CN105426824B - Hidden danger target analysis method and system based on Aerial Images - Google Patents
Hidden danger target analysis method and system based on Aerial Images Download PDFInfo
- Publication number
- CN105426824B CN105426824B CN201510753533.1A CN201510753533A CN105426824B CN 105426824 B CN105426824 B CN 105426824B CN 201510753533 A CN201510753533 A CN 201510753533A CN 105426824 B CN105426824 B CN 105426824B
- Authority
- CN
- China
- Prior art keywords
- hidden danger
- target
- visible images
- target image
- images
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/176—Urban or other man-made structures
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Processing Or Creating Images (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of hidden danger target analysis method and system based on Aerial Images, methods described includes：Pass through acquisition visible images of taking photo by plane in inspection；In the visible images, the target image containing tower bar is recognized；Insulator chain is detected in the target image containing tower bar；Hidden danger target is detected in the region of the target image above insulator chain；Insulator chain flashover impact analysis is carried out to the hidden danger target detected.The program is by the way that the target image containing tower bar is identified, choose the target image containing shaft tower, detection and analysis on the basis of the target image of screening by the detection realization of insulator chain to hidden danger target, substantial amounts of inspection Aerial Images can quickly be screened, improve to hidden danger target identification and analysis efficiency, contrast and prior art, the degree of accuracy and efficiency of the program to the discriminance analysis of hidden danger target are obviously improved.
Description
Technical field
The invention belongs to helicopter in electric inspection process technical field, more particularly, to a kind of hidden danger target based on Aerial Images
Analysis method and system.
Background technology
Application with unmanned plane/helicopter in electric inspection process field has obtained increasing accreditation, and tradition is by leader
Issued written instructions task, and patrol officer is maked an inspection tour by experience and memory along transmission line of electricity, with paper record, again by special messenger after return
The patrol mode of arrangement changes.There are some shortcomings in traditional routine inspection mode, for example：Cost of labor is high, the line walking time is long；Patrol
There is certain subjectivity to the analysis of transmission line of electricity hidden danger by manual record in inspection；Patrol officer disperses, and each collects record and patrols
Content is examined, is unfavorable for statistical analysis etc..
At present, UAV flight's Visible Light Camera carries out the Technical comparing maturation of aerial mission, for the hidden of transmission line of electricity
Suffer from identification mission, unmanned plane needs to carry the devices such as Visible Light Camera, GPS measuring instruments, and design suitable photoelectric nacelle and carry
Required loading device.Helicopter carry out power-line patrolling application time it is shorter, for helicopter power-line patrolling auxiliary equipment compared with
It is few, the patrol plan on the day of existing helicopter power line inspection auxiliary equipment generation, and the shooting of real-time reception visible light camera
Photo.Matched by manually being checked to the picture that aforesaid way is got, and with current shaft tower information of making an inspection tour, can
With hidden danger such as the Bird's Nests on manual identified power equipment, the mark and report generation of hidden danger target are realized.
Carry out unmanned plane/helicopter routing inspection and improve routing inspection efficiency to a certain extent, but be due to the figure of acquisition of taking photo by plane
As data volume greatly, the depending merely on artificial visual interpretation lookup hidden danger target of the task is weighed very much, to the standard of the discriminance analysis of hidden danger target
Exactness and less efficient is this in the urgent need to improving hidden danger target identification analysis efficiency by the method and system of intellectual analysis.
The content of the invention
Therefore, the object of the invention is providing a kind of hidden danger target analysis method and system based on Aerial Images, to solve
Take photo by plane in the prior art acquisition image data amount it is big, depend merely on artificial visual interpretation and search the task of hidden danger target and weigh very much, it is right
The degree of accuracy of the discriminance analysis of hidden danger target and less efficient defect.
The technical scheme that the present invention is provided is as follows：
A kind of hidden danger target analysis method based on Aerial Images, including：
Pass through acquisition visible images of taking photo by plane in inspection；
In the visible images, the target image containing tower bar is recognized；
Insulator chain is detected in the target image containing tower bar；
Hidden danger target is detected in the region of the target image above insulator chain；
Insulator chain flashover impact analysis is carried out to the hidden danger target detected.
Preferably, after the target image of the identification containing tower bar, in addition to：
Target image containing shaft tower is marked and stored respectively with not containing the image of tower bar.
Preferably, the target image of the identification containing tower bar, including：
Color Segmentation is carried out to the visible images；
Straight-line detection is carried out to the visible images after Color Segmentation, shaft tower edge line is detected；
Parallel lines judgement is carried out to the straight line detected；
Judge that the straightway that there are parallel lines in the range of presetted pixel is used as the part at shaft tower edge；
Preferably, the visible images are carried out with Color Segmentation, including：
Green area is obtained using threshold method in visible images；
Green area in the visible images is filtered out.
Preferably, before the target image of the identification containing tower bar, in addition to：
When obtaining visible images by taking photo by plane, while obtaining GPS information；
According to the GPS information and the shooting time information of the visible images, the coordinate letter of visible images is obtained
Breath；
According to the coordinate information, selection sets the visible images in geographic area, and recognizes wherein containing tower bar
Target image.
Preferably, before the coordinate information for obtaining visible images, in addition to；
The GPS information and visible images are stored in the memory cell of inspection device；
The GPS information and visible images are transferred to by presupposition analysis server by wired or wireless transmission method
On；
The Analysis server carries out difference operation by the shooting time to visible images, obtains the visible ray figure
The coordinate information of picture.
Preferably, insulator chain is detected in the target image containing tower bar, including：
According to the statistical nature of Hough curves, Hough curves oval in target image are sampled, the ellipse is obtained
The directional information at edge；
The oval positional information is obtained by Line segment detection in image space according to the directional information；
According to the oval directional information and positional information fitting detection insulator chain.
Preferably, hidden danger target is detected in the region of the target image above insulator chain, including：
Row threshold division is entered to the region of the target image above insulator chain；
Calculated by connected domain, and the syntople of connected domain and shaft tower recognizes hidden danger target.
Preferably, after the described pair of hidden danger target progress insulator chain flashover impact analysis detected, in addition to：
It will be seen that light image and analysis result are forwarded by network.
Corresponding to the above method, present invention also offers a kind of hidden danger targeted assay systems based on Aerial Images, including：
Visible Light Camera, for passing through acquisition visible images of taking photo by plane in inspection；
Tower bar optical sieving module, in the visible images, recognizing the target image containing tower bar；
Insulator chain detection module, for detecting insulator chain in the target image containing tower bar；
Hidden danger module of target detection, for detecting hidden danger mesh in the region of the target image above insulator chain
Mark；
Hidden danger target analysis module, for carrying out insulator chain flashover impact analysis to the hidden danger target detected.
Preferably, the tower bar optical sieving module, including：
Color Segmentation unit, for carrying out Color Segmentation to the visible images；
Straight-line detection unit, for the visible images after Color Segmentation to be carried out with straight-line detection, detection shaft tower edge
Straight line；
Parallel lines judging unit, for carrying out parallel lines judgement to the straight line detected；
Shaft tower edge judgment unit, for the straightway that judges to there are parallel lines in the range of presetted pixel as shaft tower side
A part for edge；
Tower bar recognition unit, the target image for containing tower bar according to shaft tower limb recognition.
Preferably, the Color Segmentation unit, including：
Region recognition component, for obtaining green area using threshold method in visible images；
Region filters out component, for the green area in the visible images to be filtered out.
Preferably, the hidden danger targeted assay systems based on Aerial Images, in addition to：
GPS module, for when obtaining visible images by taking photo by plane, while obtaining GPS information；
Coordinate obtaining module, for the shooting time information according to the GPS information and the visible images, acquisition can
See the coordinate information of light image；
Image selection module, for according to the coordinate information, the visible images in selection setting geographic area, and by
The tower bar optical sieving module recognizes the target image wherein containing tower bar.
Preferably, the hidden danger targeted assay systems based on Aerial Images, in addition to；
Memory cell, is arranged in inspection device, for storing the GPS information and visible images；
Transmission unit, for being transferred to the GPS information and visible images by wired or wireless transmission method
On presupposition analysis server；
Analysis server, for carrying out difference operation by the shooting time to visible images, obtains the visible ray
The coordinate information of image.
Preferably, the insulator chain detection module, including：
Directional information acquiring unit, for the statistical nature according to Hough curves, to Hough oval in target image
Curve is sampled, and obtains the directional information of the elliptical edge；
Location information acquiring unit, for obtaining described ellipse by Line segment detection in image space according to the directional information
Round positional information；
Detection unit is fitted, for according to the oval directional information and positional information fitting detection insulator chain.
