CN110910341A - Transmission line corrosion area defect detection method and device - Google Patents
Transmission line corrosion area defect detection method and device Download PDFInfo
- Publication number
- CN110910341A CN110910341A CN201910823411.3A CN201910823411A CN110910341A CN 110910341 A CN110910341 A CN 110910341A CN 201910823411 A CN201910823411 A CN 201910823411A CN 110910341 A CN110910341 A CN 110910341A
- Authority
- CN
- China
- Prior art keywords
- image
- target
- corrosion
- hardware
- transmission line
- 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.)
- Granted
Links
- 230000007547 defect Effects 0.000 title claims abstract description 65
- 238000005260 corrosion Methods 0.000 title claims abstract description 56
- 230000007797 corrosion Effects 0.000 title claims abstract description 56
- 230000005540 biological transmission Effects 0.000 title claims abstract description 45
- 238000001514 detection method Methods 0.000 title claims abstract description 34
- 238000000034 method Methods 0.000 claims abstract description 59
- 238000007689 inspection Methods 0.000 claims abstract description 37
- 230000011218 segmentation Effects 0.000 claims abstract description 23
- 238000004458 analytical method Methods 0.000 claims abstract description 18
- 238000012216 screening Methods 0.000 claims abstract description 8
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 5
- 238000012545 processing Methods 0.000 claims abstract description 5
- JEIPFZHSYJVQDO-UHFFFAOYSA-N iron(III) oxide Inorganic materials O=[Fe]O[Fe]=O JEIPFZHSYJVQDO-UHFFFAOYSA-N 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 9
- 238000003709 image segmentation Methods 0.000 claims description 5
- 239000000203 mixture Substances 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 6
- 238000012549 training Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000000903 blocking effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 239000010977 jade Substances 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000013526 transfer learning Methods 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
-
- 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
- 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
- G06F16/587—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using geographical or spatial information, e.g. location
-
- 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
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- 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
- G06T2207/30164—Workpiece; Machine component
-
- 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)
- Computer Vision & Pattern Recognition (AREA)
- Business, Economics & Management (AREA)
- Library & Information Science (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- General Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Primary Health Care (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention provides a method for detecting defects of a corrosion area of a power transmission line, which comprises the following steps: acquiring aerial images and inertial navigation information data; determining the geographic coordinate of each image according to the time information of the unmanned aerial vehicle inertial navigation data and the time information of the aerial image; detecting the hardware by using a deep convolutional neural network model, screening an image containing the hardware as a target image, and simultaneously giving out information of an external rectangular area of a hardware target; segmenting a target circumscribed rectangular area to obtain a target small image; performing graying processing on the small target image by adopting a super red method; obtaining a corrosion area by adopting a threshold value method; calculating by adopting a connected domain to obtain a corrosion defect detection image; and generating a patrol defect report. The invention has the beneficial effects that: the method has the advantages that foreground segmentation is carried out on the inspection data image, background interference is eliminated, corrosion defect detection is carried out by a super-red method, rapid detection of hardware corrosion is achieved, and the efficiency of defect analysis is improved.
Description
Technical Field
The invention belongs to the technical field of power transmission line detection, and particularly relates to a method and a device for detecting defects of a corrosion area of a power transmission line.
Background
The power transmission line contains various metal parts including hardware fittings such as a vibration damper and the like, and the power transmission line is corroded by various severe environments after being operated in the field for a long time, is easy to be corroded and damaged, and causes great harm to the safe operation of the power transmission line.
The development of modern remote sensing technology and remote sensing data acquisition mode, especially the rapid development of unmanned aerial vehicle application, provides a safe, quick and effective mode for the fine routing inspection of the small target of the power transmission line equipment, so that the image data of the small target area of the power transmission line equipment can be conveniently acquired. The method for intelligently identifying the defects of the power transmission line provides a method for intelligently identifying the defects for a large amount of image data generated by one-time inspection of the power transmission line, solves the problems of manual image reading and defect recording by using a computer vision method, greatly improves the inspection efficiency and saves labor and time.
In a thesis of 'detection of rusted areas of power transmission lines based on color and texture characteristics' (jade silence, etc., industrial control computer, 09 years 2018), a method for detecting rusted areas of power transmission lines based on color and texture characteristics is provided aiming at the problems of complex background and lack of effective rusted defect detection means of power transmission lines. Because the corrosion generated by important parts in the high-voltage transmission line is red and has more remarkable color characteristics, the corrosion and other areas are distinguished by setting the thresholds of the S component and the H component of the HIS color model, and the corrosion area and the non-corrosion area are further distinguished by analyzing the texture characteristics of the corrosion.
