when the camera 2 captures only one fixed surrounding utility pole P2, the angle and direction at the time of capture are input when one camera 2 changes direction and captures multiple different surrounding utility poles P2, and information indicating which surrounding utility pole P2 it is (right side or left side, front side or back side, etc.) is input when one image includes multiple surrounding utility poles P2.

[0032] Meanwhile, the monitoring computer 4 stores the installation positions of all utility poles P and a system diagram showing which monitoring device 3, i.e. which camera 2, is installed on which utility pole P. Furthermore, it stores which peripheral utility pole P is to be photographed when the camera 2 photographs only one fixed peripheral utility pole P, which peripheral utility pole P2 is to be photographed at what angle and direction when one camera 2 photographs multiple different peripheral utility poles P2 while changing its direction, and where on the image which peripheral utility pole P2 is located when multiple peripheral utility poles P2 are included in one image. Therefore, when abnormality information is received from the monitoring device 3, it is possible to know which peripheral utility pole P2 has occurred and what kind of abnormality has occurred based on the identification information of the peripheral utility pole P2 and the abnormality content.

[0033] Furthermore, the monitoring task 34 notifies the monitoring computer 4 of recommended measures for the surrounding utility pole P2 according to the determined abnormality. That is, recommended measures and countermeasures are stored in advance according to the abnormality content, and the recommended measures corresponding to the abnormality content determined as described above are determined and transmitted to the monitoring computer 4. For example, if it is determined that the surrounding utility pole P2 is slightly tilted, a message to reinforce the surrounding utility pole P2 is transmitted as a recommended measure, and if it is determined that the surrounding utility pole P2 is significantly tilted, a message to rebuild the surrounding utility pole P2 is transmitted as a recommended measure.

[0034] Such a monitoring task 34 uses a monitoring learning model 332 that has been machine-learned based on past performance data so that, when an image captured by the camera 2 is input, the presence or absence of an abnormality in the surrounding utility pole P2 is output. This monitoring learning model 332 is created by the learning task 35.

[0035] That is, the learning task 35 creates a monitoring learning model 332 by a known machine learning algorithm such as a neural network, using past performance data recorded and accumulated in a monitoring performance database 333. This monitoring performance database 333 is a database in which performance data including the results of a power facility monitoring and maintenance expert or knowledgeable person (such as a maintenance and inspection person of a power utility company) judging what kind of abnormality has occurred in the utility pole P based on an image of the utility pole P photographed by the camera 2 as input information is recorded and accumulated. Note that the past performance data includes data created based on an image of the utility pole P actually photographed and the results of an actual judgment made by a monitoring and maintenance expert or knowledgeable person, as well as data created in advance training, etc.

[0036] As shown in Fig. 9, this learning task 35 uses machine learning and deep learning using a neural network to create a neural network based on the performance data recorded in the monitoring performance database 333, with, for example, normal images stored in the image database 331 and the image of the judgment target captured this time as an input layer, a judgment result on the presence or absence of an abnormality in the utility pole P and what type of abnormality there is as an output layer, and an analysis process from the input layer to the output layer as an intermediate layer. Then, the learning task 35 uses the performance data of the monitoring learning model 332 as learning data to learn various parameters in the intermediate layer. That is, the learning task 35 learns various parameters in the intermediate layer so that an appropriate judgment result is output based on the image of the utility pole P captured by the camera 2.

[0037] As described above, according to this utility pole monitoring system 1, the camera 2 installed on the utility pole P1 photographs the surrounding utility pole P2 at a predetermined frequency, that is, periodically, and the presence or absence of an abnormality in the surrounding utility pole P2 is judged based on the image. If an abnormality is detected, the image is reported to the monitoring computer 4 of the power company, so that the power company can quickly and reliably know the abnormality in the utility pole P. That is, even if an investigator or the like does not go to the site, the surrounding utility pole P2 is photographed, the presence or absence of an abnormality is judged, and if an abnormality is detected, the image is reported to the monitoring computer 4, so that the power company can quickly and reliably know the abnormality in the utility pole P. In addition, since the camera 2 and the monitoring device 3 are installed on the utility pole P, there is no need to secure a separate installation location, and the system 1 can be easily and reliably constructed. Moreover, the image is reported to the monitoring computer 4 only when it is determined that an abnormality exists, and is not reported when it is determined that no abnormality exists, so that not only can the amount of communication be reduced, but also the complication of processing due to frequent reports can be reduced.

