CN105261011B - A kind of unmanned plane inspection is taken photo by plane the extracting method of insulator in complex background image - Google Patents
A kind of unmanned plane inspection is taken photo by plane the extracting method of insulator in complex background image Download PDFInfo
- Publication number
- CN105261011B CN105261011B CN201510612844.6A CN201510612844A CN105261011B CN 105261011 B CN105261011 B CN 105261011B CN 201510612844 A CN201510612844 A CN 201510612844A CN 105261011 B CN105261011 B CN 105261011B
- Authority
- CN
- China
- Prior art keywords
- insulator
- binary map
- pixel
- connected domain
- vegetation
- 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.)
- Expired - Fee Related
Links
Landscapes
- Image Analysis (AREA)
Abstract
Taken photo by plane the invention discloses a kind of unmanned plane inspection the extracting method of insulator in complex background image, raw video is converted into HSI images using H variables extraction soil first, and vegetation is extracted from raw video according to vegetation extraction formula in Ecognition and power tower is extracted by spectral value segmentation according to the conspicuousness of power tower, segmentation is then carried out to raw video using two-dimentional OSTU thresholding methods and extracts coarse extraction insulation subgraph and obtaining thick insulator binary map with vegetation, soil binary map and power tower progress difference operation;Find the principal direction of each connected domain of binary map and travel through to obtain distance distribution histogram along the method direction of principal direction, be judged as insulator if histogram has regularity, finally extract the insulation subgraph to work well.
Description
Technical field
The invention belongs to remote sensing image technical field, taken photo by plane more particularly, to a kind of unmanned plane inspection in complex background image
The extracting method of insulator.
Background technology
Transmission line of electricity is an important part in national grid, mainly undertakes the conveying of electric energy, wherein super-pressure
Basic routing line of the power transmission line as power network, the safe and reliable operation to power network play vital effect, directly affect its people
Economic stable development.Most power transmission line is all exposed in outdoor environment, experiencing for a long time expose to the weather, the thunderbolt of sleet, lighting,
The infringement of the external environment such as dirt shifting and depression, while also subject to electric load and the internal pressure of mechanical load.Above-mentioned factor
The different degrees of aging of power line components will be caused even damaged, if can not find and eliminate in time and be in these power transmission lines hidden
Suffer from, it is possible to bring bigger harm to transmission of electricity link, the safe operation to national grid forms serious potential threat.Therefore,
Inspect periodically power transmission line and be one and power transmission line safety, the important process that normally conveys is effectively ensured, be national grid to electric energy
Reliable supply and basis and the premise of safe transmission.
The maintenance model of Traditional Man periodic inspection and correction maintenance power transmission line, it is difficult to meet the daily dimension of high voltage transmission line
Requirement is repaiied, the safe operation to national grid causes huge hidden danger, has been difficult to the demand for development for meeting the current power transmission network in China.Directly
The technology of liter aircraft line walking is developed so far in China to be had more than 20 years, and power department has been carried out greatly to helicopter line walking detection technique
The experimental study of amount, some power departments have also carried out the application of helicopter line walking.It is use in helicopter line walking data acquisition more
The equipment such as comfort, DV, high-resolution telescope, visible ray video recorder are as observation instrument.Utilize helicopter
Inspection power network, danger existing for artificial work high above the ground can be not only eliminated, and the omnidirectional shooting Detection results of aircraft compare people
Work detection is more comprehensive, can in time pinpoint the problems and solve it.
Insulator is the important component of overhead transmission line, and it, which is acted on, is support wire and prevents electric current from returning ground.By
Wind, Exposure to Sunlight are subjected to for a long time in insulator, is drenched with rain, and along with oneself body mechanical fatigue, therefore insulator crushes, crack
Equivalent damage, prevent insulator from normally playing a role.Hundreds of million can be collected per hour using helicopter line walking mode
Image/video data, during identification, judging power transmission line view data, if staff judges image using naked eyes mode
In whether there is anomaly, it would be desirable to substantial amounts of manpower, material resources and financial resources, and the artificial interpretation time one grow, easily cause regarding
Feel fatigue, erroneous judgement or situation of failing to judge easily occurs, it is difficult to be accurate to find potential safety hazard existing for transmission line of electricity.Therefore using at image
The method of reason and the method for pattern-recognition, the inexorable trend that insulator is development is quickly recognized from image.At present in threshold value
During split plot design extraction insulator, it is used as the basis of segmentation just with a certain variable in HSI color spaces, not
Suitable for most insulators.Classification of remote-sensing images is carried out using simple spectral signature and easily produces the different spectrum phenomenon of jljl,
Such as insulator is similar with the color of power tower and background is complicated, then is partitioned into partial picture using spectral signature
The effect of insulator is not ideal.Exactly under this background, the identification for insulator in unmanned plane filmed image is carried out
Correlative study.
The content of the invention
It is mostly to be directed to the problem of being carried out under simple background for existing many insulator extraction algorithms, the invention provides
A kind of unmanned plane inspection is taken photo by plane the insulator extracting method in complex background image.
The technical solution adopted in the present invention is:A kind of unmanned plane inspection is taken photo by plane the extraction of insulator in complex background image
Method, it is characterised in that comprise the following steps:
Step 1:Color space conversion is carried out to original remote sensing image, by RGB video conversions into HSI images, after conversion
HSI Extraction of Image S variables, S variable two-value images are handled using two-dimentional OSTU thresholding methods, obtain soil two-value
Figure;
Step 2:Original remote sensing image is handled, obtains vegetation binary map;
Step 3:Original remote sensing image is handled using two-dimentional OSTU thresholding methods, it is preliminary after being split
Insulator binary map;
Step 4:Spectrum saliency is had in raw video according to power tower, extraction power tower two is split by spectral value
Value figure;
Step 5:According to soil binary map, vegetation binary map and power tower binary map, preliminary insulator binary map is removed
In soil, vegetation and power tower, obtain thick insulator binary map;
Step 6:Connected domain principal direction mark is carried out to thick insulator binary map;The picture of each connected domain pixel is calculated first
First shape index, the connected domain principal direction is obtained after statistics;Then calculated using the shape index of principal direction and pixel
Preserved to the main shaft pixel of the connected domain, and by its pixel coordinate;Each main shaft pixel of pixel coordinate is traveled through, in main shaft pixel
On do a normal perpendicular to principal direction, the maximum distance that normal is intersected with connected domain preserves;
Step 7:According to the size property for middle insulator of taking photo by plane, if the maximum distance obtained in step 6 exist more than 60 or
Person is less than 10 element, just removes connected domain;
Step 8:The maximum distance obtained in step 7 is further processed;According to the style characteristic for middle insulator of taking photo by plane,
The average value of maximum distance is asked for, then each numerical value of maximum distance is subtracted each other with average value, the processing causes the tooth of insulator
It is more notable to take turns feature;
Step 9:Maximum distance in step 8 after further processing is counted, calculated adjacent current farthest
Distance is more than or equal to 0, and next distance is less than 0 distance length, if the accumulation number appearance of identical distance length is big
In 3, then can be determined that the connected domain is insulator, finally gives accurate insulator binary map.
Preferably, it is to utilize RGB vegetation extraction formula in Ecognition softwares in step 2, to original remote sensing image
Handled.
Relative to prior art, the beneficial effects of the invention are as follows:Background is complicated, insulator and power tower color it is close and
In the case that shooting angle is changeable, insulator is extracted from aerial images.
Brief description of the drawings
Fig. 1:For the method flow diagram of the embodiment of the present invention.
Embodiment
Understand for the ease of those of ordinary skill in the art and implement the present invention, below in conjunction with the accompanying drawings and embodiment is to this hair
It is bright to be described in further detail, it will be appreciated that implementation example described herein is merely to illustrate and explain the present invention, not
For limiting the present invention.
See Fig. 1, a kind of unmanned plane inspection provided by the invention is taken photo by plane the extracting method of insulator in complex background image,
Comprise the following steps:
Step 1:Color space conversion is carried out to original remote sensing image, by RGB video conversions into HSI images, after conversion
HSI Extraction of Image S variables, S variable two-value images are handled using two-dimentional OSTU thresholding methods, obtain soil two-value
Figure;
Spectral value and insulator of the soil in original remote sensing image near power tower pedestal is closer to, and carries out two dimension
Soil is always taken as insulator and extracted when OSTU thresholding methods, also, due to the remote sensing shadow of unmanned plane shooting
Picture, together with insulator is always obscured with soil, the shape facility of insulator is weakened by soil, is unfavorable for insulator below
Accurate extraction.Therefore always it is no more than the feature of certain threshold value using the value of soil S variables in HSI images, from raw video
In extract soil.
Wherein it is expressed asAccording to feature of the soil in variable S, located as follows
Reason obtains soil binary map.
Step 2:Formula is extracted using RGB vegetation in Ecognition softwares, original remote sensing image is handled, obtained
Vegetation binary map;
In remote sensing image vegetation large area exist, rely solely on threshold segmentation method raw video is carried out processing can not will
Vegetation thoroughly splits.The formula that vegetation is extracted wherein in Ecognition softwares is as follows:
VI '=(2*G '-R '-B ')-(1.4*R '-G ')
Wherein
Step 3:Original remote sensing image is handled using two-dimentional OSTU thresholding methods, it is preliminary after being split
Insulator binary map;
Two-dimentional OSTU thresholding methods are the extensions of varimax between class, are asked between two one-dimensional distribution maximum kinds
Variance is extended for solving the maximum of the mark of inter _ class relationship matrix.This method is the segmentation effect of raw video compared with maximum kind
Poor method is well a lot.Therefore Selection utilization two dimension OSTU thresholding methods enter row threshold division.The master of two-dimentional OSTU thresholding methods
Want step:
(1) establish image grey level histogram and (share L gray level, each probability of occurrence is p)
pi=ni/N
(2) probability of occurrence of background and target is calculated, computational methods are as follows:
(3) inter-class variance for calculating two regions of A and B is as follows:
ω0=pA*ωA+pB*ωB;
σ2=pA*(ωA-ω0)2+pB*(ωB-ω0)2;
First expression formula calculates A and the average gray value of B area respectively;
Second expression formula calculates the global average gray value of gray level image;
3rd expression formula calculates the inter-class variance in two regions of A, B.
(4) several steps more than have calculated the inter-class variance on single gray value, therefore optimal segmentation threshold value should
It is the gray value of gray variance between the class that A and B is enabled in image.
Step 4:Spectrum saliency is had in raw video according to power tower, extraction power tower two is split by spectral value
Value figure;
Step 5:According to soil binary map, vegetation binary map and power tower binary map, preliminary insulator binary map is removed
In soil, vegetation and power tower, obtain thick insulator binary map;
In raw video, power tower is always obvious compared with other atural objects.According to the significant characteristics of power tower,
The maximum and average value of three wave bands are obtained by traveling through raw video, then with following formulas Extraction power tower two-value
Figure.
Wherein Rmax、Gmax、BmaxRepresent the maximum of tri- wave bands of RGB, Raver、Gaver、BaverIt is expressed as tri- wave bands of RGB
Average value.
Step 6:Connected domain principal direction mark is carried out to thick insulator binary map;The picture of each connected domain pixel is calculated first
First shape index, the connected domain principal direction is obtained after statistics;Then calculated using the shape index of principal direction and pixel
Preserved to the main shaft pixel of the connected domain, and by its pixel coordinate;Each main shaft pixel of pixel coordinate is traveled through, in main shaft pixel
On do a normal perpendicular to principal direction, the maximum distance that normal is intersected with connected domain preserves;
The principal direction of connected domain is mainly to be determined by the most long connected domain of the length on some direction.According to pixel
Shape index, calculates the length of each direction line of each pixel of each connected domain, the extreme length at record pixel center and
The direction of the length, direction corresponding to the extreme length of the connected domain are the principal direction of the connected domain.Main shaft element is connected domain
The direction of middle pixel center extreme length is the pixel of principal direction.
Step 7:According to the size property for middle insulator of taking photo by plane, if the maximum distance obtained in step 6 exist more than 60 or
Person is less than 10 element, just removes connected domain;
Step 8:The maximum distance obtained in step 7 is further processed;According to the style characteristic for middle insulator of taking photo by plane,
The average value of maximum distance is asked for, then each numerical value of maximum distance is subtracted each other with average value, the processing causes the tooth of insulator
It is more notable to take turns feature;
Step 9:Maximum distance in step 8 after further processing is counted, calculated adjacent current farthest
Distance is more than or equal to 0, and next distance is less than 0 distance length, if the accumulation number appearance of identical distance length is big
In 3, then can be determined that the connected domain is insulator, finally gives accurate insulator binary map.
It should be appreciated that the part that this specification does not elaborate belongs to prior art.
It should be appreciated that the above-mentioned description for preferred embodiment is more detailed, therefore can not be considered to this
The limitation of invention patent protection scope, one of ordinary skill in the art are not departing from power of the present invention under the enlightenment of the present invention
Profit is required under protected ambit, can also be made replacement or deformation, be each fallen within protection scope of the present invention, this hair
It is bright scope is claimed to be determined by the appended claims.
Claims (2)
1. a kind of unmanned plane inspection is taken photo by plane in complex background image insulator extracting method, it is characterised in that including following step
Suddenly:
Step 1:Color space conversion is carried out to original remote sensing image, by RGB video conversions into HSI images, to the HSI after conversion
Extraction of Image S variables, S variable two-value images are handled using two-dimentional OSTU thresholding methods, obtain soil binary map;
Step 2:Original remote sensing image is handled, specifically included:
Original remote sensing image is handled, obtains vegetation binary map;
Original remote sensing image is handled using two-dimentional OSTU thresholding methods, the preliminary insulator two-value after being split
Figure;
Step 3:Spectrum saliency is had in raw video according to power tower, extraction power tower two-value is split by spectral value
Figure;
Step 4:According to soil binary map, vegetation binary map and power tower binary map, remove in preliminary insulator binary map
Soil, vegetation and power tower, obtain thick insulator binary map;
Step 5:Connected domain principal direction mark is carried out to thick insulator binary map;The pixel shape of each connected domain pixel is calculated first
Shape index, the connected domain principal direction is obtained after statistics;Then calculated using the pixel shape index of principal direction and pixel
Preserved to the main shaft pixel of the connected domain, and by its pixel coordinate;Each main shaft pixel of pixel coordinate is traveled through, in main shaft pixel
On do a normal perpendicular to principal direction, the maximum distance that normal is intersected with connected domain preserves;
Step 6:According to the size property for middle insulator of taking photo by plane, if the maximum distance obtained in step 5, which exists, is more than 60 or small
In 10 element, just connected domain is removed;
Step 7:The maximum distance obtained in step 6 is further processed, its specific implementation process is:According to middle insulation of taking photo by plane
The style characteristic of son, asks for the average value of maximum distance, then each numerical value of maximum distance is subtracted each other with average value so that insulation
The gear feature of son is more notable;
Step 8:Maximum distance in step 7 after further processing is counted, calculates adjacent current maximum distance
More than or equal to 0, if identical distance length accumulation number is more than 3, judge that the connected domain for insulator, finally gives essence
True insulator binary map.
2. unmanned plane inspection according to claim 1 is taken photo by plane in complex background image insulator extracting method, its feature
It is:It is to utilize RGB vegetation extraction formula in Ecognition softwares in step 2, original remote sensing image is handled.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510612844.6A CN105261011B (en) | 2015-09-22 | 2015-09-22 | A kind of unmanned plane inspection is taken photo by plane the extracting method of insulator in complex background image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510612844.6A CN105261011B (en) | 2015-09-22 | 2015-09-22 | A kind of unmanned plane inspection is taken photo by plane the extracting method of insulator in complex background image |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105261011A CN105261011A (en) | 2016-01-20 |
CN105261011B true CN105261011B (en) | 2017-12-15 |
Family
ID=55100683
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510612844.6A Expired - Fee Related CN105261011B (en) | 2015-09-22 | 2015-09-22 | A kind of unmanned plane inspection is taken photo by plane the extracting method of insulator in complex background image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105261011B (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107369162B (en) * | 2017-07-21 | 2020-07-10 | 华北电力大学(保定) | Method and system for generating insulator candidate target area |
CN107492094A (en) * | 2017-07-21 | 2017-12-19 | 长安大学 | A kind of unmanned plane visible detection method of high voltage line insulator |
CN109636771B (en) * | 2018-10-23 | 2021-09-03 | 中国船舶重工集团公司第七0九研究所 | Flight target detection method and system based on image processing |
CN113673385B (en) * | 2021-08-06 | 2024-06-21 | 南京理工大学 | Sea surface ship detection method based on infrared image |
CN114625166B (en) * | 2022-03-03 | 2024-04-30 | 江苏方天电力技术有限公司 | Intelligent positioning method for shooting position of unmanned aerial vehicle |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104794707A (en) * | 2015-04-08 | 2015-07-22 | 国家电网公司 | Automatic extraction method of insulator chain infrared imagery disk and steel cap area |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2013134666A (en) * | 2011-12-27 | 2013-07-08 | Dainippon Screen Mfg Co Ltd | Binary image generation device, classification device, binary image generation method, and classification method |
-
2015
- 2015-09-22 CN CN201510612844.6A patent/CN105261011B/en not_active Expired - Fee Related
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104794707A (en) * | 2015-04-08 | 2015-07-22 | 国家电网公司 | Automatic extraction method of insulator chain infrared imagery disk and steel cap area |
Non-Patent Citations (2)
Title |
---|
图像处理中的绝缘子缺陷检测方法;单成 等;《中国计量学院学报》;20101231;第21卷(第4期);第297-300页,第304页 * |
形状感知的绝缘子识别与缺陷诊断;张晶晶 等;《中国图象图形学报》;20140831;第19卷(第8期);第1194-1201页 * |
Also Published As
Publication number | Publication date |
---|---|
CN105261011A (en) | 2016-01-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105261011B (en) | A kind of unmanned plane inspection is taken photo by plane the extracting method of insulator in complex background image | |
Ahmad et al. | Vegetation encroachment monitoring for transmission lines right-of-ways: A survey | |
CN102590823B (en) | Rapid extraction and reconstruction method for data power line of airborne LIDAR | |
CN111342391B (en) | Power transmission line insulator and line fault inspection method and inspection system | |
CN108986104B (en) | Insulator image segmentation method based on distance transformation and watershed marking algorithm | |
CN109300118B (en) | High-voltage power line unmanned aerial vehicle inspection method based on RGB image | |
Chen et al. | The application of the tasseled cap transformation and feature knowledge for the extraction of coastline information from remote sensing images | |
CN112435208B (en) | Insulator region detection method for significant edge guidance | |
CN104849287A (en) | Composite insulator contamination degree non-contact detection method | |
CN103529362A (en) | Perception based insulator recognition and defect diagnosis method | |
CN106326808A (en) | Method for detecting bird nests in power transmission line poles based on unmanned plane images | |
CN110334623B (en) | Method for extracting collapsing information based on Sentinel-2A satellite remote sensing image | |
CN104951754A (en) | Sophisticated crop classifying method based on combination of object oriented technology and NDVI (normalized difference vegetation index) time series | |
CN114266958A (en) | Cloud platform based mangrove remote sensing rapid and accurate extraction method | |
CN106250835A (en) | Bird's Nest recognition methods on the transmission line of electricity of feature based identification | |
Hu et al. | Information extraction of subsided cultivated land in high-groundwater-level coal mines based on unmanned aerial vehicle visible bands | |
CN112270234B (en) | Power transmission line insulation sub-target identification method based on aerial image | |
CN109697410A (en) | A kind of remote sensing Objects recognition method of overhead transmission line covering area | |
CN106228192B (en) | A kind of transmission line forest fire domain identification method analyzed based on double-threshold cluster | |
Jiang et al. | Developing a new red band–SEVI–blue band (RSB) enhancement method for recognition the extra-high-voltage transmission line corridor in green mountains | |
CN104778469B (en) | Automatic extraction method of sun outline | |
Sai et al. | Texture-based forest segmentation in satellite images | |
Xu et al. | Change detection based on IR-MAD model for GF-5 remote sensing imagery | |
Wang et al. | Canopy extraction and height estimation of trees in a shelter forest based on fusion of an airborne multispectral image and photogrammetric point cloud | |
Yue et al. | The extraction of water information based on SPOT5 image using object-oriented method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20171215 Termination date: 20180922 |