CN106780483A - Many continuous casting billet end face visual identifying systems and centre coordinate acquiring method - Google Patents
Many continuous casting billet end face visual identifying systems and centre coordinate acquiring method Download PDFInfo
- Publication number
- CN106780483A CN106780483A CN201710013250.2A CN201710013250A CN106780483A CN 106780483 A CN106780483 A CN 106780483A CN 201710013250 A CN201710013250 A CN 201710013250A CN 106780483 A CN106780483 A CN 106780483A
- Authority
- CN
- China
- Prior art keywords
- continuous casting
- casting billet
- image
- face
- billet end
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30116—Casting
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
A kind of method for obtaining continuous casting billet feature and extracting continuous casting billet centre coordinate for continuous casting billet labeling system Land use models identification image procossing, for final machine vision labeling system of setting up establishes technical foundation, the hardware that continuous casting billet center coordinate extraction method of the present invention based on machine vision is used includes：CCD camera, LED light source, optical filtering is provided with the computer of image storage and processing routine, scaling board.With industrial production is actual be combined theories of vision by this method, and theories of vision is applied in continuous casting billet production, has set up a set of automatic identification continuous casting billet, automatically extracts the automatic station-keeping system of continuous casting billet end face center.Continuous casting billet labeling efficiency, labeling precision are not only effectively increased, labeling labour cost is reduced, while advancing application of the vision system in actual industrial production.
Description
Technical field
The present invention relates to a kind of object center coordinate extraction method of view-based access control model positioning, vision positioning is based especially on
Many continuous casting billet end face visual identifying systems and centre coordinate acquiring method.
Background technology
Steel industry is a column support type industry for country, but " greatly without strong " is always that Iron and Steel Enterprises in China is plucked
Cap.Search to the bottom reason, lack technological innovation, it is a weight for restricting Iron and Steel Enterprises in China competitiveness that production technology falls behind
Want factor.Continuous casting billet has a wide range of applications as the raw material of shaped steel, wire rod etc. in national product life, and it has product
The features such as added value is high, Industrial Correlation is wide.Continuous casting billet production efficiency is thus improved, reducing continuous casting billet production cost turns into urgently
The problem of solution.Because continuous casting billet product category is numerous, difference is huge between various products, can effectively be distinguished using tag identifier
Various model continuous casting billets.But current steel mill's labeling link generally uses manual work, not only labour intensity is big, high cost, and
Labeling error rate is higher, seriously governs continuous casting billet production efficiency.
With the fast development of modern manufacturing industry, it is desirable to which labeling system must have fast speed, high precision, automation high etc.
Feature, this is accomplished by researching and developing a set of automatic labeling system for continuous casting billet.Continuous casting billet end face center is extracted as automatic labeling
System labeling provides important posture information.Machine vision is high, non-with positioning precision as a kind of emerging measurement means
Contact measurement, measurement range are wide, require low feature to measurement external environment condition.Therefore by machine vision applications in continuous casting billet end face
Center extraction is extracted to continuous casting billet end face center and is significant.Especially in the steel industry produced in enormous quantities, machine vision
Application undoubtedly to improve productivity ratio, reduce patch error rate, improve labeling precision have significant impetus.With machine vision as base
Plinth, continuous casting billet end face is recognized by pattern-recognition, image procossing and then continuous casting billet centre coordinate is extracted, automatic as continuous casting billet
The effective ways that the labeling system centre of location is obtained.Because bias light is complicated in steel mill's work on the spot environment, it is to detection easily
System forms larger interference, and some systems are to protrude object by installing light source additional in vision system, but the method can not
Weaken natural light, the interference of operating illumination light, therefore existing labeling vision positioning system cannot effectively process complex background light and do
Disturb.
The content of the invention
Based on object above, the present invention proposes a kind of the present invention is directed to propose one kind utilizes mould for continuous casting billet labeling system
Formula identification image procossing obtains continuous casting billet feature and the method for extracting continuous casting billet centre coordinate, is finally to set up machine vision labeling
System establishes technical foundation.The hardware that continuous casting billet center coordinate extraction method of the present invention based on machine vision is used includes：
CCD camera, LED light source, optical filtering is provided with the computer of image storage and processing routine, scaling board, it is characterised in that LED
Light source is LED monochrome annular array of light sources, LED light source center with can fixed placement CCD camera endoporus, CCD camera is positioned over
LED light source center, and face continuous casting billet end face, it is ensured that treat labeling continuous casting billet end face in CCD camera coverage, it is ensured that even
In CCD camera focal range, optical filtering is that can use the upper and lower deviation 15nm models of LED light source wavelength by the system to casting blank end surface
The narrow bandpass optical filtering for enclosing, is coordinated using screw thread and is screwed on CCD camera camera lens, is provided with the meter of image storage and processing routine
Calculation machine is arranged in and does not block the position that CCD camera gathers image, and CCD camera and light source pass through communication cable and be provided with figure respectively
As the computer of storage and processing routine links together, scaling board is circular array target, and scaling board is arranged in and continuous casting billet
On the perpendicular of end face alignment, and ensure scaling board perpendicular to horizontal plane.
The step of continuous casting billet end face center coordinate extraction method based on machine vision：
（1）Ensure that continuous casting billet end face is vertical with CCD camera camera lens optical axis, gather continuous casting billet end view drawing picture to be collected, and to gained
Image carries out image gray processing treatment；
（2）Image binaryzation treatment is carried out using fixed threshold to image, gray level image is changed into bianry image, prominent continuous casting billet
The feature of end face；
（3）Go unless ROI (region of interest) area image, filter due in background window, reflective thing to continuous casting
The influence that the identification of base end face brings；
（4）Removal ROI (region of interest) region inner area is too small, cross picture of large image scale, original image is tried one's best only
Retain continuous casting billet end surface features, reduce subsequent arithmetic amount；
（5）Designed size isSquare structure element, and morphology carried out to the image for obtaining with this structural element open
The size of computing, wherein n is set according to continuous casting billet size；
（6）Rim detection is carried out to the characteristics of image after treatment；
（7）To resulting image border using the tube method Corner Detection detection continuous casting billet end face four for the system design
Angle point；
（8）Length-width ratio according to feature object in image judges that each image-region is single continuous casting billet end face or many with continuous casting
Base end face, and acquired results are marked；
（9）Centre coordinate the Fitting Calculation is carried out to single continuous casting billet end face in image, many continuous casting billet end faces in image are carried out
Centre coordinate is estimated, and optimizes treatment with continuous casting billet end face center to many of estimation；
（10）Asked between continuous casting billet end face center image coordinate and world coordinates using Delaunay Triangulation standardization
Transformation relation, asks for continuous casting billet end face center world coordinates, and preserve resulting continuous casting billet according to resulting transformation relation
End face center world coordinates.
With industrial production is actual be combined theories of vision by this method, theories of vision is applied in continuous casting billet production, group
A set of automatic identification continuous casting billet has been built, the automatic station-keeping system of continuous casting billet end face center has been automatically extracted.The not only company of effectively increasing
Strand labeling efficiency, labeling precision, reduce labeling labour cost, while advancing vision system in actual industrial production
Using.
Brief description of the drawings
Fig. 1 is the general structure schematic diagram of the inventive method；
Fig. 2 is the extraction continuous casting billet end face center coordinate flow chart of the method for the present invention.
Specific embodiment
CCD camera, light source, optical filtering, be provided with image storage and image processing function computer.CCD light sources are ring
Shape monochromatic source, camera by annular light source endoporus with combination of light sources together with.Optical filtering is the band corresponding with light source wave band
Logical optical filtering, optical filtering is screwed normal diameter filter, can be coordinated by screw thread and is installed on camera lens.Guarantee light source,
Plane where optical filtering, CCD camera camera lens is parallel to each other, it is ensured that CCD camera optical axis is vertical with plane where continuous casting billet end face, protects
Card light source range of exposures covers all continuous casting billet end faces, it is ensured that continuous casting billet end face is in CCD camera focal range, it is ensured that be provided with
Image is stored and the computer of image processing function does not produce influence to CCD collection continuous casting billet end view drawing pictures.By light source and installation
There is image to store to be connected together by data wire with the computer of image processing function, CCD camera is deposited with image is provided with
Storage is connected together with the computer of image processing function by data wire.During whole vision positioning, it is only necessary to carry out
Once demarcate.During demarcation, scaling board with continuous casting billet plane at grade, and ensures that scaling board is vertical with optical axis, demarcates
End can just be taken away with scaling board.