In a paper research on a helicopter routing inspection transmission line rust defect identification method (Zhanghou, university of maritime, 2009), aiming at the complexity of aerial images and the rust characteristics, a color image with rust is converted into a gray image, different image graying methods are compared, the ultrared method provided by the paper is applied to perform graying on the rust image with a good effect, aiming at errors in a grayed image threshold segmentation method, a plurality of rust gray images are applied to perform least square fitting, a segmentation threshold range of rust defects is determined according to fitting data, the segmentation threshold range is used for constraining a threshold determined by a maximum inter-class variance method to segment the image, the segmented rust region is subjected to morphological processing, geometric characteristics of the segmentation region are extracted, the segmented non-rust region is removed, and the rust region is marked as red in the color image, and the corroded part is determined through geometric characteristic analysis, and the power transmission line corrosion defect identification system is realized.
However, in the prior art, the complex background interference part is not removed, hardware targets such as a vibration damper in the power transmission line are not considered, whether hardware parts exist in a large number of aerial images is not judged, corrosion detection is directly performed on the basis of the images, target image screening of a large number of data is not achieved, a target area is rapidly and accurately positioned, and therefore the power transmission line is greatly damaged in safe operation.
Disclosure of Invention
In view of the above, the present invention aims to provide a method for detecting defects in a corrosion area of a power transmission line, so as to solve the above-mentioned problems.
In order to achieve the purpose, the technical scheme of the invention is realized as follows:
a method for detecting defects of a corrosion area of a power transmission line comprises the following steps:
A. acquiring an aerial image and inertial navigation information data obtained by the unmanned aerial vehicle inspection;
B. b, obtaining time information of unmanned aerial vehicle inertial navigation data and time information of aerial images through the data obtained in the step A, determining geographic coordinates of each image, and randomly selecting a power transmission line section to be analyzed according to the geographic coordinates;
C. c, detecting the hardware fitting on the image obtained in the step B by using a deep convolutional neural network model, screening the image containing the hardware fitting as a target image, and simultaneously giving out the information of an external rectangular area of the hardware fitting target;
D. c, segmenting the target circumscribed rectangular area obtained in the step C to obtain a target small image;
E. d, performing graying processing on the small target image obtained in the step D by adopting a hyper-red method;
F. obtaining a rust area in the gray-scale image obtained in the step E by adopting a threshold value method;
G. d, calculating a connected domain of the corrosion area in the step F to obtain a corrosion defect detection image;
H. deleting the part of the image obtained in the step G, which occupies a small area of the target image, and the part with a large length-width ratio;
I. generating a patrol defect report;
J. and forwarding the original patrol data and the patrol defect report stored in the patrol.
Further, the process of determining the geographic coordinates of each image in step B is as follows: and numbering the aerial images according to the time sequence, and corresponding the aerial images to inertial navigation information according to a time axis to obtain a matching list of inertial navigation data and images, so that the geographic coordinate of each image is determined according to the inertial navigation information.
Furthermore, in the step D, the target circumscribed rectangular region is used as an initial region, 50 pixels are respectively expanded in four directions, namely, the upper direction, the lower direction, the left direction and the right direction, of a frame of the target circumscribed rectangular region, and then the target circumscribed rectangular region is divided.
Further, the segmentation process in the step D is as follows: and D, adopting a grabcut image segmentation method, taking the target circumscribed rectangular region detected in the step C as a foreground target, taking all pixels outside the region as a background, and modeling the target and the background by using a full covariance Gaussian mixture model with k being 5 Gaussian components to finish segmentation.
Further, in the step E, a target small image obtained in the step D is reflected by a super red method, and a gray level image is obtained by calculating a color component formula 2R-G-B.
In addition, based on the power transmission line corrosion area defect detection method, the invention also provides a power transmission line corrosion area defect detection device, which comprises the following steps:
the data acquisition unit is used for acquiring aerial images and inertial navigation information data acquired by the unmanned aerial vehicle inspection;
the aerial image positioning unit is used for determining the geographic coordinates of each image in the data acquisition unit;
the hardware target detection unit is used for screening the images containing the hardware in the aerial image positioning unit and giving the information of the external rectangular area of the hardware target;
the hardware fitting region foreground target segmentation unit is used for segmenting a target circumscribed rectangular region obtained by the hardware fitting target detection unit to obtain a target small image;
the hardware target corrosion analysis unit is used for carrying out corrosion analysis on the small target image obtained by the hardware region foreground target segmentation unit to obtain a patrol defect report;
and the routing inspection data and corrosion defect analysis report forwarding unit is used for forwarding the original routing inspection data and the routing inspection defect report.
Compared with the prior art, the method for detecting the defects of the rusty area of the power transmission line has the following advantages:
(1) according to the method for detecting the defects of the corrosion area of the power transmission line, the geographic position coordinates in the flight path of the unmanned aerial vehicle correspond to the aerial images through the technology of matching the recording time of the inertial navigation data with the image shooting time, so that the images can be quickly positioned on the actual power transmission line on the basis of not splicing a large number of aerial images, on one hand, the speed of finding the defects is improved, on the other hand, the data is inquired and stored in a mode of matching with the inertial navigation information, the data storage mode is standardized, and the effective management of a large amount of aerial data is facilitated; meanwhile, foreground segmentation is carried out on the inspection data image, background interference is eliminated, corrosion defect detection is carried out by using a super red method, rapid detection of corrosion of hardware fittings such as a vibration damper is realized, and the efficiency of defect analysis is improved;
(2) the method for detecting the defects of the corrosion area of the power transmission line can realize hardware target detection such as a vibration damper and target image selection on the inspected image, support quick analysis and corrosion defect marking on the hardware corrosion defects such as the vibration damper of the inspected image, automatically position a tower section according to the matching relation of the image and inertial navigation, and generate a defect report by combining the corrosion defect analysis result. Original inspection data such as images, telemetering information and the like stored in inspection and inspection defect reports are forwarded through network transmission equipment or data lines;
(3) aiming at the defect of eliminating the complicated background interference part, the method for detecting the defects of the corrosion area of the power transmission line adopts a Grabcut image segmentation method to extract the parts aiming at the part area, eliminates the interference of most background factors, and utilizes color characteristics, geometric characteristics and the like to further eliminate the background, so that the corrosion detection result maximally reflects the part.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate an embodiment of the invention and, together with the description, serve to explain the invention and not to limit the invention. In the drawings:
fig. 1 is a block diagram of a structure of a device for detecting defects in a rusted area of a power transmission line according to an embodiment of the present invention.
Detailed Description
It should be noted that the embodiments and features of the embodiments of the present invention may be combined with each other without conflict.
The present invention will be described in detail below with reference to the embodiments with reference to the attached drawings.
A method for detecting defects of a corrosion area of a power transmission line comprises the following steps:
A. acquiring an aerial image and inertial navigation information data obtained by the unmanned aerial vehicle inspection;
B. b, acquiring the data obtained in the step A, determining the geographic coordinates of each image according to the time information of the unmanned aerial vehicle inertial navigation data and the time information of the aerial image, and randomly selecting the power transmission line section to be analyzed according to the geographic coordinates;
C. c, detecting the hardware fitting on the image obtained in the step B by using a deep convolutional neural network model, screening the image containing the hardware fitting as a target image, and simultaneously giving out the information of an external rectangular area of the hardware fitting target;
D. c, segmenting the target circumscribed rectangular area obtained in the step C to obtain a target small image;
E. d, performing graying processing on the small target image obtained in the step D by adopting a hyper-red method;
F. obtaining a rust area in the gray-scale image obtained in the step E by adopting a threshold value method;
G. d, calculating a connected domain of the corrosion area in the step F to obtain a corrosion defect detection image;
H. deleting the part of the image obtained in the step G, which occupies a small area of the target image, and the part with a large length-width ratio;
I. generating a patrol defect report;
J. and forwarding original routing inspection data such as routing inspection stored aerial images and remote measurement information and routing inspection defect reports.
The process of determining the geographic coordinates of each image in the step B comprises the following steps: and numbering the aerial images according to the time sequence, and corresponding the aerial images to inertial navigation information according to a time axis to obtain a matching list of inertial navigation data and images, so that the geographic coordinate of each image is determined according to the inertial navigation information.
