Information processing device and information processing method

Abstract

To provide an information processing apparatus which outputs information on tools for preventing animal damage according to the situation of power distribution equipment damaged by animals.SOLUTION: An information processing apparatus includes: an acquisition unit for acquiring imaging data acquired by imaging power distribution equipment damaged by animals; and an output unit for outputting information on tools for preventing animal damage which are to be installed on the power distribution equipment for preventing animal damage, from the imaging data acquired by the acquisition unit, using a learned model. The learned model is a learned machine learning model configured to output correct answer data when input data is input based on training data obtained by associating the input data including imaging data acquired by imaging power distribution equipment damaged by animals, with correct answer data including information on tools for preventing animal damage installed on the power distribution equipment damaged by animals.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present invention relates to an information processing device and an information processing method. [Background technology]

[0002] When birds such as sparrows land on the electric wires of a power distribution facility, the power distribution facility and surrounding residential areas, houses, vehicles, etc. may suffer bird damage from bird droppings, etc. To prevent such bird damage, bird damage prevention devices are installed on the power distribution facility (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2020-092629 A Summary of the Invention [Problem to be solved by the invention]

[0004] In order to install the bird damage prevention device disclosed in Patent Document 1 on the power distribution equipment that has been damaged by birds, a worker goes to the site to check the situation and measures the extent of the bird damage, etc. After that, the worker decides on the specific product, size, quantity, etc. of the bird damage prevention device based on the results of the measurement. The worker then creates design drawings and orders the installation work.

[0005] However, the process from checking the site conditions to deciding on the specific product, size, quantity, etc. of the bird pest control equipment depends heavily on the experience of the worker, so workers with little experience may have difficulty creating appropriate design drawings.

[0006] An object of the present invention is to provide an information processing device that outputs information about animal damage prevention devices according to the status of power distribution equipment that has been damaged by animals. [Means for solving the problem]

[0007] One of the present inventions for achieving the above-mentioned object is an information processing device comprising: an acquisition unit that acquires imaging data acquired by imaging distribution equipment that has suffered animal damage; and an output unit that uses a trained model to output information about animal damage prevention equipment to be installed on the distribution equipment to prevent the animal damage from the imaging data acquired by the acquisition unit, wherein the trained model is a machine learning model that has been trained to output correct answer data when input data is input, based on teacher data that corresponds input data including imaging data acquired by imaging distribution equipment that has suffered animal damage and correct answer data including information about animal damage prevention equipment installed on the distribution equipment that has suffered animal damage.

[0008] According to this configuration, it is possible to output information on animal damage prevention equipment according to the state of the power distribution equipment that has been damaged by animals. As a result, it becomes easy to design animal damage prevention equipment to be installed in the power distribution equipment based on the output of the information processing device.

[0009] Another aspect of the present invention is the above-mentioned information processing device, wherein the information on the animal damage prevention tools includes the type and quantity of the animal damage prevention tools.

[0010] According to this configuration, it is possible to output detailed information on animal damage prevention equipment according to the state of the power distribution equipment that has been damaged by animals. As a result, it becomes easier to design animal damage prevention equipment to be installed in the power distribution equipment based on the output of the information processing device.

[0011] Another aspect of the present invention is the information processing device, wherein the correct answer data includes information on installation work of animal damage prevention devices, and the output unit further outputs the information on the installation work of the animal damage prevention devices.

[0012] According to this configuration, it is possible to output information on installation work according to the state of the power distribution facility that has been damaged by animals. As a result, arrangements for installation work can be made easily based on the output of the information processing device.

[0013] Another aspect of the present invention is the above-mentioned information processing device, wherein the information on the installation work includes at least one of the installation work cost of the animal damage prevention device, information on vehicles required for the installation work, and details of traffic regulations required for the installation work.

[0014] According to this configuration, detailed information on the installation work can be output according to the state of the power distribution facility that has been damaged by animals. As a result, arrangements for the installation work can be made easier based on the output of the information processing device.

[0015] Another aspect of the present invention is an information processing device that has a model generation unit that generates a trained model that is trained to output correct data when input data is input, based on training data that corresponds input data including imaging data obtained by photographing a distribution equipment that has been damaged by wild animals and correct data including information on animal damage prevention equipment installed in the distribution equipment that has been damaged by wild animals.