Preferably, the hidden danger module of target detection, including：
Threshold segmentation unit, row threshold division is entered for the region to the target image above insulator chain；
Hidden danger object-recognition unit, for being calculated by connected domain, and the syntople identification of connected domain and shaft tower is hidden
Suffer from target.
Preferably, the hidden danger targeted assay systems based on Aerial Images, in addition to：
Forwarding module, for will be seen that light image and analysis result are forwarded by network.
The present invention has the following advantages that compared with prior art：
In the above-mentioned hidden danger target analysis scheme based on Aerial Images, by knowing to the target image containing tower bar
Not, the target image containing shaft tower is chosen, is realized on the basis of the target image of screening by the detection of insulator chain to hidden danger
Substantial amounts of inspection Aerial Images can quickly be screened, improved to hidden danger target identification by the detection and analysis of target
And analysis efficiency, contrast and prior art, the degree of accuracy and efficiency of the program to the discriminance analysis of hidden danger target have significantly
Lifting.
Brief description of the drawings
Fig. 1 is the hidden danger target analysis method flow schematic diagram based on Aerial Images that the embodiment of the present invention one is provided；
Fig. 2 is the hidden danger targeted assay systems composition schematic diagram based on Aerial Images that the embodiment of the present invention two is provided.
Embodiment
To overcome the image data amount for acquisition of taking photo by plane in the prior art big, depend merely on artificial visual interpretation and search hidden danger target
Task is weighed very much, and the degree of accuracy and less efficient defect to the discriminance analysis of hidden danger target are based on this application provides one kind
The hidden danger target analysis method and system of Aerial Images.
The present invention will be further described below in conjunction with the accompanying drawings.
Embodiment one：
The hidden danger target analysis method flow schematic diagram based on Aerial Images shown in reference picture 1, specific steps of the present invention
Implement as follows：
Step S101, passes through acquisition visible images of taking photo by plane in inspection；
During the inspection by unmanned plane or helicopter, it can carry required by designing suitable photoelectric nacelle
Loading device, carry Visible Light Camera carry out aerial mission.Patrol task can also be arranged simultaneously, collected region of patrolling and examining
Know data, including shaft tower label etc..
Step S102, in the visible images, recognizes the target image containing tower bar；
, can be no longer by the screening to tower bar data due to during inspection, substantial amounts of view data can be got
The image for not containing tower bar is handled, substantial amounts of analysis task is reduced.
It is corresponding, in this step, the target image containing shaft tower can also respectively be carried out with not containing the image of tower bar
Mark and storage.
Step S103, insulator chain is detected in the target image containing tower bar；
Destruction due to hidden danger target (mainly Bird's Nest) to transmission line of electricity is mainly manifested in the insulation that hidden danger target is located at
Insulator chain flashover is easily caused above substring, therefore hidden danger target bearing can be determined by the detection of insulator chain first.
Hidden danger target is detected in step S104, the region of the target image above insulator chain；
After the position that insulator chain is determined in the target image containing tower bar, you can to confirm the presence of hidden danger target
Region, substantial amounts of detection workload can be reduced by carrying out perils detecting only for the region.
Step S105, insulator chain flashover impact analysis is carried out to the hidden danger target detected.
In-depth analysis is carried out to hidden danger target identification result and hidden danger target image, carried out for the hidden danger target detected
Insulator chain flashover impact analysis, and generate hidden danger target analysis report.
In above-described embodiment, when being taken photo by plane, substantial amounts of visible images can be got, in order to further reduce to figure
The workload of the analysis of picture, the method that the present embodiment is provided can also include：
Step S101a, when obtaining visible images by taking photo by plane, also obtains GPS information simultaneously；
Step S101b, according to the GPS information and the shooting time information of the visible images, obtains visible ray figure
The coordinate information of picture；
Step S101c is according to the coordinate information, and selection sets the visible images in geographic area, and identification wherein contains
There is the target image of tower bar.
A kind of specific implementation as follows, being provided for the present embodiment correspondence：
The GPS information and the visible light image information are loaded first；Extract GPS measurements and visible ray in inspection data
The temporal information of image shot by camera；By carrying out difference operation to the time, the coordinate information matched with every image is obtained；
The numbering of Aerial Images, generation picture number and the list of GPS information corresponding relation are extracted again；And shown on analytical equipment
GPS information is numbered with Aerial Images.
The screening based on geographical position to image realized by above-mentioned GPS information, can arbitrarily choose needs what is analyzed
Transmission line of electricity section carries out hidden danger target analysis, and the workload of reduction graphical analysis identification improves analysis efficiency.
Further, since the data analysis and process limited capacity of UAV flight's equipment, related visible images and GPS
Information can be first temporarily stored in corresponding storage device, and subsequently transfers to corresponding server to be analyzed and processed, specifically,
Before the coordinate information for obtaining visible images, it can also include；
The GPS information and visible images are stored in the memory cell of inspection device by step S101d, step；
The GPS information and visible images, are transferred to default by step S101e by wired or wireless transmission method
On Analysis server；
Step S101f, the Analysis server carries out difference operation by the shooting time to visible images, obtains institute
State the coordinate information of visible images.
A kind of specific implementation as follows, being provided for the present embodiment correspondence：
The acquisition of unmanned plane electric inspection process data, is initially arrangement patrol task, collects the given data of region of patrolling and examining, bag
Include shaft tower label etc.；The load such as UAV flight's Visible Light Camera, GPS measuring instruments are flown, and by GPS information, visible ray
Image and attitude information etc. are stored temporarily on unmanned plane in Visible Light Camera storage card and GPS measuring instrument storage cards；Unmanned plane
Flight will be temporarily stored in storage device on unmanned plane after terminating by USB connecting lines, Wi-Fi or cordless communication network transmission equipment
On information transfer in specified Analysis server equipment.The information data obtained from unmanned plane can be read into number first
Data Format Transform is carried out according to acquiring unit, then is transferred to hidden danger targeted assay systems carrying out Data Analysis Services, obtain hidden danger
The analysis report of target (such as Bird's Nest).
In the scene of electric inspection process, tower bar position is easiest to exist in hidden danger, therefore the present embodiment, in step S102,
By recognizing the target image containing tower bar, the image without tower bar can be excluded, the analysis workload of image is reduced, carries as follows
A kind of method for recognizing the target image containing tower bar has been supplied, including：
The visible images are carried out Color Segmentation by step S102a；
The color segmentation is based on geographical environment, under the varying environments such as desert area, mountain region, snowfield and field, its
Implementing for the Color Segmentation of use is also different, so that the target context of tower image is planted for green as an example, to the visible ray figure
As carrying out Color Segmentation, including：
Green area is obtained using threshold method in visible images；
Green area in the visible images is filtered out.
Specifically, under RGB color, green information integrated distribution dominates the color component of green at a position
For green component.Generally, the value of green component is more than red component and blue component, and the green component values of vegetation are small
In certain value (about 200), red component and the blue color component value rainy 150 of vegetation green, to color component formula (2B-
R-G) calculated, green area is obtained using threshold method, filtered out that (pixel value is with 0 value generation to the green area in image
For).
Visible images after Color Segmentation are carried out straight-line detection by step S102b, detect shaft tower edge line；Tool
Body can be adopted carries out Hough straight-line detections to visible images.
Step S102c, parallel lines judgement is carried out to the straight line detected；
Step S102d, judges that the straightway that there are parallel lines in the range of presetted pixel is used as the part at shaft tower edge；
The presetted pixel, can be preferably 50 pixels, can exclude ridge, the interference of power transmission line isoline.
Step S102e, the target image of tower bar is contained according to shaft tower limb recognition.
Because the hidden danger such as Bird's Nest target is mainly manifested in the top of the insulator chain of Bird's Nest foundation to the destruction of transmission line of electricity
Easily cause in insulator chain flashover, therefore the method for the present embodiment offer, hidden danger mesh can be determined by the detection of insulator chain
(such as Bird's Nest) orientation is marked, because most insulation subgraphs are corresponding into ellipticity in Aerial Images, in the step S103,
Insulator chain is detected in target image containing tower bar, can be included：
Step S103a, according to the statistical nature of Hough curves, samples to Hough curves oval in target image, takes
Obtain the directional information of the elliptical edge；
If elliptic equation is：
Then parameter has 4：Centre coordinate, length half shaft length a and b.Due to there is 4 parameters, so directly utilizing
Hough transform is needed to set up the cumulative arrays of a 4-D, and q, a and b all values will be traveled through to each p.It is bent according to Hough
The statistical nature of line, samples to oval Hough curves, obtains the directional information of elliptical edge
Step S103b, obtains the oval position by Line segment detection in image space according to the directional information and believes
Breath.
Step S103c, according to the oval directional information and positional information fitting detection insulator chain.
In order to improve the precision of detection, before step S103a, it can also include：
The pretreatment such as medium filtering and image enhaucament is carried out to the shaft tower image screened.
On the basis of step S103, the scheme of the application is entered to the insulator upper area for being easiest to occur hidden danger target
Row is concealed in target identification, step S104, and hidden danger mesh is detected in the region of the target image above insulator chain
Mark, can include：
Row threshold division is entered to the region of the target image above insulator chain；
Calculated by connected domain, and the syntople of connected domain and shaft tower recognizes hidden danger target.
In addition, for the ease of the use to analysis result, after step S105 obtains Hazard Analysis Report, can also wrap
Include：
Step S106, it will be seen that light image and analysis result are forwarded by network.
Specifically, the inspection such as image and GPS information data and Hazard Analysis Report can together be passed through into wire communication line
Cable or wireless connection are sent on inspection command system, while can also be incited somebody to action by mobile radio network transmission equipment or Wi-Fi
Hidden danger target analysis report is forwarded in hand-held inspection terminal.
In the above-mentioned hidden danger target analysis method based on Aerial Images that the present embodiment is provided, by the mesh containing tower bar
Logo image is identified, and chooses the target image containing shaft tower, passes through the inspection of insulator chain on the basis of the target image of screening
Survey and realize detection and analysis to hidden danger target, substantial amounts of inspection Aerial Images can quickly be screened, improved pair
Hidden danger target identification and analysis efficiency, contrast and prior art, the degree of accuracy and effect of the program to the discriminance analysis of hidden danger target
Rate is obviously improved.
Embodiment two：
The method provided corresponding to embodiment one, the present embodiment additionally provides a kind of hidden danger target point based on Aerial Images
Analysis system, as shown in Fig. 2 constituting figure for the module of the system, is specifically included：
Visible Light Camera 201, for passing through acquisition visible images of taking photo by plane in inspection；
Tower bar optical sieving module 202, in the visible images, recognizing the target image containing tower bar；
Insulator chain detection module 203, for detecting insulator chain in the target image containing tower bar；
Hidden danger is detected in hidden danger module of target detection 204, the region for the target image above insulator chain
Target；
Hidden danger target analysis module 205, for carrying out insulator chain flashover impact analysis to the hidden danger target detected.
In the scene of electric inspection process, tower bar position is easiest to exist in hidden danger, therefore the present embodiment, the tower bar image
Screening module 202 can exclude the image without tower bar by recognizing the target image containing tower bar, reduce the analysis work of image
Measure, specifically, the tower bar optical sieving module 202 can include：
Color Segmentation unit, for carrying out Color Segmentation to the visible images；
Straight-line detection unit, for the visible images after Color Segmentation to be carried out with straight-line detection, detection shaft tower edge
Straight line；
Parallel lines judging unit, for carrying out parallel lines judgement to the straight line detected；
Shaft tower edge judgment unit, for the straightway that judges to there are parallel lines in the range of presetted pixel as shaft tower side
A part for edge；
Tower bar recognition unit, the target image for containing tower bar according to shaft tower limb recognition.
The color segmentation is based on geographical environment, under the varying environments such as desert area, mountain region, snowfield and field, its
Implementing for the Color Segmentation of use is also different, so that the target context of tower image is planted for green as an example, the Color Segmentation list
Member, can include：
Region recognition component, for obtaining green area using threshold method in visible images；
Region filters out component, for the green area in the visible images to be filtered out.
When being taken photo by plane, substantial amounts of visible images can be got, in order to further reduce the work of the analysis to image
Measure, the hidden danger targeted assay systems based on Aerial Images can also include：
GPS module 206, for when obtaining visible images by taking photo by plane, while obtaining GPS information；
Coordinate obtaining module, for the shooting time information according to the GPS information and the visible images, acquisition can
See the coordinate information of light image；
Image selection module, for according to the coordinate information, the visible images in selection setting geographic area, and by
The tower bar optical sieving module recognizes the target image wherein containing tower bar.
Due to the data analysis and process limited capacity of UAV flight's equipment, related visible images and GPS information
It can first be temporarily stored in corresponding memory cell, and subsequently transfer to corresponding server to be analyzed and processed, therefore described base
In the hidden danger targeted assay systems of Aerial Images, it can also include；
Memory cell, is arranged in inspection device, for storing the GPS information and visible images, can be visible
Light camera memory and GPS measuring instrument storage cards.
Transmission unit, for being transferred to the GPS information and visible images by wired or wireless transmission method
On presupposition analysis server；The transmission unit enters line number by USB connecting lines, Wi-Fi or cordless communication network transmission means
According to transmission.
Analysis server, for carrying out difference operation by the shooting time to visible images, obtains the visible ray
The coordinate information of image.
Because the hidden danger such as Bird's Nest target is mainly manifested in the top of the insulator chain of Bird's Nest foundation to the destruction of transmission line of electricity
Easily cause in insulator chain flashover, therefore the system of the present embodiment offer, hidden danger mesh can be determined by the detection of insulator chain
(such as Bird's Nest) orientation is marked, corresponding insulator chain detection module 203 can include：
Directional information acquiring unit, for the statistical nature according to Hough curves, to Hough oval in target image
Curve is sampled, and obtains the directional information of the elliptical edge；
Location information acquiring unit, for obtaining described ellipse by Line segment detection in image space according to the directional information
Round positional information；
Detection unit is fitted, for according to the oval directional information and positional information fitting detection insulator chain.
The system that the present embodiment is provided carries out concealing target knowledge to the insulator upper area for being easiest to occur hidden danger target
Not, the hidden danger module of target detection 204, can include：
Threshold segmentation unit, row threshold division is entered for the region to the target image above insulator chain；
Hidden danger object-recognition unit, for being calculated by connected domain, and the syntople identification of connected domain and shaft tower is hidden
Suffer from target.
In addition, for the ease of the use to analysis result, the hidden danger targeted assay systems based on Aerial Images, also
It can include：
Forwarding module, for will be seen that light image and analysis result are forwarded by network.
Specifically, the inspection such as image and GPS information data and Hazard Analysis Report can together be passed through into wire communication line
Cable or wireless connection are sent on inspection command system, while can also be incited somebody to action by mobile radio network transmission equipment or Wi-Fi
Hidden danger target analysis report is forwarded in hand-held inspection terminal.
In the above-mentioned hidden danger targeted assay systems based on Aerial Images that the present embodiment is provided, tower bar optical sieving module is led to
Cross and the target image containing tower bar is identified, choose the target image containing shaft tower, insulator chain detection module and hidden
Suffer from module of target detection on the basis of the target image of screening by the detection of insulator chain realize to the detection of hidden danger target with
And analysis, substantial amounts of inspection Aerial Images can quickly be screened, be improved to hidden danger target identification and analysis efficiency, it is right
Than with prior art, the degree of accuracy and efficiency of the program to the discriminance analysis of hidden danger target be obviously improved.
The embodiment of each in this specification is described by the way of progressive, and what each embodiment was stressed is and other
The difference of embodiment.The present embodiment is the corresponding system embodiment of embodiment of the method one, and its similar part cross-reference is
Can, repeat no more.
The foregoing description of the disclosed embodiments, enables professional and technical personnel in the field to realize or using the present invention.
A variety of modifications to these embodiments will be apparent for those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, it is of the invention
The embodiments shown herein is not intended to be limited to, and is to fit to and principles disclosed herein and features of novelty phase one
The most wide scope caused.
Claims (11)
1. a kind of hidden danger target analysis method based on Aerial Images, it is characterised in that including：
Pass through acquisition visible images of taking photo by plane in inspection；
In the visible images, the target image containing tower bar is recognized；
Insulator chain is detected in the target image containing tower bar；
Hidden danger target is detected in the region of the target image above insulator chain；
Insulator chain flashover impact analysis is carried out to the hidden danger target detected；
Wherein, before the acquisition visible images, in addition to；
GPS information and visible images are stored in the memory cell of inspection device；
The GPS information and visible images are transferred on presupposition analysis server by wired or wireless transmission method；
The Analysis server obtains the coordinate information of the visible images according to the shooting time to visible images.
2. the hidden danger target analysis method according to claim 1 based on Aerial Images, it is characterised in that identification contains tower
After the target image of bar, in addition to：
Target image containing shaft tower is marked and stored respectively with not containing the image of tower bar.
3. the hidden danger target analysis method according to claim 1 based on Aerial Images, it is characterised in that identification contains tower
The target image of bar, including：
Color Segmentation is carried out to the visible images；
Straight-line detection is carried out to the visible images after Color Segmentation, shaft tower edge line is detected；
Parallel lines judgement is carried out to the straight line detected；
Judge that the straightway that there are parallel lines in the range of presetted pixel is used as the part at shaft tower edge；
Contain the target image of tower bar according to shaft tower limb recognition.
4. the hidden danger target analysis method according to claim 1 based on Aerial Images, it is characterised in that to described visible
Light image carries out Color Segmentation, including：
Green area is obtained using threshold method in visible images；
Green area in the visible images is filtered out.
5. the hidden danger target analysis method according to claim 1 based on Aerial Images, it is characterised in that the identification contains
Before the target image for having tower bar, in addition to：
When obtaining visible images by taking photo by plane, while obtaining GPS information；
According to the GPS information and the shooting time information of the visible images, the coordinate information of visible images is obtained；
According to the coordinate information, selection sets the visible images in geographic area, and recognizes the target wherein containing tower bar
Image.
6. the hidden danger target analysis method according to claim 1 based on Aerial Images, it is characterised in that described in insulation
Hidden danger target is detected in the region of the target image above substring, including：
Row threshold division is entered to the region of the target image above insulator chain；
Calculated by connected domain, and the syntople of connected domain and shaft tower recognizes hidden danger target.
7. a kind of hidden danger targeted assay systems based on Aerial Images, it is characterised in that including：
Visible Light Camera, for passing through acquisition visible images of taking photo by plane in inspection；
Tower bar optical sieving module, in the visible images, recognizing the target image containing tower bar；
Insulator chain detection module, for detecting insulator chain in the target image containing tower bar；
Hidden danger module of target detection, for detecting hidden danger target in the region of the target image above insulator chain；
Hidden danger target analysis module, for carrying out insulator chain flashover impact analysis to the hidden danger target detected,
Wherein, in addition to；
Memory cell, is arranged in inspection device, for storing GPS information and visible images；
Transmission unit, it is default for being transferred to the GPS information and visible images by wired or wireless transmission method
On Analysis server；
Analysis server, the coordinate information for obtaining the visible images according to the shooting time to visible images.
8. the hidden danger targeted assay systems according to claim 7 based on Aerial Images, it is characterised in that the tower bar figure
As screening module, including：
Color Segmentation unit, for carrying out Color Segmentation to the visible images；
Straight-line detection unit, for carrying out straight-line detection to the visible images after Color Segmentation, detects shaft tower edge line；
Parallel lines judging unit, for carrying out parallel lines judgement to the straight line detected；
Shaft tower edge judgment unit, for the straightway that judges to there are parallel lines in the range of presetted pixel as shaft tower edge
A part；
Tower bar recognition unit, the target image for containing tower bar according to shaft tower limb recognition.
9. the hidden danger targeted assay systems according to claim 8 based on Aerial Images, it is characterised in that described colored point
Unit is cut, including：
Region recognition component, for obtaining green area using threshold method in visible images；
Region filters out component, for the green area in the visible images to be filtered out.
10. the hidden danger targeted assay systems according to claim 7 based on Aerial Images, it is characterised in that also include：
GPS module, for when obtaining visible images by taking photo by plane, while obtaining GPS information；
Coordinate obtaining module, for the shooting time information according to the GPS information and the visible images, obtains visible ray
The coordinate information of image；
Image selection module, for according to the coordinate information, selecting the visible images in setting geographic area, and by described
Tower bar optical sieving module recognizes the target image wherein containing tower bar.
11. the hidden danger targeted assay systems according to claim 7 based on Aerial Images, it is characterised in that the hidden danger
Module of target detection, including：
Threshold segmentation unit, row threshold division is entered for the region to the target image above insulator chain；
Hidden danger object-recognition unit, for being calculated by connected domain, and the syntople of connected domain and shaft tower recognizes hidden danger mesh
Mark.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510753533.1A CN105426824B (en) | 2015-11-06 | 2015-11-06 | Hidden danger target analysis method and system based on Aerial Images |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510753533.1A CN105426824B (en) | 2015-11-06 | 2015-11-06 | Hidden danger target analysis method and system based on Aerial Images |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105426824A CN105426824A (en) | 2016-03-23 |
| CN105426824B true CN105426824B (en) | 2017-10-10 |
Family
ID=55505024
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510753533.1A Active CN105426824B (en) | 2015-11-06 | 2015-11-06 | Hidden danger target analysis method and system based on Aerial Images |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105426824B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110008845B (en) * | 2019-03-12 | 2020-01-24 | 北京市燃气集团有限责任公司 | Method and device for detecting hidden danger points of gas pipeline |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2399878A (en) * | 2003-02-25 | 2004-09-29 | Schlumberger Holdings | Testing downhole equipment |
| CN103530620A (en) * | 2013-10-30 | 2014-01-22 | 华北电力大学 | Method for identifying bird nest on electric transmission line tower |
| CN103745224A (en) * | 2013-12-24 | 2014-04-23 | 西南交通大学 | Image-based railway contact net bird-nest abnormal condition detection method |
| CN104483326A (en) * | 2014-12-19 | 2015-04-01 | 长春工程学院 | High-voltage wire insulator defect detection method and high-voltage wire insulator defect detection system based on deep belief network |
| CN105023014A (en) * | 2015-08-21 | 2015-11-04 | 马鞍山市安工大工业技术研究院有限公司 | Method for extracting tower target in unmanned aerial vehicle routing inspection power transmission line image |
-
2015
- 2015-11-06 CN CN201510753533.1A patent/CN105426824B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2399878A (en) * | 2003-02-25 | 2004-09-29 | Schlumberger Holdings | Testing downhole equipment |
| CN103530620A (en) * | 2013-10-30 | 2014-01-22 | 华北电力大学 | Method for identifying bird nest on electric transmission line tower |
| CN103745224A (en) * | 2013-12-24 | 2014-04-23 | 西南交通大学 | Image-based railway contact net bird-nest abnormal condition detection method |
| CN104483326A (en) * | 2014-12-19 | 2015-04-01 | 长春工程学院 | High-voltage wire insulator defect detection method and high-voltage wire insulator defect detection system based on deep belief network |
| CN105023014A (en) * | 2015-08-21 | 2015-11-04 | 马鞍山市安工大工业技术研究院有限公司 | Method for extracting tower target in unmanned aerial vehicle routing inspection power transmission line image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105426824A (en) | 2016-03-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105389555A (en) | Aerial-image-based hidden-danger target analysis method and system | |
| CN111342391B (en) | Power transmission line insulator and line fault inspection method and inspection system | |
| CN103996036B (en) | A kind of map data collecting method and device | |
| CN105527969B (en) | A kind of mountain garden belt investigation and monitoring method based on unmanned plane | |
| BR112018009108B1 (en) | METHOD FOR ACQUISITION AND ANALYSIS OF AERIAL IMAGES | |
| CN108881825A (en) | Rice weed monitoring unmanned system and its monitoring method based on Jetson TK1 | |
| CN112990262B (en) | Integrated solution system for monitoring and intelligent decision of grassland ecological data | |
| CN107527037A (en) | Blue-green algae identification and analysis system based on unmanned aerial vehicle remote sensing data | |
| CN104881865A (en) | Forest disease and pest monitoring and early warning method and system based on unmanned plane image analysis | |
| US9098745B2 (en) | Sampling position-fixing system | |
| CN108648169A (en) | The method and device of high voltage power transmission tower defects of insulator automatic identification | |
| US11475550B2 (en) | Information processing apparatus, information processing method, program, and information processing system | |
| CN106646137A (en) | Method of detecting defect of power transmission line, defect detecting device, and defect detecting system | |
| CN112101088A (en) | Automatic unmanned aerial vehicle power inspection method, device and system | |
| CN106570853A (en) | Shape and color integration insulator identification and defect detection method | |
| CN109300118B (en) | High-voltage power line unmanned aerial vehicle inspection method based on RGB image | |
| CN108680833A (en) | Composite insulator defect detecting system based on unmanned plane | |
| CN106326949A (en) | Bridge engineering component location, identification and inspection system based on RFID | |
| CN108334815A (en) | Method for inspecting, on off state recognition methods and the system of second power equipment | |
| CN106989783A (en) | A kind of UAS that analysis is monitored based on soil information | |
| CN105426824B (en) | Hidden danger target analysis method and system based on Aerial Images | |
| CN112802027A (en) | Target object analysis method, storage medium and electronic device | |
| CN106320173B (en) | Vehicle-mounted unmanned aerial vehicle bridge routine safety detection system and detection method | |
| CN102621419A (en) | Method for automatically recognizing and monitoring line electrical equipment based on laser and binocular vision image | |
| CN109712018A (en) | A kind of power transmission line comprehensive monitoring system |
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 |