CN113129273B - Automatic transmission line fog inspection method and system - Google Patents
Automatic transmission line fog inspection method and system Download PDFInfo
- Publication number
- CN113129273B CN113129273B CN202110343120.1A CN202110343120A CN113129273B CN 113129273 B CN113129273 B CN 113129273B CN 202110343120 A CN202110343120 A CN 202110343120A CN 113129273 B CN113129273 B CN 113129273B
- Authority
- CN
- China
- Prior art keywords
- atomization
- image
- standard
- target
- training
- 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.)
- Active
Links
- 238000007689 inspection Methods 0.000 title claims abstract description 103
- 230000005540 biological transmission Effects 0.000 title claims abstract description 60
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000000889 atomisation Methods 0.000 claims abstract description 306
- 238000003062 neural network model Methods 0.000 claims abstract description 70
- 238000012549 training Methods 0.000 claims abstract description 67
- 238000012545 processing Methods 0.000 claims description 35
- 238000002663 nebulization Methods 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 6
- 230000007547 defect Effects 0.000 description 4
- 238000010438 heat treatment Methods 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 230000008961 swelling Effects 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 238000010521 absorption reaction Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 239000012212 insulator Substances 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000003595 mist Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
-
- 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
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Library & Information Science (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Public Health (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Biophysics (AREA)
- Water Supply & Treatment (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention relates to a method and a system for automatically patrolling a transmission line in a foggy day, wherein the method comprises the steps of establishing a packet standard image library and determining expansion information of standard sample atomized images under different atomization grades; establishing a training image library for training the power transmission line; inputting the training image library into a neural network model for training to obtain model parameter information of the neural network model at different atomization grades; acquiring a standard real-time atomization image and determining an atomization grade during the inspection in the foggy days, and reading corresponding model parameter information; and inputting the acquired target atomization image into the neural network model during the inspection in the foggy day to obtain an inspection result. According to the method, the influence of foggy days on image acquisition is determined, the atomization grade is determined according to the standard real-time atomization image, the neural network model is trained by combining the training image library, and then the trained neural network model obtains an accurate inspection result, so that the accuracy of the foggy day inspection result is greatly improved, and scientific guidance is provided for accurate countermeasures taken by power departments.
Description
Technical Field
The invention relates to the electric power technology, in particular to a method and a system for automatically patrolling a transmission line in a foggy day.
Background
The transmission line is a main carrier of power transmission, and the safety of a power system is directly threatened once a fault occurs. The inspection and the maintenance of the power transmission line are carried out, potential safety hazards and fault defects are timely clear, and the method is an important guarantee for the safe operation of the power system. Traditional inspection work mainly relies on the manual work to carry out.
With the rapid development of the fields of advanced sensing technology, information communication technology, automatic control technology, artificial intelligence technology and the like in recent years, various automatic inspection devices and methods, such as unmanned aerial vehicle inspection, robot inspection and the like, appear. The automatic inspection acquires data through various sensing technologies, and data processing is carried out through an intelligent algorithm to find out defects and potential safety hazards of circuit equipment. The method mainly judges the visual defects of towers, hardware fittings, ground wires, insulators and the like. The accuracy of the routing inspection is closely related to the computing capability of the intelligent algorithm.
However, the existing intelligent algorithm is only weak artificial intelligence, and has some problems in practical application, especially when the use condition is bad, such as in the inspection under the rain and fog condition, the image collected by the system is expanded and blurred due to the absorption and scattering of water drops in the air to the optical fiber, so that the inspection result is always large in deviation, and the power department is inconvenient to take accurate countermeasures.
Disclosure of Invention
The invention aims to solve the technical problem of providing a method and a system for automatically patrolling a power transmission line in a foggy day aiming at the defects of the prior art.
The technical scheme for solving the technical problems is as follows: an automatic inspection method for a power transmission line in a foggy day comprises the following steps:
s1: establishing a standard image library containing standard sample atomization images of the standard image under different atomization levels according to a preset standard image, and determining expansion information of the standard sample atomization images under different atomization levels;
s2: establishing a training image library corresponding to training power transmission lines on sunny days and different atomization levels of weather, wherein the training images of the same target power transmission line in the training image library are correspondingly associated with the atomization training images of different atomization levels;
s3: inputting the training image library into a pre-established neural network model for training, and obtaining model parameter information corresponding to the neural network model on sunny days and weather with different atomization grades;
s4: acquiring a standard real-time atomization image corresponding to the standard graph during the inspection of the foggy day, determining the atomization level according to the standard real-time atomization image, and acquiring model parameter information of the neural network model corresponding to the atomization level;
s5: the method comprises the steps of carrying out image acquisition on a target power transmission line in the process of inspection in a foggy day, inputting an acquired target atomization image into a neural network model, carrying out image processing and analysis on the target atomization image by the neural network model according to expansion information of standard sample atomization images under different atomization levels to obtain expansion information corresponding to the target atomization image, and determining an inspection result according to the expansion information of the target atomization image.
