CN108280449A - Power equipment image collecting method based on multispectral sensor group - Google Patents
Power equipment image collecting method based on multispectral sensor group Download PDFInfo
- Publication number
- CN108280449A CN108280449A CN201810119274.0A CN201810119274A CN108280449A CN 108280449 A CN108280449 A CN 108280449A CN 201810119274 A CN201810119274 A CN 201810119274A CN 108280449 A CN108280449 A CN 108280449A
- Authority
- CN
- China
- Prior art keywords
- image
- data
- information
- sensor
- equipment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/143—Sensing or illuminating at different wavelengths
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The present invention provides the power equipment image collecting methods based on multispectral sensor group, include the following steps：1）Image information acquires：Aggregation units are formed by Multiple Source Sensor, spectral information and field data are acquired by aggregation units, by data conversion and are merged the acquisition information progress data prediction of sensor；2）Information storage and management：Type judgement is carried out to the equipment image got by separate sources, and then builds image data storage organization；3）Information type is analyzed：Image analysis interface is used to provide the interface for providing scalability based on the operation of the general images such as text, content-based image retrieval, image rough sort for subsequent image processing operation；4）As a result it exports：The judgement and output of testing result are realized by data processor.The present invention can solve in the analysis of transformer equipment multispectral image, how efficiently utilize image data problem, ensure grid equipment safe and reliable operation, have higher theoretical and practical significance.
Description
Technical field
The present invention relates to power equipment visible light application of the graphic chart method and technology fields, more particularly, to based on multispectral sensing
The power equipment image collecting method of device group.
Background technology
With the fast development of China's economy, largely access etc. becomes power grid certainty for extra-high voltage, alternating current-direct current mixing, new energy
Development trend.Currently, the power grid coverage area in China is very big, the quantity of various transformer equipments is also huge, such as power transformation
Stand, the quantity such as conducting wire, shaft tower increase significantly, and have a very wide distribution, distance it is remote.All overhauled clearly by artificial
Impossible, the requirement simultaneously for the unified management monitoring of these transformer equipments also improves therewith.On this basis, unmanned value
It is also that its safe operation management proposes new requirement to keep substation and be developed rapidly.Therefore, for improve it is unattended or
The safety of few man on duty transformer station personnel and equipment needs the operating status and information hidden danger that monitor transformer equipment in real time.Mesh
Some preceding substations are mounted with video monitoring system, it can be achieved that field device monitoring, the movement of control remote camera, digital video
The functions such as video recording.But only function for monitoring does not have image identification function, lacks the automatic identification to transformer device of transformer substation and divides
Analyse function.Operator on duty is still relied on to go to observe and analyze the image of acquisition, to analyze the operating status of transformer equipment, system
Lack the automatic identification and analytic function to transformer equipment image.It is at all the analysis to substation's image of background complexity and
The research of the method for discrimination of transformer equipment operation information is not mature enough, and improves image analysis capabilities conscientiously, become there is an urgent need to
It solves the problems, such as.
The invention of Patent No. 201410332599.9 is related to multigroup in a kind of transformer oil based on spectrum analysis
Point gas-detecting device and method, including Wavelength tunable light source, modulation circuit, gas chamber, photodetector, at image information acquisition
Circuit and computer are managed, using time division multiplexing idea, the different narrow line of control light source output center wavelength in different time sections
Wide light, each wavelength correspond to a kind of specific absorption spectral peak of gas to be detected, are obtained by the spectral absorption of the corresponding gas of analysis
Its concentration value.The invention passes through detection of gas with multiple constituents in the transformer oil of simple structure and low cost based on spectrum analysis
Device realizes the fast and reliable detection of multicomponent gas in transformer oil using time division multiplexing idea.Solves gas with various light
Detection difficult problem caused by spectrum signal is overlapping realizes the synchronous detection of multiple gases, and measurement sensitivity is high, fast response time,
Compact-sized simple, system cost is cheap.
