CN114140707A - Power grid fault inspection method - Google Patents
Power grid fault inspection method Download PDFInfo
- Publication number
- CN114140707A CN114140707A CN202111450273.2A CN202111450273A CN114140707A CN 114140707 A CN114140707 A CN 114140707A CN 202111450273 A CN202111450273 A CN 202111450273A CN 114140707 A CN114140707 A CN 114140707A
- Authority
- CN
- China
- Prior art keywords
- image
- fault
- detected
- optical image
- preset
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000007689 inspection Methods 0.000 title claims abstract description 44
- 230000003287 optical effect Effects 0.000 claims abstract description 118
- 239000000203 mixture Substances 0.000 claims abstract description 35
- 230000008569 process Effects 0.000 claims abstract description 16
- 230000005540 biological transmission Effects 0.000 claims description 20
- 230000015654 memory Effects 0.000 claims description 16
- 238000007781 pre-processing Methods 0.000 claims description 13
- 238000001914 filtration Methods 0.000 claims description 5
- 230000007613 environmental effect Effects 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 4
- 230000002787 reinforcement Effects 0.000 claims description 3
- 230000004927 fusion Effects 0.000 description 4
- 230000008439 repair process Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000006386 memory function Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a power grid fault inspection method, which comprises the following steps: the method comprises the steps that an image acquisition device is controlled to carry out inspection on a target area according to a first preset route to obtain a first optical image in the inspection process; adopting a preset priori knowledge base to carry out fault discrimination on the object to be detected in the first optical image to obtain a first fault recognition result, wherein the preset priori knowledge base comprises Gaussian mixture models of the object to be detected; controlling the image acquisition equipment to inspect the target area according to a second preset route to acquire a second optical image in the inspection process; fusing the first optical image and the second optical image to obtain a fused image, and judging a fault of an object to be detected in the fused image by adopting a preset prior knowledge base to obtain a second fault identification result; and obtaining a fault identification result according to the first fault identification result and the second fault identification result. According to the invention, the priori knowledge base is established through a small amount of samples, and the image to be detected with faults can be accurately identified.
Description
Technical Field
The invention relates to the technical field of power grid fault detection, in particular to a power grid fault inspection method.
Background
At present, a large number of power transmission networks are exposed to the natural environment and are extremely susceptible to the influence of extremely severe weather, such as typhoon, earthquake and the like. After disaster, the manual repair work of the power grid fault is limited by extreme geographical environment and spanning long-distance power transmission network, and huge labor cost and time cost are needed, even the life cost of the repair personnel for operation. Although the target detection technology of deep learning makes a major breakthrough in the field of computers, the quality of a deep learning model depends on the quantity and quality of training data to a great extent, and a data set caused in the process of emergency repair of a power transmission line after a disaster is limited, so that accurate identification of faults cannot be realized through deep learning.
Disclosure of Invention
Therefore, the technical problem to be solved by the invention is to overcome the defect that the accurate identification of the fault cannot be realized in the prior art, thereby providing a power grid fault inspection method.
The invention provides a power grid fault inspection method in a first aspect, which comprises the following steps: controlling an image acquisition device to inspect a target area according to a first preset route, and acquiring a first optical image in an inspection process, wherein the first optical image comprises an object to be detected; adopting a preset priori knowledge base to carry out fault discrimination on the object to be detected in the first optical image to obtain a first fault recognition result, wherein the preset priori knowledge base comprises Gaussian mixture models of the object to be detected; controlling the image acquisition equipment to inspect a target area according to a second preset route, acquiring a second optical image in the inspection process, wherein the first preset route and the second preset route are opposite in direction, and when the first optical image and the second optical image are acquired, the horizontal inclination angles of the image acquisition equipment are different; fusing the first optical image and the second optical image to obtain a fused image, and performing fault discrimination on an object to be detected in the fused image by adopting a preset prior knowledge base to obtain a second fault recognition result; and obtaining a fault identification result according to the first fault identification result and the second fault identification result.
Optionally, in the power grid fault inspection method provided by the present invention, the first preset route is determined through the following steps: acquiring a height map of a target area, and performing three-dimensional modeling on the target area based on the height map to form a virtual environment of the target area; determining a power transmission network topological structure in a target area according to the virtual environment, and determining an initial route according to the power transmission network topological structure; and optimizing the initial route by using reinforcement learning to obtain a first preset route.
Optionally, in the power grid fault inspection method provided by the present invention, the fault discrimination of the object to be detected in the first optical image is performed by using a preset prior knowledge base, so as to obtain a first fault recognition result, including: preprocessing the first optical image to obtain a denoising line characteristic space image; analyzing the de-noising line characteristic space image to obtain at least one candidate frame, wherein the candidate frame comprises an object to be detected; and judging the fault of the object to be detected in the candidate frame by adopting a preset prior knowledge base to obtain a first fault identification result.
Optionally, in the power grid fault inspection method provided by the present invention, preprocessing the first optical image to obtain a denoised line feature space image includes: mapping pixel points in the first optical image to a line characteristic space to obtain a communicated edge profile of each object to be detected; based on the connected edge profile, removing the environmental background noise in the first optical image by adopting a preset image filtering algorithm to obtain a denoising line characteristic space image.
Optionally, in the power grid fault inspection method provided by the present invention, the step of constructing the preset prior knowledge base includes: acquiring known images which are acquired at a plurality of angles and contain an object to be detected; preprocessing the known image to obtain a denoising feature space image of the known image; carrying out plane map Gaussian mixture model modeling on the denoising feature space image of the known image at different angles to respectively obtain a first Gaussian mixture model of each angle; and carrying out secondary modeling on the plane graph Gaussian mixture models of all the angles to obtain a second Gaussian mixture model of each object to be detected.
Optionally, in the power grid fault inspection method provided by the present invention, the preset prior knowledge base further includes a rule model of the object to be detected, and the step of constructing the preset prior knowledge base further includes: acquiring a known image containing an object to be detected; preprocessing the known image to obtain a denoising feature space image of the known image; and summarizing the denoising characteristic space image of the known image to obtain a regular model of the object to be detected.
Optionally, in the power grid fault inspection method provided by the present invention, analyzing the de-noised line feature space image to obtain at least one candidate frame includes: and determining a region with the pixel density larger than a preset value in the denoised line characteristic space image as a candidate frame, and/or fitting each pixel point in the denoised line characteristic space image, and determining an external rectangle of a curve obtained by fitting as the candidate frame.
Optionally, the power grid fault inspection method provided by the invention further comprises the step of determining the geographic coordinates of the image to be inspected in each candidate frame according to the GPS information.
A second aspect of the present invention provides a computer apparatus comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executable by the at least one processor to perform the grid fault routing method as provided in the first aspect of the invention.
A third aspect of the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to perform the grid fault routing inspection method according to the first aspect of the present invention.
The technical scheme of the invention has the following advantages:
the invention provides a power grid fault inspection method, which adopts a preset prior knowledge base to judge faults of objects to be detected in a first optical image and a second optical image, wherein the preset prior knowledge base comprises Gaussian mixture models of the objects to be detected, and because the dependency on samples is lower when the Gaussian mixture models are established, the images to be detected with faults can be accurately identified after a few samples are established in the prior period, in the method provided by the invention, an object area is inspected according to a first preset route and a second preset route respectively to obtain a first optical image and a second optical image, the first optical image is used to obtain a first fault identification result, the second optical image is used to carry out secondary verification to obtain a second fault identification result, and finally the fault identification result is obtained, and the directions of the first optical image and the second optical image are different, the horizontal inclination angles are different, so that the fault identification result obtained by combining the first optical image and the second optical image is more accurate.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a flowchart of a specific example of a power grid fault inspection method according to an embodiment of the present invention;
fig. 2 is a schematic block diagram of a specific example of a power grid fault inspection device in an embodiment of the present invention;
fig. 3 is a schematic block diagram of a specific example of a computer device in the embodiment of the present invention.