[0038] Moreover, since the recommended measures according to the abnormality of the surrounding utility pole P2 are reported to the monitoring computer 4, it becomes possible to take appropriate measures according to the abnormality of the surrounding utility pole P2 early and easily. In other words, since the recommended measures are automatically determined and reported to the monitoring computer 4, not only can the time and effort required for a person to consider measures be reduced, but also appropriate measures can be taken reliably (without variation).

[0039] Furthermore, a camera 2 is disposed on a utility pole P capable of photographing a plurality of surrounding utility poles P2, and the monitoring device 3 judges the presence or absence of an abnormality for each photographed surrounding utility pole P2. That is, the camera 2 changes direction while photographing the plurality of surrounding utility poles P2 in sequence, or photographs the plurality of surrounding utility poles P2 so as to include them in one image, and judges the presence or absence of an abnormality for each surrounding utility pole P2. This makes it possible to reliably judge the presence or absence of an abnormality for the plurality of surrounding utility poles P2 while reducing the number of cameras 2 and monitoring devices 3 disposed. In other words, it becomes possible to monitor more utility poles P while keeping costs low.

[0040] Furthermore, since the presence or absence of an abnormality in the surrounding utility pole P2 is output using the machine-learned monitoring learning model 332, it becomes possible to more accurately determine the presence or absence of an abnormality in the surrounding utility pole P2. As a result, it becomes possible to more accurately and reliably know the abnormality of the utility pole P.

[0041] Although the embodiment of the present invention has been described above in detail, the specific configuration is not limited to this embodiment, and even if there are design changes within the scope of the present invention, they are included in the present invention. For example, in the above embodiment, the identification information of the surrounding utility pole P2 determined to be abnormal by the monitoring device 3 is included in the abnormality information, but the surrounding utility pole P2 determined to be abnormal may be identified and specified in the monitoring computer 4 that receives the abnormality information, i.e., the monitoring center C. That is, shooting information (reference image) indicating which utility pole P is photographed by each camera 2 at what angle and with what background may be stored in advance, and the surrounding utility pole P2 determined to be abnormal may be identified and specified based on the image included in the abnormality information (the image of the surrounding utility pole P2 determined to be abnormal) and this shooting information. [Explanation of symbols]

[0042] 1. Street utility pole monitoring system 2. Camera (photography means) 3 Monitoring device (monitoring means) 331 Image database (image storage means) 332 Supervised Learning Model 333 Monitoring performance database 34 Monitoring tasks (monitoring methods) 35 Learning Tasks 4. Monitoring computer (designated reporting point) P utility pole P1 Electric pole with camera installed P2 Surrounding utility poles

Claims

1. An imaging means is provided on the utility pole and captures images of surrounding utility poles other than the utility pole; a monitoring means that is disposed on the utility pole and determines whether or not there is an abnormality in the surrounding utility pole based on the image captured by the photographing means, and notifies a predetermined reporting destination when it is determined that there is an abnormality; A utility pole monitoring system comprising:

2. The monitoring means notifies the predetermined reporting destination of recommended measures for the surrounding utility pole in response to the abnormality.

2. The utility pole monitoring system according to claim 1 .

3. The photographing means is disposed on a utility pole capable of photographing a plurality of the surrounding utility poles, and the monitoring means determines the presence or absence of the abnormality for each of the photographed plurality of the surrounding utility poles.

3. The utility pole monitoring system according to claim 1, wherein the utility pole monitoring system is a power pole monitoring system.

4. The monitoring means uses a monitoring learning model that has been machine-learned based on past performance data so that, when an image captured by the imaging means is input, the presence or absence of an abnormality in the surrounding utility pole is output.

4. The utility pole monitoring system according to claim 1, wherein the utility pole monitoring system is a power pole monitoring system.

Images