The disclosure of the invention of Patent No. 201510412958.6 is a kind of to be based on multispectral composite insulator detection method,
Including：Detection device is selected, under identical operating condition, visible light inspection is carried out using detection device to same composite insulator
Survey, infrared detection and ultraviolet detection, obtain the detection image of the composite insulator；It shines to the shelf depreciation of visible images
The corona point of point, the local hot spot of infrared image and ultraviolet image is compared；To same under identical operating condition
Visible images, infrared image and the ultraviolet image of the different composite insulator of the same base shaft tower of circuit are compared；For every
A composite insulator establishes multispectral Test database, according to the data of the detection of some cycles, is compared, looks for data
Go out existing data difference.The invention will be seen that light, infrared, ultraviolet three kinds of detection means organically combine, and have complementary advantages,
Live detection easily is carried out to insulator, can find composite insulator defect in time, convenient for carrying out the inspection of large area.
The invention of Patent No. 201110419576.8 provides a kind of method for detecting high optical spectrum of composite insulator, including following
Step：1）Composite insulator is imaged using hyperspectral imager, obtains the Hyperspectral imaging of composite insulator；2）To compound exhausted
The Hyperspectral imaging of edge is pre-processed, including geometric correction, radiant correction are to obtain more accurate spectral information；3）With
Specialty analysis platform carries out processing analysis, determines that the operating status of composite insulator, judgement result output/display analyse whether
Need replacing composite insulator.The invention method for detecting high optical spectrum of composite insulator can carry out contactless inspection to composite insulator
It surveys, is not required to worker Deng Ta；And it is short the time required to execute-in-place, when obtaining data with hyperspectral imager, it can obtain simultaneously more
The information of a composite insulator carries out processing analysis to multiple composite insulators simultaneously convenient for the later stage, reaches the mesh of batch detection
, meet the needs of China's composite insulator state-detection.
It, can be from by the multispectral detection and analysis such as visible light, infrared, ultraviolet and identification technology however, in practical applications
Fundamentally solve the problems, such as some present in current transformer equipment on-line monitoring：The operating parameter of important equipment needs prison in real time
It surveys, is difficult to meet requirement of real-time using manual patrol, sense of responsibility, working attitude and the mental status of floor manager seriously affect
The result of detection；The operating status of many high-tension apparatuses is difficult to be converted into electric signal, in signal conversion and transmission process easily by
Strong-electromagnetic field interferes and influences diagnosis；Equipment is detected still in the simple application level to thermograph using infrared,
It is not associated with equipment state, historical data is also difficult to memory scan；In addition, the gray scale that human eye is difficult to differentiate fine image becomes
Change, it is difficult to the degree of objective analysis transformer equipment surface defect.Therefore, multispectral image analysis and the utilization of identification technology can promote
Intelligence, automation into on-line monitoring system, improve the working efficiency of work of transformer substation personnel, obtain higher economic effect
Benefit, will be with larger practical value and application prospect.
Invention content
In view of this, in view of the deficiencies of the prior art, it is an object of the present invention to provide the electricity based on multispectral sensor group
Power equipment image collecting method can effectively avoid based on wrong report, failing to report phenomenon existing for single type collecting and detecting device, can be with
It solves in the analysis of transformer equipment multispectral image, how efficiently to utilize image data problem, ensure that grid equipment is safe and reliable
Operation has higher theoretical and practical significance.
In order to achieve the above objectives, the present invention uses following technical scheme：Power equipment shadow based on multispectral sensor group
As acquisition method, include the following steps：1）Image information acquires：Aggregation units are formed by Multiple Source Sensor, pass through aggregation units
Spectral information and field data are acquired, by data conversion and is merged the acquisition information progress data prediction of sensor；2）
Information storage and management：Type judgement is carried out to the equipment image got by separate sources, and then builds image data and deposits
Its data block is stored to corresponding data storage server, its associated description information is deposited in the form of metadata by storage structure
Store up meta data server；3）Information type is analyzed：Image analysis interface is used to be provided based on text for subsequent image processing operation
The operation of the general images such as sheet, content-based image retrieval, image rough sort provides the interface of scalability；4）As a result it exports：
The judgement and output of testing result are realized by data processor.
Further, the Multiple Source Sensor includes infrared thermal imagery sensor, ultraviolet imagery sensor, arc-detection sensing
Device, X-ray digital sensor.
Further, the spectral information includes infrared information, UV information, x-ray image information and visible images letter
Breath, the field data includes equipment GPS information and equipment nameplate information.