[0016] With this configuration, the information processing device can generate a learned model capable of outputting information on animal damage prevention devices for preventing bird damage to the power distribution equipment, based on captured image data of the power distribution equipment. Effect of the Invention

[0017] According to the present invention, it is possible to provide an information processing device that outputs information about animal damage prevention tools according to the status of power distribution equipment that has been damaged by animals. [Brief description of the drawings]

[0018] [Figure 1] FIG. 1 is a diagram showing a configuration of an information processing system 1. [Diagram 2] FIG. 2 is a diagram illustrating an example of hardware of an information processing device 2. [Diagram 3] FIG. 13 is a diagram illustrating an example of teacher data. [Figure 4] FIG. 2 is a diagram showing a captured image of the teacher data. [Diagram 5] FIG. 2 is a diagram showing functional blocks of an information processing device 2. [Figure 6]FIG. 4 is a diagram showing an example of a captured image. [Figure 7] 1 is a flowchart illustrating the process flow for generating a trained model. [Figure 8] 13 is a flowchart illustrating a process flow up to displaying an output screen 30. [Figure 9] FIG. 2 is a diagram showing an example of an output screen 30. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0019] ==First embodiment== <Information processing system> 1 is a diagram showing a schematic configuration of an information processing system 1 according to the present embodiment. The information processing system 1 includes an information processing device 2 and a mobile terminal 3. The information processing device 2 and the mobile terminal 3 are communicatively connected via a network NW. The network NW is, for example, a public line such as the Internet, or a dedicated line such as a Wide Area Network (WAN) or a Local Area Network (LAN).

[0020] The mobile terminal 3 is a portable terminal such as a smartphone, tablet terminal, or laptop computer, and is equipped with a camera. A worker carries the mobile terminal 3 to a site for patrol, inspection, or the like, and photographs the power distribution equipment. The mobile terminal 3 provides the information processing device 2 with an image of the power distribution equipment (hereinafter referred to as a "captured image"). "Power distribution equipment" is equipment for transmitting power generated at a power plant to power consumers. Power distribution equipment includes utility poles and electric wires, etc. Here, "electric wires, etc." includes not only electric wires (high voltage lines, low voltage lines, service lines, etc.) supported by utility poles, but also communication lines, overhead ground wires, wires for supporting utility poles, etc.

[0021] The captured image is not limited to one captured by the mobile terminal 3, and may be, for example, one captured by a camera managed by a power company, etc. The captured image may be provided to the information processing device 2 via a recording medium, or may be provided to the information processing device 2 via a wired or wireless communication means (such as a local area network (LAN) or universal serial bus (USB)).

[0022] The information processing device 2 generates a trained model that outputs information on animal damage prevention equipment used in the past from imaging data acquired by imaging the electric power distribution equipment damaged by animals in the past based on the teacher data described later. The information processing device 2 further uses the trained model generated to output information on animal damage prevention equipment to be prepared from imaging data acquired by newly imaging the electric power distribution equipment damaged by animals.

[0023] Here, "animal damage" refers to pollution caused by animals that come into contact with electric wires of power distribution facilities. The "animals" are not particularly limited as long as they are animals that may cause animal damage to power distribution facilities, but birds will be used as an example in the following explanation. Therefore, bird damage will be used as an example of animal damage, and bird damage prevention equipment will be used as an example of animal damage prevention equipment in the following explanation.

[0024] "Bird damage" refers to pollution caused by birds perching on electric wires of power distribution facilities. Examples of such pollution include noise caused by bird calls and bird droppings on or around the power distribution facilities. Furthermore, "bird damage prevention devices" are devices installed on power distribution facilities to prevent bird damage. Bird damage prevention devices can prevent birds from perching at the locations where they are installed. The product, size, material, etc. of bird damage prevention devices are selected according to the type of bird, the thickness of the electric wires, etc., the range affected by bird damage, the type of electric wires, etc. Alternatively, a bird damage prevention device that can be installed in a desired range can be constructed by connecting multiple bird damage prevention devices of a predetermined size.

[0025] [Hardware configuration of information processing device] 2 is a diagram showing an example of hardware of an information processing device 2. The illustrated information processing device 2 includes a processor 200, a main storage device 201, an auxiliary storage device 202, an input device 203, an output device 204, and a communication device 205. The illustrated information processing device 2 may be realized, in whole or in part, by using virtual information processing resources provided by a virtualization technology, a process space separation technology, or the like, such as a virtual server provided by a cloud system. In addition, all or in part of the functions provided by the information processing device 2 may be realized by a service provided by the cloud system via an API (Application Programming Interface), for example. The information processing system 1 may be configured by using a plurality of information processing devices 2 connected to each other so as to be able to communicate with each other.

[0026] In the figure, the processor 200 is configured using, for example, a CPU (Central Processing Unit), an MPU (Micro Processing Unit), a GPU (Graphics Processing Unit), an FPGA (Field Programmable Gate Array), an ASIC (Application Specific Integrated Circuit), an AI (Artificial Intelligence) chip, etc.

[0027] The main memory device 201 is a device that stores programs and data, and is, for example, a read only memory (ROM), a random access memory (RAM), or a non-volatile memory (NVRAM (Non Volatile RAM)).

[0028] The auxiliary storage device 202 is, for example, a solid state drive (SSD), a hard disk drive, an optical storage device (CD (Compact Disc), DVD (Digital Versatile Disc), etc.), a storage system, a reading / writing device for a recording medium such as an IC card, an SD card, or an optical recording medium, a storage area of ​​a cloud server, etc. Programs and data can be read into the auxiliary storage device 202 via a recording medium reading device or a communication device 205. The programs and data stored (memorized) in the auxiliary storage device 202 are read into the main storage device 201 as needed.

[0029] The input device 203 is an interface that accepts input from the outside, and is, for example, a keyboard, a mouse, a touch panel, a card reader, a pen-input tablet, a voice input device, or the like.

[0030] The output device 204 is an interface that outputs various information such as the process progress, the process results, etc. The output device 204 is, for example, a display device (liquid crystal monitor, LCD (Liquid Crystal Display), graphic card, etc.) that visualizes the various information described above, a device that converts the various information described above into voice (voice output device (speaker, etc.)), or a device that converts the various information described above into text (printer, etc.).

[0031] The input device 203 and the output device 204 constitute a user interface that receives and presents information to and from the user.

[0032] The communication device 205 is a device that realizes communication with other devices. The communication device 205 is a wired or wireless communication interface that realizes communication with other devices via a communication network such as the Internet, and is, for example, a NIC (Network Interface Card), a wireless communication module, a USB module, etc. The information processing device 2 is configured to input and output information to and from the mobile terminal 3 via the communication device 205.

[0033] In the information processing device 2, for example, an operating system, a file system, a DBMS (DataBase Management System) (relational database, NoSQL, etc.), a KVS (Key-Value Store), etc. may be installed.

[0034] Each function of the information processing system 1 is realized by a processor 200 of the information processing device 2 reading and executing a program stored in a main memory device 201, or by hardware (FPGA, ASIC, AI chip, etc.) constituting the information processing system 1. The information processing system 1 stores various types of information (data), for example, as a table of a database or a file managed by a file system.

[0035] [Teacher data] Fig. 3 is a diagram for explaining the teacher data 4 of this embodiment stored in the above-mentioned auxiliary storage device 202. The teacher data 4 includes a plurality of data 4a. One data 4a is arranged in one column of Fig. 3. Each of the plurality of data 4a is data based on one installation work of a bird pest prevention device.

[0036] The teacher data 4 in this embodiment is data in which captured image data 40, information 41 about bird pest control devices, and installation work information 52 are associated with each other. Here, the captured image data 40 is input data, and the information 41 about bird pest control devices and the information 42 about installation work are correct answer data.

[0037] The captured image data 40 is data that represents a captured image of the power distribution facility. A file name that identifies the captured image data 40 is shown in FIG.

[0038] 4 is a diagram in which captured image data 40 in data 4a of teacher data 4 is expressed as captured image 40a. As shown in FIG. 4, captured image 40a shows utility pole 50 included in power distribution facility 5 and a number of birds (sparrows S) perched on electric wire 51 supported by utility pole 50.

[0039] The bird pest prevention device information 41 is one or more pieces of information on bird pest prevention devices necessary to prevent bird damage to the power distribution facility. In this embodiment, the bird pest prevention device information 41 includes product information 410, an inner diameter 411, a length 412, and a quantity 413.

[0040] The product information 410 is information that identifies a specific product of bird pest prevention equipment that has been determined to be necessary. In this embodiment, the product information 410 is the product name. However, this is not limited to this, and may be a product code or the like. In this embodiment, it is assumed that the size (inner diameter, length, etc., described below) can be selected and specified for each of the product information 410 of the bird pest prevention equipment.

[0041] Inner diameter 411 is information specifying the inner diameter of the tube when the product specified in product information 410 has a tubular structure and is a product that is installed on an electric wire or the like by passing the electric wire or the like through a hollow portion (corresponding to protective equipment A and protective equipment C in FIG. 3). Length 412 and quantity 413 are information specifying the length of the product specified in product information 410 and information specifying the quantity of the product, respectively.

[0042] The installation work information 42 is information indicating a record of the work of installing a product corresponding to the product information 410 in a power distribution facility. In this embodiment, the installation work information 42 includes a work cost 420, a vehicle 421, and a traffic regulation 422.

[0043] Construction cost 420 is information on the cost required for construction. Vehicle 421 is information on the vehicle required for construction. "Vehicle information" is, for example, information that identifies special vehicles such as aerial work vehicles and crane trucks that can be used by workers. Traffic restrictions 422 is the content of traffic restrictions required for construction. If traffic restrictions are necessary for construction, the "contents of traffic restrictions" include, for example, one-way road closure, full road closure, whether traffic guides are needed, and if traffic guides are needed, the number of traffic guides, etc.

[0044] The teacher data 4 of this embodiment has been described above, but the input data and correct answer data in the teacher data 4 are not limited to the above description. For example, the input data may further include information about the surroundings of the power distribution facility. "Information about the surroundings of the power distribution facility" refers to location information such as infrastructure facilities (roads, railways, bridges, etc.), topography (rivers, mountains, seas, etc.), and trees within a predetermined range based on the power distribution facility. Installation work information 52 also depends on the environment surrounding the power distribution facility. Therefore, when information about the surroundings of the power distribution facility is included in the input data, the accuracy of the installation work information output by the trained model is improved.

[0045] [Functional blocks of information processing equipment] 5 is a diagram showing functional blocks of the information processing device 2 of this embodiment. In the information processing device 2 of this embodiment, a first acquisition unit 210, a model generation unit 211, a second acquisition unit 212, an output unit 213, and a display control unit 214 are realized by the processor 200 executing a predetermined program.

[0046] The first acquisition unit 210 acquires the teacher data 4 (FIG. 3) stored in the auxiliary storage device 202.

[0047] The model generation unit 211 generates a trained model based on the teacher data 4 acquired by the first acquisition unit 210. The trained model is a machine learning model that has been trained to output information 41 on bird pest control devices and information 42 on installation work, which are correct answer data, when captured image data 40, which is input data, is input.

[0048] The model generation unit 211 performs deep learning to learn the features of the captured image 40a, and constructs (generates) a neural network as a trained model, which takes the captured image data 40 as input and outputs information 41 on the bird pest prevention equipment and information 42 on the installation work.

[0049] The "feature amount" includes, for example, the height, thickness, and length of electric wires etc. supported by the power distribution facility, the state of bird damage, etc. The "state of bird damage, etc." refers to scenery including the type and number of birds resting on the electric wires etc., the position and range where the birds are resting, the distribution of droppings attached to the vicinity of the power distribution facility, etc.

[0050] The neural network is, for example, a CNN (Convolution Neural Network), and has an input layer that receives input of captured image data 40, an output layer that outputs information 41 on bird pest control devices and information 42 on installation work for the bird pest control devices, and an intermediate layer that extracts feature amounts. The detailed process of constructing (generating) the neural network will be described later.

[0051] The second acquisition unit 212 acquires captured image data of an image acquired by capturing an image of a power distribution facility that has been damaged by birds and for which installation work for bird damage prevention devices is scheduled. Specifically, the second acquisition unit 212 acquires the captured image data from the mobile terminal 3 via the network NW. Note that the mobile terminal 3 may store the captured image data in advance in the auxiliary storage device 202 via the network NW, and the second acquisition unit 212 may acquire the captured image data from the auxiliary storage device 202.

[0052] 6 is a diagram illustrating an example of a captured image 60a of the power distribution facility 7, which is represented by the captured image data 60 acquired by the second acquisition unit 212. As shown in FIG. 6, the captured image 60a shows a utility pole 70 included in the power distribution facility 7 and a number of birds (sparrows S) perched on a power wire 71 supported by the utility pole 70.

[0053] The output unit 213 inputs the captured image data 60 acquired by the second acquisition unit 212 into the trained model generated by the model generation unit 211, and outputs information on the bird pest prevention devices and information on the installation work of the bird pest prevention devices. The information on the bird pest prevention devices here is information on the bird pest prevention devices to be installed in the power distribution facility to prevent bird damage.

[0054] The display control unit 214 generates an output screen that explains the information on the bird pest prevention devices output by the output unit 213 and information on the installation work for the bird pest prevention devices, and causes the output device 204 to display it.

[0055] [Processing to generate a trained model] 7 is a flowchart illustrating the flow of processing up to generation of a trained model by the information processing device 2. The processing up to generation of a trained model includes step S11 and step S12.

[0056] First, in step S11, the first acquisition unit 210 of the information processing device 2 acquires the teacher data 4. The teacher data 4 shown in Fig. 3 is an example of the teacher data acquired by the first acquisition unit 210 in step S11.

[0057] Next, in step S12, the model generation unit 211 of the information processing device 2 generates a trained model by using the teacher data 4 shown in FIG.

[0058] In step S12, the input layer of the neural network (not shown) passes the input pixel values ​​to the intermediate layer. The intermediate layer passes the extracted feature values ​​to the output layer. The output layer outputs information 41 on the bird pest control device and information 42 on the installation work of the bird pest control device based on the feature values ​​output from the intermediate layer.

[0059] In this embodiment, the trained model is described as a neural network, but the trained model may be a trained model constructed using other learning algorithms, such as a Support Vector Machine (SVM), a Bayesian network, or a regression tree.

[0060] The model generation unit 211 inputs the captured image data 40, which is input data, to the input layer, and obtains information 41 on bird pest prevention devices and information 42 on installation work for the bird pest prevention devices from the output layer. The model generation unit 211 compares the results of the information 41 on bird pest prevention devices and the information 42 on installation work for the bird pest prevention devices output from the output layer with correct answer data, and optimizes parameters used in the calculation process so that the results of the information 41 on bird pest prevention devices and the information 42 on installation work for the bird pest prevention devices become correct answer data. The parameters used in the calculation process are, for example, weights (coupling coefficients) between neurons, activation function coefficients used in each neuron, and the like. There are no particular limitations on the method of optimizing the parameters, but the model generation unit 211 optimizes various parameters using, for example, an error backpropagation method.

[0061] Through the above-described processing, the processing up to generation of the trained model is completed. The generated trained model is stored in the auxiliary storage device 202.

[0062] [Processing until the output screen is displayed] 8 is a flowchart illustrating the flow of processing up to when the information processing device 2 displays the output screen. The processing up to when the output screen is displayed includes steps S21 to S23. It is assumed that a trained model is stored in the auxiliary storage device 202 before this processing.

[0063] First, in step S21, the second acquisition unit 212 acquires captured image data 60 of the power distribution facility 7 where installation work for the bird and animal prevention device is scheduled, and a trained model. At this time, the second acquisition unit 212 acquires the captured image data 60 from the mobile terminal 3 via the network NW. Furthermore, the second acquisition unit 212 acquires the trained model from the auxiliary storage device 202. The captured image 60a shown in Fig. 6 is an example of a captured image represented by the captured image data 60 acquired by the mobile terminal 3 by capturing an image of the power distribution facility.

[0064] In step S22, the output unit 213 inputs the captured image data 60 of the captured image 60a to the trained model, and outputs information on the bird pest prevention device and information on the installation work for the bird pest prevention device.

[0065] In step S23, the display control unit 214 generates an output screen 30 that explains the information 41 on the bird pest prevention devices output by the output unit 213 and information on the installation work of the bird pest prevention devices, and causes the output device 204 to display it.

[0066] 9 is a diagram showing an example of the output screen 30 in this example. A captured image 60a and a display window 30a are displayed on the output screen 30. In this example, information on the bird pest prevention device and information on the installation work of the bird pest prevention device are displayed in the display window 30a.

[0067] Based on the contents of the output screen 30, the user can determine the product, size and quantity of bird pest control devices to be installed in the power distribution facility 7, as well as the cost required for the installation work, and the vehicle and traffic regulations. Then, the user can place an order for the installation work.

[0068] The above-mentioned embodiment is for the purpose of facilitating understanding of the present invention, and is not intended to limit the present invention. Furthermore, the present invention can be modified or improved without departing from the spirit of the present invention, and it goes without saying that the present invention includes equivalents thereof.

[0069] For example, in the present embodiment, the first acquisition unit 210, the model generation unit 211, the second acquisition unit 212, the output unit 213, and the display control unit 214 are provided in the information processing device 2, but this is not limiting. For example, the first acquisition unit 210 and the model generation unit 211 may be provided in one information processing device, and the second acquisition unit 212, the output unit 213, and the display control unit 214 may be provided in another information processing device.

[0070] Even in such cases, the information processing device 2 can output information about animal damage prevention tools according to the state of the power distribution facility that has suffered animal damage. [Explanation of symbols]

[0071] 1: Information processing system 2: Information processing device 200: Processor 201: Main memory 202:Auxiliary storage device 203: Input device 204: Output device 205: Communication equipment 210:First acquisition part 211: Model generation unit 212:Second acquisition part 213: Output section 214: Display control unit 3. Mobile devices 30: Output screen 30a: Display window 4: Training data 40: Captured image data 40a: Captured image 41: Information on bird pest prevention equipment 410: Product information 411: Inner diameter 412:Length 413:Quantity 42: Installation information 420: Construction cost 421: Vehicle 422: Traffic regulations 5: Power distribution equipment 50: Electric pole 51:Electric wire 60: Captured image data 60a: Captured image 7: Power distribution equipment 70: Electric pole 71: Electric wire

Claims

1. An acquisition unit that acquires image data by capturing an image of the power distribution facility that has been damaged by an animal; An output unit that outputs information about animal damage prevention equipment to be installed in the power distribution facility to prevent the animal damage from the imaging data acquired by the acquisition unit using a trained model; Equipped with The trained model is a machine learning model that has been trained to output correct answer data when input data is input based on teacher data in which input data including image data acquired by imaging a power distribution facility that has been damaged by animals and correct answer data including information on animal damage prevention equipment installed in the power distribution facility that has been damaged by animals are associated with the input data. Information processing device.

2. 2. The information processing device according to claim 1, The information on the animal damage prevention equipment includes the type and quantity of the animal damage prevention equipment, Information processing device.

3. 3. The information processing device according to claim 2, The correct answer data includes information on the installation work of animal damage prevention equipment, The output unit further outputs information on the installation work of the animal damage prevention equipment. Information processing device.

4. 4. The information processing device according to claim 3, The information on the installation work includes at least one of the installation cost of the animal damage prevention device, information on the vehicle required for the installation work, and the content of the traffic regulations required for the installation work. Information processing device.

5. A model generation unit that generates a trained model that has been trained to output correct answer data when input data is input, based on teacher data in which input data including image data acquired by capturing an image of a power distribution facility that has been damaged by animals and correct answer data including information on animal damage prevention equipment installed in the power distribution facility that has been damaged by animals is associated with the input data. An information processing device comprising:

6. A step of acquiring imaging data by imaging the power distribution facility that has been damaged by animals; A step of outputting information on animal damage prevention equipment to be installed in the power distribution facility to prevent the animal damage from the imaging data acquired by the acquisition unit using the trained model; Including, The trained model is a machine learning model that has been trained to output correct answer data when input data is input based on teacher data in which input data including image data acquired by imaging a power distribution facility that has been damaged by animals and correct answer data including information on animal damage prevention equipment installed in the power distribution facility that has been damaged by animals are associated with the input data. Information processing methods.

7. A step of generating a trained model that has been trained to output the correct answer data when input data is input based on teacher data in which input data including image data acquired by imaging the distribution equipment damaged by animals and correct answer data including information on animal damage prevention equipment installed in the distribution equipment damaged by animals are associated with the input data. An information processing method comprising:

Images