The invention has the beneficial effects that: according to the automatic inspection method for the power transmission line in the foggy day, the expansion information of the standard sample atomized image corresponding to the preset standard image under different atomization levels is predetermined, so that the influence of the foggy day on image acquisition can be calculated, then the atomization level can be determined according to the standard real-time atomized image corresponding to the standard image during real-time inspection, meanwhile, the training image library is combined to train the neural network model, and then the trained neural network model determines the expansion information of the target atomized image, so that an accurate inspection result is obtained, the accuracy of the foggy day inspection result is greatly improved, and scientific guidance is provided for an electric power department to take accurate countermeasures.
On the basis of the technical scheme, the invention can be further improved as follows:
further: in S1, the formula for determining the expansion information of the standard sample nebulized image at different nebulization levels is as follows:
f (a) ＝n×[(l a -l 0 )/(l max -l 0 )] e
wherein f is (a) The expansion information of the standard sample atomization image under different atomization grades is obtained, and n is the total atomization grade; f. of (a) Is the current fog level; l 0 Is the standard length in the standard graph; l a The length is the display length corresponding to the standard length in the standard graph after current atomization, namely the length after expansion; l max E is a natural constant for the maximum expansion length.
The beneficial effects of the further scheme are as follows: through the display length of the standard length atomized standard length in the standard graph, the expansion influence of different atomization grades on the standard graph can be accurately calculated by combining the atomization grades, so that the subsequent neural network model can be conveniently trained and calculated, and an accurate inspection result can be obtained.
Further: the neural network model adopts a multilayer limited Boltzmann machine or a multilayer self-coding network model, and the training of the neural network model adopts layer-by-layer reverse training based on similarity.
Further: in S4, the specific implementation of determining the atomization level is as follows:
the standard graph is arranged at a fixed position, and standard real-time atomization image information of the standard graph is collected during the fog inspection;
and comparing the standard real-time atomization image information with standard sample atomization images corresponding to different atomization levels in the standard image library, and determining the atomization level during inspection.
The beneficial effects of the further scheme are as follows: the standard real-time atomization image information is compared with the standard sample atomization images corresponding to different atomization levels in the standard image library, so that the atomization level of the current inspection environment can be accurately determined according to the expansion degree of the standard image affected by atomization, and a subsequent neural network model can conveniently select corresponding model parameter information so as to accurately determine the expansion information corresponding to the target atomization image.
Further: in S5, before inputting the acquired target atomization image information into the neural network model, the method further includes the following steps:
and carrying out dark channel first-aid algorithm processing on the acquired target atomization image information so as to carry out image enhancement and restoration on the target atomization image information.
The beneficial effects of the further scheme are as follows: through inputting the target atomization image information of gathering before the neural network model, will gather target atomization image information carries out dark channel and tests algorithm processing earlier, can be right like this target atomization image information carries out image enhancement and recovers, thereby makes target atomization image information is changeed discernment, and is follow-up like this neural network model can obtain more accurate inflation information according to target atomization image information, and then obtains more accurate result of patrolling and examining.
Further: in S5, the specific method for determining the inspection result according to the expansion information of the target atomization image includes:
comparing the expansion information of the target atomization image with the expansion information of the standard sample atomization image under the same atomization level, and determining a comparison deviation value;
and determining the inspection grade of the target power transmission line according to the deviation value and a pre-deviation threshold comparison table.
The beneficial effects of the further scheme are as follows: the expansion information of the target atomization image is compared with the expansion information of the standard sample atomization image under the same atomization level, and the inspection level of the target power transmission line is determined according to the deviation value of the expansion information and the standard sample atomization image, so that the inspection level is fed back to an electric power department for monitoring, and reasonable countermeasures must be taken.
The invention also provides an automatic power transmission line foggy day inspection system which comprises a standard image library module, a training module, an atomization grade module and an acquisition, processing and analysis module;
the standard image library module is used for establishing a standard image library containing standard sample atomization images of the standard image under different atomization levels according to a preset standard image and determining expansion information of the standard sample atomization images under different atomization levels;
the training image library module is used for establishing a training image library corresponding to the training power transmission line on sunny days and different atomization grade weathers, wherein the training images of the same target power transmission line in the training image library on sunny days are correspondingly associated with the atomization training images of different atomization grades;
the training module is used for inputting the training image library into a pre-established neural network model for training and obtaining model parameter information corresponding to the neural network model in sunny days and weather with different atomization grades;
the atomization level module is used for acquiring a standard real-time atomization image corresponding to the standard graph during the inspection in the foggy days, determining the atomization level according to the standard real-time atomization image, and acquiring model parameter information of the neural network model corresponding to the atomization level;
the acquisition processing and analysis module is used for acquiring images of a target power transmission line during inspection in foggy weather, inputting the acquired target atomization images into the neural network model, carrying out image processing and analysis on the target atomization images by the neural network model according to the expansion information of the standard sample atomization images under different atomization levels to obtain expansion information corresponding to the target atomization images, and determining an inspection result according to the expansion information of the target atomization images.
According to the automatic inspection system for the power transmission line in the foggy day, the expansion information of the standard sample atomized image corresponding to the preset standard image under different atomization levels is predetermined, so that the influence of the foggy day on image acquisition can be calculated, then the atomization level can be determined according to the standard real-time atomized image corresponding to the standard image during real-time inspection, meanwhile, the training image library is combined to train the neural network model, and then the trained neural network model determines the expansion information of the target atomized image, so that an accurate inspection result is obtained, the accuracy of the foggy day inspection result is greatly improved, and scientific guidance is provided for an electric power department to take accurate countermeasures.
On the basis of the technical scheme, the invention can be further improved as follows:
further: the atomization grade module determines the specific implementation of the atomization grade according to the standard real-time atomization image as follows:
the standard graph is arranged at a fixed position, and standard real-time atomization image information of the standard graph is collected during the fog inspection;
and comparing the standard real-time atomization image information with standard sample atomization images corresponding to different atomization levels in the standard image library, and determining the atomization level during inspection.
The beneficial effects of the further scheme are as follows: the standard real-time atomization image information is compared with the standard sample atomization images corresponding to different atomization levels in the standard image library, so that the atomization level of the current inspection environment can be accurately determined according to the expansion degree of the standard image affected by atomization, and a subsequent neural network model can conveniently select corresponding model parameter information so as to accurately determine the expansion information corresponding to the target atomization image.
Further: the acquisition processing analysis module is used for carrying out dark channel first-aid algorithm processing on the acquired target atomization image information before inputting the acquired target atomization image information into the neural network model so as to carry out image enhancement and restoration processing on the target atomization image information.
The beneficial effects of the further scheme are as follows: through inputting the target atomization image information of gathering before the neural network model, will gather target atomization image information carries out dark channel and tests algorithm processing earlier, can be right like this target atomization image information carries out image enhancement and recovers, thereby makes target atomization image information is changeed discernment, and is follow-up like this neural network model can obtain more accurate inflation information according to target atomization image information, and then obtains more accurate result of patrolling and examining.
Further: the acquisition, processing and analysis module determines the specific implementation of the inspection result according to the expansion information of the target atomization image as follows:
comparing the expansion information of the target atomization image with the expansion information of the standard sample atomization image under the same atomization level, and determining a comparison deviation value;
and determining the inspection grade of the target power transmission line according to the deviation value and a pre-deviation threshold comparison table.
The beneficial effects of the further scheme are as follows: the expansion information of the target atomization image is compared with the expansion information of the standard sample atomization image under the same atomization level, and the inspection level of the target power transmission line is determined according to the deviation value of the expansion information and the standard sample atomization image, so that the inspection level is fed back to an electric power department for monitoring, and reasonable countermeasures must be taken.
Drawings
Fig. 1 is a schematic flow chart of an automatic transmission line inspection method in a foggy day according to an embodiment of the present invention;
fig. 2 is a schematic structural view of an automatic power transmission line inspection system in a foggy day according to an embodiment of the present invention.
Detailed Description
The principles and features of this invention are described below in conjunction with the following drawings, which are set forth by way of illustration only and are not intended to limit the scope of the invention.
The main effect of fog on image acquisition is the absorption and scattering of light by water droplets, causing blurring and swelling of the image. The inspection of the power transmission line has the particularity that the line and the target of the inspection are fixed, and the inspection has some special properties compared with the general foggy day image enhancement. Therefore, the invention can calculate the influence of the foggy day on image acquisition by predetermining the expansion information of the standard sample atomized image corresponding to the preset standard image line under different atomization levels, and processes the image by adopting the pre-trained neural network aiming at the influence, thereby obtaining the accurate inspection result.