The deep convolutional neural network model establishing process in the step C is as follows: establishing a hardware fitting sample database, and marking the position information of the target image; training the model through a training set and a test set obtained by labeling data; and after the model is trained, carrying out image detection test, and finally giving out the external rectangular area information of the hardware target. During model training, the detection capability of the model on hardware fittings of different sizes is improved by randomly introducing multi-scale images; the generalization capability of the models in different scenes is improved by means of random illumination change, image rotation and mirror image, random partial blocking and the like; and the model is enabled to reach a global optimization state through methods such as model transfer learning and parameter fine tuning.
Wherein, the hardware objects such as the stockbridge dampers and the like occupy less in the inspection image, the detection is typical small-target detection, the small-target detection has the difficulty that the carried information is less, the characteristics of a target prediction layer are not obvious after a large amount of convolution and pooling operations, and the position regression and classification are difficult, so that the deep convolution neural network model in the step C is established in a multi-scale characteristic fusion mode, the result obtained after the characteristics of different characteristic layers are fused is used for prediction, a multi-layer characteristic diagram is used, each characteristic diagram is used for carrying out new prediction, the characteristic diagram of each layer is obtained on the basis, the small characteristic diagram is sampled by a top-down method and then fused with the next characteristic diagram, the prediction is carried out after the fusion, and by combining with fast-rcnn, ssd, the semantic information of the high layer and the position information of the low layer are fused, the effect is obvious in the aspect of small target detection.
And D, taking the target circumscribed rectangular area as an initial area, expanding 50 pixels in the upper, lower, left and right directions of a frame of the target circumscribed rectangular area respectively, and then dividing.
The segmentation process in the step D is as follows: and D, adopting a grabcut image segmentation method, taking the target circumscribed rectangular region detected in the step C as a foreground target, taking all pixels outside the region as a background, and modeling the target and the background by using a full covariance Gaussian mixture model with k being 5 Gaussian components to finish segmentation.
And E, reacting the target small image obtained in the step D by adopting a super red method, and calculating a color component formula 2R-G-B to obtain a gray level image.
In addition, based on the method for detecting the defects of the corrosion area of the power transmission line, as shown in fig. 1, the invention also provides a device for detecting the defects of the corrosion area of the power transmission line, which comprises the following steps:
the data acquisition unit is used for acquiring aerial images and inertial navigation information data acquired by the unmanned aerial vehicle inspection;
the aerial image positioning unit is used for determining the geographic coordinates of each image in the data acquisition unit;
the hardware target detection unit is used for screening the images containing the hardware in the aerial image positioning unit and giving the information of the external rectangular area of the hardware target;
the hardware fitting region foreground target segmentation unit is used for segmenting a target circumscribed rectangular region obtained by the hardware fitting target detection unit to obtain a target small image;
the hardware target corrosion analysis unit is used for carrying out corrosion analysis on the small target image obtained by the hardware region foreground target segmentation unit to obtain a patrol defect report;
and the routing inspection data and corrosion defect analysis report forwarding unit is used for forwarding the original routing inspection data and the routing inspection defect report.
The working process of the embodiment is as follows:
before the unmanned aerial vehicle performs the power inspection task, an inspection line needs to be arranged, known data such as a tower machine account of an area to be inspected are collected, after the inspection is started, aerial image data and inertial navigation information data are collected through the data acquisition unit, the inertial navigation information data and the aerial image data are temporarily stored in a visible light camera storage card and an inertial navigation equipment storage card, and after the unmanned aerial vehicle finishes the flight task, the aerial image data are connected through 4G transmission equipment or a data line and transmitted to an aerial image positioning unit;
the aerial image positioning unit loads data stored in the data acquisition unit, numbers aerial images according to a time sequence, and corresponds the aerial images to inertial navigation information according to a time axis to obtain a matching list of the inertial navigation data and the images, so that the geographic coordinate of each image is determined according to the inertial navigation information, the image number of the corresponding geographic coordinate position is displayed, and an operator can randomly select a power transmission line section to be analyzed according to line characteristics to perform corrosion defect detection;
the hardware target detection unit adopts a deep convolution neural network model to detect the hardware from the image in the aerial image positioning unit, screens the image containing the hardware as a target image, and simultaneously provides the information of an external rectangular area of the hardware target;
the hardware region foreground target segmentation unit takes a target external rectangular region detected by the hardware target detection unit as an initial region, 50 pixels are respectively expanded in the upper, lower, left and right directions of a frame of the target external rectangular region, segmentation is carried out, a grabcut image segmentation method is adopted during segmentation, the target external rectangular region is taken as a foreground target, all pixels outside the region are taken as a background, a full covariance mixed Gaussian model with k being 5 Gaussian components is used for modeling the target and the background, and segmentation is completed;
the hardware target corrosion analysis unit is used for carrying out corrosion analysis on the small target image obtained by the hardware region foreground target segmentation unit, and the analysis process is as follows: because the rust of the objects such as hardware and the like shows certain red characteristics, the image after the foreground object is segmented can be reflected by adopting the ultrared method, red information is intensively distributed at one part in an RGB color space, a color component of the dominant red is a red component, generally, the value of the red component is larger than that of a green component and that of a blue component, a gray image is obtained by calculating a color component formula 2R-G-B (wherein R represents the red component, G represents the green component, and B represents the blue component), because the ultrared method has better stability for the influence of image illumination, brightness and the like, a rust area is obtained by adopting a threshold value method on the basis, and finally, the part with small area occupying the target image proportion and the part with large length-width ratio are deleted by calculating a connected domain, and are mostly the shadow part of the target boundary, finally, obtaining the rust defect condition of the segmented small target image and generating a patrol defect report;
and finally, original routing inspection data such as routing inspection images, telemetering information and the like and routing inspection defect reports stored in the routing inspection data and corrosion defect analysis report forwarding unit are forwarded to a decision command console and a routing inspection handheld terminal, so that routing inspection command operation and line maintenance personnel operation are facilitated.
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, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (6)
1. A method for detecting defects of a corrosion area of a power transmission line is characterized by comprising the following steps:
A. acquiring an aerial image and inertial navigation information data obtained by the unmanned aerial vehicle inspection;
B. b, acquiring the data obtained in the step A, determining the geographic coordinate of each image according to the time information of the unmanned aerial vehicle inertial navigation data and the time information of the aerial image, and randomly selecting the power transmission line road section to be analyzed according to the geographic coordinate;
C. c, detecting the hardware fitting on the image obtained in the step B by using a deep convolutional neural network model, screening the image containing the hardware fitting as a target image, and simultaneously giving out the information of an external rectangular area of the hardware fitting target;
D. c, segmenting the target circumscribed rectangular area obtained in the step C to obtain a target small image;
E. d, performing graying processing on the small target image obtained in the step D by adopting a hyper-red method;
F. obtaining a rust area in the gray-scale image obtained in the step E by adopting a threshold value method;
G. d, calculating a connected domain of the corrosion area in the step F to obtain a corrosion defect detection image;
H. deleting the part of the image obtained in the step G, which occupies a small area of the target image, and the part with a large length-width ratio;
I. generating a patrol defect report;
J. and forwarding the original patrol data and the patrol defect report stored in the patrol.
2. The method for detecting the defects of the corrosion area of the power transmission line according to claim 1, wherein the process of determining the geographic coordinates of each image in the step B is as follows: and numbering the aerial images according to the time sequence, and corresponding the aerial images to inertial navigation information according to a time axis to obtain a matching list of inertial navigation data and images, so that the geographic coordinate of each image is determined according to the inertial navigation information.
3. The method for detecting the defects of the corrosion area of the power transmission line according to claim 1, wherein the method comprises the following steps: and D, taking the target circumscribed rectangular region as an initial region, expanding 50 pixels in the upper, lower, left and right directions of the frame of the target circumscribed rectangular region respectively, and then dividing.
4. The method for detecting the defects of the corrosion area of the power transmission line according to claim 3, wherein the segmentation process in the step D is as follows: and D, adopting a grabcut image segmentation method, taking the circumscribed rectangular region of the target detected in the step C as a foreground target, taking all pixels outside the region as a background, and modeling the target and the background by using a full covariance Gaussian mixture model with k as 5 Gaussian components to finish segmentation.
5. The method for detecting the defects of the corrosion area of the power transmission line according to claim 1, wherein the method comprises the following steps: and E, reacting the target small image obtained in the step D by adopting a super red method, and calculating a color component formula 2R-G-B to obtain a gray level image.
6. A device based on the power transmission line corrosion area defect detection method of any one of claims 1 to 5, is characterized by comprising the following steps:
the data acquisition unit is used for acquiring aerial images and inertial navigation information data acquired by the unmanned aerial vehicle inspection;
the aerial image positioning unit is used for determining the geographic coordinates of each image in the data acquisition unit;
the hardware target detection unit is used for screening the images containing the hardware in the aerial image positioning unit and providing the information of the circumscribed rectangular area of the hardware target;
the hardware fitting region foreground target segmentation unit is used for segmenting a target circumscribed rectangular region obtained by the hardware fitting target detection unit to obtain a target small image;
the hardware target corrosion analysis unit is used for carrying out corrosion analysis on the small target image obtained by the hardware region foreground target segmentation unit to obtain a patrol defect report;
and the routing inspection data and corrosion defect analysis report forwarding unit is used for forwarding the original routing inspection data and the routing inspection defect report.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910823411.3A CN110910341B (en) | 2019-09-02 | 2019-09-02 | Method and device for detecting defects of rusted areas of power transmission line |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910823411.3A CN110910341B (en) | 2019-09-02 | 2019-09-02 | Method and device for detecting defects of rusted areas of power transmission line |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110910341A true CN110910341A (en) | 2020-03-24 |
| CN110910341B CN110910341B (en) | 2023-05-30 |
Family
ID=69814476
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910823411.3A Active CN110910341B (en) | 2019-09-02 | 2019-09-02 | Method and device for detecting defects of rusted areas of power transmission line |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110910341B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112090869A (en) * | 2020-09-08 | 2020-12-18 | 临沂大学 | Method and device for cleaning rusty spot by using laser of unmanned aerial vehicle and unmanned aerial vehicle |
| CN113191210A (en) * | 2021-04-09 | 2021-07-30 | 杭州海康威视数字技术股份有限公司 | Image processing method, device and equipment |
| CN113744179A (en) * | 2021-01-25 | 2021-12-03 | 云南电网有限责任公司德宏供电局 | Hardware defect detection method on power transmission and distribution line |
| CN114581692A (en) * | 2022-03-06 | 2022-06-03 | 扬州孚泰电气有限公司 | Vibration damper fault detection method and system based on intelligent pattern recognition |
| CN115376219A (en) * | 2022-06-28 | 2022-11-22 | 广州番禺电缆集团有限公司 | Cable insulation layer damage inspection device and method, electronic equipment and storage medium |
| CN115436384A (en) * | 2022-11-07 | 2022-12-06 | 国网山东省电力公司荣成市供电公司 | Distribution box surface defect detection system and method based on unmanned aerial vehicle image |
| CN115760858A (en) * | 2023-01-10 | 2023-03-07 | 东南大学附属中大医院 | Kidney pathological section cell identification method and system based on deep learning |
| CN117274723A (en) * | 2023-11-22 | 2023-12-22 | 国网智能科技股份有限公司 | Target identification method, system, medium and equipment for power transmission inspection |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103941745A (en) * | 2014-03-07 | 2014-07-23 | 国家电网公司 | Movable substation and working method for unmanned aerial vehicle electric transmission line inspection |
| CN106326808A (en) * | 2015-06-23 | 2017-01-11 | 上海深邃智能科技有限公司 | Method for detecting bird nests in power transmission line poles based on unmanned plane images |
| CN106643790A (en) * | 2016-09-26 | 2017-05-10 | 南方电网科学研究院有限责任公司 | Photoelectric pod resolution detection device and detection method |
| CN107703959A (en) * | 2017-10-31 | 2018-02-16 | 成都意町工业产品设计有限公司 | A kind of crusing robot control system and method based on unmanned plane |
| CN108470141A (en) * | 2018-01-27 | 2018-08-31 | 天津大学 | Insulator recognition methods in a kind of distribution line based on statistical nature and machine learning |
| JP2019009919A (en) * | 2017-06-26 | 2019-01-17 | 株式会社東芝 | Patrol inspection support system and patrol inspection support control program |
-
2019
- 2019-09-02 CN CN201910823411.3A patent/CN110910341B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103941745A (en) * | 2014-03-07 | 2014-07-23 | 国家电网公司 | Movable substation and working method for unmanned aerial vehicle electric transmission line inspection |
| CN106326808A (en) * | 2015-06-23 | 2017-01-11 | 上海深邃智能科技有限公司 | Method for detecting bird nests in power transmission line poles based on unmanned plane images |
| CN106643790A (en) * | 2016-09-26 | 2017-05-10 | 南方电网科学研究院有限责任公司 | Photoelectric pod resolution detection device and detection method |
| JP2019009919A (en) * | 2017-06-26 | 2019-01-17 | 株式会社東芝 | Patrol inspection support system and patrol inspection support control program |
| CN107703959A (en) * | 2017-10-31 | 2018-02-16 | 成都意町工业产品设计有限公司 | A kind of crusing robot control system and method based on unmanned plane |
| CN108470141A (en) * | 2018-01-27 | 2018-08-31 | 天津大学 | Insulator recognition methods in a kind of distribution line based on statistical nature and machine learning |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112090869A (en) * | 2020-09-08 | 2020-12-18 | 临沂大学 | Method and device for cleaning rusty spot by using laser of unmanned aerial vehicle and unmanned aerial vehicle |
| CN113744179A (en) * | 2021-01-25 | 2021-12-03 | 云南电网有限责任公司德宏供电局 | Hardware defect detection method on power transmission and distribution line |
| CN113191210A (en) * | 2021-04-09 | 2021-07-30 | 杭州海康威视数字技术股份有限公司 | Image processing method, device and equipment |
| CN113191210B (en) * | 2021-04-09 | 2023-08-29 | 杭州海康威视数字技术股份有限公司 | Image processing method, device and equipment |
| CN114581692A (en) * | 2022-03-06 | 2022-06-03 | 扬州孚泰电气有限公司 | Vibration damper fault detection method and system based on intelligent pattern recognition |
| CN115376219A (en) * | 2022-06-28 | 2022-11-22 | 广州番禺电缆集团有限公司 | Cable insulation layer damage inspection device and method, electronic equipment and storage medium |
| CN115436384A (en) * | 2022-11-07 | 2022-12-06 | 国网山东省电力公司荣成市供电公司 | Distribution box surface defect detection system and method based on unmanned aerial vehicle image |
| CN115760858A (en) * | 2023-01-10 | 2023-03-07 | 东南大学附属中大医院 | Kidney pathological section cell identification method and system based on deep learning |
| CN117274723A (en) * | 2023-11-22 | 2023-12-22 | 国网智能科技股份有限公司 | Target identification method, system, medium and equipment for power transmission inspection |
| CN117274723B (en) * | 2023-11-22 | 2024-03-26 | 国网智能科技股份有限公司 | Target identification method, system, medium and equipment for power transmission inspection |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110910341B (en) | 2023-05-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110910341A (en) | Transmission line corrosion area defect detection method and device | |
| CN109977813B (en) | Inspection robot target positioning method based on deep learning framework | |
| US20210319561A1 (en) | Image segmentation method and system for pavement disease based on deep learning | |
| CN109816024B (en) | Real-time vehicle logo detection method based on multi-scale feature fusion and DCNN | |
| US10621717B2 (en) | System and method for image-based target object inspection | |
| CN111784685A (en) | Power transmission line defect image identification method based on cloud edge cooperative detection | |
| CN103279765B (en) | Steel wire rope surface damage detection method based on images match | |
| CN109117836A (en) | Text detection localization method and device under a kind of natural scene based on focal loss function | |
| CN108764082A (en) | A kind of Aircraft Targets detection method, electronic equipment, storage medium and system | |
| CN112528979B (en) | Transformer substation inspection robot obstacle distinguishing method and system | |
| CN114627021A (en) | Point cloud and deep learning based defect detection method and system | |
| US20220327684A1 (en) | Method and device for detecting mechanical equipment parts | |
| CN108648210B (en) | Rapid multi-target detection method and device under static complex scene | |
| WO2020047316A1 (en) | System and method for training a damage identification model | |
| CN115578326A (en) | Road disease identification method, system, equipment and storage medium | |
| CN113295142B (en) | Terrain scanning analysis method and device based on FARO scanner and point cloud | |
| CN107767366B (en) | A kind of transmission line of electricity approximating method and device | |
| EP3526768B1 (en) | Processing digital image to remove unwanted portion | |
| CN113447902A (en) | Sea surveillance radar target identification method based on machine learning | |
| CN108021868A (en) | A kind of quick highly reliable circular target detection recognition method | |
| CN114226262A (en) | Flaw detection method, flaw classification method and flaw detection system | |
| CN111325076B (en) | Aviation ground building extraction method based on fusion of U-net and Seg-net networks | |
| Niblock et al. | Fast model-based feature matching technique applied to airport lighting | |
| CN110298287A (en) | A kind of power equipment automatic identifying method based on deep learning | |
| CN116363435B (en) | Remote sensing image target detection system and method based on deep learning |
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 |