Detailed Description
The technical solutions of the present invention will be described clearly and completely with reference to the accompanying drawings, and it should be understood that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the description of the present invention, it should be noted that the terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In addition, the technical features involved in the different embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
The embodiment of the invention provides a power grid fault inspection method, as shown in fig. 1, comprising the following steps:
step S11: and controlling the image acquisition equipment to patrol the target area according to a first preset route, and acquiring a first optical image in the patrol process, wherein the first optical image comprises an object to be detected.
In an optional embodiment, the target area includes a plurality of objects to be detected, which need to detect whether a fault occurs, and the image to be detected includes, but is not limited to, a tower, a power transmission line, and the like.
In an optional embodiment, the image acquisition device may be an aircraft with an image acquisition function, such as an unmanned aerial vehicle, and since the method provided by the embodiment of the present invention is used for detecting a power grid fault, the coverage area of the object to be detected is wide, the target area is large, and the image of each object to be detected in the target area can be rapidly acquired by the unmanned aerial vehicle.
In an alternative embodiment, the first optical image may be a video image or a set of optical images taken consecutively.
In an optional embodiment, the first preset route may be preset before the image acquisition device starts to acquire the image, may be generated by the image acquisition device according to the distribution of the objects to be detected in the target area during the acquisition process, and may be obtained by adjusting a preset path according to the distribution of the objects to be detected during the image acquisition process by the image acquisition device.
In an optional embodiment, when the image acquisition device inspects the target area according to the first preset route, the image acquisition device acquires an image by keeping a certain distance and angle with the object to be detected.
Step S12: and adopting a preset prior knowledge base to judge the fault of the object to be detected in the first optical image to obtain a first fault identification result, wherein the preset prior knowledge base comprises Gaussian mixture models of the objects to be detected.
The gaussian mixture model is a generative model that assumes that the samples obey a gaussian mixture distribution. When a gaussian model is established for a tower, it is assumed that there are k different types of towers in the sample, i.e., the model targets k clusters. Its parameters are only k μ and σ. Only the Gaussian mixture model is used, and the assumption that the data in the samples are subjected to Gaussian mixture distribution is applied, so that the data dependence on modeling of a small number of tower samples is reduced.
In an optional embodiment, if the first fault identification result determines that the object to be detected has a fault, the first fault identification result further includes a fault type and a confidence of the object to be detected.
In an optional embodiment, when the fault is determined, if it is detected that the object to be detected has the fault, the method provided in the embodiment of the present invention further includes determining, by combining with the GPS information, the world coordinate of the object to be detected having the fault.
Step S13: and controlling the image acquisition equipment to patrol the target area according to a second preset route, acquiring a second optical image in the patrol process, wherein the first preset route and the second preset route are opposite in direction, and when the first optical image and the second optical image are acquired, the horizontal inclination angles of the image acquisition equipment are different.
In an alternative embodiment, the first optical image is a plan view and the second optical pattern is a 45 ° side view.
In an alternative embodiment, the image capture device acquires the first optical image and the second optical image at the same frame rate.
In an alternative embodiment, the first predetermined route and the second predetermined route have the same passing path and opposite directions.
In an alternative embodiment, the second optical image may be a video image or a set of optical images taken consecutively.
In an optional embodiment, in the process of acquiring the first optical image, the image acquisition device simultaneously starts a route memory function, records a first preset route actually traveled in the process of acquiring the first optical image, and acquires the second optical image in the process of returning according to the original route of the first preset route, so that the same object to be detected contained in the first optical image and the second optical image is ensured, and the same position of the image to be detected in the first optical image and the second optical image is ensured, so that the efficiency is higher when the fault identification is carried out on the fusion image obtained by fusing the first optical image and the second optical image subsequently, and the identification result is more accurate.
Step S14: and fusing the first optical image and the second optical image to obtain a fused image, and performing fault discrimination on the object to be detected in the fused image by adopting a preset prior knowledge base to obtain a second fault recognition result.
In an alternative embodiment, the first optical image and the second optical image are fused to obtain a plan view and a 45 ° side view of the same image to be detected.
In an alternative embodiment, the gaussian mixture model used in performing step S12 is a plane gaussian mixture model, and the gaussian mixture model used in performing step S14 is obtained by performing quadratic modeling based on the plane gaussian mixture model.
In the embodiment of the invention, the angles when the first optical image is acquired and the second optical image is acquired are different, so that the second fault identification results obtained according to the first optical image and the second optical image are fused to realize complementation, and a more accurate fault identification result can be obtained.
Step S15: and obtaining a fault identification result according to the first fault identification result and the second fault identification result.
In an optional embodiment, the manner of determining the fault result may be that when any one of the first fault identification result and the second fault identification result determines that the object to be detected has a fault, it is determined that the object to be detected has a fault, and when both the first fault identification result and the second fault identification result determine that the object to be detected does not have a fault, it is determined that the object to be detected does not have a fault.
The power grid fault inspection method provided by the embodiment of the invention adopts the preset prior knowledge base to judge the fault of the object to be detected in the first optical image and the second optical image, the preset prior knowledge base comprises the Gaussian mixture model of each object to be detected, and because the dependency on the sample is lower when the Gaussian mixture model is established, the image to be detected with the fault can be accurately identified after the prior knowledge base is established by a small number of samples in the early stage, and in the method provided by the embodiment of the invention, the target area is inspected according to the first preset route and the second preset route respectively to obtain the first optical image and the second optical image, the first fault identification result is obtained by using the first optical image, the second fault identification result is obtained by using the second optical image for secondary verification, the fault identification result is finally obtained, and the directions of the first optical image and the second optical image are different, the horizontal inclination angles are different, so that the fault identification result obtained by combining the first optical image and the second optical image is more accurate.
In an alternative embodiment, the first predetermined route is determined by:
firstly, a height map of a target area is obtained, and three-dimensional modeling is carried out on the target area based on the height map to form a virtual environment of the target area.
In an alternative embodiment, a satellite map may be used to obtain a height map of the target area.
In an alternative embodiment, the target area is modeled three-dimensionally by a HeightMap, which is a gray scale map with gray scale values of 0-255, with the gray scale value of an image point reflecting the height of the point in the original relief.
And then, determining a power transmission network topological structure in the target area according to the virtual environment, and determining an initial route according to the power transmission network topological structure.
In an alternative embodiment, when determining the topology of the transmission network, points in the topology of the transmission network are determined, which are significantly different from surrounding features in a grey scale map due to the fact that the height of the main grid tower base is about 80-120 meters. The specific expression may be an abnormal point of a continuous value-taking image on the height map, and therefore, the height difference between the point and the surrounding area can be obtained through conversion, and the point is inferred to be a tower footing, namely, a point in a graph structure. After determining the points in the graph structure, edges of the graph structure are determined according to expert knowledge, thereby forming the power transmission network topology.
In an alternative embodiment, the initial route determined from the grid topology can pass through every point in the grid topology.
And finally, optimizing the initial route by utilizing reinforcement learning to obtain a first preset route. When the image acquisition equipment acquires the first optical image according to the first preset line, the image acquisition equipment moves at a certain distance and angle with the line to be detected in the target area.
In an optional embodiment, in step S12, the method for determining a fault of the object to be detected in the first optical image by using a preset prior knowledge base to obtain a first fault identification result specifically includes:
firstly, preprocessing a first optical image to obtain a denoising line characteristic space image.
In an alternative embodiment, the preprocessing of the first optical image includes extracting line features of the first optical image, forming connected contours, and performing background filtering on the first optical image to distinguish foreground and background scenes, and for example, distinguishing foreground and background scenes of a picture by a machine learning algorithm, and setting the background scene as empty.
And then, analyzing the de-noising line characteristic space image to obtain at least one candidate frame, wherein the candidate frame comprises the object to be detected.
In an alternative embodiment, the candidate frames of different types of objects to be detected are calculated in different ways.
In an optional embodiment, the method for obtaining the candidate box includes: and determining the area with the pixel density larger than the preset value in the denoising line feature space image as a candidate frame, and exemplarily, if the object to be detected is a tower, determining the candidate frame by using the method.
In an optional embodiment, the method for obtaining the candidate box includes: and fitting each pixel point in the de-noising line characteristic space image, and determining an external rectangle of a curve obtained by fitting as a candidate frame.
And finally, carrying out fault discrimination on the object to be detected in the candidate frame by adopting a preset prior knowledge base to obtain a first fault recognition result.
In an optional embodiment, besides the gaussian mixture model, the prior knowledge base further includes other various fault discrimination models, for different objects to be detected, different fault discrimination models are used for fault recognition, when the objects to be detected are subjected to fault recognition, the objects to be detected are firstly classified, the models of the objects to be detected in the prior knowledge base are judged, if the models corresponding to the objects to be detected are gaussian mixture models, the models are distinguished according to the probability of the gaussian mixture models, a first fault recognition result is obtained, and for towers, illustratively, the mixed gaussian models are used for fault recognition; if the model corresponding to the object to be detected is a rule model, training a classifier according to rules so as to obtain a first fault identification result, and exemplarily, for the power transmission line, performing fault identification by using the rule model.
In an optional embodiment, the step S14 of adopting a preset prior knowledge base to perform fault determination on the object to be detected in the fusion image specifically includes:
firstly, the second optical image is preprocessed to obtain a denoising line feature space image, and for details, reference is made to the description of preprocessing the first optical image in the above embodiment, which is not described herein again.
Then, the de-noised line feature space image is analyzed to obtain at least one candidate frame, where the candidate frame includes the object to be detected, and the detailed content refers to the description of determining the candidate frame in the first optical image in the above embodiment, which is not described herein again.
Finally, a preset priori knowledge base is adopted to perform fault discrimination on the object to be detected in the candidate frame to obtain a second fault recognition result, and for details, reference is made to the description of the embodiment for recognizing the fault of the object to be detected in the first optical image, and details are not repeated here.
In an optional embodiment, the step of preprocessing the first optical image to obtain a denoised line feature space image specifically includes:
firstly, mapping pixel points in a first optical image to a line feature space to obtain a connected edge profile of each object to be detected.
In an optional embodiment, after the pixel points in the first optical image are mapped to the line feature space, an edge profile is obtained, and in order to prevent the line feature pixel points from being broken and discontinuous, corresponding line segments are connected together through a preset rule to form a communicated edge profile.
And then, based on the connected edge profile, removing the environmental background noise in the optical image by adopting a preset image filtering algorithm to obtain a denoising line characteristic space image.
In an optional embodiment, the step of preprocessing the second optical image to obtain a denoised line feature space image specifically includes:
firstly, mapping pixel points in the second optical image to a line feature space to obtain a connected edge contour of each object to be detected, for details, refer to the description of obtaining the connected edge contour according to the first optical image in the above embodiment, and no further description is given here.
And then, based on the connected edge profile, removing the environmental background noise in the optical image by adopting a preset image filtering algorithm to obtain a denoising line characteristic space image.
In an optional embodiment, the step of constructing the predetermined a priori knowledge base includes:
firstly, acquiring known images which are acquired at a plurality of angles and contain an object to be detected;
secondly, the known image is preprocessed to obtain a de-noised feature space image of the known image, and details of the steps for obtaining the de-noised feature space image of the first optical image in the above embodiment are included, which are not described herein again.
And then, carrying out plane map Gaussian mixture model modeling on the denoising feature space image of the known image at different angles to respectively obtain a first Gaussian mixture model of each angle. In an optional embodiment, when fault recognition is performed on the object to be detected in the first optical image and the second optical image, fault recognition is performed through the first gaussian mixture model.
And finally, performing secondary modeling fusion on the plane graph Gaussian mixture models of all angles to obtain second Gaussian mixture models of all objects to be detected. In an optional embodiment, when fault identification is performed based on the first optical image or the second optical image, fault identification is performed by using a first gaussian mixture model, and when fault identification is performed based on a fused image obtained by fusing the first optical image and the second optical image, fault identification is performed by using a second gaussian mixture model.
In an optional embodiment, during secondary modeling, the same object to be detected and pictures at different angles are mapped into clusters in two models, and the two clusters are regarded as the type of the object to be detected, so that the corresponding relation between different model clusters and clusters is completed, and fusion is realized.
In an optional embodiment, the preset prior knowledge base further includes a rule model of the object to be detected, and the step of constructing the preset prior knowledge base further includes:
firstly, acquiring a known image containing an object to be detected;
then, the known image is preprocessed to obtain a de-noised feature space image of the known image, and details of the steps of obtaining the de-noised feature space image of the first optical image in the above embodiment are included, which are not described herein again.
And finally, generalizing the denoising characteristic space image of the known image to obtain a regular model of the object to be detected.
In an alternative embodiment, the rule model may be established for the power line. Aiming at different power transmission lines, simulating the radian of the power transmission line by using a plurality of different natural profiles, recording parameters of the power transmission line, and regarding the power transmission line as a rule, namely considering that the type of the power transmission line should obey the style of the profile, thereby obtaining a rule model of the power transmission line.
An embodiment of the present invention provides a power grid fault inspection device, as shown in fig. 2, including:
the first inspection module 21 is configured to control the image acquisition device to inspect the target area according to a first preset route, and acquire a first optical image in the inspection process, where the first optical image includes an object to be detected, and details of the first optical image refer to the description of step S11 in the foregoing embodiment, and are not described herein again.
The first fault identifying module 22 is configured to perform fault discrimination on the object to be detected in the first optical image by using a preset prior knowledge base to obtain a first fault identifying result, where the preset prior knowledge base includes a gaussian mixture model of each object to be detected, and the detailed content refers to the description of step S12 in the foregoing embodiment, and is not described herein again.
The second inspection module 23 is configured to control the image acquisition device to inspect the target area according to a second preset route, and acquire a second optical image in the inspection process, where the first preset route and the second preset route are opposite in direction, and when the first optical image and the second optical image are acquired, the horizontal tilt angles of the image acquisition device are different, for details, refer to the description of step S13 in the foregoing embodiment, and are not described herein again.
The second fault identifying module 24 is configured to fuse the first optical image and the second optical image to obtain a fused image, and perform fault discrimination on the object to be detected in the fused image by using a preset prior knowledge base to obtain a second fault identifying result, where details are described in the above embodiment in relation to step S14, and are not described herein again.
The fault determining module 25 is configured to obtain a fault identification result according to the first fault identification result and the second fault identification result, refer to the description of step S15 in the foregoing embodiment for details, which are not described herein again.
An embodiment of the present invention provides a computer device, as shown in fig. 3, the computer device mainly includes one or more processors 31 and a memory 32, and one processor 31 is taken as an example in fig. 3.
The computer device may further include: an input device 33 and an output device 34.
The processor 31, the memory 32, the input device 33 and the output device 34 may be connected by a bus or other means, and fig. 3 illustrates the connection by a bus as an example.
The processor 31 may be a Central Processing Unit (CPU). The Processor 31 may also be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, or combinations thereof. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The memory 32 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to use of the power grid fault inspection device, and the like. Further, the memory 32 may include high speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory 32 optionally includes a memory remotely located from the processor 31, and these remote memories may be connected to the grid fault inspection device via a network. The input device 33 may receive user input of a calculation request (or other numeric or alphanumeric information) and generate key signal inputs relating to the grid fault inspection device. The output device 34 may include a display device such as a display screen for outputting the calculation result.