As shown in fig. 1, a method for automatically patrolling a transmission line in a foggy day includes the following steps:
s1: establishing a standard image library containing standard sample atomization images of the standard image under different atomization levels according to a preset standard image, and determining expansion information of the standard sample atomization images under different atomization levels;
s2: establishing a training image library corresponding to training power transmission lines on sunny days and different atomization levels of weather, wherein the training images of the same target power transmission line in the training image library are correspondingly associated with the atomization training images of different atomization levels;
s3: inputting the training image library into a pre-established neural network model for training, and obtaining model parameter information corresponding to the neural network model on sunny days and weather with different atomization grades;
s4: acquiring a standard real-time atomization image corresponding to the standard graph during the inspection in the foggy days, determining the atomization level according to the standard real-time atomization image, and acquiring model parameter information of the neural network model corresponding to the atomization level;
s5: the method comprises the steps of carrying out image acquisition on a target power transmission line in the process of inspection in a foggy day, inputting an acquired target atomization image into a neural network model, carrying out image processing and analysis on the target atomization image by the neural network model according to expansion information of standard sample atomization images under different atomization levels to obtain expansion information corresponding to the target atomization image, and determining an inspection result according to the expansion information of the target atomization image.
The invention has the beneficial effects that: according to the automatic inspection method for the power transmission line in the foggy day, the expansion information of the standard sample atomized images corresponding to the preset standard image lines under different atomization levels is predetermined, so that the influence of the foggy day on image acquisition can be calculated, then the atomization levels can be determined according to the standard real-time atomized images corresponding to the standard images during real-time inspection, meanwhile, the training image library is combined to train the neural network model, and then the trained neural network model determines the expansion information of the target atomized images, so that an accurate inspection result is obtained, the accuracy of the foggy day inspection result is greatly improved, and scientific guidance is provided for an electric power department to take accurate countermeasures.
In one or more embodiments of the present invention, in S1, the formula for determining the swelling information of the standard sample nebulized image at different nebulization levels is as follows:
f (a) ＝n×[(l a -l 0 )/(l max -l 0 )] e
wherein f is (a) The expansion information of the standard sample atomization image under different atomization grades is obtained, and n is the total atomization grade; f. of (a) Is the current fog level; l 0 Is the standard length in the standard graph; l a The length is the display length corresponding to the standard length in the standard graph after current atomization, namely the length after expansion; l max E is a natural constant for the maximum expansion length.
Through the display length of the standard length atomized standard length in the standard graph, the expansion influence of different atomization grades on the standard graph can be accurately calculated by combining the atomization grades, so that the subsequent neural network model can be conveniently trained and calculated, and an accurate inspection result can be obtained.
In one or more embodiments of the invention, the neural network model adopts a multilayer restricted boltzmann machine or a multilayer self-coding network model, and the training of the neural network model adopts layer-by-layer reverse training based on similarity.
In one or more embodiments of the present invention, in S4, the determining the atomization level is implemented by:
the standard graph is arranged at a fixed position, and standard real-time atomization image information of the standard graph is collected during the fog inspection;
and comparing the standard real-time atomization image information with standard sample atomization images corresponding to different atomization levels in the standard image library, and determining the atomization level during inspection.
The standard real-time atomization image information is compared with the standard sample atomization images corresponding to different atomization levels in the standard image library, so that the atomization level of the current inspection environment can be accurately determined according to the expansion degree of the standard image affected by atomization, and a subsequent neural network model can conveniently select corresponding model parameter information so as to accurately determine the expansion information corresponding to the target atomization image.
It should be noted that in S4, when the standard real-time atomization image information of the standard graph is collected, interval sampling can be carried out at regular and irregular intervals, so that the atomization level in the inspection environment can be reflected more accurately, and a more accurate inspection result can be obtained.
Optionally, in one or more embodiments of the present invention, in S5, before inputting the acquired target atomization image information into the neural network model, the method further includes the following steps:
and carrying out dark channel first-aid algorithm processing on the acquired target atomization image information so as to carry out image enhancement and restoration on the target atomization image information.
Through inputting the target atomization image information of gathering before the neural network model, will gather target atomization image information carries out dark channel and tests algorithm processing earlier, can be right like this target atomization image information carries out image enhancement and recovers, thereby makes target atomization image information is changeed discernment, and is follow-up like this neural network model can obtain more accurate inflation information according to target atomization image information, and then obtains more accurate result of patrolling and examining.
In one or more embodiments of the present invention, in S5, the specific method for determining the inspection result according to the swelling information of the target atomization image includes:
comparing the expansion information of the target atomization image with the expansion information of the standard sample atomization image under the same atomization level, and determining a comparison deviation value;
and determining the inspection grade of the target power transmission line according to the deviation value and a pre-deviation threshold comparison table.
The expansion information of the target atomization image is compared with the expansion information of the standard sample atomization image under the same atomization level, and the inspection level of the target power transmission line is determined according to the deviation value of the expansion information and the standard sample atomization image, so that the inspection level is fed back to an electric power department for monitoring, and reasonable countermeasures must be taken.
In the embodiment of the present invention, the standard graph may be a plane or a three-dimensional mark, or may be a transformed display, and the embodiment of the present invention is not limited to this, but the standard graph used for establishing the standard image library and the standard graph corresponding to the acquired standard real-time atomization image need to be kept as the same graph.
As shown in fig. 2, the invention also provides an automatic power transmission line inspection system in fog days, which comprises a standard image library module, a training module, an atomization level module and an acquisition, processing and analysis module;
the standard image library module is used for establishing a standard image library containing standard sample atomization images of the standard image under different atomization levels according to a preset standard image and determining expansion information of the standard sample atomization images under different atomization levels;
the training image library module is used for establishing a training image library corresponding to the training power transmission line on sunny days and different atomization grade weathers, wherein the training images of the same target power transmission line in the training image library on sunny days are correspondingly associated with the atomization training images of different atomization grades;
the training module is used for inputting the training image library into a pre-established neural network model for training and obtaining model parameter information corresponding to the neural network model in sunny days and weather with different atomization grades;
the atomization level module is used for acquiring a standard real-time atomization image corresponding to the standard graph during the inspection in the foggy days, determining the atomization level according to the standard real-time atomization image, and acquiring model parameter information of the neural network model corresponding to the atomization level;
and the acquisition processing and analysis module is used for acquiring images of the target power transmission line during the inspection in the foggy weather, inputting the acquired target atomization images into the neural network model, carrying out image processing and analysis on the target atomization images by the neural network model according to the expansion information of the standard sample atomization images under different atomization grades to obtain expansion information corresponding to the target atomization images, and determining the inspection result according to the expansion information of the target atomization images.
According to the automatic inspection system for the power transmission line in the foggy day, the expansion information of the standard sample atomized image corresponding to the preset standard image under different atomization levels is predetermined, so that the influence of the foggy day on image acquisition can be calculated, then the atomization level can be determined according to the standard real-time atomized image corresponding to the standard image during real-time inspection, meanwhile, the training image library is combined to train the neural network model, and then the trained neural network model determines the expansion information of the target atomized image, so that an accurate inspection result is obtained, the accuracy of the foggy day inspection result is greatly improved, and scientific guidance is provided for an electric power department to take accurate countermeasures.
In one or more embodiments of the present invention, the atomization level module determines the atomization level according to the standard real-time atomization image by:
the standard graph is arranged at a fixed position, and standard real-time atomization image information of the standard graph is collected during the fog inspection;
and comparing the standard real-time atomization image information with standard sample atomization images corresponding to different atomization levels in the standard image library, and determining the atomization level during inspection.
The standard real-time atomization image information is compared with the standard sample atomization images corresponding to different atomization levels in the standard image library, so that the atomization level of the current inspection environment can be accurately determined according to the expansion degree of the standard image affected by atomization, a subsequent neural network model can conveniently select corresponding model parameter information, and the expansion information corresponding to the target atomization image can be accurately determined.
Optionally, in one or more embodiments of the present invention, before inputting the acquired target atomization image information into the neural network model, the acquisition processing and analyzing module is further configured to perform dark channel advanced verification algorithm processing on the acquired target atomization image information, so as to perform image enhancement and restoration processing on the target atomization image information.
Through inputting the target atomization image information of gathering before the neural network model, will gather target atomization image information carries out dark channel and tests algorithm processing earlier, can be right like this target atomization image information carries out image enhancement and recovers, thereby makes target atomization image information is changeed discernment, and is follow-up like this neural network model can obtain more accurate inflation information according to target atomization image information, and then obtains more accurate result of patrolling and examining.
In one or more embodiments of the present invention, the acquiring, processing and analyzing module determines, according to the expansion information of the target atomization image, a specific implementation of the inspection result as follows:
comparing the expansion information of the target atomization image with the expansion information of the standard sample atomization image under the same atomization level, and determining a comparison deviation value;
and determining the inspection grade of the target power transmission line according to the deviation value and a pre-deviation threshold comparison table.
The expansion information of the target atomization image is compared with the expansion information of the standard sample atomization image under the same atomization level, and the inspection level of the target power transmission line is determined according to the deviation value of the expansion information and the standard sample atomization image, so that the inspection level is fed back to an electric power department for monitoring, and reasonable countermeasures must be taken.
In practice, the specific device for collecting the target atomization image information is an unmanned aerial vehicle carrying an anti-atomization lens; the antifogging lens eliminates fog of the lens by arranging an automatic heating device near the lens; the control mode of the automatic heating device is carried out by regularly emitting light with anti-reflection wavelength from the inside of the lens to the lens and calculating the emissivity; if the lens is not atomized, the emitted light has low emissivity due to the antireflection effect, and if the lens forms water mist, the light can be reflected back to the inside of the lens; when the intensity of the reflected light exceeds a threshold value, a heating device is started to eliminate lens fog; and when the reflected light is lower than the threshold value, the heating device is closed, so that the sampling quality in the foggy days is further ensured. The image processing and analysis of the target atomization image are completed through an onboard processor of the unmanned aerial vehicle, or the target atomization image is wirelessly sent to a ground receiving station by the unmanned aerial vehicle to be processed, and the target atomization image can be flexibly selected according to different conditions in practice.
In addition, in the embodiment of the invention, the unmanned aerial vehicle is provided with the lighting equipment so as to improve the illumination intensity in foggy days and be beneficial to improving the definition of image information.
According to the method and the system for automatically inspecting the transmission line in the foggy day, the image processing in the transmission line inspection is different from the image processing in other fields; the method is characterized in that images needing to be processed in power transmission line inspection are relatively fixed (a route and equipment to be inspected are kept unchanged for a long time), and a scheme of pre-training an intelligent system model is adopted, so that the data processing speed in the inspection process is greatly increased; the method for quantitatively grading variable and complex atomization is adopted, a plurality of different atomization levels are trained on a neural network model in advance, the difficulty and complexity of the model are reduced on the premise of ensuring the quality of the model, therefore, the calculated amount in actual work is reduced, and meanwhile, the fog level is continuously tracked to ensure the correct selection of the model.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and should not be taken as limiting the scope of the present invention, which is intended to cover any modifications, equivalents, improvements, etc. within the spirit and scope of the present invention.
Claims (9)
1. An automatic power transmission line inspection method in foggy days is characterized by comprising the following steps:
s1: establishing a standard image library containing standard sample atomization images of the standard image under different atomization levels according to a preset standard image, and determining expansion information of the standard sample atomization images under different atomization levels;
s2: establishing a training image library corresponding to training power transmission lines on sunny days and different atomization levels of weather, wherein the training images of the same target power transmission line in the training image library are correspondingly associated with the atomization training images of different atomization levels;
s3: inputting the training image library into a pre-established neural network model for training, and obtaining model parameter information corresponding to the neural network model on sunny days and weather with different atomization grades;
s4: acquiring a standard real-time atomization image corresponding to the standard graph during the inspection in the foggy days, determining the atomization level according to the standard real-time atomization image, and acquiring model parameter information of the neural network model corresponding to the atomization level;
s5: acquiring images of a target power transmission line during inspection in a foggy day, inputting the acquired target atomization images into the neural network model, carrying out image processing and analysis on the target atomization images by the neural network model according to expansion information of the standard sample atomization images under different atomization levels to obtain expansion information corresponding to the target atomization images, and determining an inspection result according to the expansion information of the target atomization images;
in S1, the formula for determining the expansion information of the standard sample nebulized image at different nebulization levels is as follows:
f (a) ＝n×[(l a -l 0 )/(l max -l 0 )] e
wherein f is (a) The expansion information of the standard sample atomization image under different atomization grades is obtained, and n is the total atomization grade; f. of (a) Is the current fog level; l 0 Is the standard length in the standard graph; l a The length is the display length corresponding to the standard length in the standard graph after current atomization, namely the length after expansion; l max E is a natural constant for the maximum expansion length.
2. The automatic fog-day patrol method for the power transmission line according to claim 1, characterized in that: the neural network model adopts a multilayer limited Boltzmann machine or a multilayer self-coding network model, and the training of the neural network model adopts layer-by-layer reverse training based on similarity.
3. The automatic night patrol method for the power transmission line in the foggy day according to claim 1, which is characterized in that: in S4, the specific implementation of determining the atomization level is as follows:
the standard graph is arranged at a fixed position, and standard real-time atomization image information of the standard graph is collected during the fog inspection;
and comparing the standard real-time atomization image information with standard sample atomization images corresponding to different atomization levels in the standard image library, and determining the atomization level during inspection.
4. The automatic night patrol method for the power transmission line in the foggy day according to any one of claims 1 to 3, which is characterized in that: in S5, before inputting the acquired target atomization image information into the neural network model, the method further includes the following steps:
and carrying out dark channel first-aid algorithm processing on the acquired target atomization image information so as to carry out image enhancement and restoration on the target atomization image information.
5. The automatic night patrol method for the power transmission line in the foggy day according to any one of claims 1 to 3, which is characterized in that: in S5, the specific method for determining the inspection result according to the expansion information of the target atomization image includes:
comparing the expansion information of the target atomization image with the expansion information of the standard sample atomization image under the same atomization level, and determining a comparison deviation value;
and determining the inspection grade of the target power transmission line according to the deviation value and a pre-deviation threshold comparison table.
6. The utility model provides an automatic system of patrolling in transmission line fog day which characterized in that: comprises a standard image library module, a training module, an atomization grade module and an acquisition processing analysis module;
the standard image library module is used for establishing a standard image library containing standard sample atomization images of the standard image under different atomization levels according to a preset standard image and determining expansion information of the standard sample atomization images under different atomization levels;
the training image library module is used for establishing a training image library corresponding to the training power transmission line on sunny days and different atomization grade weathers, wherein the training images of the same target power transmission line in the training image library on sunny days are correspondingly associated with the atomization training images of different atomization grades;
the training module is used for inputting the training image library into a pre-established neural network model for training and obtaining model parameter information corresponding to the neural network model in sunny days and weather with different atomization grades;
the atomization level module is used for acquiring a standard real-time atomization image corresponding to the standard graph during the inspection in the foggy days, determining the atomization level according to the standard real-time atomization image, and acquiring model parameter information of the neural network model corresponding to the atomization level;
the acquisition processing and analysis module is used for acquiring images of a target power transmission line during inspection in foggy weather, inputting the acquired target atomization images into the neural network model, carrying out image processing and analysis on the target atomization images by the neural network model according to the expansion information of the standard sample atomization images under different atomization levels to obtain expansion information corresponding to the target atomization images, and determining an inspection result according to the expansion information of the target atomization images;
the standard image library module determines the expansion information formula of the standard sample atomization image under different atomization levels as follows:
f (a) ＝n×[(l a -l 0 )/(l max -l 0 )] e
wherein, f (a) The expansion information of the standard sample atomization image under different atomization grades is obtained, and n is the total atomization grade; f. of (a) Is the current fog level; l 0 Is the standard length in the standard graph; l a The length is the display length corresponding to the standard length in the standard graph after current atomization, namely the length after expansion; l max E is a natural constant for the maximum expansion length.
7. The automatic patrol system for the fog days of the power transmission line according to claim 6, characterized in that: the atomization grade module determines the atomization grade according to the standard real-time atomization image, and the specific implementation is as follows:
the standard graph is arranged at a fixed position, and standard real-time atomization image information of the standard graph is collected during the fog inspection;
and comparing the standard real-time atomization image information with standard sample atomization images corresponding to different atomization levels in the standard image library, and determining the atomization level during inspection.
8. The automatic night patrol system for the power transmission line in the foggy day according to claim 7, characterized in that: the acquisition processing analysis module is used for carrying out dark channel first-aid algorithm processing on the acquired target atomization image information before inputting the acquired target atomization image information into the neural network model so as to carry out image enhancement and restoration on the target atomization image information.
9. The automatic night patrol system for the power transmission line in the foggy day according to any one of claims 6 to 8, which is characterized in that: the acquisition, processing and analysis module determines the specific implementation of the inspection result according to the expansion information of the target atomization image as follows:
comparing the expansion information of the target atomization image with the expansion information of the standard sample atomization image under the same atomization level, and determining a comparison deviation value;
and determining the inspection grade of the target power transmission line according to the deviation value and a pre-deviation threshold comparison table.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110343120.1A CN113129273B (en) | 2021-03-30 | 2021-03-30 | Automatic transmission line fog inspection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110343120.1A CN113129273B (en) | 2021-03-30 | 2021-03-30 | Automatic transmission line fog inspection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113129273A CN113129273A (en) | 2021-07-16 |
| CN113129273B true CN113129273B (en) | 2023-04-07 |
Family
ID=76774781
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110343120.1A Active CN113129273B (en) | 2021-03-30 | 2021-03-30 | Automatic transmission line fog inspection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113129273B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109241831A (en) * | 2018-07-26 | 2019-01-18 | 东南大学 | A kind of greasy weather at night visibility classification method based on image analysis |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI423166B (en) * | 2009-12-04 | 2014-01-11 | Huper Lab Co Ltd | Method for determining if an input image is a foggy image, method for determining a foggy level of an input image and cleaning method for foggy images |
| CN103903008B (en) * | 2014-03-26 | 2018-01-19 | 国家电网公司 | A kind of method and system of the mist grade based on image recognition transmission line of electricity |
| CN103927523B (en) * | 2014-04-24 | 2017-01-18 | 东南大学 | Fog level detection method based on longitudinal gray features |
| CN104240239B (en) * | 2014-09-01 | 2018-05-01 | 西南交通大学 | A kind of detection method based on road image detection local section haze weather |
| CN109741322A (en) * | 2019-01-08 | 2019-05-10 | 南京蓝绿物联科技有限公司 | A kind of visibility measurement method based on machine learning |
| CN110866593B (en) * | 2019-11-05 | 2022-09-23 | 西南交通大学 | Highway severe weather identification method based on artificial intelligence |
| CN111275027B (en) * | 2020-03-26 | 2023-07-04 | 赛特斯信息科技股份有限公司 | Method for realizing expressway foggy day detection early warning treatment |
| CN112365476B (en) * | 2020-11-13 | 2023-12-08 | 南京信息工程大学 | Fog day visibility detection method based on double-channel depth network |
| CN112183788B (en) * | 2020-11-30 | 2021-03-30 | 华南理工大学 | Domain adaptive equipment operation detection system and method |
-
2021
- 2021-03-30 CN CN202110343120.1A patent/CN113129273B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109241831A (en) * | 2018-07-26 | 2019-01-18 | 东南大学 | A kind of greasy weather at night visibility classification method based on image analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113129273A (en) | 2021-07-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108037770B (en) | Unmanned aerial vehicle power transmission line inspection system and method based on artificial intelligence | |
| WO2020199538A1 (en) | Bridge key component disease early-warning system and method based on image monitoring data | |
| CN108956640B (en) | Vehicle-mounted detection device and detection method suitable for power distribution line inspection | |
| CN112633535A (en) | Photovoltaic power station intelligent inspection method and system based on unmanned aerial vehicle image | |
| CN107728136A (en) | A kind of airfield runway monitoring foreign bodies and removing guiding system and method | |
| CN112327906A (en) | Intelligent automatic inspection system based on unmanned aerial vehicle | |
| CN110378865A (en) | A kind of greasy weather visibility intelligence hierarchical identification method and system under complex background | |
| CN111539355A (en) | Photovoltaic panel foreign matter detection system and detection method based on deep neural network | |
| CN103808723A (en) | Exhaust gas blackness automatic detection device for diesel vehicles | |
| CN114240868A (en) | Unmanned aerial vehicle-based inspection analysis system and method | |
| CN110910440B (en) | Power transmission line length determination method and system based on power image data | |
| CN106570948A (en) | Transformer substation intelligent meter reading system provided with portable device | |
| CN115620239B (en) | Point cloud and video combined power transmission line online monitoring method and system | |
| CN113988573A (en) | Risk judgment method, system and medium for routing inspection unmanned aerial vehicle based on power system | |
| WO2022247597A1 (en) | Papi flight inspection method and system based on unmanned aerial vehicle | |
| CN112508911A (en) | Rail joint touch net suspension support component crack detection system based on inspection robot and detection method thereof | |
| CN105447431B (en) | A kind of docking aircraft method for tracking and positioning and system based on machine vision | |
| CN113129273B (en) | Automatic transmission line fog inspection method and system | |
| CN116086547B (en) | Contact net icing detection method based on infrared imaging and meteorological monitoring | |
| CN113111714A (en) | Risk prevention and control system applied to power transmission field operation | |
| CN117389294A (en) | Partial discharge detection system and method for airborne solar blind ultraviolet camera | |
| CN116978139A (en) | Unmanned aerial vehicle intelligent inspection system and method based on Beidou technology | |
| CN107194923B (en) | Ultraviolet image diagnosis method for defect inspection of contact network power equipment | |
| CN114442658B (en) | Automatic inspection system for power transmission and distribution line unmanned aerial vehicle and operation method thereof | |
| CN113763326B (en) | Pantograph detection method based on Mask scanning R-CNN network |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |