CN109063764A - A kind of judgment method of disconnecting switch closing operation in place based on machine vision - Google Patents
A kind of judgment method of disconnecting switch closing operation in place based on machine vision Download PDFInfo
- Publication number
- CN109063764A CN109063764A CN201810834617.1A CN201810834617A CN109063764A CN 109063764 A CN109063764 A CN 109063764A CN 201810834617 A CN201810834617 A CN 201810834617A CN 109063764 A CN109063764 A CN 109063764A
- Authority
- CN
- China
- Prior art keywords
- disconnecting link
- edge line
- disconnecting
- image
- angle
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
Abstract
The judgment method of disconnecting switch closing operation in place based on machine vision that the present invention relates to a kind of.Deep learning is combined with classical machine vision method, realizes that the accurate positioning of disconnecting link and disconnecting link angle are analyzed in real time.Disconnecting link image is trained first with deep learning, using trained model realization disconnecting link Primary Location, and obtains disconnecting link state；Shape, size and the surrounding objects feature for extracting disconnecting link image, further determine that disconnecting link position；Part-area edge line where extracting disconnecting link according to disconnecting link position, obtains all parallel edge line set；Edge line direction and foreground image direction are calculated, Clutter edge line is rejected, obtains disconnecting link edge line set；Angle is opened and closed using edge line computation disconnecting link, realizes the real-time judge of disconnecting link folding condition.Method processing speed of the invention is fast, and detection accuracy is high, practical application value with higher.
Description
Technical field
The judgment method of disconnecting switch closing operation in place based on machine vision that the present invention relates to a kind of.
Background technique
Disconnecting switch closing operation is the important component in sequence operation, and sequence operation is also known as procedure operation, it is
Refer in one or more grid switching operation tasks, equipment state does not need artificial judgement confirmation item by item when operation, is by scheduling
System or monitoring system of electric substation carry out a series of operation of consecutives automatically.It is anti-error with putting into operation for intelligent substation
System is using the comprehensive integrated five-preventing system from integration and applies the intelligent protection of plate mode with pressure, it can be achieved that second protection
Pressing plate is remotely controlled operation from system backstage comprehensive, this lays a good foundation for the application of sequence.The confirmation whether disconnecting link closes in place is
Realize a key sequence basis, disconnecting link, which is opened one's mouth, to be closed not will cause overheat or edge of a knife melting welding sternly, is easy to cause great accident, be threatened
The safety of power system device and staff.Therefore whether disconnecting link opening and closing is required in place it is very high, but at present it is most of
Substation is still to judge whether disconnecting link closes in place using artificial observation mode, and in low-angle, it is easy to appear accidentally for this
Difference, while artificial observation wastes a large amount of human resources, there is also some potential safety problemss, therefore, current artificial observation
It is not able to satisfy the demand of substation's practical operation.
At present due to the particularity and diversity of transformer substation switch equipment, the domestic research for the direction is gone back substantially at present
It is to belong to blank stage, current main research method is as follows:
1, existing research only positions disconnecting link by simple straight-line detection, edge analysis and SHAPE DETECTION, such algorithm
Affected by environment larger, stability is poor；
2, existing research only for part there is the disconnecting link of text information or specific shape disconnecting link to be identified, solution has not yet been formed
The certainly universal solution of various types of switches equipment state identification；
3, existing method stability is poor, and anti-interference ability is not strong, and there is biggish false detection rates.
Summary of the invention
The judgment method of disconnecting switch closing operation in place based on machine vision that the purpose of the present invention is to provide a kind of,
This method can be realized the accurate positioning of disconnecting link, and can overcome the influence of external environment, realize the real-time of disconnecting link folding condition
Detection and analysis.
To achieve the above object, the technical scheme is that a kind of disconnecting switch closing operation based on machine vision
Judgment method in place, includes the following steps:
Step S1, it first with deep learning algorithm, is trained by collection in worksite great amount of samples, obtains training pattern；
Step S2, it is based on deep learning model, realizes the Primary Location and state judgement of disconnecting link, state is broadly divided into disconnecting link herein
Arm contact and disconnecting link arm separate two states；
Step S3, disconnecting link characteristics of image is extracted, shape, size and surrounding objects feature including disconnecting link further determine that knife
Gate position；
Step S4, realize that disconnecting link is accurately positioned using step S2 and step S3；
Step S5, disconnecting link position coordinates are obtained according to disconnecting link location information, and extracts the edge line of disconnecting link place regional area；
Step S6, in disconnecting link motion process, the angle between disconnecting link two-arm edge line is calculated, is classified according to angle and deep learning
As a result disconnecting link state is judged；
Step S7, the real-time status for intuitively observing disconnecting link for the convenience of the user can be realized aobvious in disconnecting link video in the detection process
Show disconnecting link edge line, disconnecting link angle and folding condition.
In an embodiment of the present invention, the method for training pattern is obtained in the step S1 are as follows:
As training sample, main includes without angle, different shapes for S11, the image first under each state of collection in worksite disconnecting link
The disconnecting link image of state, different shape and varying environment, to guarantee that the model that training obtains has stronger adaptability；
S12, disconnecting link image is marked according to the different conditions of disconnecting link；
S13, model training is carried out using Faster-RCNN method, obtains training pattern；
S14, model training is carried out using SSD method, obtains training pattern.
In an embodiment of the present invention, disconnecting link characteristics of image is extracted in the step S3, shape, size including disconnecting link,
Color and surrounding objects feature, the method for further determining that disconnecting link position are as follows:
S31, the Primary Location position for obtaining disconnecting link according to depth model positioning result first, then regional area in the position
It is interior that edge analysis is carried out to disconnecting link image, obtain disconnecting link profile；
S32, significant position is occupied due to disconnecting link in the picture, it is possible to which rejecting according to known disconnecting link dimension information can not
For the profile of disconnecting link；
Object features near S33, detection disconnecting link regional area, further reject interference disconnecting link position；
S34, determining disconnecting link position is obtained based on step S32 and step S33.
In an embodiment of the present invention, the method for disconnecting link edge line is extracted in the step S5 are as follows:
S51, first by disconnecting link image procossing be gray level image, and to gray level image carry out Gaussian Blur to reduce picture noise
Interference；
The gradient value of each pixel and direction in image after S52, calculating noise reduction, carry out the gradient value of each pixel non-
Maximum inhibits, and obtains image border point set；
S53, the edge array that image is obtained using dual threshold method, and carry out edge connection；
S54, all edge line angles are calculated, obtains all parallel edge line set；
S55, all parallel edge line set of traversal, reject Clutter edge line, obtain disconnecting link edge line set:
(1) background and foreground image direction are calculated；
(2) edge line direction is calculated；
(3) Clutter edge line is rejected according to foreground image direction, obtains disconnecting link edge line set；
S56, all disconnecting link edge lines are obtained.
In an embodiment of the present invention, the angle between disconnecting link two-arm edge line is calculated in the step S6, and according to angle
The method that disconnecting link state is judged are as follows:
S61, it is closed by reaching in disconnecting link or by closing in open procedure, is obtained the edge line of disconnecting link two-arm；
Angle between S62, calculating two-arm edge line；
S63, disconnecting link state is divided into three kinds, determines that disconnecting link is conjunction state, disconnecting link edge line when disconnecting link edge line angle is less than 3 degree
Angle is to determine that disconnecting link is empty conjunction state when 3-15 is spent, and disconnecting link edge line angle determines that disconnecting link is open state when being greater than 15 degree；
S64, according to disconnecting link two-arm angle real-time display disconnecting link open/close status.
Compared to the prior art, the invention has the following advantages:
(1) deep learning algorithm and classical learning algorithm are combined, realizes the accurate positioning of disconnecting link；In general,
Deep learning algorithm is used alone to obtain preferable locating effect, needs a large amount of training sample, and is used alone
Classical machine vision method, algorithm can be by external interferences, and stability is not good enough, and by deep learning algorithm and classical learning algorithm
It can combine to improve algorithm stability, while reducing training sample number, improve algorithm operation efficiency；
(2) using background image and foreground image directional information and disconnecting link edge line directional information, the external world can effectively be rejected
Edge line interference, realizes the accurate detection of disconnecting link edge line；
(3) speed of service is fast, and detection accuracy is high.
Detailed description of the invention
Fig. 1 is the method for the present invention flow chart.
Fig. 2 is the opposite opened disconnecting link with " conjunction " state.
Fig. 3 is the opposite opened disconnecting link with " void is closed " state.
Fig. 4 is the opposite opened disconnecting link with "On" state.
Specific embodiment
With reference to the accompanying drawing, technical solution of the present invention is specifically described.
As shown in Figure 1, the present invention provides a kind of judgement side of disconnecting switch closing operation in place based on machine vision
Method includes the following steps:
Step S1, it first with deep learning algorithm, is trained by collection in worksite great amount of samples, obtains training pattern；
Step S2, it is based on deep learning model, realizes the Primary Location and state judgement of disconnecting link, state is broadly divided into disconnecting link herein
Arm contact and disconnecting link arm separate two states；
Step S3, disconnecting link characteristics of image is extracted, shape, size and surrounding objects feature including disconnecting link further determine that knife
Gate position；
Step S4, realize that disconnecting link is accurately positioned using step S2 and step S3；
Step S5, disconnecting link position coordinates are obtained according to disconnecting link location information, and extracts the edge line of disconnecting link place regional area；
Step S6, in disconnecting link motion process, the angle between disconnecting link two-arm edge line is calculated, is classified according to angle and deep learning
As a result disconnecting link state is judged；
Step S7, the real-time status for intuitively observing disconnecting link for the convenience of the user can be realized aobvious in disconnecting link video in the detection process
Show disconnecting link edge line, disconnecting link angle and folding condition.
The method of training pattern is obtained in the step S1 are as follows:
As training sample, main includes without angle, different shapes for S11, the image first under each state of collection in worksite disconnecting link
The disconnecting link image of state, different shape and varying environment, to guarantee that the model that training obtains has stronger adaptability；
S12, disconnecting link image is marked according to the different conditions of disconnecting link；
S13, model training is carried out using Faster-RCNN method, obtains training pattern；
S14, model training is carried out using SSD method, obtains training pattern.
Disconnecting link characteristics of image is extracted in the step S3, shape, size, color and surrounding objects including disconnecting link are special
Sign, the method for further determining that disconnecting link position are as follows:
S31, the Primary Location position for obtaining disconnecting link according to depth model positioning result first, then regional area in the position
It is interior that edge analysis is carried out to disconnecting link image, obtain disconnecting link profile；
S32, significant position is occupied due to disconnecting link in the picture, it is possible to which rejecting according to known disconnecting link dimension information can not
For the profile of disconnecting link；
Object features near S33, detection disconnecting link regional area, further reject interference disconnecting link position；
S34, determining disconnecting link position is obtained based on step S32 and step S33.
The method of disconnecting link edge line is extracted in the step S5 are as follows:
S51, first by disconnecting link image procossing be gray level image, and to gray level image carry out Gaussian Blur to reduce picture noise
Interference；
The gradient value of each pixel and direction in image after S52, calculating noise reduction, carry out the gradient value of each pixel non-
Maximum inhibits, and obtains image border point set；
S53, the edge array that image is obtained using dual threshold method, and carry out edge connection；
S54, all edge line angles are calculated, obtains all parallel edge line set；
S55, all parallel edge line set of traversal, reject Clutter edge line, obtain disconnecting link edge line set:
(1) background and foreground image direction are calculated；
(2) edge line direction is calculated；
(3) Clutter edge line is rejected according to foreground image direction, obtains disconnecting link edge line set；
S56, all disconnecting link edge lines are obtained.
In an embodiment of the present invention, the angle between disconnecting link two-arm edge line is calculated in the step S6, and according to angle
The method that disconnecting link state is judged are as follows:
S61, it is closed by reaching in disconnecting link or by closing in open procedure, is obtained the edge line of disconnecting link two-arm；
Angle between S62, calculating two-arm edge line；
S63, disconnecting link state is divided into three kinds, determines that disconnecting link is conjunction state, disconnecting link edge line when disconnecting link edge line angle is less than 3 degree
Angle is to determine that disconnecting link is empty conjunction state when 3-15 is spent, and disconnecting link edge line angle determines that disconnecting link is open state when being greater than 15 degree；
S64, according to disconnecting link two-arm angle real-time display disconnecting link open/close status.
The following are specific implementation processes of the invention.
The opposite opened disconnecting link of one kind " conjunction " state as shown in Figure 2, a kind of disconnecting switch combined floodgate behaviour based on machine vision
Make judgment method in place, the calculating of accurate positioning and disconnecting link folding angle including disconnecting link can be realized disconnecting link folding condition
Real-time detection and analysis；
Further, above-mentioned disconnecting link, which is accurately positioned, first combines deep learning algorithm and classical machine vision method, gram
The accurate positioning of disconnecting link is realized in the interference for taking outside environmental elements.The disconnecting link stated is accurately positioned first with based on caffe frame
Ssd method carry out model training, obtain disconnecting link Primary Location result；Disconnecting link feature is extracted again, utilizes the size of disconnecting link, shape
Etc. information further determine that disconnecting link position, overcome the interference of outside environmental elements, realize the accurate positioning of disconnecting link.
Further, a large amount of training samples of collection in worksite first utilize the ssd method based on caffe frame to carry out model
Training obtains training pattern；
Further, the deep learning model obtained based on ssd method realizes the Primary Location and state judgement of disconnecting link, herein
State is broadly divided into disconnecting link arm contact and disconnecting link arm separates two states；
Further, disconnecting link characteristics of image is extracted, shape, size and surrounding objects feature including disconnecting link further determine that
Disconnecting link position, disconnecting link occupies significant position in the picture in practical application, has larger size, and extracting after disconnecting link feature can be compared with
It is fast to obtain disconnecting link position；
Further, disconnecting link position coordinates are obtained according to disconnecting link location information, and extracts disconnecting link place using canny detection algorithm
The edge line of regional area；
Further, in disconnecting link motion process, the angle between disconnecting link two-arm edge line is calculated, according to angle and deep learning point
Class result judges disconnecting link state；
Further, the real-time status for intuitively observing disconnecting link for the convenience of the user, can realize in disconnecting link video in the detection process
Show disconnecting link edge line, disconnecting link angle and folding condition.
The method that the present invention realizes the Primary Location of disconnecting link, the image first under each state of collection in worksite disconnecting link is as instruction
Practice sample, mainly include the disconnecting link image for not having to angle, different conditions, different shape and varying environment, to guarantee that training obtains
Model have stronger adaptability；Further, disconnecting link image is marked according to the different conditions of disconnecting link, is then utilized
Ssd method based on depth frame caffe carries out model training, and the preliminary fixed of disconnecting link may be implemented using trained model
Position.
The present invention extracts disconnecting link characteristics of image, and shape, size and surrounding objects information including disconnecting link further determine that
Disconnecting link position, it is characterised in that: the Primary Location position of disconnecting link is obtained first, to disconnecting link in regional area then in the position
Image carries out edge analysis, obtains disconnecting link profile；It further, can not be disconnecting link according to the rejecting of known disconnecting link dimension information
Profile；Detect the object information near disconnecting link regional area, the position of the disconnecting link further determined that.
The present invention extracts disconnecting link edge line, is first gray level image by disconnecting link image procossing, and carries out to gray level image high
This fuzzy interference to reduce picture noise；Further, the gradient value of each pixel and side in the image after calculating noise reduction
To carrying out non-maxima suppression to the gradient value of each pixel, obtain image border point set；Further, using dual threashold
Value method obtains the edge array of image, and carries out edge connection；Further, all edge line angles are calculated, are obtained all
Parallel edge line set；Further, all parallel edge line set are traversed, Clutter edge line is rejected, obtain disconnecting link edge line
Set.
The present invention judges disconnecting link state, in disconnecting link by reaching conjunction or by closing in open procedure, obtains disconnecting link two-arm
Edge line；Further, the angle between two-arm edge line is calculated；Further, disconnecting link state is divided into three kinds, such as Fig. 2-4
It is shown, determine that disconnecting link is " conjunction " state when disconnecting link edge line angle is less than 3 degree, disconnecting link edge line angle determines knife when spending for 3-15
Lock is " void is closed " state, and disconnecting link edge line angle determines that disconnecting link is "On" state when being greater than 15 degree；Further, according to disconnecting link two
Arm angle real-time display disconnecting link open/close status.
The above are preferred embodiments of the present invention, all any changes made according to the technical solution of the present invention, and generated function is made
When with range without departing from technical solution of the present invention, all belong to the scope of protection of the present invention.
Claims (5)
1. a kind of judgment method of disconnecting switch closing operation in place based on machine vision, which is characterized in that including walking as follows
It is rapid:
Step S1, it first with deep learning algorithm, is trained by collection in worksite great amount of samples, obtains training pattern；
Step S2, it is based on deep learning model, realizes the Primary Location and state judgement of disconnecting link, state is broadly divided into disconnecting link herein
Arm contact and disconnecting link arm separate two states；
Step S3, disconnecting link characteristics of image is extracted, shape, size and surrounding objects feature including disconnecting link further determine that knife
Gate position；
Step S4, realize that disconnecting link is accurately positioned using step S2 and step S3；
Step S5, disconnecting link position coordinates are obtained according to disconnecting link location information, and extracts the edge line of disconnecting link place regional area；
Step S6, in disconnecting link motion process, the angle between disconnecting link two-arm edge line is calculated, is classified according to angle and deep learning
As a result disconnecting link state is judged；
Step S7, the real-time status for intuitively observing disconnecting link for the convenience of the user can be realized aobvious in disconnecting link video in the detection process
Show disconnecting link edge line, disconnecting link angle and folding condition.
2. a kind of judgment method of disconnecting switch closing operation in place based on machine vision according to claim 1,
It is characterized in that, the method for training pattern is obtained in the step S1 are as follows:
As training sample, main includes without angle, different shapes for S11, the image first under each state of collection in worksite disconnecting link
The disconnecting link image of state, different shape and varying environment, to guarantee that the model that training obtains has stronger adaptability；
S12, disconnecting link image is marked according to the different conditions of disconnecting link；
S13, model training is carried out using Faster-RCNN method, obtains training pattern；
S14, model training is carried out using SSD method, obtains training pattern.
3. a kind of judgment method of disconnecting switch closing operation in place based on machine vision according to claim 1,
It is characterized in that, disconnecting link characteristics of image is extracted in the step S3, shape, size, color and surrounding objects including disconnecting link are special
Sign, the method for further determining that disconnecting link position are as follows:
S31, the Primary Location position for obtaining disconnecting link according to depth model positioning result first, then regional area in the position
It is interior that edge analysis is carried out to disconnecting link image, obtain disconnecting link profile；
S32, significant position is occupied due to disconnecting link in the picture, it is possible to which rejecting according to known disconnecting link dimension information can not
For the profile of disconnecting link；
Object features near S33, detection disconnecting link regional area, further reject interference disconnecting link position；
S34, determining disconnecting link position is obtained based on step S32 and step S33.
4. a kind of judgment method of disconnecting switch closing operation in place based on machine vision according to claim 1,
It is characterized in that, the method for disconnecting link edge line is extracted in the step S5 are as follows:
S51, first by disconnecting link image procossing be gray level image, and to gray level image carry out Gaussian Blur to reduce picture noise
Interference；
The gradient value of each pixel and direction in image after S52, calculating noise reduction, carry out the gradient value of each pixel non-
Maximum inhibits, and obtains image border point set；
S53, the edge array that image is obtained using dual threshold method, and carry out edge connection；
S54, all edge line angles are calculated, obtains all parallel edge line set；
S55, all parallel edge line set of traversal, reject Clutter edge line, obtain disconnecting link edge line set:
(1) background and foreground image direction are calculated；
(2) edge line direction is calculated；
(3) Clutter edge line is rejected according to foreground image direction, obtains disconnecting link edge line set；
S56, all disconnecting link edge lines are obtained.
5. a kind of judgment method of disconnecting switch closing operation in place based on machine vision according to claim 1,
It is characterized in that, the angle between disconnecting link two-arm edge line is calculated in the step S6, and judge disconnecting link state according to angle
Method are as follows:
S61, it is closed by reaching in disconnecting link or by closing in open procedure, is obtained the edge line of disconnecting link two-arm；
Angle between S62, calculating two-arm edge line；
S63, disconnecting link state is divided into three kinds, determines that disconnecting link is conjunction state, disconnecting link edge line when disconnecting link edge line angle is less than 3 degree
Angle is to determine that disconnecting link is empty conjunction state when 3-15 is spent, and disconnecting link edge line angle determines that disconnecting link is open state when being greater than 15 degree；
S64, according to disconnecting link two-arm angle real-time display disconnecting link open/close status.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810834617.1A CN109063764A (en) | 2018-07-26 | 2018-07-26 | A kind of judgment method of disconnecting switch closing operation in place based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810834617.1A CN109063764A (en) | 2018-07-26 | 2018-07-26 | A kind of judgment method of disconnecting switch closing operation in place based on machine vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109063764A true CN109063764A (en) | 2018-12-21 |
Family
ID=64835697
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810834617.1A Pending CN109063764A (en) | 2018-07-26 | 2018-07-26 | A kind of judgment method of disconnecting switch closing operation in place based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109063764A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110348380A (en) * | 2019-07-10 | 2019-10-18 | 无锡地铁集团有限公司 | A kind of visualization grounding switch state video recognition system and method |
| CN111028225A (en) * | 2019-12-12 | 2020-04-17 | 深圳集智数字科技有限公司 | Method and device for monitoring opening degree of valve handle |
| CN111046943A (en) * | 2019-12-09 | 2020-04-21 | 国网智能科技股份有限公司 | Method and system for automatically identifying state of isolation switch of transformer substation |
| CN111126253A (en) * | 2019-12-23 | 2020-05-08 | 国网福建省电力有限公司 | Knife switch state detection method based on image recognition |
| CN111178395A (en) * | 2019-12-12 | 2020-05-19 | 平高集团有限公司 | Isolation switch state identification method and device |
| CN111191641A (en) * | 2019-09-21 | 2020-05-22 | 南京鑫和汇通电子科技有限公司 | Method for accurately detecting motion state of split type disconnecting link |
| CN111597868A (en) * | 2020-01-08 | 2020-08-28 | 浙江大学 | SSD-based substation disconnecting switch state analysis method |
| CN111723796A (en) * | 2019-03-20 | 2020-09-29 | 天津美腾科技有限公司 | Power distribution cabinet power-on and power-off state identification method and device based on machine vision |
| CN112036402A (en) * | 2020-08-26 | 2020-12-04 | 济南信通达电气科技有限公司 | Split type disconnecting link state identification method and device |
| CN112508019A (en) * | 2020-12-16 | 2021-03-16 | 国网江苏省电力有限公司检修分公司 | GIS isolation/grounding switch state detection method and system based on image recognition |
| CN112036402B (en) * | 2020-08-26 | 2024-04-26 | 济南信通达电气科技有限公司 | Split type disconnecting link state identification method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102509100A (en) * | 2011-10-26 | 2012-06-20 | 山东电力研究院 | Knife switch connecting-disconnecting reliability recognition method based on image pattern recognition |
| CN106326901A (en) * | 2016-08-30 | 2017-01-11 | 南京鑫和汇通电子科技有限公司 | Water stain image recognition based on edge point self-similarity and TEDS system |
| CN106650788A (en) * | 2016-11-16 | 2017-05-10 | 合肥普望电子有限责任公司 | Isolation switch fault detection method |
| CN107257161A (en) * | 2017-06-20 | 2017-10-17 | 安徽南瑞继远电网技术有限公司 | A kind of transformer station's disconnecting link remote control auxiliary check method and system based on state recognition algorithm |
| US20180120019A1 (en) * | 2016-11-02 | 2018-05-03 | Lg Electronics Inc. | Refrigerator |
-
2018
- 2018-07-26 CN CN201810834617.1A patent/CN109063764A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102509100A (en) * | 2011-10-26 | 2012-06-20 | 山东电力研究院 | Knife switch connecting-disconnecting reliability recognition method based on image pattern recognition |
| CN106326901A (en) * | 2016-08-30 | 2017-01-11 | 南京鑫和汇通电子科技有限公司 | Water stain image recognition based on edge point self-similarity and TEDS system |
| US20180120019A1 (en) * | 2016-11-02 | 2018-05-03 | Lg Electronics Inc. | Refrigerator |
| CN106650788A (en) * | 2016-11-16 | 2017-05-10 | 合肥普望电子有限责任公司 | Isolation switch fault detection method |
| CN107257161A (en) * | 2017-06-20 | 2017-10-17 | 安徽南瑞继远电网技术有限公司 | A kind of transformer station's disconnecting link remote control auxiliary check method and system based on state recognition algorithm |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111723796A (en) * | 2019-03-20 | 2020-09-29 | 天津美腾科技有限公司 | Power distribution cabinet power-on and power-off state identification method and device based on machine vision |
| CN110348380B (en) * | 2019-07-10 | 2024-01-23 | 无锡地铁集团有限公司 | Visual grounding disconnecting link state video identification system and method |
| CN110348380A (en) * | 2019-07-10 | 2019-10-18 | 无锡地铁集团有限公司 | A kind of visualization grounding switch state video recognition system and method |
| CN111191641B (en) * | 2019-09-21 | 2024-04-09 | 南京鑫和汇通电子科技有限公司 | Accurate detection method for motion state of split type disconnecting link |
| CN111191641A (en) * | 2019-09-21 | 2020-05-22 | 南京鑫和汇通电子科技有限公司 | Method for accurately detecting motion state of split type disconnecting link |
| CN111046943A (en) * | 2019-12-09 | 2020-04-21 | 国网智能科技股份有限公司 | Method and system for automatically identifying state of isolation switch of transformer substation |
| CN111178395B (en) * | 2019-12-12 | 2023-04-07 | 平高集团有限公司 | Isolation switch state identification method and device |
| CN111028225B (en) * | 2019-12-12 | 2020-09-01 | 深圳集智数字科技有限公司 | Method and device for monitoring opening degree of valve handle |
| CN111028225A (en) * | 2019-12-12 | 2020-04-17 | 深圳集智数字科技有限公司 | Method and device for monitoring opening degree of valve handle |
| CN111178395A (en) * | 2019-12-12 | 2020-05-19 | 平高集团有限公司 | Isolation switch state identification method and device |
| CN111126253A (en) * | 2019-12-23 | 2020-05-08 | 国网福建省电力有限公司 | Knife switch state detection method based on image recognition |
| CN111597868A (en) * | 2020-01-08 | 2020-08-28 | 浙江大学 | SSD-based substation disconnecting switch state analysis method |
| CN112036402A (en) * | 2020-08-26 | 2020-12-04 | 济南信通达电气科技有限公司 | Split type disconnecting link state identification method and device |
| CN112036402B (en) * | 2020-08-26 | 2024-04-26 | 济南信通达电气科技有限公司 | Split type disconnecting link state identification method and device |
| CN112508019A (en) * | 2020-12-16 | 2021-03-16 | 国网江苏省电力有限公司检修分公司 | GIS isolation/grounding switch state detection method and system based on image recognition |
| CN112508019B (en) * | 2020-12-16 | 2024-02-27 | 国网江苏省电力有限公司检修分公司 | GIS isolation/grounding switch state detection method and system based on image recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109063764A (en) | A kind of judgment method of disconnecting switch closing operation in place based on machine vision | |
| CN101389004B (en) | Moving target classification method based on on-line study | |
| CN104697768B (en) | A kind of moving contact in high voltage breaker kinetic characteristic method of testing | |
| CN105404867B (en) | A kind of substation isolating-switch state identification method of view-based access control model | |
| CN105868722A (en) | Identification method and system of abnormal power equipment images | |
| CN109670430A (en) | A kind of face vivo identification method of the multiple Classifiers Combination based on deep learning | |
| CN106339657B (en) | Crop straw burning monitoring method based on monitor video, device | |
| CN105512666A (en) | River garbage identification method based on videos | |
| CN106446869A (en) | Automatic detection method for state of isolating switch based on image intelligent recognition algorithm | |
| CN107977639A (en) | A kind of face definition judgment method | |
| CN109145756A (en) | Object detection method based on machine vision and deep learning | |
| CN109271844A (en) | Electrical cabinet electrical symbol recognition methods based on OpenCV | |
| CN113903081A (en) | Visual identification artificial intelligence alarm method and device for images of hydraulic power plant | |
| CN105894701A (en) | Large construction vehicle identification and alarm method for preventing external damage to transmission lines | |
| CN109117764A (en) | Using the method for color threshold method identification target object region electrical symbol in power monitoring | |
| CN107133592A (en) | Electricity substation is using infrared thermal imaging and the human body target feature detection algorithm of visual light imaging technological incorporation | |
| CN113052894B (en) | Door opening and closing state detection method and system based on image semantic segmentation | |
| CN106127812A (en) | A kind of passenger flow statistical method of non-gate area, passenger station based on video monitoring | |
| CN108335294A (en) | The power distribution room abnormality image-recognizing method of complex condition | |
| CN112183472A (en) | Method for detecting whether test field personnel wear work clothes or not based on improved RetinaNet | |
| CN107330914A (en) | Face position method for testing motion and device and vivo identification method and system | |
| CN102682291B (en) | A kind of scene demographic method, device and system | |
| CN105163076B (en) | The substation breakers state video intelligent parser mapped using transversal scanning | |
| CN107358151A (en) | A kind of eye motion detection method and device and vivo identification method and system | |
| CN112395900A (en) | Fatigue driving state detection algorithm based on YOLOv3 algorithm |
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 |
Application publication date: 20181221 |
|
| RJ01 | Rejection of invention patent application after publication |