The embodiment of the invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions can execute the power grid fault inspection method in any method embodiment. The storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a Flash Memory (Flash Memory), a Hard Disk (Hard Disk Drive, abbreviated as HDD) or a Solid State Drive (SSD), etc.; the storage medium may also comprise a combination of memories of the kind described above.
It should be understood that the above examples are only for clarity of illustration and are not intended to limit the embodiments. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. And obvious variations or modifications therefrom are within the scope of the invention.
Claims (10)
1. A power grid fault inspection method is characterized by comprising the following steps:
controlling an image acquisition device to inspect a target area according to a first preset route, and acquiring a first optical image in an inspection process, wherein the first optical image comprises an object to be detected;
adopting a preset prior knowledge base to carry out fault discrimination on the object to be detected in the first optical image to obtain a first fault recognition result, wherein the preset prior knowledge base comprises Gaussian mixture models of the objects to be detected;
controlling an image acquisition device to inspect a target area according to a second preset route, and acquiring a second optical image in the inspection process, wherein the first preset route and the second preset route are opposite in direction, and when the first optical image and the second optical image are acquired, the horizontal inclination angles of the image acquisition device are different;
fusing the first optical image and the second optical image to obtain a fused image, and judging a fault of the object to be detected in the fused image by adopting a preset prior knowledge base to obtain a second fault identification result;
and obtaining a fault identification result according to the first fault identification result and the second fault identification result.
2. The grid fault inspection method according to claim 1, wherein the first preset route is determined by:
acquiring a height map of the target area, and performing three-dimensional modeling on the target area based on the height map to form a virtual environment of the target area;
determining a power transmission network topological structure in the target area according to the virtual environment, and determining an initial route according to the power transmission network topological structure;
and optimizing the initial route by utilizing reinforcement learning to obtain the first preset route.
3. The power grid fault inspection method according to claim 1, wherein fault discrimination is performed on the object to be detected in the first optical image by using a preset prior knowledge base to obtain a first fault recognition result, and the method comprises the following steps:
preprocessing the first optical image to obtain a denoising line characteristic space image;
analyzing the de-noising line characteristic space image to obtain at least one candidate frame, wherein the candidate frame comprises an object to be detected;
and adopting the preset prior knowledge base to judge the fault of the object to be detected in the candidate frame to obtain the first fault identification result.
4. The power grid fault inspection method according to claim 3, wherein the preprocessing is performed on the first optical image to obtain a de-noised line characteristic space image, and the method comprises the following steps:
mapping pixel points in the first optical image to a line characteristic space to obtain a communicated edge profile of each object to be detected;
based on the connected edge profile, removing the environmental background noise in the first optical image by adopting a preset image filtering algorithm to obtain the denoising line characteristic space image.
5. The power grid fault inspection method according to claim 1, wherein the step of constructing the preset prior knowledge base comprises:
acquiring known images which are acquired at a plurality of angles and contain an object to be detected;
preprocessing the known image to obtain a denoising feature space image of the known image;
carrying out plane map Gaussian mixture model modeling on the denoising feature space image of the known image at different angles to respectively obtain a first Gaussian mixture model of each angle;
and carrying out secondary modeling on the plane graph Gaussian mixture models of all the angles to obtain a second Gaussian mixture model of each object to be detected.
6. The power grid fault inspection method according to claim 5, wherein the preset prior knowledge base further includes a rule model of an object to be detected, and the step of constructing the preset prior knowledge base further includes:
acquiring a known image containing an object to be detected;
preprocessing the known image to obtain a denoising feature space image of the known image;
and summarizing the denoising characteristic space image of the known image to obtain a regular model of the object to be detected.
7. The power grid fault inspection method according to claim 3, wherein the analyzing the de-noised line feature space image to obtain at least one candidate frame comprises:
determining the area with the pixel density larger than a preset value in the denoising line feature space image as the candidate frame, and/or,
and fitting each pixel point in the denoising line characteristic space image, and determining an external rectangle of a curve obtained by fitting as the candidate frame.
8. The power grid fault inspection method according to claim 3, further comprising:
and determining the geographic coordinates of the image to be detected in each candidate frame according to the GPS information.
9. A computer device, comprising:
at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to perform the grid fault routing method of any one of claims 1-8.
10. A computer-readable storage medium storing computer instructions for causing a computer to perform the grid fault routing inspection method according to any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111450273.2A CN114140707A (en) | 2021-11-30 | 2021-11-30 | Power grid fault inspection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111450273.2A CN114140707A (en) | 2021-11-30 | 2021-11-30 | Power grid fault inspection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114140707A true CN114140707A (en) | 2022-03-04 |
Family
ID=80386471
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111450273.2A Pending CN114140707A (en) | 2021-11-30 | 2021-11-30 | Power grid fault inspection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114140707A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116935245A (en) * | 2023-06-16 | 2023-10-24 | 国网山东省电力公司金乡县供电公司 | Long-distance communication power grid unmanned aerial vehicle inspection system and method |
-
2021
- 2021-11-30 CN CN202111450273.2A patent/CN114140707A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116935245A (en) * | 2023-06-16 | 2023-10-24 | 国网山东省电力公司金乡县供电公司 | Long-distance communication power grid unmanned aerial vehicle inspection system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210319561A1 (en) | Image segmentation method and system for pavement disease based on deep learning | |
| AU2021201579B2 (en) | System and method for detecting features in aerial images using disparity mapping and segmentation techniques | |
| CN111353413B (en) | Low-missing-report-rate defect identification method for power transmission equipment | |
| KR102094341B1 (en) | System for analyzing pot hole data of road pavement using AI and for the same | |
| CN109615611B (en) | Inspection image-based insulator self-explosion defect detection method | |
| Duque et al. | Bridge deterioration quantification protocol using UAV | |
| CN109086668B (en) | Unmanned aerial vehicle remote sensing image road information extraction method based on multi-scale generation countermeasure network | |
| Kwak et al. | Automatic representation and reconstruction of DBM from LiDAR data using Recursive Minimum Bounding Rectangle | |
| CN111784685A (en) | Power transmission line defect image identification method based on cloud edge cooperative detection | |
| US11893538B1 (en) | Intelligent system and method for assessing structural damage using aerial imagery | |
| CN113822247B (en) | Method and system for identifying illegal building based on aerial image | |
| KR102346676B1 (en) | Method for creating damage figure using the deep learning-based damage image classification of facility | |
| CN114519819B (en) | Remote sensing image target detection method based on global context awareness | |
| CN115909096A (en) | Unmanned aerial vehicle cruise pipeline hidden danger analysis method, device and system | |
| CN114140707A (en) | Power grid fault inspection method | |
| US20220083774A1 (en) | Method and system for asset inspection using unmanned aerial vehicles | |
| RU2612571C1 (en) | Method and system for recognizing urban facilities | |
| CN116051980B (en) | Building identification method, system, electronic equipment and medium based on oblique photography | |
| KR102521825B1 (en) | Automatic determination system for determinating abnormal area of structure damage based on point cloud | |
| CN115797310A (en) | Method for determining inclination angle of photovoltaic power station group string and electronic equipment | |
| CN113514053B (en) | Method and device for generating sample image pair and method for updating high-precision map | |
| CN115482277A (en) | Social distance risk early warning method and device | |
| CN114219073A (en) | Method and device for determining attribute information, storage medium and electronic device | |
| KR20170108552A (en) | Information system for analysis of waterfront structure damage | |
| CN113505643B (en) | Method and related device for detecting violation target |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |