Instrument identification method for inspection robot of power supply and distribution system and related equipment

Abstract

The invention provides an instrument identification method of a power supply and distribution system inspection robot. The method comprises the following steps: acquiring first image data of a to-be-identified instrument; instrument detection processing is carried out on the first image data through a trained instrument detection model, and whether the to-be-identified instrument is in a normal operation state is determined; if the to-be-recognized instrument is in the normal operation state, second image data of the to-be-recognized instrument are obtained, reading recognition processing is conducted on the second image data through the trained instrument reading recognition model, reading data of the to-be-recognized instrument are obtained, and the inspection robot is controlled to output the reading data; determining whether the power supply and distribution system is abnormal or not based on the output reading data, and if yes, sending an early warning prompt to a user; acquiring historical data of the power supply and distribution system, performing trend analysis on the power supply and distribution system in combination with the historical data, determining whether equipment in the power supply and distribution system is damaged continuously, and if the equipment is damaged continuously, arranging maintenance and / or replacement.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to an instrument identification method for a power supply and distribution system inspection robot and related equipment. Background Art

[0002] As the source and a vital component of the information industry, instrumentation helps people better acquire and monitor data. It is widely used in production and life, medical equipment, industrial production, transportation, and other areas. Currently, many devices are equipped with digital instruments to monitor their operating status. Current inspection methods primarily rely on manual inspections, which suffer from low reading efficiency, large reading errors, and the inability to promptly detect power supply and distribution system anomalies, leading to equipment damage caused by system failures. Summary of the Invention

[0003] An embodiment of the present invention provides an instrument identification method for a power supply and distribution system inspection robot, aiming to solve the problems of low efficiency of manual inspection readings, large reading errors, and failure to timely detect abnormalities in the power supply and distribution system in existing inspection methods, resulting in equipment damage caused by system failures. The method obtains first image data of an instrument to be identified, performs instrument detection processing on the first image data through a trained instrument detection model, and determines whether the instrument to be identified is in a normal operating state. When the instrument to be identified is in a normal operating state, the method obtains second image data of the instrument to be identified, performs reading recognition processing on the second image data through a trained instrument reading recognition model, and obtains reading data of the instrument to be identified. The inspection robot is controlled to output the reading data, and based on the output reading data, determines whether there is an abnormality in the power supply and distribution system. If there is an abnormality in the power supply and distribution system, an early warning prompt is issued to the user, and historical data of the power supply and distribution system is obtained, and the power supply and distribution system is inspected in combination with the historical data. The system conducts trend analysis to determine whether the equipment in the power supply and distribution system is continuously deteriorating. If the equipment is continuously deteriorating, repair and / or replacement is arranged. Computer vision technology is used to collect and analyze images of various instruments in the power supply and distribution system, and deep learning algorithms are used to accurately identify instrument readings. Through data analysis, the power supply and distribution system is monitored and warned in real time, which can promptly detect abnormalities in the power supply and distribution system and avoid equipment damage caused by system failures. The stability and reliability of the system are improved, and the existing inspection methods are solved. The problems of low efficiency of manual inspection readings, large reading errors, and failure to promptly detect abnormalities in the power supply and distribution system, resulting in equipment damage caused by system failures.

[0004] In a first aspect, an embodiment of the present invention provides an instrument identification method for a power supply and distribution system inspection robot, the method comprising the following steps:

[0005] Acquire first image data of the instrument to be identified;

[0006] Performing instrument detection processing on the first image data using a trained instrument detection model to determine whether the instrument to be identified is in a normal operating state;

[0007] If the instrument to be identified is in a normal operating state, obtaining second image data of the instrument to be identified;

[0008] Performing reading recognition processing on the second image data using a trained instrument reading recognition model to obtain reading data of the instrument to be recognized, and controlling the inspection robot to output the reading data;

[0009] Based on the output reading data, determining whether there is an abnormality in the power supply and distribution system, and if there is an abnormality in the power supply and distribution system, issuing an early warning prompt to the user;

[0010] Obtain historical data of the power supply and distribution system, perform trend analysis on the power supply and distribution system in combination with the historical data, determine whether equipment in the power supply and distribution system continues to deteriorate, and if so, arrange for repair and / or replacement.

[0011] Optionally, before performing instrument detection processing on the first image data using the trained instrument detection model to determine whether the instrument is in a normal operating state, the method further includes:

[0012] Obtaining an instrument image training set and an untrained instrument detection model, wherein the instrument image training set includes sample instrument images and annotated data corresponding to the sample instrument images;

[0013] The sample instrument image is input into the untrained instrument detection model for training, and the parameters of the model are adjusted according to the labeled data corresponding to the sample instrument image. After the training is completed, a trained instrument detection model is obtained.

[0014] Optionally, performing instrument detection processing on the first image data using a trained instrument detection model to determine whether the instrument is in a normal operating state includes:

[0015] preprocessing the first image data to obtain preprocessed first image data;

[0016] Performing instrument recognition processing on the pre-processed first image data using a trained instrument detection model to identify the instrument type of the instrument to be identified;

[0017] Based on the instrument type, it is determined whether the instrument to be identified is in a normal operating state.

[0018] Optionally, before performing reading recognition processing on the second image data using the trained instrument reading recognition model to obtain the instrument reading data, the method further includes:

[0019] Obtaining a meter reading image training set and an untrained meter reading recognition model, wherein the meter reading image training set includes sample meter reading images and annotated data corresponding to the sample meter reading images;

[0020] The sample meter reading image is input into the untrained meter reading recognition model for training, and the parameters of the model are adjusted according to the labeled data corresponding to the sample meter reading image. After the training is completed, a trained meter reading recognition model is obtained.

[0021] Optionally, performing reading recognition processing on the second image data using a trained instrument reading recognition model to obtain the reading data of the instrument to be recognized includes:

[0022] Performing pointer recognition processing on the second image data using a trained instrument reading recognition model to determine the pointer position of the instrument to be recognized;

[0023] Perform reading recognition processing on the pointer position to obtain reading data of the instrument to be recognized.

[0024] Optionally, the controlling the inspection robot to output the reading data includes:

[0025] Performing digital conversion on the reading data to obtain digital data;

[0026] Based on the digital data, the inspected robot is controlled to output the digital data.

[0027] Optionally, after performing instrument detection processing on the first image data using the trained instrument detection model to determine whether the instrument to be identified is in a normal operating state, the method further includes:

[0028] If the instrument is in an abnormal operating state, abnormal data in the first image data is determined; based on the abnormal data, the inspection robot is controlled to report the abnormal data to a superior.

[0029] In a second aspect, an embodiment of the present invention further provides an instrument identification device for a power supply and distribution system inspection robot, the instrument identification device for the power supply and distribution system inspection robot comprising:

[0030] A first acquisition module, for first image data of the instrument to be identified;

[0031] A first processing module is configured to perform instrument detection processing on the first image data using a trained instrument detection model to determine whether the instrument to be identified is in a normal operating state;

[0032] A second acquisition module is used to acquire second image data of the instrument to be identified if the instrument to be identified is in a normal operating state;

[0033] The second processing module performs reading recognition processing on the second image data using a trained instrument reading recognition model to obtain reading data of the instrument to be recognized, and controls the inspection robot to output the reading data;

[0034] a first determining module configured to determine whether an abnormality exists in the power supply and distribution system based on the output reading data, and to issue an early warning to a user if an abnormality exists in the power supply and distribution system;

[0035] The second determination module is used to obtain historical data of the power supply and distribution system, perform trend analysis on the power supply and distribution system based on the historical data, and determine whether the equipment in the power supply and distribution system continues to deteriorate. If the equipment continues to deteriorate, arrange for repair and / or replacement.

[0036] In a third aspect, an embodiment of the present invention provides an electronic device comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the steps of the instrument identification method for the power supply and distribution system inspection robot provided in an embodiment of the present invention are implemented.

[0037] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the instrument identification method of the power supply and distribution system inspection robot provided in the embodiment of the invention are implemented.

[0038] In an embodiment of the present invention, first image data of an instrument to be identified is obtained; instrument detection processing is performed on the first image data using a trained instrument detection model to determine whether the instrument to be identified is in a normal operating state; if the instrument to be identified is in a normal operating state, second image data of the instrument to be identified is obtained; reading recognition processing is performed on the second image data using a trained instrument reading recognition model to obtain reading data of the instrument to be identified, and the inspection robot is controlled to output the reading data. Based on the output reading data, it is determined whether there is an abnormality in the power supply and distribution system. If there is an abnormality in the power supply and distribution system, an early warning prompt is issued to the user; historical data of the power supply and distribution system is obtained, and trend analysis is performed on the power supply and distribution system in combination with the historical data to determine whether the equipment in the power supply and distribution system continues to deteriorate. If the equipment continues to deteriorate, maintenance and / or replacement is arranged. The present invention uses computer vision technology to collect and analyze images of various instruments in the power supply and distribution system, and uses deep learning algorithms to accurately identify instrument readings. Through data analysis, the power supply and distribution system is monitored and warned in real time, which can promptly detect abnormalities in the power supply and distribution system, avoid damage to equipment due to system failures, improve the stability and reliability of the system, and solve the problems of low efficiency of manual inspection readings, large reading errors, and failure to promptly detect abnormalities in the power supply and distribution system due to system failures in existing inspection methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0040] Figure 1 This is a flow chart of an instrument identification method for a power supply and distribution system inspection robot provided by an embodiment of the present invention;

[0041] Figure 2 This is a schematic structural diagram of an instrument identification device for a power supply and distribution system inspection robot provided by an embodiment of the present invention;

[0042] Figure 3 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0044] like Figure 1 As shown, Figure 1 This is a flow chart of an instrument identification method for a power supply and distribution system inspection robot provided by an embodiment of the present invention. The instrument identification method for a power supply and distribution system inspection robot includes the following steps:

[0045] 101. Acquire first image data of a meter to be identified.

[0046] In an embodiment of the present invention, the above-mentioned meter recognition method of the power supply and distribution system inspection robot can be applied to a server, and the server is in communication with the power distribution system inspection robot. The above-mentioned power distribution system inspection robot includes an image acquisition device, which is used to acquire image data.

[0047] The above-mentioned instruments to be identified may be instruments such as voltmeters, ammeters, and power meters in the power supply and distribution system.

[0048] The first image data can be understood as image information about the meter's appearance and status. This first image data is acquired in real time by the image acquisition equipment of the power distribution system inspection robot. This first image data includes image data such as the meter's appearance, location, and status.

[0049] 102. Perform instrument detection processing on the first image data using the trained instrument detection model to determine whether the instrument to be identified is in a normal operating state.

[0050] In an embodiment of the present invention, the trained instrument detection model may be an instrument detection model constructed based on deep learning or machine learning, such as LeNet-5, ResNet, etc.

[0051] The above-mentioned trained instrument detection model is obtained by training an untrained instrument detection model using an instrument image training set. The above-mentioned instrument image training set includes sample instrument images and annotated data corresponding to the sample instrument images. The above-mentioned sample instrument images can be understood as pictures of instruments (such as voltmeters, ammeters, power meters, etc.) in various states and conditions; the annotated data corresponding to the above-mentioned sample instrument images can be understood as relevant information or labels of the instrument images, describing the content in the picture, such as the type and state of the instrument. The above-mentioned untrained instrument detection model can be an instrument detection model built based on deep learning or machine learning, such as LeNet-5, ResNet, etc. The above-mentioned training can be supervised training. Supervised training refers to a machine learning technology that uses labeled training data to train the model so that the model learns the mapping relationship between input features and output labels, thereby being able to predict and classify new and unseen data.

[0052] The trained instrument detection model can automatically identify and classify instruments, locate their positions, and determine whether the instruments are in normal operation.

[0053] The above-mentioned instrument detection process can be understood as a process of analyzing the operating status of the instrument to determine whether the instrument is operating normally.

[0054] It should be noted that by performing instrument detection processing on the first image data of the instrument and judging the operating status of the instrument, intuitive and comprehensive monitoring of the instrument status of the power supply and distribution system can be achieved, making it convenient for operation and maintenance personnel to grasp the system operation status in a timely manner.

[0055] 103. If the instrument to be identified is in a normal operating state, obtain second image data of the instrument to be identified.

[0056] In the embodiment of the present invention, when the instrument to be identified is in a normal operating state, the second image data of the instrument to be identified is collected in real time by an image collection device.

[0057] The second image data mentioned above can be understood as image data within the instrument panel, and the second image data includes digital information of the instrument and pointer position, etc.

[0058] 104. Perform reading recognition processing on the second image data using the trained instrument reading recognition model to obtain reading data of the instrument to be recognized, and control the inspection robot to output the reading data.

[0059] In an embodiment of the present invention, the trained meter reading recognition model is obtained by training an untrained meter reading recognition model using a training set of meter reading images. The untrained meter reading recognition model can be a meter reading recognition model constructed based on deep learning or machine learning, such as YOLOv5s, YOLOv8x-pose, etc. The meter reading image training set includes sample meter reading images and annotated data corresponding to the sample meter reading images. The sample meter reading images contain readings of various instruments, such as voltage meters, ammeters, and power meters. The annotated data refers to the annotated data that adds structured labels or annotations to the meter reading images so that they can be understood and used by the machine learning model. The training can be supervised training, which refers to the process of training a model using labeled training data to enable the model to learn the mapping relationship between input features and output labels, thereby enabling it to predict and classify new, unseen data.

[0060] The trained instrument reading recognition model can accurately identify and interpret the instrument readings in the image.

[0061] The above reading recognition process can be understood as a process of extracting and recognizing digital information on the meter from the second image data.

[0062] The above reading data can be understood as data describing the reading of the instrument, including parameter data such as voltage, current, and temperature.

[0063] The above output can be understood as the process of transmitting reading data to superiors or recipients, such as voice output, visual output, etc.

[0064] In an embodiment of the present invention, the present invention collects and analyzes images of various instruments in the power supply and distribution system, and accurately identifies instrument readings through a deep learning algorithm, thereby solving the problems of low efficiency, large errors, and high risks of manual inspection readings in existing inspection methods.

[0065] 105. Based on the output reading data, determine whether there is an abnormality in the power supply and distribution system. If there is an abnormality in the power supply and distribution system, issue an early warning prompt to the user.

[0066] In an embodiment of the present invention, the above readings include parameter data such as voltage, current, and temperature.

[0067] Specifically, the meter reading data can be compared with the normal meter parameter range. If it exceeds the normal range, it is determined that there is an abnormality in the power supply and distribution system, and a warning can be issued to the user so that the user can take timely measures to deal with it.

[0068] 106. Obtain historical data on the power supply and distribution system, conduct trend analysis on the power supply and distribution system based on the historical data, and determine whether equipment in the power supply and distribution system is continuously deteriorating. If the equipment is continuously deteriorating, arrange for repair and / or replacement.

[0069] In the embodiment of the present invention, the historical data of the power supply and distribution system can be understood as the historical operation data of the equipment in the power supply and distribution system, including the use time, operation status, fault records and other data of the equipment.

[0070] The above trend analysis can be understood as an analysis process of identifying development trends and patterns in the data by analyzing historical operating data in the power supply and distribution system.

[0071] Specifically, by analyzing the usage time, operating status, fault records and other data of the equipment in the power supply and distribution system and combining them with the current reading data, it is possible to determine whether the equipment in the power supply and distribution system is continuously deteriorating. For example, if the equipment has a high failure rate, high maintenance frequency, and rapid performance degradation, and the current reading data exceeds the normal parameter range, and there is an abnormality in the power supply and distribution system, it is determined that the equipment in the power supply and distribution system is continuously deteriorating.

[0072] The above-mentioned continuous deterioration can be understood as a gradual decline in the performance of equipment in the power supply and distribution system, such as reduced efficiency and increased failure rate.

[0073] Furthermore, when the equipment continues to deteriorate, the equipment can be repaired and / or replaced, reducing the risk of equipment failure.

[0074] In an embodiment of the present invention, first image data of an instrument to be identified is obtained; instrument detection processing is performed on the first image data using a trained instrument detection model to determine whether the instrument to be identified is in a normal operating state; if the instrument to be identified is in a normal operating state, second image data of the instrument to be identified is obtained; reading recognition processing is performed on the second image data using a trained instrument reading recognition model to obtain reading data of the instrument to be identified, and the inspection robot is controlled to output the reading data. Based on the output reading data, it is determined whether there is an abnormality in the power supply and distribution system. If there is an abnormality in the power supply and distribution system, an early warning prompt is issued to the user; historical data of the power supply and distribution system is obtained, and trend analysis is performed on the power supply and distribution system in combination with the historical data to determine whether the equipment in the power supply and distribution system continues to deteriorate. If the equipment continues to deteriorate, maintenance and / or replacement is arranged. The present invention uses computer vision technology to collect and analyze images of various instruments in the power supply and distribution system, and uses deep learning algorithms to accurately identify instrument readings. Through data analysis, the power supply and distribution system is monitored and warned in real time, which can promptly detect abnormalities in the power supply and distribution system, avoid damage to equipment due to system failures, improve the stability and reliability of the system, and solve the problems of low efficiency of manual inspection readings, large reading errors, and failure to promptly detect abnormalities in the power supply and distribution system due to system failures in existing inspection methods.

[0075] It is understandable that in the specific implementation of this application, related data such as image data, knowledge data, user data, etc. are involved. When the embodiments in this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data, as well as the training, deployment and calling of algorithm models, must comply with relevant laws, regulations and standards of relevant countries and regions.

[0076] Optionally, before performing instrument detection processing on the first image data through a trained instrument detection model to determine whether the instrument is in a normal operating state, an instrument image training set and an untrained instrument detection model can also be obtained; the sample instrument image is input into the untrained instrument detection model for training, and the model parameters are adjusted according to the labeled data corresponding to the sample instrument image. After the training is completed, a trained instrument detection model is obtained.

[0077] In an embodiment of the present invention, the instrument image training set includes sample instrument images and annotation data corresponding to the sample instrument images. The sample instrument images include images of instruments (such as voltmeters, ammeters, and power meters) in various states and conditions. The annotation data corresponding to the sample instrument images can be understood as relevant information or labels describing the contents of the image, such as the type and state of the instrument.

[0078] The untrained instrument detection model may be an instrument detection model built based on deep learning or machine learning, such as LeNet-5, ResNet, etc.

[0079] The above training can be supervised training. Supervised training refers to a machine learning technology that uses labeled training data to train a model so that the model learns the mapping relationship between input features and output labels, thereby being able to predict and classify new, unseen data.

[0080] Parameter adjustment is the process of optimizing model performance by adjusting parameters such as weights and biases within the model. During training, the model parameters are optimized using labeled data corresponding to sample instrument images to achieve better prediction or decision-making capabilities.

[0081] The trained instrument detection model can automatically identify and classify instruments, locate their positions, and determine whether the instruments are in normal operation.

[0082] Optionally, in the step of performing instrument detection processing on the first image data through the trained instrument detection model to determine whether the instrument is in a normal operating state, the first image data can be preprocessed to obtain preprocessed first image data; instrument recognition processing is performed on the preprocessed first image data through the trained instrument detection model to identify the instrument type of the instrument to be identified; based on the instrument type, it is determined whether the instrument to be identified is in a normal operating state.

[0083] In the embodiment of the present invention, the first image data may be understood as image information of the appearance and status of the instrument, including image data such as the appearance, position, and status of the instrument.

[0084] The above-mentioned preprocessing can be understood as a process of performing image denoising, image enhancement, and other processing on the first image data. The above-mentioned image denoising refers to reducing noise in the image and improving the signal-to-noise ratio of the image; the above-mentioned image enhancement can be understood as enhancing the contrast and details of the image to make the image clearer.

[0085] The trained instrument detection model is obtained by training an untrained instrument detection model using an instrument image training set. The trained instrument detection model can automatically identify and classify instruments, locate their positions, and determine whether the instruments are in normal operation.

[0086] The above-mentioned instrument identification process can be understood as a process of identifying the type of instrument.

[0087] The above-mentioned instrument types include voltmeters, ammeters, power meters, etc.

[0088] In one possible embodiment, when the identified meter type is a voltmeter, whether the voltmeter is in normal operation can be determined based on the on / off state of the voltmeter's switch and whether its appearance is deformed, damaged, or leaking. If the voltmeter is powered on and its appearance is not deformed, damaged, or leaking, the voltmeter is determined to be in normal operation; if the voltmeter is powered on and its appearance is deformed, damaged, or leaking, the voltmeter is determined to be in abnormal operation; if the voltmeter is powered on and its appearance is deformed, damaged, or leaking, the voltmeter is determined to be in abnormal operation, etc.

[0089] In another possible embodiment, when the identified meter type is a power meter, whether the power meter is in normal operation can be determined based on the power meter's switch status and whether its appearance is deformed, damaged, or leaking. If the power meter is powered on and its appearance is not deformed, damaged, or leaking, the power meter is determined to be in normal operation; if the power meter is powered on and its appearance is deformed, damaged, or leaking, the power meter is determined to be in abnormal operation; if the power meter is powered on and its appearance is deformed, damaged, or leaking, the power meter is determined to be in abnormal operation, etc.

[0090] Optionally, before the step of performing reading recognition processing on the second image data through the trained instrument reading recognition model to obtain the instrument reading data, a meter reading image training set and an untrained instrument reading recognition model can also be obtained; the sample instrument reading image is input into the untrained instrument reading recognition model for training, and the model parameters are adjusted according to the labeled data corresponding to the sample instrument reading image. After the training is completed, a trained instrument reading recognition model is obtained.

[0091] In an embodiment of the present invention, the instrument reading image training set includes sample instrument reading images and annotated data corresponding to the sample instrument reading images. The sample instrument reading images include various instrument readings, such as voltmeters, ammeters, and power meters. The annotated data refers to the addition of structured labels or annotations to the instrument reading images, making them understandable and usable by the machine learning model.

[0092] The untrained meter reading recognition model may be a meter reading recognition model constructed based on deep learning or machine learning, for example, YOLOv5s, YOLOv8x-pose, etc.

[0093] The above training can be supervised training. Supervised training refers to a machine learning technology that uses labeled training data to train a model so that the model learns the mapping relationship between input features and output labels, thereby being able to predict and classify new, unseen data.

[0094] Parameter adjustment is the process of optimizing model performance by adjusting parameters such as weights and biases within the model. During training, the model parameters are optimized using labeled data corresponding to sample meter reading images to achieve better prediction or decision-making capabilities.

[0095] The trained instrument reading recognition model can accurately identify and interpret the instrument readings in the image.

[0096] Optionally, in the step of performing reading recognition processing on the second image data through the trained instrument reading recognition model to obtain the reading data of the instrument to be identified, the second image data can be subjected to pointer recognition processing through the trained instrument reading recognition model to determine the pointer position of the instrument to be identified; and reading recognition processing can be performed on the pointer position to obtain the reading data of the instrument to be identified.

[0097] In an embodiment of the present invention, the trained meter reading recognition model is obtained by training an untrained meter reading recognition model using a training set of meter reading images, and can accurately recognize and interpret meter readings in images.

[0098] The second image data mentioned above can be understood as image data within the instrument panel, and the second image data includes digital information of the instrument and pointer position, etc.

[0099] The above-mentioned pointer recognition process can be understood as a recognition process of analyzing the position of the pointer in the second image by using a trained instrument reading recognition model.

[0100] The above reading recognition process can be understood as a recognition process for determining the specific value pointed to by the pointer position.

[0101] The above reading data can be understood as data describing the reading of the instrument, including parameter data such as voltage, current, and temperature.

[0102] Optionally, in the step of controlling the inspection robot to output the reading data, the reading data may be digitally converted to obtain digital data; and based on the digital data, the inspection robot may be controlled to output the digital data.

[0103] In the embodiment of the present invention, the digital conversion can be understood as a process of converting reading data into digital signals or data, for example, converting the reading data of a voltmeter into digital data representing the current voltage.

[0104] The above output can be understood as the process of transmitting reading data to superiors or recipients, such as voice output, visual output, etc.

[0105] Optionally, after performing instrument detection processing on the first image data through the trained instrument detection model to determine whether the instrument to be identified is in a normal operating state, if the instrument is in an abnormal operating state, abnormal data in the first image data is determined; based on the abnormal data, the inspection robot is controlled to report the abnormal data to the superior.

[0106] In the embodiment of the present invention, when the meter is in an abnormal operating state, abnormal data in the first image data may be determined.

[0107] The above-mentioned abnormal data includes the appearance image data of the instrument, for example, when the instrument is in the off state, the appearance is not deformed, damaged, leaked, etc.; when the instrument is in the on state, the appearance is deformed, damaged, leaked, etc.; when the power meter is in the off state, the appearance is deformed, damaged, leaked, etc.

[0108] Specifically, the abnormal data in the first image data may be presented to a superior in a visual manner.

[0109] In the embodiment of the present invention, the present invention reports abnormal data of the instrument to the superior, which can facilitate the superior operation and maintenance personnel to timely understand the operation status of the instrument and make scientific decisions in a timely manner.

[0110] like Figure 2 As shown, an embodiment of the present invention provides an instrument identification device for a power supply and distribution system inspection robot, the instrument identification device for the power supply and distribution system inspection robot comprising:

[0111] A first acquisition module 201 is used to acquire first image data of a meter to be identified;

[0112] A first processing module 202 is configured to perform instrument detection processing on the first image data using a trained instrument detection model to determine whether the instrument to be identified is in a normal operating state;

[0113] The second acquisition module 203 is configured to acquire second image data of the instrument to be identified if the instrument to be identified is in a normal operating state;

[0114] The second processing module 204 is configured to perform reading recognition processing on the second image data using a trained instrument reading recognition model to obtain reading data of the instrument to be recognized, and control the inspection robot to output the reading data;

[0115] A first determining module 205 is configured to determine whether there is an abnormality in the power supply and distribution system based on the output reading data, and to issue an early warning to a user if there is an abnormality in the power supply and distribution system;

[0116] The second determination module 206 is used to obtain historical data of the power supply and distribution system, perform trend analysis on the power supply and distribution system in combination with the historical data, and determine whether equipment in the power supply and distribution system continues to deteriorate. If equipment continues to deteriorate, arrange for repair and / or replacement.

[0117] Optionally, the device further includes:

[0118] a third acquisition module, configured to acquire an instrument image training set and an untrained instrument detection model, wherein the instrument image training set includes sample instrument images and annotated data corresponding to the sample instrument images;

[0119] The first training module is used to input the sample instrument image into the untrained instrument detection model for training, adjust the parameters of the model according to the labeled data corresponding to the sample instrument image, and obtain a trained instrument detection model after the training is completed.

[0120] Optionally, the first processing module includes:

[0121] a first processing submodule, configured to preprocess the first image data to obtain preprocessed first image data;

[0122] A second processing submodule is configured to perform instrument recognition processing on the pre-processed first image data using a trained instrument detection model to identify the instrument type of the instrument to be recognized;

[0123] The determination submodule is configured to determine whether the to-be-identified instrument is in a normal operating state based on the instrument type.

[0124] Optionally, the device further includes:

[0125] a fourth acquisition module, configured to acquire a meter reading image training set and an untrained meter reading recognition model, wherein the meter reading image training set includes sample meter reading images and annotated data corresponding to the sample meter reading images;

[0126] The second training module is used to input the sample meter reading image into the untrained meter reading recognition model for training, adjust the parameters of the model according to the labeled data corresponding to the sample meter reading image, and obtain a trained meter reading recognition model after the training is completed.

[0127] Optionally, the second processing module includes:

[0128] a third processing submodule, configured to perform pointer recognition processing on the second image data using a trained instrument reading recognition model to determine the pointer position of the instrument to be recognized;

[0129] The fourth processing submodule is used to perform reading recognition processing on the pointer position to obtain reading data of the instrument to be recognized.

[0130] Optionally, the second processing module includes:

[0131] a fifth processing submodule, configured to digitally convert the reading data to obtain digital data;

[0132] The output submodule is used to control the inspected robot to output the digitized data based on the digitized data.

[0133] Optionally, the device further includes:

[0134] a determination module, configured to determine abnormal data in the first image data if the instrument is in an abnormal operating state;

[0135] The third processing module is used to control the inspection robot to report the abnormal data to a superior based on the abnormal data.

[0136] like Figure 3 As shown, an embodiment of the present invention further provides an electronic device, including a processor, and the processor can execute any of the above-mentioned instrument identification methods of the power supply and distribution system inspection robot.

[0137] Specifically, the method includes a processor 301, a memory 302, and a computer program stored in the memory 302 and capable of running on the processor 301 for executing the meter identification method of the power supply and distribution system inspection robot, wherein:

[0138] The processor 301 runs the computer program of the meter identification method of the power supply and distribution system inspection robot stored in the memory 302 and performs the following steps:

[0139] Acquire first image data of the instrument to be identified;

[0140] Performing instrument detection processing on the first image data using a trained instrument detection model to determine whether the instrument to be identified is in a normal operating state;

[0141] If the instrument to be identified is in a normal operating state, obtaining second image data of the instrument to be identified;

[0142] Performing reading recognition processing on the second image data using a trained instrument reading recognition model to obtain reading data of the instrument to be recognized, and controlling the inspection robot to output the reading data;

[0143] Based on the output reading data, determining whether there is an abnormality in the power supply and distribution system, and if there is an abnormality in the power supply and distribution system, issuing an early warning prompt to the user;

[0144] Obtain historical data of the power supply and distribution system, perform trend analysis on the power supply and distribution system in combination with the historical data, determine whether equipment in the power supply and distribution system continues to deteriorate, and if so, arrange for repair and / or replacement.

[0145] Optionally, before performing instrument detection processing on the first image data using the trained instrument detection model to determine whether the instrument is in a normal operating state, the method executed by the processor 301 further includes:

[0146] Obtaining an instrument image training set and an untrained instrument detection model, wherein the instrument image training set includes sample instrument images and annotated data corresponding to the sample instrument images;

[0147] The sample instrument image is input into the untrained instrument detection model for training, and the parameters of the model are adjusted according to the labeled data corresponding to the sample instrument image. After the training is completed, a trained instrument detection model is obtained.

[0148] Optionally, the processor 301 performs instrument detection processing on the first image data using a trained instrument detection model to determine whether the instrument is in a normal operating state, including:

[0149] preprocessing the first image data to obtain preprocessed first image data;

[0150] Performing instrument recognition processing on the pre-processed first image data using a trained instrument detection model to identify the instrument type of the instrument to be identified;

[0151] Based on the instrument type, it is determined whether the instrument to be identified is in a normal operating state.

[0152] Optionally, before performing reading recognition processing on the second image data using the trained instrument reading recognition model to obtain the instrument reading data, the method executed by the processor 301 further includes:

[0153] Obtaining a meter reading image training set and an untrained meter reading recognition model, wherein the meter reading image training set includes sample meter reading images and annotated data corresponding to the sample meter reading images;

[0154] The sample meter reading image is input into the untrained meter reading recognition model for training, and the parameters of the model are adjusted according to the labeled data corresponding to the sample meter reading image. After the training is completed, a trained meter reading recognition model is obtained.

[0155] Optionally, the processor 301 performs reading recognition processing on the second image data using a trained instrument reading recognition model to obtain the reading data of the instrument to be recognized, including:

[0156] Performing pointer recognition processing on the second image data using a trained instrument reading recognition model to determine the pointer position of the instrument to be recognized;

[0157] Perform reading recognition processing on the pointer position to obtain reading data of the instrument to be recognized.

[0158] Optionally, the controlling of the inspection robot to output the reading data executed by the processor 301 includes:

[0159] Performing digital conversion on the reading data to obtain digital data;

[0160] Based on the digital data, the inspected robot is controlled to output the digital data.

[0161] Optionally, after performing instrument detection processing on the first image data using the trained instrument detection model to determine whether the instrument to be identified is in a normal operating state, the method executed by the processor 301 further includes:

[0162] If the instrument is in an abnormal operating state, abnormal data in the first image data is determined; based on the abnormal data, the inspection robot is controlled to report the abnormal data to a superior.

[0163] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the various processes of the instrument identification method of the power supply and distribution system inspection robot provided in the embodiment of the present invention are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0164] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0165] The above disclosure is merely a preferred embodiment of the present invention and certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A method for identifying instruments of a power supply and distribution system inspection robot, characterized in that: The method comprises the following steps: Acquire first image data of the instrument to be identified; Performing instrument detection processing on the first image data using a trained instrument detection model to determine whether the instrument to be identified is in a normal operating state; If the instrument to be identified is in a normal operating state, obtaining second image data of the instrument to be identified; Performing reading recognition processing on the second image data using a trained instrument reading recognition model to obtain reading data of the instrument to be recognized, and controlling the inspection robot to output the reading data; Based on the output reading data, determining whether there is an abnormality in the power supply and distribution system, and if there is an abnormality in the power supply and distribution system, issuing an early warning prompt to the user; Obtain historical data of the power supply and distribution system, perform trend analysis on the power supply and distribution system in combination with the historical data, determine whether equipment in the power supply and distribution system continues to deteriorate, and if so, arrange for repair and / or replacement.

2. The instrument identification method for a power supply and distribution system inspection robot according to claim 1, characterized in that: Before performing instrument detection processing on the first image data using the trained instrument detection model to determine whether the instrument is in a normal operating state, the method further includes: Obtaining an instrument image training set and an untrained instrument detection model, wherein the instrument image training set includes sample instrument images and annotated data corresponding to the sample instrument images; The sample instrument image is input into the untrained instrument detection model for training, and the parameters of the model are adjusted according to the labeled data corresponding to the sample instrument image. After the training is completed, a trained instrument detection model is obtained.

3. The instrument identification method for a power supply and distribution system inspection robot according to claim 2, characterized in that: The performing instrument detection processing on the first image data using the trained instrument detection model to determine whether the instrument is in a normal operating state includes: preprocessing the first image data to obtain preprocessed first image data; Performing instrument recognition processing on the pre-processed first image data using a trained instrument detection model to identify the instrument type of the instrument to be identified; Based on the instrument type, it is determined whether the instrument to be identified is in a normal operating state.

4. The instrument identification method for a power supply and distribution system inspection robot according to claim 3, characterized in that: Before performing reading recognition processing on the second image data using the trained instrument reading recognition model to obtain the instrument reading data, the method further includes: Obtaining a meter reading image training set and an untrained meter reading recognition model, wherein the meter reading image training set includes sample meter reading images and annotated data corresponding to the sample meter reading images; The sample meter reading image is input into the untrained meter reading recognition model for training, and the parameters of the model are adjusted according to the labeled data corresponding to the sample meter reading image. After the training is completed, a trained meter reading recognition model is obtained.

5. The instrument identification method for a power supply and distribution system inspection robot according to claim 4, characterized in that: The step of performing reading recognition processing on the second image data using the trained instrument reading recognition model to obtain the reading data of the instrument to be recognized includes: Performing pointer recognition processing on the second image data using a trained instrument reading recognition model to determine the pointer position of the instrument to be recognized; Perform reading recognition processing on the pointer position to obtain reading data of the instrument to be recognized.

6. The instrument identification method for a power supply and distribution system inspection robot according to claim 5, characterized in that: The controlling the inspection robot to output the reading data includes: Performing digital conversion on the reading data to obtain digital data; Based on the digital data, the inspected robot is controlled to output the digital data.

7. The instrument identification method for a power supply and distribution system inspection robot according to claim 1, characterized in that: After performing instrument detection processing on the first image data using the trained instrument detection model to determine whether the instrument to be identified is in a normal operating state, the method further includes: If the meter is in an abnormal operating state, determining abnormal data in the first image data; Based on the abnormal data, the inspection robot is controlled to report the abnormal data to a superior.

8. An instrument identification device for a power supply and distribution system inspection robot, characterized in that: The instrument identification device of the power supply and distribution system inspection robot includes: A first acquisition module, for first image data of the instrument to be identified; A first processing module is configured to perform instrument detection processing on the first image data using a trained instrument detection model to determine whether the instrument to be identified is in a normal operating state; A second acquisition module is used to acquire second image data of the instrument to be identified if the instrument to be identified is in a normal operating state; a second processing module, configured to perform reading recognition processing on the second image data using a trained instrument reading recognition model to obtain reading data of the instrument to be recognized, and control the inspection robot to output the reading data; a first determining module configured to determine whether an abnormality exists in the power supply and distribution system based on the output reading data, and to issue an early warning to a user if an abnormality exists in the power supply and distribution system; The second determination module is used to obtain historical data of the power supply and distribution system, perform trend analysis on the power supply and distribution system based on the historical data, and determine whether the equipment in the power supply and distribution system continues to deteriorate. If the equipment continues to deteriorate, arrange for repair and / or replacement.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the instrument identification method of the power supply and distribution system inspection robot are implemented as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the instrument identification method for a power supply and distribution system inspection robot according to any one of claims 1 to 7.

Images