CN110688919A - Intelligent line patrol fault identification method - Google Patents
Intelligent line patrol fault identification method Download PDFInfo
- Publication number
- CN110688919A CN110688919A CN201910874396.5A CN201910874396A CN110688919A CN 110688919 A CN110688919 A CN 110688919A CN 201910874396 A CN201910874396 A CN 201910874396A CN 110688919 A CN110688919 A CN 110688919A
- Authority
- CN
- China
- Prior art keywords
- image
- output result
- recognition
- training
- line patrol
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an intelligent line patrol fault identification method, and relates to the field of electric power line patrol fault identification; which comprises the following steps of 1: dividing an image acquired by an unmanned aerial vehicle into N regions, and then performing image enhancement to obtain a preprocessed image, wherein N is a positive integer greater than 1; step 2: classifying the images into a test set and a training set, inputting the training set into a constructed recognition network to obtain an output result, and filtering influence factors in the output result by using height difference and depth information based on the output result to finish training; and step 3: and inputting the test set into the trained recognition network to obtain a recognition result. According to the invention, the image is partitioned, the partitioned data is subjected to image enhancement, and the recognition process is trained to consider influence factors, so that the image recognition speed and accuracy are improved, automatic recognition and rapid line patrol fault recognition are facilitated, and the workload of workers is greatly reduced.
Description
Technical Field
The invention relates to the field of power line patrol fault identification, in particular to an intelligent line patrol fault identification method.
Background
With the interconnection of large power grids and the continuous expansion of the scale of the power grids, the safety and stability of the operation of the power grids are concerned widely. On one hand, the basic level routing work is often organized with reference to operational experience. However, as the operating environment of the line changes, the operating experience also changes. On the other hand, with the continuous extension of the operation line, the contradiction between the increase of the line patrol workload and the shortage of the line patrol personnel is increasingly prominent. Therefore, the refined line patrol of the power transmission line is realized, the operation and maintenance efficiency of the power transmission line is improved, and the power transmission line is widely concerned by power operation units.
At present, each level of line operation and inspection departments mainly adopt a manual inspection mode and assist advanced technologies such as helicopters, unmanned planes and the like to carry out cooperative operation so as to improve the operation efficiency. However, this method requires manual interpretation of the multi-source image, which is a huge workload and often causes the situations of missing judgment and passing through empirical judgment, and the main reasons for the above problems are: on one hand, the current manual inspection mode still stays at a unilateral informatization stage of manually recording/inputting line operation and defect fault data, the operation condition of a power transmission channel is complex, and operation and maintenance personnel have limited information which can be acquired by judging the operation of equipment and the channel condition on site, so that risk assessment deviation is easily caused; on the other hand, the line inspection operation lacks a unified standard flow, and the inspection work can be freely performed, so that the conditions of insufficient inspection, missing important inspection items and the like can be caused. In view of the hidden trouble brought to the stable operation of the power grid by the insufficient informatization degree and the standard loss of the line inspection, an intelligent line inspection fault identification method is urgently needed at present so as to improve the efficiency and the reliability of the inspection operation of the power transmission line.
Disclosure of Invention
The invention aims to: the invention provides an intelligent line patrol fault identification method, which utilizes the image enhancement and the training result filtering influence factors by regions to improve the identification precision and speed.
The technical scheme adopted by the invention is as follows:
an intelligent line patrol fault identification method comprises the following steps:
step 1: dividing an image acquired by an unmanned aerial vehicle into N regions, and then performing image enhancement to obtain a preprocessed image, wherein N is a positive integer greater than 1;
step 2: classifying the images into a test set and a training set, inputting the training set into a constructed recognition network to obtain an output result, and filtering influence factors in the output result by using height difference and depth information based on the output result to finish training;
and step 3: and inputting the test set into the trained recognition network to obtain a recognition result.
Preferably, the step 1 comprises the steps of:
step 1.1: dividing an image acquired by an unmanned aerial vehicle into N areas according to data flow;
step 1.2: and performing spatial conversion, filtering and transformation on the image in each region to complete image enhancement and obtain a preprocessed image.
Preferably, the step 2 comprises the steps of:
step 2.1: dividing the preprocessed image into a test set and a training set according to a proportion;
step 2.2: constructing an identification network comprising an input layer, a convolutional layer and a full-connection layer, and inputting a training set into the identification network to map to obtain an output result;
step 2.3: filtering ground influence factors in the output result by using the height difference and the depth information based on the output result to obtain a training output result, and finishing training; the influence factors include ground clutter and ground towers.
Preferably, said step 1.2 comprises the steps of:
step 1.2.1: converting the preprocessed image from an RGB space to a YUV space, and extracting an intensity channel Y of the YUV space as a gray image I;
step 1.2.2: carrying out iterative filtering on the gray level image I by using a local extremum filter, and taking a filtering result as an illumination component L of the image;
step 1.2.3: separating a reflection component R from the gray level image I according to the illumination component L;
step 1.2.4: and carrying out gamma conversion on the illumination component L, and then reconstructing the illumination component L and the reflection component R to obtain an enhanced image.
In summary, due to the adoption of the technical scheme, the invention has the beneficial effects that:
1. according to the invention, the image is partitioned, the partitioned data is subjected to image enhancement, and the recognition process is trained to consider influence factors, so that the image recognition speed and accuracy are improved, automatic recognition and rapid line patrol fault recognition are facilitated, and the workload of workers is greatly reduced;
2. according to the invention, a large amount of collected data is partitioned, and image enhancement is carried out on all partitioned data, so that the image processing speed and the image identification accuracy are improved;
3. according to the image enhancement method, local extremum filtering is utilized, texture information is effectively filtered, detail information is highlighted, the image enhancement is carried out through space conversion, the distribution conditions such as illumination are considered, the image brightness is improved, and the light and shade relation of the image is coordinated;
4. according to the invention, influence factors such as ground stray lines on the electric power patrol line are filtered through depth information and height difference during training, so that the identification accuracy is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
FIG. 1 is a flow chart of a method of the present invention;
fig. 2 is a flow chart of image enhancement according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the detailed description and specific examples, while indicating the preferred embodiment of the invention, are intended for purposes of illustration only and are not intended to limit the scope of the invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without making any creative effort, shall fall within the protection scope of the present invention.
It is noted that relational terms such as "first" and "second," and the like, may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The features and properties of the present invention are described in further detail below with reference to examples.
Example 1
The existing manual inspection mode still stays at a unilateral informatization stage of manually recording/inputting line operation and defect fault data, the operation condition of a power transmission channel is complex, and operation and maintenance personnel have limited information which can be obtained by judging the operation of equipment and the channel condition on site, so that risk assessment deviation is easily caused; the line inspection operation lacks unified standard flow, and the free performance of the work of patrolling and examining can cause the not in place situation such as the inspection, important inspection project disappearance, consequently patrols and examines the hidden danger that informationization degree is not enough and standard disappearance brought for the electric wire netting steady operation to the line, and this application provides an intelligence and patrols line fault identification method to improve the efficiency and the reliability of transmission line inspection operation, the details are as follows:
an intelligent line patrol fault identification method comprises the following steps:
step 1: dividing an image acquired by an unmanned aerial vehicle into N regions, and then performing image enhancement to obtain a preprocessed image, wherein N is a positive integer greater than 1;
step 2: classifying the images into a test set and a training set, inputting the training set into a constructed recognition network to obtain an output result, and filtering influence factors in the output result by using height difference and depth information based on the output result to finish training;
and step 3: and inputting the test set into the trained recognition network to obtain a recognition result.
The step 1 comprises the following steps:
step 1.1: dividing an image acquired by an unmanned aerial vehicle into N areas according to data flow;
step 1.2: and performing spatial conversion, filtering and transformation on the image in each region to complete image enhancement and obtain a preprocessed image.
And partitioning a large amount of acquired data, and performing image enhancement on all partitioned data to improve the image processing speed and the image identification accuracy.
Step 1.2 comprises the following steps:
step 1.2.1: converting the preprocessed image from an RGB space to a YUV space, and extracting an intensity channel Y of the YUV space as a gray image I;
step 1.2.2: carrying out iterative filtering on the gray level image I by using a local extremum filter, and taking a filtering result as an illumination component L of the image;
step 1.2.3: separating a reflection component R from the gray level image I according to the illumination component L;
step 1.2.4: and carrying out gamma conversion on the illumination component L, and then reconstructing the illumination component L and the reflection component R to obtain an enhanced image.
The image enhancement utilizes local extremum filtering, effectively filters texture information, highlights detail information, considers the distribution conditions such as illumination and the like through space conversion, improves the image brightness, and coordinates the light and shade relation of the image.
The step 2 comprises the following steps:
step 2.1: dividing the preprocessed image into a test set and a training set according to a proportion;
step 2.2: constructing an identification network comprising an input layer, a convolutional layer and a full-connection layer, and inputting a training set into the identification network to map to obtain an output result;
step 2.3: filtering ground influence factors in the output result by using the height difference and the depth information based on the output result to obtain a training output result, and finishing training; the influence factors include ground clutter and ground towers.
In the specific implementation: the unmanned aerial vehicle acquires the images of the line patrol, wherein the images comprise targets such as a tower, a lead, a ground wire, a drainage wire, an insulator and the like, the line patrol fault needs to identify a corresponding correct target, and the identified fault target is prompted; for example, identification of insulators: identifying the insulator by using an identification network, wherein the identification comprises the steps of constructing a training sample set, training, storing an identification model, detecting by using a sliding window, and then linearly fitting a candidate frame to determine an insulator marking area in an image; for the identification of the wire: because the wire presents straight line and penetration characteristics in the image, prewitt operator edge extraction is carried out on the image; analyzing the edge direction information and carrying out clustering processing; carrying out pixel region growing operation on each clustering center; determining the position of the lead according to the length information; the wire foreign body defect identification rule comprises: analyzing the smoothness and consistency of the gray level image on the surface of the lead to detect broken strands or foreign matter defects; on condition that the parallel conductor group is identified, the presence or absence of a defect on the ground conductor is diagnosed group by group.
Training the recognition network by using a back propagation algorithm and a gradient descent method, adjusting the weight value of the recognition network through the cost function feedback between the characteristic vector and the actual value of each type of target, and repeating iteration until the cost function is smaller than a set threshold value to obtain the network with the optimal weight value. Finding out high-rise ground objects mixed on the ground and targets mixed on the high-rise objects through feature extraction, finding out targets which are mistakenly identified as power lines by using depth information in a cuboid division mode, setting the length of an XY fixed side to be 5 meters, and limiting a Z axis (similar to a infinitesimal method); then, the lowest point value (the minimum value of Z) in the cuboid is obtained, the Z coordinate value of the point which is identified as the power line in the cuboid is differentiated from the lowest point value, the points with the difference value smaller than the threshold value are filtered (the threshold value is set as 10), the threshold value is the size and the height of the object obtained through measurement, and after the training is completed, the test set is input into the identification network to obtain the final identification result. Influence factors such as ground stray lines on the electric power patrol line are filtered through depth information and height difference during training, and the identification accuracy rate is improved.
According to the invention, the image is partitioned, the partitioned data is subjected to image enhancement, and the recognition process is trained to consider influence factors, so that the image recognition speed and accuracy are improved, automatic recognition and rapid line patrol fault recognition are facilitated, and the workload of workers is greatly reduced.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (4)
1. An intelligent line patrol fault identification method is characterized by comprising the following steps: the method comprises the following steps:
step 1: dividing an image acquired by an unmanned aerial vehicle into N regions, and then performing image enhancement to obtain a preprocessed image, wherein N is a positive integer greater than 1;
step 2: classifying the images into a test set and a training set, inputting the training set into a constructed recognition network to obtain an output result, and filtering influence factors in the output result by using height difference and depth information based on the output result to finish training;
and step 3: and inputting the test set into the trained recognition network to obtain a recognition result.
2. The intelligent line patrol fault identification method according to claim 1, characterized in that: the step 1 comprises the following steps:
step 1.1: dividing an image acquired by an unmanned aerial vehicle into N areas according to data flow;
step 1.2: and performing spatial conversion, filtering and transformation on the image in each region to complete image enhancement and obtain a preprocessed image.
3. The intelligent line patrol fault identification method according to claim 1, characterized in that: the step 2 comprises the following steps:
step 2.1: dividing the preprocessed image into a test set and a training set according to a proportion;
step 2.2: constructing an identification network comprising an input layer, a convolutional layer and a full-connection layer, and inputting a training set into the identification network to map to obtain an output result;
step 2.3: filtering ground influence factors in the output result by using the height difference and the depth information based on the output result to obtain a training output result, and finishing training; the influence factors include ground clutter and ground towers.
4. The intelligent line patrol fault identification method according to claim 2, characterized in that: the step 1.2 comprises the following steps:
step 1.2.1: converting the preprocessed image from an RGB space to a YUV space, and extracting an intensity channel Y of the YUV space as a gray image I;
step 1.2.2: carrying out iterative filtering on the gray level image I by using a local extremum filter, and taking a filtering result as an illumination component L of the image;
step 1.2.3: separating a reflection component R from the gray level image I according to the illumination component L;
step 1.2.4: and carrying out gamma conversion on the illumination component L, and then reconstructing the illumination component L and the reflection component R to obtain an enhanced image.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910874396.5A CN110688919A (en) | 2019-09-17 | 2019-09-17 | Intelligent line patrol fault identification method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910874396.5A CN110688919A (en) | 2019-09-17 | 2019-09-17 | Intelligent line patrol fault identification method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110688919A true CN110688919A (en) | 2020-01-14 |
Family
ID=69109338
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910874396.5A Pending CN110688919A (en) | 2019-09-17 | 2019-09-17 | Intelligent line patrol fault identification method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110688919A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113052941A (en) * | 2021-03-15 | 2021-06-29 | 中国能源建设集团江苏省电力设计院有限公司 | Regional power facility state analysis method and image acquisition method |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102590823A (en) * | 2012-01-06 | 2012-07-18 | 中国测绘科学研究院 | Rapid extraction and reconstruction method for data power line of airborne LIDAR |
CN106504362A (en) * | 2016-10-18 | 2017-03-15 | 国网湖北省电力公司检修公司 | Power transmission and transformation system method for inspecting based on unmanned plane |
CN108960134A (en) * | 2018-07-02 | 2018-12-07 | 广东容祺智能科技有限公司 | A kind of patrol UAV image mark and intelligent identification Method |
CN109712127A (en) * | 2018-12-21 | 2019-05-03 | 云南电网有限责任公司电力科学研究院 | A kind of electric transmission line fault detection method for patrolling video flowing for machine |
-
2019
- 2019-09-17 CN CN201910874396.5A patent/CN110688919A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102590823A (en) * | 2012-01-06 | 2012-07-18 | 中国测绘科学研究院 | Rapid extraction and reconstruction method for data power line of airborne LIDAR |
CN106504362A (en) * | 2016-10-18 | 2017-03-15 | 国网湖北省电力公司检修公司 | Power transmission and transformation system method for inspecting based on unmanned plane |
CN108960134A (en) * | 2018-07-02 | 2018-12-07 | 广东容祺智能科技有限公司 | A kind of patrol UAV image mark and intelligent identification Method |
CN109712127A (en) * | 2018-12-21 | 2019-05-03 | 云南电网有限责任公司电力科学研究院 | A kind of electric transmission line fault detection method for patrolling video flowing for machine |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113052941A (en) * | 2021-03-15 | 2021-06-29 | 中国能源建设集团江苏省电力设计院有限公司 | Regional power facility state analysis method and image acquisition method |
CN113052941B (en) * | 2021-03-15 | 2022-06-10 | 中国能源建设集团江苏省电力设计院有限公司 | Regional power facility state analysis method and image acquisition method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108037133B (en) | Intelligent electric power equipment defect identification method and system based on unmanned aerial vehicle inspection image | |
WO2022077605A1 (en) | Wind turbine blade image-based damage detection and localization method | |
Sohn et al. | Automatic powerline scene classification and reconstruction using airborne lidar data | |
CN108734689B (en) | Method for detecting scattered strands of conducting wires based on region growth | |
CN106326808B (en) | Method for detecting bird nest on power transmission line tower based on unmanned aerial vehicle image | |
CN103442209B (en) | Video monitoring method of electric transmission line | |
CN108680833B (en) | Composite insulator defect detection system based on unmanned aerial vehicle | |
CN105718842B (en) | A kind of machine vision detection method of power transmission line line-broken malfunction | |
CN113205063A (en) | Visual identification and positioning method for defects of power transmission conductor | |
CN112435208B (en) | Insulator region detection method for significant edge guidance | |
CN111145159B (en) | Method and device for extracting inspection key component points | |
CN114187511A (en) | Intelligent monitoring method for power transmission panoramic big data | |
CN111899219A (en) | Image identification method and system for power transmission line machine patrol | |
CN111105398A (en) | Transmission line component crack detection method based on visible light image data | |
CN112508911A (en) | Rail joint touch net suspension support component crack detection system based on inspection robot and detection method thereof | |
CN112101138A (en) | Bridge inhaul cable surface defect real-time identification system and method based on deep learning | |
CN110490261B (en) | Positioning method for power transmission line inspection image insulator | |
CN114663672A (en) | Method and system for detecting corrosion of steel member of power transmission line tower | |
CN108537792B (en) | Power defect image identification method based on convolutional neural network | |
CN110688919A (en) | Intelligent line patrol fault identification method | |
CN117635905A (en) | Intelligent monitoring method for electric energy meter attachment quality based on image recognition algorithm | |
CN112683916A (en) | Method and device for identifying missing or mounting error of small hardware fittings of current collecting line tower | |
CN112033393A (en) | Three-dimensional route planning method and device based on laser radar point cloud data | |
CN115641448A (en) | Method for rapidly identifying height of tree barrier under power line based on unmanned aerial vehicle remote sensing | |
CN116071299A (en) | Insulator RTV spraying defect detection method and system |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200114 |