Further, the data prediction is simultaneously by infrared thermal imagery sensor, ultraviolet imagery sensor, arc-detection
Sensor, X-ray digital sensor the image collected signal are via optical fiber, by data collecting card and data processor into line number
According to acquisition, data are handled.
Further, the data processor includes the PC terminals for being equipped with client software, the client software packet
Labview softwares are included, for realizing output spectrum image information recognition result.
Further, described image analysis interface includes PLC Logic control modules, DSP digital signal processor and network
Port, described image analysis interface are connect with data processor.
Further, the image data storage management structure include client, describe image information name node and
Equipment image data is used for carrying out image data and name node by providing the file interface of a standard for client
Interactive access.
The beneficial effects of the invention are as follows：
The method of the present invention can be widely applied to representative power equipment visible light spectrum recognition research, including transformer, arrester,
The classification and identification of the equipment such as capacitive apparatus and disconnecting switch and research equipment cosmetic variation detection technique, including accumulated snow, ice
Icepro, foreign matter lodge etc., can be by studying the multi-level image based on deep neural network/target's feature-extraction method and being based on
The collection of illustrative plates sorting technique of global and local characteristics of image builds the multispectral detection acquisition technique frame of power equipment, establishes power grid
Equipment image collection standardized work system, the image of power equipment is acquired by using different types of sensor, then
Classifying type, efficient storage, setting general-purpose search strategy complete the sequence of operations such as image access, access, and carry out phase to image
How efficiently pass is handled, and ancillary equipment identification is detected with detection, can not only solve in the analysis of transformer equipment multispectral image,
Using image data problem, while new acquisition means also are provided with diagnostic data for transformer equipment detection detection, ensures power grid
Equipment safety reliability service has higher theoretical and practical significance.
Description of the drawings
Fig. 1 is to be based on multispectral sensor group power equipment image collection operation structure figure.
Fig. 2 is the Organization Chart of image data storage management structure of the present invention.
Specific implementation mode
The invention will be further described with reference to the accompanying drawings and examples.
As shown in Figure 1 and Figure 2, the present invention is mainly based upon video/image storage, classification, knows method for distinguishing, carries out current
All kinds of video/images analysis and research work of grid equipment, are standardized by the image datas such as infrared to grid equipment, ultraviolet
Acquisition and arrangement study the video/image storage based on data processing centre, analysis, calculate, retrieve key technology, and structure is suitable
For the technological frame of grid equipment image collection, in conjunction with grid equipment operating status, by unstructured datas such as video/images
It is converted, provides basic condition for further support grid equipment state evaluation and analysis, invention core is to build electricity
Net equipment image collection standardized work system and research image data storage and management technology.
The basis of the present invention is that structure grid equipment image collection standardized work system, equipment image Collecting operation are
It completes the equipment identification subsequently based on image, mark, the basis of the operations such as equipment detection detection.In order to adapt to grid equipment image
The features such as big data quantity, multi-source, builds grid equipment image collection standardized work system.The system is by using different type
Sensor acquire the image of power equipment, then classifying type, efficient storage, setting general-purpose search strategy, complete image and visit
It the sequence of operations such as asks, access, and relevant treatment is carried out to image, ancillary equipment identification is detected with detection.Body as shown in Figure 1
Architecture function module.
1）Image collection module based on multispectral sensor：The module is detected according to the equipment subsequently based on multi-source image
The concrete applications demands such as detection identification, shoot different types of equipment image data, by using IR thermal imaging inspection technology, purple
Outer imager corona and arc-detection and X-ray digital flat panel direct imaging technology, obtain including infrared, ultraviolet, x-ray image,
The specific manifestations form such as visible images, and obtain the information such as GPS and equipment nameplate；
2）The storage of equipment image data, management module：The module is mainly responsible for the equipment image to being got by separate sources
Type judgement is carried out, and then its data block is stored to corresponding data storage server, by its associated description information with first number
According to form storage to meta data server, which mainly uses the number based on image data storage management structured file system
According to storage strategy, complete paired data it is efficient access, it is general retrieval etc. operations；
3）Image analysis interface：Within the system, image analysis interface is provided based on text for subsequent image processing operation, is based on
The operation of the general images such as image retrieval, the image rough sort of content provides the interface of scalability.
4）Data result exports：Realize that the judgement and output of testing result, data processor include peace by data processor
PC terminals equipped with client software, the client software include Labview softwares, for realizing output spectrum image information
Recognition result.LabVIEW is similar to C and BASIC development environments in commercial Application environment, but other computer languages are all
Code is generated using text based language, and LabVIEW uses graphical author language G to write program, the journey of generation
Sequence is the form of block diagram, be particularly well suited to data processing of the present invention, exploitation measures or the software type of control system, be integrated with
Electrical engineer's rapid build is various using required all tools.
The present invention apply when another big innovative point be to study the application technology based on data storage management tactic pattern,
Data storage management structure is the fault-tolerant sexual system of a height, is convenient for large-scale commercial use.And data storage management knot
Structure is capable of providing the data access of high-throughput, be very suitable for large-scale equipment image real time transfer according to the present invention, point
The application scenarios of analysis.
Data storage management structure is also just specially image based on the image analysing computer of multispectral sensor group by the present invention
Change handle, image data storage management structured file system basic framework as shown in Fig. 2, its basic framework mainly by client,
The name node and equipment image data composition of image information are described.System provides the file interface of a standard for client,
For interacting access to image data and name node.Name node manages image data in entire Hadoop clusters
Storage management structural system NameSpace.Pass through the name node to image information in heartbeat mechanism equipment image data meeting period
Report the file data information of oneself.If it find that certain node delay machine, then remove the node, the number on the node is backed up again
According to on other nodes.Finally each node in Hadoop clusters is the node that data are calculated and stored, each
Working node is all an image data node, is responsible for actual data storage, oneself is reported to the name node of image information
Storage information.File as unit of the block of fixed size to be stored in image data storage management structure, default size
For 64MB.User uploads to very big image data in image data storage management structure, and file is divided into several
Then block is stored on different image data nodes with certain algorithm.
In practical applications, it when carrying out basic data acquisition, generally requires after being acquired to multispectral image information
It pre-processes, the pretreatment in the present invention refers to image classification and merges, that is, needs the infrared image of equipment, ultraviolet image, X
Ray image and visible images are merged, and various information are made full use of.Image co-registration is not simply to be superimposed, what it was generated
It is the image that new implication has more value informations, main thought is that have redundancy by multiple by using certain algorithm
The fusion of the source images of characteristic and complementary information becomes the new image of a width so that the image after fusion can maximally utilise mutually
Information is mended, redundancy is reduced, to obtain comprehensibility and higher clarity, the detection further to carry out electrical equipment provides
More effective information.
Multi-source image information can be divided into from level：Pixel-level, feature level and decision level, therefore according to residing for fusion treatment
Different phase, the processing of Multi-Source Image Data Fusion can usually carry out in three different levels：Pixel-level fusion, feature level melt
It closes, decision level fusion.Since power transmission and transforming equipment local environment is complicated, image collection precision is inadequate, Pixel-level fusion operation amount compared with
Greatly.And decision level fusion mainly focuses on subjective demand and ignores other element, is not suitable for sentencing for power equipment complex situation
It is disconnected.And power equipment morphological feature is apparent, poor morphology respectively has feature away from larger between different type equipment.So according to acquisition
Electrical equipment image the characteristics of, be mainly based upon in this project feature-based fusion carry out electrical equipment checkout and diagnosis.
Feature level image co-registration is the fusion treatment process of middle layer.Feature level image co-registration is utilized from each sensor map
The characteristic information extracted in the raw information of picture carries out comprehensive analysis and fusion treatment, not only may be used by feature level image co-registration
To increase the possibility for extracting characteristic information from image, it is also possible to obtain some useful compound characteristics.It is characterized in from image
Pixel Information in be abstracted and extract.It is so-called to be mainly characterized by by carrying out space or temporal segmentation to image data
What equal processing obtained, and compound characteristics are obtained by the synthesis to existing each feature.It is extracted from image and for melting
The characteristic feature information of conjunction has edge, angle point, texture, similar luminance region, similar depth of field region etc..When in certain circumstances
Specific region in, when multi-sensor image all has similar feature, illustrate that the possibility of these feature physical presence is very big,
The accuracy of detection of this feature is also greatly improved simultaneously.The feature obtained after fusion treatment may be the comprehensive of various characteristics of image
It closes, if the edge after merging is the synthesis for the edge section that different sensors detect, it is also possible to a kind of feature of completely new type.
A kind of geometry of feature, direction, position and time range etc. are all that this feature and other features are matched
The importance of accurate fusion.In feature-based fusion image, the requirement to image registration is not as good as Pixel-level fusion is to registration
It is required that it is so stringent, therefore each imaging sensor can be distributed in different platform.The set transform of feature can be used for image spy
The registration of sign.The precision that feature detection can be increased by the image co-registration of feature level, can be with using the compound characteristics after fusion
Improve detection performance.
When carrying out fusion treatment, the concrete form and content of main feature information of concern melt with multi-sensor image
The application purpose and occasion of conjunction are closely related, such as are examined merging to be detected with other sensors using infrared imaging sensor
When disconnected, the difference for finding out target and background is first had to, clarification of objective is extracted, followed by each feature is compared,
Selection, fusion, it is final to carry out decision and identifying processing.In the detection detection of target, the extraction of target signature is crucial.For
The target signature of extraction mainly has target temperature and target gray distribution characteristics, target shape feature, Target Motion Character, target
Statistical Distribution Characteristics, image sequence characteristic and variation characteristic etc..Meanwhile during identification is embodied, it is necessary to pay attention to
Point target（Distant object, target is small, and pixel is few）And close-in target（It can be imaged, object pixel is more）The linking of processing and
Identification problem.
Feature level image co-registration can be divided into target status information fusion and target property fusion.Feature level target status information
Fusion is mainly used in multi-sensor target tracking field, and fusion treatment mainly realizes parameter correlation and state vector estimation.
The fusion of feature level target property is exactly the joint objective identification of characteristic layer, and the phase of pattern-recognition is still used in fusion method
Pass technology only must carry out relevant treatment to feature before fusion treatment, characteristic vector is classified and integrated.More attach most importance to
It wants, in fields such as pattern-recognition, image procossing and computer visions, people gather feature extraction and feature based
Class problem conducts in-depth research, and can be used for reference there are many method.In fusion process, blur method, nerve net are commonly used
The tools such as network technology, clustering method and Kalman filter.
Finally illustrate, the above examples are only used to illustrate the technical scheme of the present invention and are not limiting, this field is common
Other modifications or equivalent replacement that technical staff makes technical scheme of the present invention, without departing from technical solution of the present invention
Spirit and scope, be intended to be within the scope of the claims of the invention.
Claims (7)
1. the power equipment image collecting method based on multispectral sensor group, which is characterized in that include the following steps：1）Image
Information collection：Aggregation units are formed by Multiple Source Sensor, spectral information and field data are acquired by aggregation units, by data
It converts and merges the acquisition information progress data prediction of sensor；2）Information storage and management：To being obtained by separate sources
The equipment image got carries out type judgement, and then builds image data storage organization, by its data block storage to corresponding number
According to storage server, its associated description information is stored in the form of metadata and arrives meta data server；3）Information type is analyzed：
Image analysis interface is used to be provided based on text, content-based image retrieval, image rough sort for subsequent image processing operation
Equal general images operation provides the interface of scalability；4）As a result it exports：By data processor realize testing result judgement and
Output.
2. the power equipment image collecting method based on multispectral sensor group as described in claim 1, it is characterised in that：Institute
It includes infrared thermal imagery sensor, ultraviolet imagery sensor, arc-detection sensor, X-ray digital sensor to state Multiple Source Sensor.
3. the power equipment image collecting method based on multispectral sensor group as described in claim 1, it is characterised in that：Institute
It includes infrared information, UV information, x-ray image information and visible light image information to state spectral information, and the field data includes
Equipment GPS information and equipment nameplate information.
4. the power equipment image collecting method based on multispectral sensor group as described in claim 1, it is characterised in that：Institute
It is simultaneously to pass infrared thermal imagery sensor, ultraviolet imagery sensor, arc-detection sensor, X-ray number to state data prediction
Sensor the image collected signal carries out data acquisition via optical fiber, by data collecting card and data processor, is carried out to data
Processing.
5. the power equipment image collecting method based on multispectral sensor group as claimed in claim 4, it is characterised in that：Institute
It includes the PC terminals for being equipped with client software to state data processor, and the client software includes Labview softwares, is used for real
Existing output spectrum image information recognition result.
6. the power equipment image collecting method based on multispectral sensor group as claimed in claim 5, it is characterised in that：Institute
It includes PLC Logic control modules, DSP digital signal processor and the network port to state image analysis interface, and described image analysis connects
Mouth is connect with data processor.
7. the power equipment image collecting method based on multispectral sensor group as described in claim 1, it is characterised in that：Institute
State image data storage management structure include client, describe image information name node and equipment image data, by for
Client provides the file interface of a standard, for interacting access to image data and name node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810119274.0A CN108280449A (en) | 2018-02-06 | 2018-02-06 | Power equipment image collecting method based on multispectral sensor group |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810119274.0A CN108280449A (en) | 2018-02-06 | 2018-02-06 | Power equipment image collecting method based on multispectral sensor group |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108280449A true CN108280449A (en) | 2018-07-13 |
Family
ID=62807694
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810119274.0A Pending CN108280449A (en) | 2018-02-06 | 2018-02-06 | Power equipment image collecting method based on multispectral sensor group |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108280449A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110110765A (en) * | 2019-04-23 | 2019-08-09 | 四川九洲电器集团有限责任公司 | A kind of multisource data fusion target identification method based on deep learning |
| CN111207838A (en) * | 2020-03-11 | 2020-05-29 | 中南大学 | Molten iron temperature measuring device based on special infrared spectrum wave band |
| CN112445240A (en) * | 2020-11-26 | 2021-03-05 | 广东电网有限责任公司 | Automatic line patrol method and automatic line patrol unmanned aerial vehicle |
| CN113449767A (en) * | 2021-04-29 | 2021-09-28 | 国网浙江省电力有限公司嘉兴供电公司 | Multi-image fusion transformer substation equipment abnormity identification and positioning method |
| CN114414051A (en) * | 2022-03-06 | 2022-04-29 | 浙江大学 | Staring type hyperspectral three-light conjugate acquisition and fusion imaging system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102169017A (en) * | 2010-12-25 | 2011-08-31 | 江西九江供电公司 | Online monitoring system for infrared thermal imaging of converting station |
| CN102646200A (en) * | 2012-03-08 | 2012-08-22 | 武汉大学 | Image classifying method and system for self-adaption weight fusion of multiple classifiers |
| CN103323460A (en) * | 2013-06-03 | 2013-09-25 | 深圳供电局有限公司 | Insulator detection method and device based on visible light image |
| WO2014183259A1 (en) * | 2013-05-14 | 2014-11-20 | 中国科学院自动化研究所 | Full-color and multi-spectral remote sensing image fusion method |
| CN104216989A (en) * | 2014-09-09 | 2014-12-17 | 广东电网公司中山供电局 | Method for storing transmission line integrated data based on HBase |
| CN106596579A (en) * | 2016-11-15 | 2017-04-26 | 同济大学 | Insulator contamination condition detection method based on multispectral image information fusion |
| CN107368490A (en) * | 2016-05-12 | 2017-11-21 | 中国移动通信集团河北有限公司 | Data processing method and device |
-
2018
- 2018-02-06 CN CN201810119274.0A patent/CN108280449A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102169017A (en) * | 2010-12-25 | 2011-08-31 | 江西九江供电公司 | Online monitoring system for infrared thermal imaging of converting station |
| CN102646200A (en) * | 2012-03-08 | 2012-08-22 | 武汉大学 | Image classifying method and system for self-adaption weight fusion of multiple classifiers |
| WO2014183259A1 (en) * | 2013-05-14 | 2014-11-20 | 中国科学院自动化研究所 | Full-color and multi-spectral remote sensing image fusion method |
| CN103323460A (en) * | 2013-06-03 | 2013-09-25 | 深圳供电局有限公司 | Insulator detection method and device based on visible light image |
| CN104216989A (en) * | 2014-09-09 | 2014-12-17 | 广东电网公司中山供电局 | Method for storing transmission line integrated data based on HBase |
| CN107368490A (en) * | 2016-05-12 | 2017-11-21 | 中国移动通信集团河北有限公司 | Data processing method and device |
| CN106596579A (en) * | 2016-11-15 | 2017-04-26 | 同济大学 | Insulator contamination condition detection method based on multispectral image information fusion |
Non-Patent Citations (3)
| Title |
|---|
| ZHOU Y ET AL.: "Application research of the multifractal wavelet model in pattern recogition", 《PROCEEDING OF SPIE-THE INTERNATIONAL SOCIETY FOR OPTICAL ENGINEERING》 * |
| 孙艳凤: "基于云平台的风电组故障诊断数据处理技术研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
| 罗俊: "遥感图像融合方法及其在目标波谱反演中的应用研究", 《中国优秀博硕士学位论文全文数据库（硕士） 信息科技辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110110765A (en) * | 2019-04-23 | 2019-08-09 | 四川九洲电器集团有限责任公司 | A kind of multisource data fusion target identification method based on deep learning |
| CN111207838A (en) * | 2020-03-11 | 2020-05-29 | 中南大学 | Molten iron temperature measuring device based on special infrared spectrum wave band |
| CN112445240A (en) * | 2020-11-26 | 2021-03-05 | 广东电网有限责任公司 | Automatic line patrol method and automatic line patrol unmanned aerial vehicle |
| CN113449767A (en) * | 2021-04-29 | 2021-09-28 | 国网浙江省电力有限公司嘉兴供电公司 | Multi-image fusion transformer substation equipment abnormity identification and positioning method |
| CN113449767B (en) * | 2021-04-29 | 2022-05-17 | 国网浙江省电力有限公司嘉兴供电公司 | Multi-image fusion transformer substation equipment abnormity identification and positioning method |
| CN114414051A (en) * | 2022-03-06 | 2022-04-29 | 浙江大学 | Staring type hyperspectral three-light conjugate acquisition and fusion imaging system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108280449A (en) | Power equipment image collecting method based on multispectral sensor group | |
| Davari et al. | Intelligent diagnosis of incipient fault in power distribution lines based on corona detection in UV-visible videos | |
| CN110598736B (en) | Power equipment infrared image fault positioning, identifying and predicting method | |
| CN108037133B (en) | Intelligent electric power equipment defect identification method and system based on unmanned aerial vehicle inspection image | |
| CN112734692B (en) | Defect identification method and device for power transformation equipment | |
| CN108389137A (en) | Power fault detection early warning system based on infared spectrum technology | |
| CN102157023B (en) | Information system for self-help declaring and supervising of entry/exit passengers | |
| CN102811343B (en) | Intelligent video monitoring system based on behavior recognition | |
| CN102565625B (en) | Method for intelligently diagnosing thermal defects of high-voltage transmission line based on infrared image | |
| CN109242439A (en) | Feature extraction recognition methods based on substation equipment associated data | |
| CN108344931B (en) | Power equipment safety analysis system based on ultraviolet spectrum technology | |
| CN103247039B (en) | A kind of high voltage cable live detection method based on composite vision | |
| CN113920097B (en) | Power equipment state detection method and system based on multi-source image | |
| Najafi et al. | Fault diagnosis of electrical equipment through thermal imaging and interpretable machine learning applied on a newly-introduced dataset | |
| CN109034272A (en) | A kind of transmission line of electricity heat generating components automatic identifying method | |
| CN102915432A (en) | Method and device for extracting vehicle-bone microcomputer image video data | |
| CN106548211A (en) | Transformer station's point for measuring temperature automatic identifying method and device based on infrared image | |
| Manninen et al. | Multi-stage deep learning networks for automated assessment of electricity transmission infrastructure using fly-by images | |
| CN202815869U (en) | Vehicle microcomputer image and video data extraction apparatus | |
| Mohan et al. | An ensemble deep learning approach for air quality estimation in Delhi, India | |
| CN112734637B (en) | Thermal infrared image processing method and system for monitoring temperature of lead | |
| CN115880472A (en) | Intelligent diagnosis and analysis system for electric power infrared image data | |
| CN108491888A (en) | Environmental monitoring hyperspectral data spectrum section selection method based on morphological analysis | |
| Zou et al. | Research progresses and trends of power line extraction based on machine learning | |
| Xia et al. | A multi-target detection based framework for defect analysis of electrical equipment |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180713 |