1st, IMAQ and gray processing
During IMAQ, picture format use .BMP forms because .BMP forms are with the image lower left corner as the origin of coordinates, origin to
Right horizontal direction is x positive directions, and the upward vertical direction of origin is y-axis positive direction, and the image lower left corner meets typically for the origin of coordinates
Mathematical coordinates system, is easy to later stage calculating to process.The secondary original image of collection one, it can be found that connecting with a large amount of interference in original image
The ambient noise of casting blank end surface identification, it is therefore desirable to remove ambient noise, to recognize continuous casting billet end face, and then extracts continuous casting billet
Centre coordinate.The .BMP format-patterns for collecting are triple channel image, and suitable weight a is selected to each passage, and b, c are by image
It is processed as single channel gray level image, wherein a+b+c=1.
2nd, image binaryzation
Binary conversion treatment is carried out to gray level image using fixed threshold.It is single because system employs monochromatic LED light source plus optical filtering
Color LED light source can effectively protrude continuous casting billet end surface features to be measured and optical filtering corresponding with monochromatic LED light source wavelength can be filtered effectively
Fall the interference of picture quality of the background miscellaneous light to collecting.Therefore the image obtained by the system receives extraneous natural light and illumination light
Interference is few, can obtain high-quality bianry image using fixed threshold and improves system running speed.
3rd, go unless ROI region
ROI is region of interest.ROI region is area-of-interest, by the graphical analysis to gathering, continuous casting
Base end surface features concentrate on picture centre region, i.e. ROI region.The speck region being connected with image border passes through for natural light
The speck that window or factory building breakage are projected.I.e. non-ROI region, by the light wave that non-ROI region light source is sent is included
All wavelengths wave bands, filter is decayed to its intensity, but can not filter completely.Non- ROI region in identification image, and
Each region is marked.The image region brightness value of mark is set to 0, is filtered because natural light is damaged through window or factory building
The speck that place projects.
4th, it is excessive in removal ROI region to cross zonule
This step is actually the interior non-continuous casting billet end face image district not being adhered with continuous casting billet end face of ROI region in removal image
Domain.Due to light source irradiation, the hot spot interference of continuous casting billet end face region is inevitably introduced, and can not be effective by binaryzation
Eliminate.Estimate that continuous casting billet end face may occupy number of pixels N by testing, statistics ROI region each white portion number of pixels,
Think to be continuous casting billet end face if number of pixels is in scope m, think to be chaff element if number of pixels is not in scope m
Element.ROI interference elements can effectively be removed by the method, accuracy of detection, subsequent treatment speed is improve.
5th, morphology opening operation
Morphology opening operation is that the characteristic area obtained after above-mentioned treatment is first corroded into reflation, and small thing is eliminated so as to play
Body, the effect for substantially not changing at very thin point its area while separating objects, the border of smooth larger object.And pass through
Setting size isSquare structure element, may be such that in image that characteristic element carries out corrosion by structural element shape swollen
Swollen, the hot spot being connected with characteristic element so as to obtain smooth border and removal corner point is done to characteristic element Corner Detection
Disturb, improve Corner Detection precision.
6th, rim detection
Using edge feature in the processed image of Canny operators detection.Edge definition is have most in gradient direction by Canny
The point of big Grad, and introduce non-maximum suppression concept.Non- maximum suppression gradient by prediction first derivative direction first,
Then it approximate to 45 degree multiple, last compare the method for gradient amplitude in the gradient direction predicted to realize.By making
Can be very good to extract the edge feature that single pixel is linked with Canny operators, and reach good required precision.
7th, Corner Detection
By to current, motion state is analyzed in pipeline, when pipeline is unchanged along water movement direction or changes smaller
When, current are relatively steady, and when pipeline along water (flow) direction produce acute variation when, current become it is disorderly and unsystematic and to pipeline shape
Into certain impact, pipeline is set to produce vibrations.Tube method Corner Detection according to the characteristics of water flow in pipeline, when current to pipeline produce
Impact point is defined as angle point during raw impact.The continuous casting billet end face border that will be detected is with image pixel from left to right from upper
It is starting point O to lower first bright spot, along clockwise direction stores Close edges coordinate successively.Continuous casting billet end face border is regarded as
Current in pipeline, the symmetrical pipeline border of continuous casting billet end face border width identical is arranged parallel on border both sides, and pipeline is wide
Degree, wherein h is distance of the continuous casting billet end face border to pipeline border.O is set to " current " starting point, in the past n point
Fitting direction obtains the inceptive direction of " current ".When " current " are along inceptive direction " flowing " and when producing slight change, " pipeline "
Change with the change of " current ", and postpone t pixel unit, and work as " current " acute variation and punching is produced to " pipeline " border
When hitting, then it is judged to angle point.If the continuous casting billet end face boundary pixel that shock point is produced to " pipeline " border is i-th border picture
Element, then angle point is the i-th-h boundary pixel
8th, mark whether to be single continuous casting billet end face
Continuous casting billet end face width Wr and height Hr are calculated according to the four angular coordinate values tried to achieve.It is according to actual continuous casting billet end face
Foursquare feature, setting side degree threshold value e, definition.Wherein Wr be continuous casting billet end view drawing as number of pixels on border width, Hr is
Continuous casting billet end view drawing is as number of pixels on brim height.If e is close to 1, prove that required region is single continuous casting billet end face.If
E is much larger than 1, then prove have many continuous casting billet end faces to be adhered together on image.And required area can be tried to achieve according to the value of e
Domain continuous casting billet radical.Finally pair single continuous casting billet end face for determining is marked.
9th, the determination of continuous casting billet end face center
For single continuous casting billet end face, using the four angle point methods that diagonal angle point finds intersection two-by-two of continuous casting billet end face tried to achieve
Try to achieve single continuous casting billet end face center coordinate.When many continuous casting billet end faces are adhered, according to first continuous casting billet of left end
End face center is away from element left hand edge is adhered, continuous casting billet end face center level for Wr is tried to achieve at a distance of being adhered continuous casting billet end face two-by-two
In the every center abscissa of continuous casting billet, and according to continuous casting billet central point for left end two-end-point ordinate average value tries to achieve the company of being adhered
Every continuous casting billet centre coordinate initial value in casting blank end surface.And process alignment error and camera due to placing the workbench of continuous casting billet
Alignment error, may cause continuous casting billet end face on image and non-level.Therefore tried to achieve first with least square fitting
Many continuous casting billet end bottom fitted straight lines of edges L1, then cross straight line L2 of each continuous casting billet end face center work perpendicular to L1,
And try to achieve and be apart from L1 on straight line L2Continuous casting billet end face center optimization coordinate.The single continuous casting billet end tried to achieve will be stored
Continuous casting billet end face center coordinate is adhered after face centre coordinate and optimization.
10th, conversion from pixel coordinate system to world coordinate system
Asking for for continuous casting billet end face center coordinate is sat in the world to obtain continuous casting billet center physical location, i.e. continuous casting billet
Coordinate value under mark system.So, it is necessary to by its image coordinate after coordinate value of the continuous casting billet center in image coordinate system is tried to achieve
World coordinates is converted into, this is accomplished by demarcating constructed vision system.
Continuous casting billet end face lies substantially on same horizontal plane in labeling, therefore can be plane where continuous casting billet end face
Z faces are defined as, the two-dimentional relation of the coordinate Yu world coordinate system that only need uncalibrated image need to be only demarcated during demarcation.Vision system is used
Delaunay Triangulation standardization based on planar array round dot carries out vision calibration.According to the local principle that do not distort, ask
Each triangular interpolation value coefficient mi, ni are obtained, according to the interpolation value coefficient tried to achieve, using triangle where central point in practice
Apex coordinate tries to achieve two-dimensional coordinate under central point world coordinate system.Interpolation value coefficient is calculated as follows：
（1）
（2）
WhereinCentered on put pixel coordinate、、Centered on put where triangle
Summit pixel coordinate.
The triangular apex real coordinate position according to where the interpolation value coefficient and central point tried to achieve, tries to achieve central point
Actual two-dimensional coordinate value.Actual two-dimensional coordinate is calculated as follows：
（3）
（4）
WhereinCentered on put world coordinate system under coordinate,、、Centered on put where
Coordinate under triangular apex world coordinate system.
This scaling method step is as follows：
1）This scaling method select demarcation board size be, target is for 7 rows multiply 82 row radiusesCircle
Shape array.The position of fixed CCD camera keeps constant, and plane target drone is placed perpendicular to ground first, and the upper and lower both sides of target are put down
Row is in horizontal plane and to ensure that plane target drone and continuous casting billet end face are generally aligned in the same plane interior and can be comprising all continuous casting billet end faces.Clap
Target image is taken the photograph, each round dot center of circle image coordinate is tried to achieve, is divided using Delaunay Triangulation and tried to achieve in each triangle
Interpolation coefficient；
2）Define the origin of the world coordinates of the plane target drone upper left angle point, and world coordinates Z faces on plane target drone, Z axis
Perpendicular to the upper left corner；
3）Continuous casting billet center two-dimensional world coordinate is asked for by the continuous casting billet center pixel coordinate for processing.
Through checking computations, after the world coordinates of the continuous casting billet end face center point that will be obtained is transferred to labelling machines people, labelling machines
People presses this world coordinates labeling, can reach good labeling effect, meets and requires.
Claims (1)
1. it is a kind of for continuous casting billet labeling system Land use models identification image procossing to obtain continuous casting billet feature and extract continuous casting billet
The heart sits calibration method, is finally to set up machine vision labeling system to establish technical foundation, continuous casting of the present invention based on machine vision
The hardware that base center coordinate extraction method is used includes：CCD camera, LED light source, optical filtering is provided with image storage and place
The computer and scaling board of reason program, it is characterised in that LED light source is LED monochrome annular array of light sources, and LED light source center carries
Can fixed placement CCD camera endoporus, CCD camera is positioned over LED light source center, and faces continuous casting billet end face, it is ensured that treat labeling
Continuous casting billet end face is in CCD camera coverage, it is ensured that in CCD camera focal range, optical filtering is to lead to continuous casting billet end face
The narrow bandpass optical filtering that the system uses the upper and lower deviation 15nm scopes of LED light source wavelength is crossed, is coordinated using screw thread and is screwed in CCD phases
On machine camera lens, the computer for being provided with image storage and processing routine is arranged in the position for not blocking CCD camera collection image,
CCD camera and light source are linked together by communication cable with the computer for being provided with image storage and processing routine respectively, are marked
Fixed board is circular array target, and scaling board is arranged on the perpendicular alignd with continuous casting billet end face, and ensures that scaling board is vertical
In horizontal plane, methods described comprises the following steps：
（1）Ensure that continuous casting billet end face is vertical with CCD camera camera lens optical axis, gather continuous casting billet end view drawing picture to be collected, and to gained
Image carries out image gray processing treatment；
（2）Image binaryzation treatment is carried out using fixed threshold to image, gray level image is changed into bianry image, prominent continuous casting billet
The feature of end face；
（3）Go unless ROI (region of interest) area image, filter due in background window, reflective thing to continuous casting
The influence that the identification of base end face brings；
（4）Removal ROI (region of interest) region inner area is too small, cross picture of large image scale, original image is tried one's best only
Retain continuous casting billet end surface features, reduce subsequent arithmetic amount；
（5）Designed size is the square structure element of n × n, and carries out morphology to the image for obtaining with this structural element and open
The size of computing, wherein n is set according to continuous casting billet size；
（6）Rim detection is carried out to the characteristics of image after treatment；
（7）To resulting image border using the tube method Corner Detection detection continuous casting billet end face four for the system design
Angle point；
（8）Length-width ratio according to feature object in image judges that each image-region is single continuous casting billet end face or many with continuous casting
Base end face, and acquired results are marked；
（9）Centre coordinate the Fitting Calculation is carried out to single continuous casting billet end face in image, many continuous casting billet end faces in image are carried out
Centre coordinate is estimated, and optimizes treatment with continuous casting billet end face center to many of estimation；
（10）Asked between continuous casting billet end face center image coordinate and world coordinates using Delaunay Triangulation standardization
Transformation relation, asks for continuous casting billet end face center world coordinates, and preserve resulting continuous casting billet according to resulting transformation relation
End face center world coordinates.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710013250.2A CN106780483B (en) | 2017-01-09 | 2017-01-09 | More continuous casting billet end face visual identifying systems and centre coordinate acquiring method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710013250.2A CN106780483B (en) | 2017-01-09 | 2017-01-09 | More continuous casting billet end face visual identifying systems and centre coordinate acquiring method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106780483A true CN106780483A (en) | 2017-05-31 |
| CN106780483B CN106780483B (en) | 2019-05-07 |
Family
ID=58950490
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710013250.2A Active CN106780483B (en) | 2017-01-09 | 2017-01-09 | More continuous casting billet end face visual identifying systems and centre coordinate acquiring method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106780483B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109141506A (en) * | 2018-06-28 | 2019-01-04 | 深圳奥比中光科技有限公司 | Multi-functional calibration system |
| CN109191525A (en) * | 2018-09-03 | 2019-01-11 | 佛亚智能装备（苏州）有限公司 | A kind of deviation pre-alert method and device |
| CN109784331A (en) * | 2019-01-08 | 2019-05-21 | 河北科技大学 | Bar section tagging scheme and character picture antidote based on index point |
| CN109775055A (en) * | 2019-01-08 | 2019-05-21 | 河北科技大学 | The bundled rods end face label missing of view-based access control model detects and error measurement method |
| CN109945842A (en) * | 2018-06-11 | 2019-06-28 | 河北科技大学 | Bundled round steel end face label missing detection and labeling error analysis method |
| CN111489367A (en) * | 2020-06-24 | 2020-08-04 | 浙江大学 | Positioning method for automatically selecting feasible coding area of hub two-dimensional code |
| WO2021037909A1 (en) | 2019-08-30 | 2021-03-04 | Primetals Technologies Germany Gmbh | A locating method and a locator system for locating a billet in a stack of billets |
| CN113996499A (en) * | 2021-09-27 | 2022-02-01 | 普洛赛斯(苏州)软件科技有限公司 | Intelligent positioning system for automatic industrial dispensing robot |
| CN116449041A (en) * | 2023-06-14 | 2023-07-18 | 江苏金恒信息科技股份有限公司 | Continuous casting blank sampling system and method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9157865B2 (en) * | 2010-05-03 | 2015-10-13 | United Technologies Corporation | Machine tool—based, optical coordinate measuring machine calibration device |
| CN105405134A (en) * | 2015-11-05 | 2016-03-16 | 山东科技大学 | Camera calibration mark center coordinate extraction method |
-
2017
- 2017-01-09 CN CN201710013250.2A patent/CN106780483B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9157865B2 (en) * | 2010-05-03 | 2015-10-13 | United Technologies Corporation | Machine tool—based, optical coordinate measuring machine calibration device |
| CN105405134A (en) * | 2015-11-05 | 2016-03-16 | 山东科技大学 | Camera calibration mark center coordinate extraction method |
Non-Patent Citations (1)
| Title |
|---|
| 黄风山等: "成捆圆钢机器人贴标系统图像识别方法", 《光电工程》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109945842A (en) * | 2018-06-11 | 2019-06-28 | 河北科技大学 | Bundled round steel end face label missing detection and labeling error analysis method |
| CN109945842B (en) * | 2018-06-11 | 2020-12-04 | 河北科技大学 | Method for detecting label missing and analyzing labeling error of end face of bundled round steel |
| CN109141506A (en) * | 2018-06-28 | 2019-01-04 | 深圳奥比中光科技有限公司 | Multi-functional calibration system |
| CN109191525A (en) * | 2018-09-03 | 2019-01-11 | 佛亚智能装备（苏州）有限公司 | A kind of deviation pre-alert method and device |
| CN109784331A (en) * | 2019-01-08 | 2019-05-21 | 河北科技大学 | Bar section tagging scheme and character picture antidote based on index point |
| CN109775055A (en) * | 2019-01-08 | 2019-05-21 | 河北科技大学 | The bundled rods end face label missing of view-based access control model detects and error measurement method |
| WO2021037909A1 (en) | 2019-08-30 | 2021-03-04 | Primetals Technologies Germany Gmbh | A locating method and a locator system for locating a billet in a stack of billets |
| US11068755B2 (en) | 2019-08-30 | 2021-07-20 | Primetals Technologies Germany Gmbh | Locating method and a locator system for locating a billet in a stack of billets |
| CN111489367A (en) * | 2020-06-24 | 2020-08-04 | 浙江大学 | Positioning method for automatically selecting feasible coding area of hub two-dimensional code |
| CN113996499A (en) * | 2021-09-27 | 2022-02-01 | 普洛赛斯(苏州)软件科技有限公司 | Intelligent positioning system for automatic industrial dispensing robot |
| CN116449041A (en) * | 2023-06-14 | 2023-07-18 | 江苏金恒信息科技股份有限公司 | Continuous casting blank sampling system and method |
| CN116449041B (en) * | 2023-06-14 | 2023-09-05 | 江苏金恒信息科技股份有限公司 | Continuous casting blank sampling system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106780483B (en) | 2019-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106780483A (en) | Many continuous casting billet end face visual identifying systems and centre coordinate acquiring method | |
| CN111507976B (en) | Defect detection method and system based on multi-angle imaging | |
| CN104112269B (en) | A kind of solar battery laser groove parameter detection method and system based on machine vision | |
| CN106570863A (en) | Detection method and device for power transmission line | |
| CN102654464A (en) | Copper strip surface defect detection system based on multi-feature fuzzy recognition | |
| CN103454285A (en) | Transmission chain quality detection system based on machine vision | |
| CN114331986A (en) | Dam crack identification and measurement method based on unmanned aerial vehicle vision | |
| CN102221559A (en) | Online automatic detection method of fabric defects based on machine vision and device thereof | |
| CN105865329A (en) | Vision-based acquisition system for end surface center coordinates of bundles of round steel and acquisition method thereof | |
| CN111539927B (en) | Detection method of automobile plastic assembly fastening buckle missing detection device | |
| CN106780526A (en) | A kind of ferrite wafer alligatoring recognition methods | |
| CN103308525B (en) | A kind of online test method for metal wire rod production and device | |
| CN114494210A (en) | Plastic film production defect detection method and system based on image processing | |
| CN106709529B (en) | Visual detection method for photovoltaic cell color difference classification | |
| CN113379712A (en) | Steel bridge bolt disease detection method and system based on computer vision | |
| CN104637067A (en) | Method for detecting defect of textured surface | |
| CN109684914A (en) | Based on unmanned plane image intelligent identification Method | |
| CN115775236A (en) | Surface tiny defect visual detection method and system based on multi-scale feature fusion | |
| CN115018846A (en) | AI intelligent camera-based multi-target crack defect detection method and device | |
| CN115797314A (en) | Part surface defect detection method, system, equipment and storage medium | |
| CN115656182A (en) | Sheet material point cloud defect detection method based on tensor voting principal component analysis | |
| CN115018785A (en) | Hoisting steel wire rope tension detection method based on visual vibration frequency identification | |
| CN114387262A (en) | Nut positioning detection method, device and system based on machine vision | |
| CN113570587A (en) | Photovoltaic cell broken grid detection method and system based on computer vision | |
| CN117079125A (en) | Kiwi fruit pollination flower identification method based on improved YOLOv5 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |