CN110516551A - A kind of line walking positional shift identifying system, method and the unmanned plane of view-based access control model - Google Patents
A kind of line walking positional shift identifying system, method and the unmanned plane of view-based access control model Download PDFInfo
- Publication number
- CN110516551A CN110516551A CN201910686677.8A CN201910686677A CN110516551A CN 110516551 A CN110516551 A CN 110516551A CN 201910686677 A CN201910686677 A CN 201910686677A CN 110516551 A CN110516551 A CN 110516551A
- Authority
- CN
- China
- Prior art keywords
- power line
- line
- unmanned plane
- positional shift
- identification
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/0088—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/12—Target-seeking control
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Abstract
The present invention provides a kind of line walking positional shift identifying systems of view-based access control model, comprising: feature extraction network, power line identification module, electric tower and component identification module；Various pieces cooperate, and execute current location identification in real time in unmanned plane line walking, and issue signal when unmanned plane deviates line walking target.Feature extraction network is responsible for extracting the features at different levels of image, is used for subsequent identification work.Power line identification module only needs class label in training, greatly reduces the workload of sample mark.Electric tower and component identification module are based on algorithm of target detection, start to identify unit type herein when power line identification module recognizes power line change in shape, determine whether unmanned plane current location deviates according to the size and location of identification frame.A kind of recognition methods and unmanned plane are provided simultaneously.The present invention realizes the position identification to patrol UAV, and issues signal in positional shift, can be used for the work such as auxiliary operation or unmanned plane automatic running on transmisson line.
Description
Technical field
The present invention relates to electric system and air vehicle technique field, specifically it is a kind of for Utilities Electric Co. nobody
Line walking positional shift identifying system, method and the unmanned plane of the view-based access control model of machine line walking.
Background technique
Electric system overhead transmission line scale is big, surrounding enviroment are complicated, climate change multiterminal.For the safety for guaranteeing electric system
Even running, the generation prevented accident need to carry out regular visit.In view of complicated meteorology and geographical conditions, harsh inspection
It is not only time-consuming and laborious to only rely on manual inspection for environment, but also tour density is lower, is difficult to meet the requirements.Unmanned plane has manipulation
Simply, the advantages that reaction speed is sensitive, flight is flexible, cruise duration is long, low in cost, has become that Utilities Electric Co. is important to patrol
Inspection tool.Transmission line of electricity inspection is carried out using unmanned plane, the efficiency and quality of transmission line of electricity O&M is not only increased, can also reduce
Labor intensity and cost ensure the personal safety of walking operation personnel.
Unmanned plane relies on staff's remote control operation at present, not only has higher requirements to the operation level of operator, and
And since the electromagnetic interference near power circuit is more serious.It is possible that the failures such as fluctuation of service, or even operation failure,
Cause line walking equipment, the even damage of transmission line of electricity.If can assign unmanned plane certain autonomous judgement, deviating
Line walking position, or can will be expected to promote the stabilization of line walking work from main regulation self-position when passing by close apart from power transmission line
Property and safety.The program also can be used as a part of the following unmanned plane automatic running on transmisson line technology simultaneously.
Recognizer based on deep learning generally requires largely have the sample of mark for model training, and insulator is prevented
The device areas such as shake hammer are larger, and shape is also relatively stable, relatively easy to mark.And the mark of transmission line of electricity is then different from insulator
The mark of equal devices, the characteristics of according to transmission line of electricity shape and size, the mark for generally requiring Pixel-level just can satisfy requirement.
This proposes higher requirement to mark personnel, also substantially increases the workload of mark.If recognizer can be improved,
The requirement marked to sample is reduced, then can greatly reduce mark cost, the offset of unmanned plane line walking position can also be promoted
The application of recognizer.
Currently without the explanation or report for finding technology similar to the present invention, it is also not yet collected into money similar both at home and abroad
Material.
Summary of the invention
Aiming at the above shortcomings existing in the prior art, the object of the present invention is to provide a kind of line walking positions of view-based access control model
Offset identifying system, method and unmanned plane, the identifying system, method and unmanned plane determine unmanned plane in conjunction with strong and weak supervised learning
Current location, and signal is issued when positional shift occurs for unmanned plane, realize the identification of unmanned plane line walking positional shift.
The present invention is achieved by the following technical solutions.
According to an aspect of the invention, there is provided a kind of line walking positional shift identifying system of view-based access control model, including spy
Sign extracts network, power line identification module and electric tower and component identification module；Wherein:
The feature extraction network, the image acquired when inputting unmanned plane line walking, extracts the characteristic informations at different levels of image, uses
In the identification work of power line identification module and electric tower and component identification module；
The power line identification module, according to the characteristic informations at different levels that feature extraction network extracts, in input picture
Power line is identified；When power line is not present or power line interrupts, electric tower and component identification module are switched to；
The electric tower and component identification module identify that the size and location of frame determine nobody according to non-electrical line of force component
Whether machine current location deviates.
Preferably, the feature extraction network uses convolutional neural networks.
Preferably, the power line identification module is by the way of Weakly supervised study, by whether containing power line this
Class label is trained module.
In the present invention, the Weakly supervised mode refers to: do not needed in marker samples it is very detailed must mark, further
Ground, if marking the detail location of power line in picture, to supervise by force, if only marking in the figure whether contain power line, for
It is Weakly supervised, it is unsupervised if not marking.
Preferably, the method for the training are as follows: give a label to every picture in training set, label is to contain electricity
The line of force does not contain power line, then picture and label is inputted convolutional neural networks together and is trained, the resulting volume of training
Product neural network, which has, to be categorized an image as containing power line or without the ability of power line.
Preferably, the power line identification module carries out the process of power line identification are as follows:
S1 classifies to each sub-regions of input picture by the way of sliding window, if being classified as unregulated power
Line region, the region zero setting, if being classified as executing S2 there are power line region and calculating the provincial characteristics figure；
S2, comprehensive characteristics extract the characteristic informations at different levels that network extracts, by the result and fixed-size Gauss after synthesis
Nuclear phase is multiplied to arrive final characteristic pattern；
According to whole input picture of the procedure ergodic of S1 and S2, that is, it may recognize that the position of power line in input picture；
Wherein:
The sliding window, Gaussian kernel and characteristic pattern size are all the same.
Preferably, the S2 specifically:
If F1 to F5 be characterized extract network extract characteristic informations at different levels, by F5 do bilinear interpolation rise sampling after with F4
In conjunction with obtaining median M4；By M4 in conjunction with F3, median M3 is obtained；By M3 in conjunction with F2, median M2 is obtained；By M2 with
F1 is combined, and obtains median M1；M1 is multiplied with Gaussian kernel, obtains final characteristic pattern M0；
Wherein, the combination refers to that two figures make matrix multiplication.
Preferably, the electric tower and component identification module are being detected based on deep learning using algorithm of target detection
It to after non-electrical line of force component, identifies the variety of components and position, obtains the size and location of identification frame, compare and be previously set
Threshold value determines whether unmanned plane current location is reasonable, and issues signal in positional shift.
According to the second aspect of the invention, a kind of line walking positional shift recognition methods of view-based access control model is provided, including
Following steps:
Feature extraction: the image acquired when according to unmanned plane line walking extracts the characteristic informations at different levels of image, is used for power line
Identification and the identification of electric tower and component；
Power line identification: according to the characteristic informations at different levels of extraction, the power line in input picture is identified；Work as electric power
When line is not present or power line interrupts, electric tower and component identification are switched to；
Electric tower and component identification: identify that the size and location of frame determine unmanned plane present bit according to non-electrical line of force component
It sets and whether deviates.
Preferably, the process of the power line identification are as follows:
S1 classifies to each sub-regions of input picture by the way of sliding window, if being classified as unregulated power
Line region, the region zero setting, if being classified as executing S2 there are power line region and calculating the provincial characteristics figure；
S2, comprehensive characteristics extract the characteristic informations at different levels that network extracts, by the result and fixed-size Gauss after synthesis
Nuclear phase is multiplied to final characteristic pattern: setting F1 to F5 and is characterized the characteristic informations at different levels for extracting network extraction, F5 is done bilinearity
After interpolation liter sampling in conjunction with F4, median M4 is obtained；By M4 in conjunction with F3, median M3 is obtained；By M3 in conjunction with F2, obtain
Median M2；By M2 in conjunction with F1, median M1 is obtained；M1 is multiplied with Gaussian kernel, obtains final characteristic pattern M0；The knot
Conjunction refers to that two figures make matrix multiplication；
According to whole input picture of the procedure ergodic of S1 and S2, that is, it may recognize that the position of power line in input picture；
Wherein:
The sliding window, Gaussian kernel and characteristic pattern size are all the same.
Preferably, the electric tower and the process of component identification are as follows: after detecting non-electrical line of force component, identification should
Variety of components and position obtain the size and location of identification frame, compare the threshold value being previously set, determine that unmanned plane current location is
It is no reasonable, and signal is issued in positional shift.
According to the third aspect of the present invention, a kind of unmanned plane is provided, the unmanned plane is equipped with any of the above-described institute
The line walking positional shift identifying system for the view-based access control model stated.
Compared with prior art, the invention has the following beneficial effects:
Line walking positional shift identifying system, method and the unmanned plane of view-based access control model provided by the invention, reduce to training
The mark requirement of sample, facilitates the stability and safety that promote line walking work, while it is automatic also to can be used as the following unmanned plane
A part of line walking technology.
Detailed description of the invention
Upon reading the detailed description of non-limiting embodiments with reference to the following drawings, other feature of the invention,
Objects and advantages will become more apparent upon:
Fig. 1 is the structural schematic diagram of feature extraction network provided in the embodiment of the present invention；
Fig. 2 is the process schematic diagram that power line identification module provided in the embodiment of the present invention extracts characteristic pattern；
Fig. 3 is the unmanned plane line walking positional shift recognition methods flow chart of view-based access control model provided by the present invention.
Specific embodiment
Elaborate below to the embodiment of the present invention: the present embodiment carries out under the premise of the technical scheme of the present invention
Implement, the detailed implementation method and specific operation process are given.It should be pointed out that those skilled in the art
For, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to guarantor of the invention
Protect range.
The embodiment of the invention provides a kind of line walking positional shift identifying systems of view-based access control model, comprising: feature extraction net
Network, power line identification module and electric tower and component identification module.Wherein:
The feature extraction network is mainly made of a convolutional neural networks, and the convolutional neural networks extract image
Features at different levels are used for subsequent identification work.The structure of feature extraction network is as shown in Figure 1.
The power line identification module, is trained module using Weakly supervised learning method, and mentioned using preceding feature
The extracted features at different levels of network are taken to identify the power line in image.In CNN (convolutional neural networks) architecture,
High-level characteristic pattern includes more information relevant to prediction, but resolution ratio is lower.Although low level characteristic pattern has compared with high score
Resolution, but it include it is less to the relevant information of prediction.The module combines the characteristic pattern of intermediate convolutional layer, obtains and prediction phase
The high resolution information of pass, and these information are used for the positioning of power line.To input picture using sliding window to each
Subregion is classified, and sliding step is the half of window size.If being classified as the region of unregulated power line, the region zero setting,
To reduce the interference of background.If being classified as the region there are power line, its characteristic pattern M0 is calculated.Fig. 2 is that module extracts feature
Scheme the process of M0.F1 therein to F5 is outputs at different levels, due to the relationship of pond layer, it is therefore desirable to which F5 is done bilinear interpolation liter
Can just be combined with F4 after sampling, in conjunction with process make matrix multiplication for two figures, obtain median M4.It is subsequent similarly, until
Obtain M1.To emphasize that central part responds, inhibit the response of fringe region, by M1 and the Gaussian kernel having a size of sliding window size
Be multiplied, obtain to the end with sliding window characteristic pattern M0 of the same size.According to whole image of this procedure ergodic, that is, it may recognize that figure
The position of power line as in.If being interrupted in image there is no power circuit or power circuit, switch to electric tower and
Component identification module is identified.
The electric tower and component identification module identify all kinds of devices on transmission line of electricity using algorithm of target detection,
The output result of identification is the type and size for the component for including in image.Threshold value can be previously set in the module, be greater than the upper limit or
Abnormal signal can be issued less than lower limit.Type of device and threshold size can adjust accordingly according to actual needs.
The embodiment of the present invention provides a kind of unmanned plane line walking positional shift identifying system of above-mentioned view-based access control model simultaneously
Recognition methods includes the following steps:
Feature extraction: the image acquired when according to unmanned plane line walking extracts the characteristic informations at different levels of image, is used for power line
Identification and the identification of electric tower and component；
Power line identification: according to the characteristic informations at different levels of extraction, the power line in input picture is identified；Work as electric power
When line is not present or power line interrupts, electric tower and component identification are switched to；
Electric tower and component identification: identify that the size and location of frame determine unmanned plane present bit according to non-electrical line of force component
It sets and whether deviates.
The process of the power line identification are as follows:
S1 classifies to each sub-regions of input picture by the way of sliding window, if being classified as unregulated power
Line region, the region zero setting, if being classified as executing S2 there are power line region and calculating the provincial characteristics figure；
S2, comprehensive characteristics extract the characteristic informations at different levels that network extracts, by the result and fixed-size Gauss after synthesis
Nuclear phase is multiplied to arrive final characteristic pattern.
The S2 specifically: set F1 to F5 and be characterized the characteristic informations at different levels for extracting network extraction, F5 is done into bilinearity and is inserted
After value liter sampling in conjunction with F4, median M4 is obtained；By M4 in conjunction with F3, median M3 is obtained；By M3 in conjunction with F2, obtain
Between value M2；By M2 in conjunction with F1, median M1 is obtained；M1 is multiplied with Gaussian kernel, obtains final characteristic pattern M0；The combination
Refer to that two figures make matrix multiplication.
According to whole input picture of the procedure ergodic of S1 and S2, that is, it may recognize that the position of power line in input picture.
Wherein:
The sliding window, Gaussian kernel and characteristic pattern size are all the same.
The electric tower and the process of component identification are as follows: after detecting non-electrical line of force component, identify the component kind
Class and position obtain the size and location of identification frame, compare the threshold value being previously set, determine whether unmanned plane current location closes
Reason, and signal is issued in positional shift.
Unmanned plane line walking positional shift identifying system and recognition methods, the embodiment of the present invention based on the above view-based access control model are same
When provide a kind of unmanned plane, the unmanned plane is equipped with the line walking positional shift identification of view-based access control model described in any of the above embodiments
System and/or the line walking positional shift recognition methods for executing view-based access control model described in any of the above embodiments.
Below with reference to a specific application example, the technical solution in the above embodiment of the present invention is described in further detail.
It comprising power line and does not include each 1000 of power line, use using the power line data collection for having class label
In training power line identification module.Using the transmission line of electricity device data collection with position and class label, insulator class and tower
Frame class each 300 are used as training set, training electric tower and component identification module.
Power line identification module using 1024 × 1024 fixed sizes input, if the non-size of input picture to input into
Row size change over.Sliding window is set as 128 × 128, and step-length is 64 pixels, M1 and the Gauss having a size of 128 × 128 anyhow
Nuclear phase is multiplied to characteristic pattern M0 for identification.Line walking photo size is 4000 × 3000 pixels, electric tower and component identification
The identification frame size threshold value upper and lower limit that module sets wherein insulator is respectively 3200 × 800,1200 × 300.The identification of pylon
Frame size threshold value upper and lower limit is respectively 2000 × 3000,1200 × 2000.
When using unmanned plane line walking, start to identify after unmanned plane reaches designated position.Power line is known when flight along the line
Other module real-time perfoming detection, when detecting that power line is not present or power line interrupts, switches to electric tower and portion
Part identification module determines whether unmanned plane current location is reasonable according to recognition result.If unreasonable, return current location it is excessively close or
Signal too far.
Line walking positional shift identifying system, method and the unmanned plane for the view-based access control model that the above embodiment of the present invention provides,
Middle identifying system includes: feature extraction network, power line identification module, electric tower and component identification module；Various pieces phase
Mutually cooperation executes the work of current location identification in real time in unmanned plane line walking, and issues when unmanned plane deviates line walking target
Signal.Feature extraction network is made of a convolutional neural networks, is responsible for extracting the features at different levels of image, is used for subsequent identification
Work.Power line identification module only needs class label in training, is not necessarily to location tags, greatly reduces the work of sample mark
Amount.Electric tower and component identification module are based on algorithm of target detection, recognize electric power wire shaped in power line identification module and become
Start to detect when change, identify unit type herein, whether determines unmanned plane current location according to the size and location of identification frame
Rationally.Line walking positional shift identifying system, method and the unmanned plane for the view-based access control model that the above embodiment of the present invention provides, realize
Position identification to patrol UAV, and signal is issued in positional shift, it can be used for auxiliary operation or unmanned plane automatic running on transmisson line
Equal work.
Specific embodiments of the present invention are described above.It is to be appreciated that the invention is not limited to above-mentioned
Particular implementation, those skilled in the art can make various deformations or amendments within the scope of the claims, this not shadow
Ring substantive content of the invention.
Claims (10)
1. a kind of line walking positional shift identifying system of view-based access control model, which is characterized in that know including feature extraction network, power line
Other module and electric tower and component identification module；Wherein:
The feature extraction network, the image acquired when inputting unmanned plane line walking, extracts the characteristic informations at different levels of image, for electricity
The identification work of line of force identification module and electric tower and component identification module；
The power line identification module, according to the characteristic informations at different levels that feature extraction network extracts, to the electric power in input picture
Line is identified；When power line is not present or power line interrupts, electric tower and component identification module are switched to；
The electric tower and component identification module identify that the size and location of frame determine that unmanned plane is worked as according to non-electrical line of force component
Whether front position deviates.
2. the line walking positional shift identifying system of view-based access control model according to claim 1, which is characterized in that the feature mentions
Take network using convolutional neural networks.
3. the line walking positional shift identifying system of view-based access control model according to claim 1, which is characterized in that the power line
Identification module is by the way of Weakly supervised study, by whether being trained containing this class label of power line to module.
4. the line walking positional shift identifying system of view-based access control model according to claim 1, which is characterized in that the power line
The process of identification module progress power line identification are as follows:
S1 classifies to each sub-regions of input picture by the way of sliding window, if being classified as unregulated power line area
Domain, the region zero setting, if being classified as executing S2 there are power line region and calculating the provincial characteristics figure；
S2, comprehensive characteristics extract network extract characteristic informations at different levels, by after synthesis result and fixed-size Gauss nuclear phase
It is multiplied to arrive final characteristic pattern；
According to whole input picture of the procedure ergodic of S1 and S2, that is, it may recognize that the position of power line in input picture；
Wherein:
The sliding window, Gaussian kernel and characteristic pattern size are all the same.
5. the line walking positional shift identifying system of view-based access control model according to claim 4, which is characterized in that the S2 is specific
Are as follows:
If F1 to F5, which is characterized, extracts the characteristic informations at different levels that network extracts, F5 is done and is tied after bilinear interpolation liter samples with F4
It closes, obtains median M4；By M4 in conjunction with F3, median M3 is obtained；By M3 in conjunction with F2, median M2 is obtained；By M2 and F1
In conjunction with obtaining median M1；M1 is multiplied with Gaussian kernel, obtains final characteristic pattern M0；
Wherein, the combination refers to that two figures make matrix multiplication.
6. the line walking positional shift identifying system of view-based access control model according to claim 1, which is characterized in that the power tower
Frame and component identification module are based on deep learning, and using algorithm of target detection, after detecting non-electrical line of force component, identification should
Variety of components and position obtain the size and location of identification frame, compare the threshold value being previously set, determine that unmanned plane current location is
It is no reasonable, and signal is issued in positional shift.
7. a kind of line walking positional shift recognition methods of view-based access control model, which comprises the steps of:
Feature extraction: the image acquired when according to unmanned plane line walking extracts the characteristic informations at different levels of image, identifies for power line
It is identified with electric tower and component；
Power line identification: according to the characteristic informations at different levels of extraction, the power line in input picture is identified；When power line not
In the presence of or power line interrupt when, switch to electric tower and component identification；
Electric tower and component identification: identify that the size and location of frame determine that unmanned plane current location is according to non-electrical line of force component
No offset.
8. the line walking positional shift recognition methods of view-based access control model according to claim 7, which is characterized in that the power line
The process of identification are as follows:
S1 classifies to each sub-regions of input picture by the way of sliding window, if being classified as unregulated power line area
Domain, the region zero setting, if being classified as executing S2 there are power line region and calculating the provincial characteristics figure；
S2, comprehensive characteristics extract network extract characteristic informations at different levels, by after synthesis result and fixed-size Gauss nuclear phase
It is multiplied to final characteristic pattern: set F1 to F5 be characterized extract network extract characteristic informations at different levels, F5 is done into bilinear interpolation
After liter sampling in conjunction with F4, median M4 is obtained；By M4 in conjunction with F3, median M3 is obtained；By M3 in conjunction with F2, centre is obtained
Value M2；By M2 in conjunction with F1, median M1 is obtained；M1 is multiplied with Gaussian kernel, obtains final characteristic pattern M0；The combination is
Refer to that two figures make matrix multiplication；
According to whole input picture of the procedure ergodic of S1 and S2, that is, it may recognize that the position of power line in input picture；
Wherein:
The sliding window, Gaussian kernel and characteristic pattern size are all the same.
9. the line walking positional shift recognition methods of view-based access control model according to claim 7, which is characterized in that the power tower
Frame and the process of component identification are as follows: after detecting non-electrical line of force component, identify the variety of components and position, obtain identification frame
Size and location, compare the threshold value being previously set, determine whether unmanned plane current location reasonable, and issue in positional shift
Signal.
10. a kind of unmanned plane, which is characterized in that the unmanned plane is equipped with described in any one of claims 1 to 6 based on view
The line walking positional shift identifying system of feel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910686677.8A CN110516551B (en) | 2019-07-29 | 2019-07-29 | Vision-based line patrol position deviation identification system and method and unmanned aerial vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910686677.8A CN110516551B (en) | 2019-07-29 | 2019-07-29 | Vision-based line patrol position deviation identification system and method and unmanned aerial vehicle |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110516551A true CN110516551A (en) | 2019-11-29 |
| CN110516551B CN110516551B (en) | 2023-04-07 |
Family
ID=68623650
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910686677.8A Active CN110516551B (en) | 2019-07-29 | 2019-07-29 | Vision-based line patrol position deviation identification system and method and unmanned aerial vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110516551B (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120016538A1 (en) * | 2010-06-21 | 2012-01-19 | Waite James W | Uav power line position and load parameter estimation |
| CN105959625A (en) * | 2016-05-04 | 2016-09-21 | 北京博瑞爱飞科技发展有限公司 | Method and device of controlling unmanned plane tracking shooting |
| CN106504233A (en) * | 2016-10-18 | 2017-03-15 | 国网山东省电力公司电力科学研究院 | Image electric power widget recognition methodss and system are patrolled and examined based on the unmanned plane of Faster R CNN |
| US20170371353A1 (en) * | 2016-06-23 | 2017-12-28 | Qualcomm Incorporated | Automatic Tracking Mode For Controlling An Unmanned Aerial Vehicle |
| CN107544541A (en) * | 2017-09-18 | 2018-01-05 | 南方科技大学 | A kind of unmanned aerial vehicle (UAV) control method and system |
| CN107729808A (en) * | 2017-09-08 | 2018-02-23 | 国网山东省电力公司电力科学研究院 | A kind of image intelligent acquisition system and method for power transmission line unmanned machine inspection |
| CN107833213A (en) * | 2017-11-02 | 2018-03-23 | 哈尔滨工业大学 | A kind of Weakly supervised object detecting method based on pseudo- true value adaptive method |
| CN108009515A (en) * | 2017-12-14 | 2018-05-08 | 杭州远鉴信息科技有限公司 | A kind of power transmission line positioning identifying method of the unmanned plane image based on FCN |
| CN108229587A (en) * | 2018-02-06 | 2018-06-29 | 贵州电网有限责任公司 | A kind of autonomous scan method of transmission tower based on aircraft floating state |
| CN108334844A (en) * | 2018-02-06 | 2018-07-27 | 贵州电网有限责任公司 | A kind of automatic tracking method along the line of polling transmission line |
| CN109255776A (en) * | 2018-07-23 | 2019-01-22 | 中国电力科学研究院有限公司 | A kind of transmission line of electricity split pin defect automatic identifying method |
| CN109471447A (en) * | 2018-12-14 | 2019-03-15 | 国网冀北电力有限公司检修分公司 | Navigation of Pilotless Aircraft method, apparatus, unmanned plane and data readable storage devices |
| CN109949361A (en) * | 2018-12-16 | 2019-06-28 | 内蒙古工业大学 | A kind of rotor wing unmanned aerial vehicle Attitude estimation method based on monocular vision positioning |
-
2019
- 2019-07-29 CN CN201910686677.8A patent/CN110516551B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120016538A1 (en) * | 2010-06-21 | 2012-01-19 | Waite James W | Uav power line position and load parameter estimation |
| CN105959625A (en) * | 2016-05-04 | 2016-09-21 | 北京博瑞爱飞科技发展有限公司 | Method and device of controlling unmanned plane tracking shooting |
| US20170371353A1 (en) * | 2016-06-23 | 2017-12-28 | Qualcomm Incorporated | Automatic Tracking Mode For Controlling An Unmanned Aerial Vehicle |
| CN106504233A (en) * | 2016-10-18 | 2017-03-15 | 国网山东省电力公司电力科学研究院 | Image electric power widget recognition methodss and system are patrolled and examined based on the unmanned plane of Faster R CNN |
| CN107729808A (en) * | 2017-09-08 | 2018-02-23 | 国网山东省电力公司电力科学研究院 | A kind of image intelligent acquisition system and method for power transmission line unmanned machine inspection |
| CN107544541A (en) * | 2017-09-18 | 2018-01-05 | 南方科技大学 | A kind of unmanned aerial vehicle (UAV) control method and system |
| CN107833213A (en) * | 2017-11-02 | 2018-03-23 | 哈尔滨工业大学 | A kind of Weakly supervised object detecting method based on pseudo- true value adaptive method |
| CN108009515A (en) * | 2017-12-14 | 2018-05-08 | 杭州远鉴信息科技有限公司 | A kind of power transmission line positioning identifying method of the unmanned plane image based on FCN |
| CN108229587A (en) * | 2018-02-06 | 2018-06-29 | 贵州电网有限责任公司 | A kind of autonomous scan method of transmission tower based on aircraft floating state |
| CN108334844A (en) * | 2018-02-06 | 2018-07-27 | 贵州电网有限责任公司 | A kind of automatic tracking method along the line of polling transmission line |
| CN109255776A (en) * | 2018-07-23 | 2019-01-22 | 中国电力科学研究院有限公司 | A kind of transmission line of electricity split pin defect automatic identifying method |
| CN109471447A (en) * | 2018-12-14 | 2019-03-15 | 国网冀北电力有限公司检修分公司 | Navigation of Pilotless Aircraft method, apparatus, unmanned plane and data readable storage devices |
| CN109949361A (en) * | 2018-12-16 | 2019-06-28 | 内蒙古工业大学 | A kind of rotor wing unmanned aerial vehicle Attitude estimation method based on monocular vision positioning |
Non-Patent Citations (2)
| Title |
|---|
| 史晋涛 等: "基于样本扩充的 Faster R-CNN 电网异物监测技术", 《电网技术》 * |
| 陈崟松 等: "基于视觉的无人机定点飞行控制技术研究", 《电子设计工程》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110516551B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109376605B (en) | Electric power inspection image bird-stab-prevention fault detection method | |
| CN103400156B (en) | Based on the High Resolution SAR image Ship Detection of CFAR and rarefaction representation | |
| CN112884931B (en) | Unmanned aerial vehicle inspection method and system for transformer substation | |
| CN105404867B (en) | A kind of substation isolating-switch state identification method of view-based access control model | |
| CN108596886A (en) | Aerial Images insulator based on deep learning falls piece fault rapid detecting method | |
| CN105023014A (en) | Method for extracting tower target in unmanned aerial vehicle routing inspection power transmission line image | |
| CN108648169A (en) | The method and device of high voltage power transmission tower defects of insulator automatic identification | |
| CN112818768B (en) | Transformer substation reconstruction and extension violation behavior intelligent identification method based on meta-learning | |
| CN110084165A (en) | The intelligent recognition and method for early warning of anomalous event under the open scene of power domain based on edge calculations | |
| CN105426905A (en) | Robot barrier identification method based on gradient histogram and support vector machine | |
| CN103177241B (en) | The method utilizing video image processing technology location power transmission line spacer | |
| CN101799968A (en) | Detection method and device for oil well intrusion based on video image intelligent analysis | |
| CN101908214B (en) | Moving object detection method with background reconstruction based on neighborhood correlation | |
| CN108009551A (en) | Suitable for the power knife switch division position state identification method of electric operating robot | |
| CN102609730A (en) | Method based on contact proportion of dynamic contact and static contact and used for distinguishing switching-on and switching-off reliability of disconnecting link | |
| CN114782626A (en) | Transformer substation scene mapping and positioning optimization method based on laser and vision fusion | |
| CN110298330A (en) | A kind of detection of transmission line polling robot monocular and localization method | |
| CN107016360A (en) | The object detection method that electricity substation is merged based on behavioral characteristics and region | |
| CN107240113B (en) | A kind of semi-automatic water body scope extracting method based on special sections line | |
| CN116245871A (en) | Power transmission line abnormal target detection method based on improved YOLOX algorithm | |
| CN113902792A (en) | Building height detection method and system based on improved RetinaNet network and electronic equipment | |
| CN109509169A (en) | The pressing plate condition checkout gear and method of learning distance metric based on nearest neighbour classification | |
| CN112542800A (en) | Method and system for identifying transmission line fault | |
| CN109977884B (en) | Target following method and device | |
| Liu et al. | A method for detecting power lines in UAV aerial images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |