CN109752627A - A method of the ca bin socket state automatic identification based on machine vision - Google Patents
A method of the ca bin socket state automatic identification based on machine vision Download PDFInfo
- Publication number
- CN109752627A CN109752627A CN201910053588.XA CN201910053588A CN109752627A CN 109752627 A CN109752627 A CN 109752627A CN 201910053588 A CN201910053588 A CN 201910053588A CN 109752627 A CN109752627 A CN 109752627A
- Authority
- CN
- China
- Prior art keywords
- image
- socket
- plug wire
- bin
- automatic 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.)
- Pending
Links
Abstract
The ca bin socket state automatic identification method based on machine vision that the invention discloses a kind of, the parameter raw data collection including establishing actuating and image recognition;Camera fixed-focus position is determined according to the detection result of metal probe, obtain the plug wire face image of ca bin and pre-processed, handled by gray processing, median filtering, obtain that feature is obvious and the image containing less noise;The profile and important feature of image are analyzed, fitting and identification marking position is made, actual position and angular deviation is calculated according to distance in figure;Data interaction is controlled by preset program, industrial robot is in place and starts to operate, while avoiding the exception during plug wire by the feedback control of damping induction.The present invention can be realized quick, accurate calculation to tap position and angular deviation, overcome socket feature it is unobvious and design it is lack of standardization caused by image procossing or the problem of feature recognition failures.
Description
Technical field
The invention belongs to applied technical field of the machine vision in automatic detection, more particularly to one kind to be based on machine vision
Ca bin socket state automatic identification method.
Background technique
Feed line automatization terminal equipment (FTU) is mounted in the intelligent terminal on switchgear house or feeder line, have telemetering,
Remote signalling, fault detection capability are able to detect the alternate of key node, ground fault etc. on route, are that be related to power distribution network reliable
The basic key equipment of operation and faulted-phase judgment.Functional performance detection is carried out to ca bin, to guarantee equipment
Stable operation, have deep meaning.In order to adapt to the full-automatic testing process of ca bin, detected in functional performance
It before implementation, needs to carry out FTU plug wire, power on the practical operation situation with analog machine.As detection device quantity increases, people
The disadvantages of easy fault of work plug wire operation, low efficiency, highlights, therefore researching and designing is suitable for the socket state automatic identification of batch jobs
System is the key that entire detection system realizes automation.Inventors have found that existing plug wire means are mostly artificial plug wire, automatically
Change the equipment that plug wire means are chiefly used in unified standard, but the standard ununified due to the plug wire face exterior structure of FTU, the operation
It is difficult to realize automate.Machine vision technique characterized by computer picture understands can adapt to different visual contexts, compared with
Solve the problems, such as that tap position is skimble-scamble well.
Machine vision has been applied to automatic detection field at present, but the research of the context of detection in Distribution Network Equipment is seldom.
Widely applied character machining and defects detection are limited for the reference of the more accurate socket identification of FTU.The boat of FTU
Outlet structure analysis complexity and plug wire efficiency contradiction, decide the image procossing of machine vision should have both rapidity and
The characteristics of real-time.
Summary of the invention
The ca bin socket state automatic identification method based on machine vision that the invention proposes a kind of, for sentencing
The specific location and socket angle of each socket in plug wire face of disconnected ca bin simultaneously carry out accurate plug wire operation.
To achieve the goals above, the present invention adopts the following technical scheme:
A kind of ca bin socket state based on machine vision disclosed in one or more embodiments from
Dynamic recognition methods, comprising:
(1) the location parameter raw data set of actuating and image recognition is established;
(2) camera fixed-focus position is determined according to the plug wire face depth of detection, obtains the plug wire face image of ca bin
And it is pre-processed;
(3) flag bit identification and contour fitting are carried out, position and the angu-lar deviation of socket are calculated separately;
(4) plug wire movement is executed according to intended flow, the exception during feedback control monitoring plug wire is damped according to plug wire
And it alarms.
Further, in the step (1), the position coordinates of the ca bin plug wire mouth of each supplier is obtained, are taken
The weighted mean of three-dimensional coordinate acquires socket image as Robotic Manipulator camera and executes the basic parameter of plug wire movement.
Further, the plug wire face image of ca bin is obtained in the step (2), specifically:
1. obtaining the plug wire face depth information of ca bin using metal probe, start plug wire process;
2. opening light source, camera lens are positioned with light source to the first spigot;
3. camera acquisition image is simultaneously preprocessed, enters in next step after confirmation image definition is errorless, otherwise examine again
Camera fixed-focus effect;
4. repeating 2. 3. to walk, until the image information collecting of all sockets finishes;
5. terminating Image Acquisition process, probe measurement result next time is waited.
Further, image is pre-processed specifically:
Gray processing processing is carried out to image, obtains the grayscale image of each socket;
Noise reduction process is carried out using the method for median filtering;
Distortion adjustment is carried out, camera and appliance outlet the relative position error are eliminated;
Camera calibration is carried out, determines that the positional relationship in image is corresponding with spatial relation in kind, acquirement is used for
The image that position deviation calculates;
Judge characteristics of image flag bit discrimination degree, chooses whether to handle using color enhancement;
Feature identifies selected flag bit, and information useless in contour fitting removal image is carried out using conic section, is obtained
The image calculated for angular deviation;
Above step is repeated until all socket image procossings finish.
Further, the corresponding relationship for determining pixel and actual object characteristic point in figure is converted by coordinate, with experiment
Method the character pair point in above-mentioned two formula is found respectively, so as to find out corresponding transition matrix, that is, know in image
The relationship of positional relationship and real space position.
Further, using the conic section of standard to the socket contour fitting of image, edge inspection is carried out to image first
It surveys, seeks the position of edge pixel, then with circle fitting socket profile.
Further, additional markers are carried out to ca bin socket, that is, the becket on piece at socket edge sprays
Facilitate the circular indicia of identification, the corresponding former socket flag bit of angle, which can obtain completely under diffusing reflection light source
Identification.
Further, in the step (3), the tap position center of the profile center of comparative silhouette fitting and former setting
Deviation coordinate value feeds back to plug wire robot control terminal to make amendment to the operation of its plug wire, and final realization socket state is automatic
The tap position of identification determines;
Characteristic point and picture centre reference line angulation are calculated, feeds back to plug wire robot control terminal to its plug wire
Amendment is made in operation, the final determination for realizing socket state automatic identification socket angle.
Further, the position for calculating socket, specifically:
The center of tap position in figure is obtained using contour fitting result;
Compare initial socket center and real image tap position center, obtains deviation by coordinate of unit pixel
Coordinate value;
According to camera calibration as a result, the relationship of picture point and actual object spatial position is determined, to obtain real space
The grid deviation value of position.
Further, according to plug wire damp feedback control monitoring plug wire during exception and alarm, specifically: industrial machine
Resistive force sensor built in device people determines probability distribution of the normal plug wire in the process by resistance by experimental method, sets resistance
Threshold value, and the control information that the information that will exceed threshold value is terminated as the warning message of plug wire operation exception and operation.
Compared with prior art, the beneficial effects of the present invention are:
Integrated optimization image processing flow prolongs Shen Kaifa in favor of automatic operation, and it is automatic to efficiently solve socket state
Position angle character recognition and computational problem when identification improve the gentle overall efficiency of Automated water of equipment testing process,
In addition, using for reference meaning with good for customizing application comprising the equipment detection based on machine vision including context of methods
Justice.
The advanced a theory method of the disclosure is applied in the actually detected assembly line of ca bin.In view of artificial
This method and the efficiency of artificial plug wire have been carried out following comparison by cost and working strength: by quantifying each index to totality
The benefit of target, to compare the superiority and inferiority of two methods of artificial plug wire and socket state automatic identification, comparing result is as shown in table 1.
1 plug wire efficiency comparative of table
Detailed description of the invention
The accompanying drawings constituting a part of this application is used to provide further understanding of the present application, and the application's shows
Meaning property embodiment and its explanation are not constituted an undue limitation on the present application for explaining the application.
Fig. 1 is the ca bin socket state automatic identification step schematic diagram based on machine vision in embodiment one;
Fig. 2 is the control flow chart that socket state automatic identification is used in embodiment one;
Fig. 3 is the control flow chart that image procossing is used in embodiment one.
Specific embodiment
It is noted that described further below be all exemplary, it is intended to provide further instruction to the application.Unless another
It indicates, all technical and scientific terms that the present invention uses have logical with the application person of an ordinary skill in the technical field
The identical meanings understood.
It should be noted that term used herein above is merely to describe specific embodiment, and be not intended to restricted root
According to the illustrative embodiments of the application.As used herein, unless the context clearly indicates otherwise, otherwise singular
Also it is intended to include plural form, additionally, it should be understood that, when in the present specification using term "comprising" and/or " packet
Include " when, indicate existing characteristics, step, operation, device, component and/or their combination.
Embodiment one
A kind of ca bin socket state based on machine vision disclosed in one or more embodiments from
It is dynamic to know method for distinguishing, can the image of each socket of bell-type ca bin to different manufacturers carry out processing analysis, identification mark
Know position and calculates separately the deviation of tap position and angle.Specific method is as depicted in figs. 1 and 2, comprising the following steps:
(1) the location parameter initial data set for being used for basic operation is established;
The position coordinates for obtaining the ca bin plug wire mouth of each supplier take the weighted mean of three-dimensional coordinate as not
Image Acquisition and plug wire before adjusted operate foundation, data weighting mean value formula are as follows:
Wherein Z representation space coordinate value, KiIndicate the corresponding position coordinate value of the plug wire mouth of i-th of supplier's product, WiTable
Show the estimated output of corresponding product, n indicates total output.
The location parameter provided in this manner will be as Robotic Manipulator camera acquisition socket image and execution plug wire
The basic parameter of movement, it is contemplated that there are certain nargin for the field range that camera obtains, and can guarantee to obtain suitable for dividing substantially
The image of analysis.
(2) using probe in detecting plug wire face depth and to operate camera in place, using diffusing reflection light source, each socket is successively acquired
Image;
The plug wire face depth information of ca bin is obtained using metal probe, the data that metal probe detects will
Key message as camera fixed-focus.
After camera and its respective members reach designated position, image is carried out by the annular diffusing reflection formula light source in front of camera
Acquisition.The light source mainly by the cyclic annular LED light direct irradiation that is centered around around camera on spherical diffuse-reflective material back reflection
It is formed, the disadvantage of the natural light deficiency of the invagination feature generation because of socket is compensated for using this light source.
According to the Image Acquisition process preset, whole socket Image Acquisition work an of ca bin is completed,
Its process is described as follows:
6. receiving the plug wire face depth information of designated equipment, start plug wire process;
7. opening light source, camera lens are positioned with light source to the first spigot;
8. camera acquisition image is simultaneously preprocessed, enters in next step after confirmation image definition is errorless, otherwise examine again
Camera fixed-focus effect;
9. repeating 2. 3. to walk, until the image information collecting of all 5 sockets finishes;
10. terminating Image Acquisition process, probe measurement result next time is waited.
(3) image is pre-processed, simplify subsequent processing difficulty and reduces picture noise, carry out flag bit identification and wheel
Exterior feature fitting, further prominent image effective information, calculates separately position and the angu-lar deviation of socket;
The image collected is pre-processed, its purpose is that simplify the complexity of image, the characteristic point of prominent image,
It is set to be conducive to subsequent upper layer operating procedure, described image processing step process is as shown in figure 3, its process includes:
1. judging whether image acquires, after Image Acquisition, image processing flow is triggered, calls gray processing process flow, it is right
Image carries out gray processing processing.
2. calling median filtering algorithm, the noise in image is removed;
3. calling distortion adjustment algorithm, correcting image;
4. carrying out camera calibration, determine that the positional relationship in image is corresponding with spatial relation in kind;
5. judging characteristics of image flag bit discrimination degree, determine whether the image can be used for deviation calculating;
6. feature identifies selected flag bit, information useless in contour fitting removal image is carried out using conic section;
7. determining profile center, calculating position deviation;
8. next socket image procossing and calculating are carried out, until all sockets are disposed.
The preprocess method is specific as follows:
1. carrying out gray processing processing to image, the grayscale image of each socket is obtained, is entered step 2. to 4.;
Grayscale image is obtained using weighted mean method, the image parameter of RGB mode is transformed to single gray-value image and is joined
Number, conversion formula are as follows:
Gray=α1G+α2B+α3R
Wherein, Gray indicates that gray value, G indicate green fundametal component value in formula, and B indicates blue fundametal component value, and R is indicated
Red fundametal component value.α1To α3Indicate weighted value.
2. the method using median filtering carries out noise reduction process;
Using quick median filter method, overcome the problems, such as that linear filtering bring details is fuzzy, the principle is as follows
When occurring impulsive noise in image, that is, when occurring being superimposed noise on the image in the form of black-white point, use
Median filter effectively removes this noise.The size that a neighborhood of pixels is arranged first is m × m, as filter
Working range;Secondly, obtaining the intermediate value in some neighborhood of pixels using ranking method;Finally, isolating relative to its neighborhood
Pixel is brighter or darker and region is less than m2/ 2 isolated pixel race.
Particularly, using relatively simple midpoint filter calculate maximum value in filter enclosing region and minimum value it
Between midpoint, formula is as follows:
Wherein, f (x, y) is pixel midrange, SxyFor the region that filter includes, g (s, t) is filter enclosing region
Pixel point set.
3. carrying out distortion adjustment, camera and appliance outlet the relative position error are eliminated;
By the picture point coordinate relationship before and after lens distortion, according to actual needs i.e. rectifiable distortion, the relationship is by diameter
It is generated to distortion and centrifugal distortion collective effect, can be described as following formula:
Wherein, (x, y) indicates the figure point coordinate after correction, (xd, yd) indicate figure point coordinate to be corrected, λ1, λ2, λ3It indicates
Coefficient of radial distortion, γ1, γ2Indicate centrifugal distortion coefficient.The distortion correction of image can Adjustment precision according to specific circumstances, with
The requirement for adapting to integral operation speed, takes default methods, ignores λ2, λ3Influence, thus it is ensured that the real-time of image procossing.
4. carrying out camera calibration, determines that the positional relationship in image is corresponding with spatial relation in kind, obtain use
In the image that position deviation calculates, it is labeled as ImageG, is entered step 5. to 6.;
The corresponding relationship for determining pixel and actual object characteristic point in figure is converted by coordinate, is experimentally distinguished
Character pair point in above-mentioned two formula is found, so as to find out corresponding transition matrix, that is, knows that image slices vegetarian refreshments and material object are right
It should be related to.
5. judging characteristics of image flag bit discrimination degree, choose whether to handle using color enhancement;
6. feature identifies selected flag bit, information useless in contour fitting removal image is carried out using conic section, is taken
It must be used for the image of angular deviation calculating, be labeled as ImageC, repeat above step until all socket image procossings finish.
Additional markers are carried out to ca bin socket, the effect of the label is that substitution socket buckle is sentenced as angle
Disconnected flag bit avoids distinct device socket because of the gradient that is recessed is different unrecognized situation.Socket additional flag position
Specifically, the becket on piece spraying at socket edge facilitates the circular indicia of identification, the corresponding former socket flag bit of angle should
Flag bit can be identified completely under diffusing reflection light source.This method establishes unified identification feature, using uniquely sentencing
Other condition identification feature position.
Using the conic section of standard to the socket contour fitting of image, edge detection is carried out to image first, seeks side
The position of edge pixel, secondly with circle fitting socket profile.
The image calculated for position and angular deviation is obtained using above-mentioned image processing method, which has data volume
It is small, feature is prominent, is easy to the characteristics of analyzing.
The deviation coordinate value at the profile center of comparative silhouette fitting and the tap position center of former setting, feeds back to plug wire machine
To make amendment to the operation of its plug wire, the final tap position for realizing socket state automatic identification determines device people control terminal;It calculates
Characteristic point and picture centre reference line angulation feed back to plug wire robot control terminal and are repaired with making to the operation of its plug wire
Just, the determination of socket state automatic identification socket angle is finally realized.
The calculation method of position deviation is as follows: obtaining the center of tap position in figure using contour fitting result;Compare just
Beginning socket center and real image tap position center, obtain the coordinate value of deviation by coordinate of unit pixel;According to phase
Machine calibration result, determines the relationship of picture point Yu actual object spatial position, to obtain the grid deviation of real space position
Value
According to following formula and calculate physical location deviation:
Wherein, SxWith SyFor actual position deviation;β is the conversion coefficient of pixel and actual size, generally in camera mark
It is determined during fixed;Px' and PxRespectively practical and home position coordinate.Regulation deviation is that timing plug wire position deviates to the right,
It is deviated to the left when being negative.
(4) position and angular deviation, data when being obtained practical operation using camera calibration technology are uploaded and control machine
People executes plug wire movement according to intended flow, and damping feedback control according to plug wire prevents plug wire operation exception.
According to the exception during damping feedback control monitoring plug wire and alarm.Resistive force sensor built in industrial robot,
Probability distribution of the normal plug wire in the process by resistance is determined by experimental method, sets resistance threshold value, and will exceed threshold value
The control information that information is terminated as the warning message of plug wire operation exception and operation.
Above-mentioned, although the foregoing specific embodiments of the present invention is described with reference to the accompanying drawings, not protects model to the present invention
The limitation enclosed, those skilled in the art should understand that, based on the technical solutions of the present invention, those skilled in the art are not
Need to make the creative labor the various modifications or changes that can be made still within protection scope of the present invention.
Claims (10)
1. a kind of ca bin socket state automatic identification method based on machine vision characterized by comprising
(1) the location parameter raw data set of actuating and image recognition is established;
(2) camera fixed-focus position is determined according to the plug wire face depth of detection, the plug wire face image for obtaining ca bin is gone forward side by side
Row pretreatment;
(3) flag bit identification and contour fitting are carried out, position and the angu-lar deviation of socket are calculated separately;
(4) plug wire movement is executed according to intended flow, according to the exception and report during plug wire damping feedback control monitoring plug wire
It is alert.
2. a kind of ca bin socket state automatic identification method based on machine vision as described in claim 1,
It is characterized in that, in the step (1), obtains the position coordinates of the ca bin plug wire mouth of each supplier, take three-dimensional coordinate
Weighted mean as Robotic Manipulator camera acquire socket image and execute plug wire movement basic parameter.
3. a kind of ca bin socket state automatic identification method based on machine vision as described in claim 1,
It is characterized in that, the plug wire face image of ca bin is obtained in the step (2), specifically:
1. obtaining the plug wire face depth information of ca bin using metal probe, start plug wire process;
2. opening light source, camera lens are positioned with light source to the first spigot;
3. camera acquisition image is simultaneously preprocessed, enters in next step after confirmation image definition is errorless, otherwise examine camera again
Fixed-focus effect;
4. repeating 2. 3. to walk, until the image information collecting of all sockets finishes;
5. terminating Image Acquisition process, probe measurement result next time is waited.
4. a kind of ca bin socket state automatic identification method based on machine vision as described in claim 1,
It is characterized in that, image is pre-processed specifically:
Gray processing processing is carried out to image, obtains the grayscale image of each socket;
Noise reduction process is carried out using the method for median filtering;
Distortion adjustment is carried out, camera and appliance outlet the relative position error are eliminated;
Camera calibration is carried out, determines that the positional relationship in image is corresponding with spatial relation in kind, obtains and be used for position
The image that deviation calculates;
Judge characteristics of image flag bit discrimination degree, chooses whether to handle using color enhancement;
Feature identifies selected flag bit, and information useless in contour fitting removal image is carried out using conic section, and acquirement is used for
The image that angular deviation calculates;
Above step is repeated until all socket image procossings finish.
5. a kind of ca bin socket state automatic identification method based on machine vision as claimed in claim 4,
It is characterized in that, the corresponding relationship for determining pixel and actual object characteristic point in figure is converted by coordinate, is experimentally divided
The character pair point in above-mentioned two formula is not found, so as to find out corresponding transition matrix, that is, knows the positional relationship in image
With the relationship of real space position.
6. a kind of ca bin socket state automatic identification method based on machine vision as claimed in claim 4,
It is characterized in that, using the conic section of standard to the socket contour fitting of image, edge detection is carried out to image first, seeks side
The position of edge pixel, then with circle fitting socket profile.
7. a kind of ca bin socket state automatic identification method based on machine vision as claimed in claim 4,
It is characterized in that, additional markers is carried out to ca bin socket, that is, the becket on piece spraying at socket edge facilitates identification
Circular indicia, the corresponding former socket flag bit of angle, which can be identified completely under diffusing reflection light source.
8. a kind of ca bin socket state automatic identification method based on machine vision as described in claim 1,
It is characterized in that, in the step (3), the deviation coordinate at the tap position center of the profile center of comparative silhouette fitting and former setting
Value feeds back to plug wire robot control terminal to make amendment to the operation of its plug wire, and final realization socket state automatic identification is inserted
Mouth position determines;
Characteristic point and picture centre reference line angulation are calculated, feeds back to plug wire robot control terminal to operate to its plug wire
Amendment is made, the final determination for realizing socket state automatic identification socket angle.
9. a kind of ca bin socket state automatic identification method based on machine vision as claimed in claim 8,
It is characterized in that, the position for calculating socket, specifically:
The center of tap position in figure is obtained using contour fitting result;
Compare initial socket center and real image tap position center, obtains the coordinate of deviation by coordinate of unit pixel
Value;
According to camera calibration as a result, the relationship of picture point and actual object spatial position is determined, to obtain real space position
Grid deviation value.
10. a kind of ca bin socket state automatic identification method based on machine vision as claimed in claim 8,
Be characterized in that, according to plug wire damp feedback control monitoring plug wire during exception and alarm, specifically: built in industrial robot
Resistive force sensor determines probability distribution of the normal plug wire in the process by resistance by experimental method, sets resistance threshold value, and will
The control information that information beyond threshold value is terminated as the warning message of plug wire operation exception and operation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910053588.XA CN109752627A (en) | 2019-01-21 | 2019-01-21 | A method of the ca bin socket state automatic identification based on machine vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910053588.XA CN109752627A (en) | 2019-01-21 | 2019-01-21 | A method of the ca bin socket state automatic identification based on machine vision |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109752627A true CN109752627A (en) | 2019-05-14 |
Family
ID=66405870
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910053588.XA Pending CN109752627A (en) | 2019-01-21 | 2019-01-21 | A method of the ca bin socket state automatic identification based on machine vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109752627A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110967530A (en) * | 2019-10-15 | 2020-04-07 | 国网山东省电力公司电力科学研究院 | Full-automatic wire plugging device with depth probe |
CN111537518A (en) * | 2020-05-25 | 2020-08-14 | 珠海格力智能装备有限公司 | Method and device for detecting defects of capacitor terminal, storage medium and processor |
CN112001888A (en) * | 2020-07-21 | 2020-11-27 | 浙江大学 | Impeller running state identification method and device |
CN113534136A (en) * | 2020-04-22 | 2021-10-22 | 郑州宇通客车股份有限公司 | Method and system for detecting left children in car |
CN114902647A (en) * | 2020-01-20 | 2022-08-12 | 日立安斯泰莫株式会社 | Image correction device and image correction method |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN202512438U (en) * | 2012-02-19 | 2012-10-31 | 陶重犇 | Moving robot SLAM platform for fish-eye camera |
CN106182004A (en) * | 2016-08-01 | 2016-12-07 | 上海交通大学 | The method of the industrial robot automatic pin hole assembling that view-based access control model guides |
CN106299972A (en) * | 2016-07-28 | 2017-01-04 | 上海交通大学 | Multiduty wiring automatic plug system and control method |
CN106504225A (en) * | 2016-09-27 | 2017-03-15 | 深圳增强现实技术有限公司 | A kind of recognition methodss of regular polygon and device |
CN108493740A (en) * | 2018-03-19 | 2018-09-04 | 苏州香农智能科技有限公司 | A kind of electric cabinet automatic routing system based on machine vision |
CN109193265A (en) * | 2018-10-29 | 2019-01-11 | 深圳市飞研智能科技有限公司 | A kind of circuit docking facilities and charging pile |
-
2019
- 2019-01-21 CN CN201910053588.XA patent/CN109752627A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN202512438U (en) * | 2012-02-19 | 2012-10-31 | 陶重犇 | Moving robot SLAM platform for fish-eye camera |
CN106299972A (en) * | 2016-07-28 | 2017-01-04 | 上海交通大学 | Multiduty wiring automatic plug system and control method |
CN106182004A (en) * | 2016-08-01 | 2016-12-07 | 上海交通大学 | The method of the industrial robot automatic pin hole assembling that view-based access control model guides |
CN106504225A (en) * | 2016-09-27 | 2017-03-15 | 深圳增强现实技术有限公司 | A kind of recognition methodss of regular polygon and device |
CN108493740A (en) * | 2018-03-19 | 2018-09-04 | 苏州香农智能科技有限公司 | A kind of electric cabinet automatic routing system based on machine vision |
CN109193265A (en) * | 2018-10-29 | 2019-01-11 | 深圳市飞研智能科技有限公司 | A kind of circuit docking facilities and charging pile |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110967530A (en) * | 2019-10-15 | 2020-04-07 | 国网山东省电力公司电力科学研究院 | Full-automatic wire plugging device with depth probe |
CN110967530B (en) * | 2019-10-15 | 2021-10-08 | 国网山东省电力公司电力科学研究院 | Use method of full-automatic wire plugging device with depth probe |
CN114902647A (en) * | 2020-01-20 | 2022-08-12 | 日立安斯泰莫株式会社 | Image correction device and image correction method |
CN113534136A (en) * | 2020-04-22 | 2021-10-22 | 郑州宇通客车股份有限公司 | Method and system for detecting left children in car |
CN113534136B (en) * | 2020-04-22 | 2023-07-28 | 宇通客车股份有限公司 | Method and system for detecting left-over children in vehicle |
CN111537518A (en) * | 2020-05-25 | 2020-08-14 | 珠海格力智能装备有限公司 | Method and device for detecting defects of capacitor terminal, storage medium and processor |
CN112001888A (en) * | 2020-07-21 | 2020-11-27 | 浙江大学 | Impeller running state identification method and device |
CN112001888B (en) * | 2020-07-21 | 2023-12-29 | 浙江大学 | Impeller running state identification method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109752627A (en) | A method of the ca bin socket state automatic identification based on machine vision | |
CN107590835B (en) | Mechanical arm tool quick-change visual positioning system and positioning method in nuclear environment | |
CN110850723B (en) | Fault diagnosis and positioning method based on transformer substation inspection robot system | |
CN105528789B (en) | Robot visual orientation method and device, vision calibration method and device | |
CN105548216B (en) | A kind of semi-finished product battery apparent visual detection method | |
CN103776841B (en) | Synthetic leather automatic defect detecting device and detection method | |
CN104573713B (en) | A kind of transformer Infrared image recognition based on image texture characteristic | |
CN111753712A (en) | Method, system and equipment for monitoring safety of power production personnel | |
CN109001224A (en) | Welding seam detection method and detection device | |
CN108380509A (en) | LED lamp panel sorting based on machine vision and detecting system | |
CN112347887A (en) | Object detection method, object detection device and electronic equipment | |
CN103149212A (en) | On-line quality detection system of button cell | |
CN109390636A (en) | Sutomatic winding machine pole piece deviation correction method, electronic equipment, storage medium and system | |
CN105631893A (en) | Method and device for detecting whether capacitor is correctly mounted through photographing | |
CN110211110A (en) | A kind of detection method of scantling, system and terminal device | |
CN109225941B (en) | Automatic detection and sorting system and method for internal thread tapping condition | |
CN113985830A (en) | Feeding control method and device for sealing nail, electronic equipment and storage medium | |
CN104931907B (en) | Digital display electrical measuring amount instrument quality group's check system based on machine vision | |
CN112847321A (en) | Industrial robot visual image recognition system based on artificial intelligence | |
CN109460757A (en) | Seal location recognition method and device | |
CN106841226B (en) | Cable orientation detection and correction system | |
WO2019000640A1 (en) | Modularized vision system manufacturing method | |
CN111242057A (en) | Product sorting system, method, computer device and storage medium | |
CN205607888U (en) | Magnetic ring detection device | |
CN113706455A (en) | Rapid detection method for damage of 330kV cable porcelain bottle sleeve |
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: 20190514 |