CN110245701A - A kind of electric power line detecting method based on unmanned plane image - Google Patents
A kind of electric power line detecting method based on unmanned plane image Download PDFInfo
- Publication number
- CN110245701A CN110245701A CN201910502613.8A CN201910502613A CN110245701A CN 110245701 A CN110245701 A CN 110245701A CN 201910502613 A CN201910502613 A CN 201910502613A CN 110245701 A CN110245701 A CN 110245701A
- Authority
- CN
- China
- Prior art keywords
- line
- image
- straightway
- electric power
- straight
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 16
- 230000005540 biological transmission Effects 0.000 claims abstract description 26
- 238000001514 detection method Methods 0.000 claims abstract description 16
- 230000005611 electricity Effects 0.000 claims abstract description 14
- 230000003993 interaction Effects 0.000 claims description 4
- 230000007423 decrease Effects 0.000 claims 1
- 230000009021 linear effect Effects 0.000 abstract description 4
- 239000000284 extract Substances 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 4
- 238000007689 inspection Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000005260 corrosion Methods 0.000 description 3
- 230000007797 corrosion Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004643 material aging Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000011946 reduction process Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- 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
-
- H02J13/0013—
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Image Analysis (AREA)
Abstract
A kind of electric power line detecting method based on unmanned plane image, LSD line segment detection is carried out to high partial image first, straightway potential in input picture region is detected and analyzed, extract doubtful power line region, secondly transmission line of electricity target candidate in image is obtained using the algorithm that positional relationship between unsupervised gaussian clustering method and line segment clusters characterized by the angle of line segment, operator finally based on line count in image, filter out transmission line of electricity target false in image, it is that linear LSD algorithm detects potential straightway in image that the present invention passes through time complexity first, then all straightways are clustered using unsupervised mixed Gaussian clustering method, the neighborhood relationships reused between straightway are removed apart from too far straightway, estimating for the line density in image is finally defined to come Straight-line target false in image is filtered, the detection accuracy of algorithm is further improved.
Description
Technical field
The invention belongs to power industry transmission line of electricity electric power line detecting method technologies.
Background technique
It is the major way for realizing long distance delivery electric power energy using high pressure, extra-high voltage transmission, so power transmission line
Road has become and its important energy resource supply line.After long range super-pressure large capacity transmission route is significantly extended, maintenance is super
The safe and stable operation of ultra-high-tension power transmission line is the solid foundation of economic construction rapid growth, and the failure for reducing transmission system is to subtract
The premise that few power outage occurs.Reduce transmission line of electricity system safe operation failure just must inspection density and guarantee patrol
The quality of inspection.
Power line is that the quality of main component its operating status in transmission line of electricity system is directly related to entire power grid system
The safety and stability of system.The erection direction of transmission line of electricity and the inspection direction of unmanned plane are consistent, therefore can pass through unmanned plane
The image of shooting determines the heading of unmanned plane after carrying out transmission line faultlocating, to realize the independent navigation of unmanned plane.
Power transmission line long-term work under the natural environment in field, by solarization, drench with rain, the extreme natural environments such as thunder and lightning are influenced；Also by
To continued mechanical tension, the influence of electrical flashover and own material aging.Transmission line of electricity is caused the defects of stranded, corrosion occur latent
The safe and stable operation for endangering network system.There are many scholars to mention in today that laser point cloud data application is gradually popularized
Go out through point cloud data and has detected transmission line of electricity, but since laser point cloud data dot density excessive cannot lead to this method
It cannot effectively detect transmission line of electricity.
With the progress of computer hardware technology and intelligent algorithm theory, so that the polling transmission line of automation becomes
Reality.In the picture real-time accurate detection go out overhead transmission line be realize power transmission line it is stranded, corrosion the defects of automatically detection and
Realize that unmanned plane independent navigation such as keeps in obscurity at the bases of key technologies.At present in digital image processing field, the master of straight-line detection
Flow Technique is hough transformation and its derivative various algorithms, and such Algorithms T-cbmplexity and space complexity are higher,
Hardly possible of being good in the application scenarios of high-definition picture reaches the requirement of real-time.
Summary of the invention
The present invention uses a kind of Algorithms T-cbmplexity for linear line detection algorithm LSD, using class mixed Gaussian point
The thought of class device gets straight line in image to carry out cluster to the straightway in image.To the great amount of images of unmanned plane shooting
After data are analyzed, it is found that true transmission line of electricity is by a group appearance, it is impossible to the overhead power transmission line of sub-thread occur, therefore
Define estimating to filter out single false target and improve the accuracy of power line detection for line density in image.
A kind of electric power line detecting method based on unmanned plane image, firstly, carrying out the inspection of LSD straightway to high partial image
It surveys, secondly, using the calculation of positional relationship cluster between unsupervised gaussian clustering method and line segment characterized by the angle of line segment
Method obtains transmission line of electricity target candidate in image, and the operator finally based on line count in image filters out false in image
Transmission line of electricity target；
The operator of line count of the present invention defines estimating to filter out single false target for line density in image
To improve the accuracy of power line detection；Assuming that all straight lines in image be all with the presence of interaction force, and this make
Firmly decay with distance, the isolated electric power line target in such position is less than dense distribution by the effect of other power lines
Straight line, false straight-line target is filtered out by this criterion；Its line density is defined as follows:
Wherein, x indicates the distance between two straight lines, and what f (x) was indicated is the active force between two straight lines.
It is that linear LSD algorithm detects potential straightway in image that the present invention passes through time complexity first, then
All straightways are clustered using unsupervised mixed Gaussian clustering method, reuse the neighborhood relationships between straightway
Removal finally defines estimating to filter straight line mesh false in image for the line density in image apart from too far straightway
Mark, further improves the detection accuracy of algorithm.
Detailed description of the invention
Fig. 1 is straight-line detection flow chart of the present invention.
Specific embodiment
1, straight line is detected based on LSD
LSD is that the line detection algorithm based on partial analysis is faster than hough transformation in arithmetic speed, in high score
Under the scene for detecting straight line in resolution image, LSD line detection algorithm has more applicability.The step of LSD line detection algorithm
It is summarised as follows:
(1) down-sampled to input picture progress Gaussian template convolution, effect is to carry out smooth and noise reduction process to image；
(2) gradient magnitude of each pixel and direction in image are calculated, the ladder of image is calculated using sobel operator
Degree, extracts the biggish point of change of gradient in image using thresholding, these will be evenly distributed on around insulation subregion.
The formula that gradient calculates is as follows:
(3) pixel is ranked up by gradient magnitude；
(4) linearity region growth analysis；
(5) NFA value calculates, and calculates the straight line enhancing processing of dot density and progress rectangular area in class；
2, unsupervised mixed Gaussian cluster judges straight line
After getting all straightways in image, not the information of priori come judge those straightways be belong to it is a certain
Straight line, therefore need to find potentially possible group of straightway combination in alignment using Unsupervised clustering algorithm.It is straight by lsd
The available four parameter, that is, x1, y1, x2 to straightway of line drawing algorithm, what y2 was represented is the starting point of straightway in the picture
And terminal point coordinate.The slope k of straightway can be calculated using these information, it in this way can be right using this information of slope
Initial straightway group, which cluster finding, may wherein organize straight straightway combination.The slope of each straight line is certain
It is the Gaussian Profile that μ standard deviation is σ that mean value is obeyed in fluctuation in range, as follows:
One slope of selection initializes a distribution from candidate straightway chained list, then selects one from remaining chained list
A slope k is compared with already existing lineal layout.If in 3 σ that current straight slope falls into some distribution
Think that current line segment belongs to this straight line；If in so selection on a plurality of matching line segments apart from that the smallest straight line, and
Current straightway is included into this straight line.
3, false target filters out in image
Since the background being imaged in actual scene is more complicated, wherein there may be the interference of other linears
Target such as highway, ridge etc..It is not in the high-tension bus-bar that sub-thread is set up from the point of view of the great amount of images analysis to overhead power transmission line
In field, therefore need to inhibit these false targets.It is single in image to filter out that the present invention proposes the concept of line count
One false straight-line target, it is assumed that all straight lines in image be all with the presence of interaction force, and this active force with
Distance and decay, the electric power line target that such position isolates is acted on by other power lines and is less than the straight of dense distribution
Line can filter out false straight-line target by this criterion.Line density is defined as follows:
Wherein, x indicates the distance between two straight lines, and what f (x) was indicated is the active force between two straight lines.It is all straight
Since the symmetry of interaction constitutes a upper triangular matrix, calculation amount is reduced to original 0.5 for active force between line
Times.
The result shows that this algorithm can identify the position of power line in high-definition picture, operational efficiency be can achieve
The requirement of real-time.To a technology for realize unmanned plane independent navigation flight and power line it is stranded, corrosion, foreign matter the defects of
Detection has very big application value.
Claims (2)
1. a kind of electric power line detecting method based on unmanned plane image, it is characterised in that: firstly, being carried out to high partial image
LSD line segment detection, secondly, being closed characterized by the angle of line segment using position between unsupervised gaussian clustering method and line segment
The algorithm of system's cluster obtains transmission line of electricity target candidate in image, and the operator finally based on line count in image filters out figure
The false transmission line of electricity target as in.
2. a kind of electric power line detecting method based on unmanned plane image according to the right 1, it is characterised in that: institute
The operator for stating line count defines estimating to filter out single false target and improve power line detection for line density in image
Accuracy；Assuming that all straight lines in image be all with the presence of interaction force, and this active force with distance and decline
Subtract, the isolated electric power line target in such position is acted on the straight line for being less than dense distribution by other power lines, passes through this
Criterion filters out false straight-line target；Its line density is defined as follows:
Wherein, x indicates the distance between two straight lines, and what f (x) was indicated is the active force between two straight lines.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910502613.8A CN110245701A (en) | 2019-06-11 | 2019-06-11 | A kind of electric power line detecting method based on unmanned plane image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910502613.8A CN110245701A (en) | 2019-06-11 | 2019-06-11 | A kind of electric power line detecting method based on unmanned plane image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110245701A true CN110245701A (en) | 2019-09-17 |
Family
ID=67886651
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910502613.8A Pending CN110245701A (en) | 2019-06-11 | 2019-06-11 | A kind of electric power line detecting method based on unmanned plane image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110245701A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112926556A (en) * | 2021-04-28 | 2021-06-08 | 上海大学 | Aerial photography power transmission line strand breaking identification method and system based on semantic segmentation |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050135680A1 (en) * | 2003-12-23 | 2005-06-23 | Carnegie Mellon University & Honda Motor Co., Ltd. | Polyhedron recognition system |
| CN102393961A (en) * | 2011-06-29 | 2012-03-28 | 山东电力研究院 | Computer vision-based real-time detecting and tracking method for electric power transmission circuit of aircraft |
| CN103761524A (en) * | 2014-01-17 | 2014-04-30 | 电子科技大学 | Image-based linear target recognition and extraction method |
| CN104156933A (en) * | 2013-05-13 | 2014-11-19 | 哈尔滨点石仿真科技有限公司 | Image registering method based on optical flow field |
| CN104867134A (en) * | 2015-05-04 | 2015-08-26 | 国家电网公司 | Identification method for transmission line tower inspection by unmanned aerial vehicle |
| CN106778822A (en) * | 2016-11-25 | 2017-05-31 | 西安电子科技大学 | Graph line detection method based on funnel conversion |
| CN108089185A (en) * | 2017-03-10 | 2018-05-29 | 南京沃杨机械科技有限公司 | The unmanned air navigation aid of agricultural machinery perceived based on farm environment |
| CN108919838A (en) * | 2018-08-27 | 2018-11-30 | 河海大学常州校区 | A kind of unmanned plane transmission line of electricity automatic tracking method based on binocular vision |
| CN109145905A (en) * | 2018-08-29 | 2019-01-04 | 河海大学常州校区 | A kind of transmission line of electricity accessory detection method of view-based access control model conspicuousness |
| CN109522852A (en) * | 2018-11-22 | 2019-03-26 | 中国科学院长春光学精密机械与物理研究所 | Man-made target detecting method, device and equipment based on optical remote sensing image |
-
2019
- 2019-06-11 CN CN201910502613.8A patent/CN110245701A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050135680A1 (en) * | 2003-12-23 | 2005-06-23 | Carnegie Mellon University & Honda Motor Co., Ltd. | Polyhedron recognition system |
| CN102393961A (en) * | 2011-06-29 | 2012-03-28 | 山东电力研究院 | Computer vision-based real-time detecting and tracking method for electric power transmission circuit of aircraft |
| CN104156933A (en) * | 2013-05-13 | 2014-11-19 | 哈尔滨点石仿真科技有限公司 | Image registering method based on optical flow field |
| CN103761524A (en) * | 2014-01-17 | 2014-04-30 | 电子科技大学 | Image-based linear target recognition and extraction method |
| CN104867134A (en) * | 2015-05-04 | 2015-08-26 | 国家电网公司 | Identification method for transmission line tower inspection by unmanned aerial vehicle |
| CN106778822A (en) * | 2016-11-25 | 2017-05-31 | 西安电子科技大学 | Graph line detection method based on funnel conversion |
| CN108089185A (en) * | 2017-03-10 | 2018-05-29 | 南京沃杨机械科技有限公司 | The unmanned air navigation aid of agricultural machinery perceived based on farm environment |
| CN108919838A (en) * | 2018-08-27 | 2018-11-30 | 河海大学常州校区 | A kind of unmanned plane transmission line of electricity automatic tracking method based on binocular vision |
| CN109145905A (en) * | 2018-08-29 | 2019-01-04 | 河海大学常州校区 | A kind of transmission line of electricity accessory detection method of view-based access control model conspicuousness |
| CN109522852A (en) * | 2018-11-22 | 2019-03-26 | 中国科学院长春光学精密机械与物理研究所 | Man-made target detecting method, device and equipment based on optical remote sensing image |
Non-Patent Citations (3)
| Title |
|---|
| HENG ZHANG 等: "Detecting Power Lines in UAV Images with Convolutional Features and Structured Constraints", 《REMOTE SENSE》 * |
| 杨理践 等: "基于电磁超声导波的铝板裂纹缺陷检测方法", 《仪器仪表学报》 * |
| 郭斯羽 等: "结合Hough变换与改进最小二乘法的直线检测", 《计算机科学》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112926556A (en) * | 2021-04-28 | 2021-06-08 | 上海大学 | Aerial photography power transmission line strand breaking identification method and system based on semantic segmentation |
| CN112926556B (en) * | 2021-04-28 | 2023-05-02 | 上海大学 | Semantic segmentation-based aerial photography transmission line broken strand identification method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112199993B (en) | Method for identifying transformer substation insulator infrared image detection model in any direction based on artificial intelligence | |
| Jiang et al. | A contour angle orientation for power equipment infrared and visible image registration | |
| Pal et al. | Real‐time condition monitoring of substation equipment using thermal cameras | |
| CN111814686A (en) | Vision-based power transmission line identification and foreign matter invasion online detection method | |
| Zhang et al. | A robust pantograph–catenary interaction condition monitoring method based on deep convolutional network | |
| CN107179479B (en) | Visible light image-based transmission conductor strand scattering defect detection method | |
| CN105912977B (en) | Lane line detection method based on point clustering | |
| CN104835152A (en) | Processing method and system of power transmission line inspection images | |
| Prasad et al. | LBP-HF features and machine learning applied for automated monitoring of insulators for overhead power distribution lines | |
| CN112131924A (en) | Transformer substation equipment image identification method based on density cluster analysis | |
| CN110245701A (en) | A kind of electric power line detecting method based on unmanned plane image | |
| Shen et al. | An automatic framework for pylon detection by a hierarchical coarse-to-fine segmentation of powerline corridors from UAV LiDAR point clouds | |
| Munir et al. | An automated method for individual wire extraction from power line corridor using lidar data | |
| Xuan et al. | Intelligent identification method of insulator defects based on CenterMask | |
| CN116452604B (en) | Complex substation scene segmentation method, device and storage medium | |
| Steiger et al. | Automatic detection of transmission towers | |
| CN112418226A (en) | Method and device for identifying opening and closing states of fisheyes | |
| Sun et al. | Transmission line tower fault identification based on image processing technology | |
| Xia et al. | Insulator recognition based on mathematical morphology and Bayesian segmentation | |
| Zhang et al. | Insulator segmentation algorithm based on k-means | |
| Zhang et al. | Automatic Classification of Transmission Lines Based on Edge Detection Filtering Algorithm and Laser Point Cloud Data Processing. | |
| Guan et al. | A visual saliency based railway intrusion detection method by UAV remote sensing image | |
| Potnuru et al. | Image processing and machine learning applied for condition monitoring of 11-kV power distribution line insulators using curvelet and LTP features | |
| Li et al. | A model-driven approach for fast modeling of three-dimensional laser point cloud in large substation | |
| Luo et al. | RETRACTED CHAPTER: Power Equipment Defect Detection Algorithm 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 | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20231229 |
|
| AD01 | Patent right deemed abandoned |