CN107138431B - A kind of components identification method for separating and system based on machine vision - Google Patents
A kind of components identification method for separating and system based on machine vision Download PDFInfo
- Publication number
- CN107138431B CN107138431B CN201710393224.7A CN201710393224A CN107138431B CN 107138431 B CN107138431 B CN 107138431B CN 201710393224 A CN201710393224 A CN 201710393224A CN 107138431 B CN107138431 B CN 107138431B
- Authority
- CN
- China
- Prior art keywords
- components
- big
- manipulator
- classification
- machine vision
- 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.)
- Active
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/02—Measures preceding sorting, e.g. arranging articles in a stream orientating
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
- B07C5/361—Processing or control devices therefor, e.g. escort memory
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
- B07C5/361—Processing or control devices therefor, e.g. escort memory
- B07C5/362—Separating or distributor mechanisms
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
- B07C5/38—Collecting or arranging articles in groups
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C2501/00—Sorting according to a characteristic or feature of the articles or material to be sorted
- B07C2501/0063—Using robots
Landscapes
- Sorting Of Articles (AREA)
- Manipulator (AREA)
Abstract
The invention discloses a kind of, and the components based on machine vision identify method for separating and system, gather appearance detection and identification sorting, select two manipulators, one is used as Image Acquisition, another is integrated with camera and a variety of suckers, it for the guidance positioning to small components and draws, can be quickly and accurately positioned and grab tested part.It can be realized the pick to various different sized vehicle components, applicability is wider.And standard machinery hand is reduced for the difficulty of different components crawl programming, strong operability.
Description
Technical field
The present invention relates to components to identify field, identifies method for separating in particular to a kind of components based on machine vision
And system.
Background technique
With the rapid development of industry, the amount of labour is increasing.In the production process of large batch of components sorting, use
The quality efficiency that manpower mode carries out the sorting of components is low and precision is not high, it is difficult to meet industrial requirements.And machine vision
Detection method then can large batch of sorting components product, substantially increase the degree of automation of production efficiency and production.
Currently, the components method of discrimination based on robot vision predominantly disposes industrial phase in conveyor belt two sides or top
Machine acquires image by take pictures to the components transmitted on a moving belt, or grabs camera by fixture, manipulator and advance
Row shooting, then by main control computer carry out part diagram picture pretreatment and feature extraction, and with template storehouse matching, to complete
The identification of components type.
Domestic colleges and universities and R&D institution have for the related patents of the parts testing method based on machine vision:
Application No. is 201510607834.3 patent application " fine components defect intelligent on-line detection method " uses
Conveyer belt transmits components to shooting area, then the method based on machine vision is cooperated to complete detection, only uses conveyer belt, fine zero
The imaging effect of part is unsatisfactory, easily causes leakage bat or image blur；
Application No. is a kind of 201510973877.3 patent application " small miniature Axle Parts sizes based on machine vision
On-line measuring device " carries out shot detection to components using the mobile camera of removable camera pedestal, and emphasis is list
The control of piece machine；
Application No. is a kind of 201210124176.9 patent application " component surface defects detections based on machine vision
Method and device " cooperates the method based on machine vision to complete detection, only uses using shooting before fixture clamping components to camera
Fixture has considerable restraint for the size of components, and fixture will cause imaging interference.If you need to the inspection for carrying out each size components
It surveys, then needs to change fixture, increase cost；
Application No. is 201610365672.1 patent application " a kind of multi-robot refuse classification control systems " using biography
Send band transmission rubbish to shooting area, processing is classified, after classified by multiple manipulators, the dynamic shooting on conveyer belt at
Picture, it is undesirable for the imaging effect of small rubbish.
Summary, only shoots components on a moving belt with camera, and for small components, imaging area is small,
The problems such as components sideslip and the abrasion of itself existing for conveyer belt, scuffing, overlap joint cracking, it is easy to cause industrial camera
Small components are leaked and claps or claps not congruent problem, at the same under conveyer belt motion state for the shooting effect of small components influence compared with
Greatly.It is only grabbed with mechanical paw or fixture and is shot to camera, the components of different size ranges then needed different big
Small mechanical paw or fixture, increases cost.Simultaneously components imaging area it is small, mechanical paw or fixture crawl can to its at
As interfering, also component surface can be caused to damage.
Summary of the invention
The present invention provides a kind of measurement and crawl is accurate, base of strong operability aiming at the deficiencies in the prior art
Method for separating and system are identified in the components of machine vision.
To achieve the goals above, the components based on machine vision designed by the present invention identify method for separating, special
Different place is, comprising the following steps:
S1 collecting sample part diagram picture carries out feature extraction to sample part diagram picture and establishes components template library；
S2 judges whether components to be detected reach detection zone, and judges the attribute of components, specifically includes big
It is small, magnetic and non magnetic, wherein small components refer to that size justifies the components being formed by within area in diameter for 200mm,
And weight is less than 1000N, remaining is big components；
S3 carries out Image Acquisition and feature extraction to components to be detected: big components directly carry out Image Acquisition and spy
Sign is extracted；Magnetic small components carry out Image Acquisition and feature extraction after drawing using magnetic chuck；Non magnetic small components are adopted
Image Acquisition and feature extraction are carried out after being drawn with vacuum chuck again；Wherein, Image Acquisition uses the machinery equipped with industrial camera
Hand is realized；The absorption of small components is realized using another manipulator equipped with magnetic chuck, vacuum chuck and industrial camera；
S4 matches part diagram picture to be detected with template library image, confirms components type and quality；
S5 will confirm that the components of type and quality are delivered to corresponding storing region, complete the classification of components.
Further, quality is detected for guarantee, the sample components and components to be detected in the step S1 and S3
Image Acquisition condition and feature extraction mode are all the same.
Further, it is additionally provided with annular light source on the manipulator of Image Acquisition in the step S3, also for guarantor
Card detection quality.
Still further, the detailed process of the step S5 are as follows: qualified big components reach big components with conveyer belt
Specification area, and transmitted by corresponding big components sort pass band；Qualified small components are directly put by another manipulator
It sets；Defect ware reaches end with conveyer belt and is recycled.
A kind of components identification separation system based on machine vision, is characterized in that including main control computer, passes
Send band, sensor, No.1 manipulator, No. two manipulators, small components classification area, classification push sensor, classification push photoelectricity
Sensor, big components sort pass band are set with area to be tested and big components processing region on the conveyer belt, described
Sensor setting is arranged in area to be tested, the No.1 manipulator and No. two manipulators in conveyor belt two sides, is respectively positioned on to be checked
Region is surveyed, for the No.1 manipulator for acquiring components image information, No. two manipulators are small by zero for drawing and putting
Component；The big components processing region is located at area to be tested downstream, and the big components processing region includes classification push
Sensor, classification push photoelectric sensor and big components sort pass band, the classification push sensor, classification push photoelectricity
Sensor is arranged in conveyer belt side, and the big components sort pass band is located at the conveyer belt other side, pushes photoelectricity with classification
Sensor is corresponding, the conveyer belt, sensor, No.1 manipulator, No. two manipulators, classification push sensor and classification push light
Electric transducer is controlled by main control computer.
Preferably, the transmission belt end is additionally provided with components recycling and processing device.
The present invention has the advantages that
1, the present invention can gather appearance detection and identification sorting, wherein appearance detection can judge components quality whether
There are problems, make up the deficiency of human eye, and functional integration is high.Simultaneously this method design mating mechanism, can be realized to it is various not
With the pick of sized vehicle components, applicability is wider.
2, the present invention selects two manipulators, and one is used as Image Acquisition, another is integrated with camera and a variety of suckers, uses
It in the guidance positioning to small components and draws, can be quickly and accurately positioned and grab tested part.
3, the present invention is directly acquired large parts image, and for small parts, imaging area is small, with suction
Disk draws alignment cameras and carries out quiet shooting, and shooting effect is better than on a moving belt carrying out it dynamic shooting, and sucker draws zero
Part will not be imaged it and interfere, and crawl will not be caused to damage to its surface, and reduce standard machinery hand for difference
The difficulty of components crawl programming, strong operability.
Detailed description of the invention
Fig. 1 is that the present invention is based on the overall workflow figures that the components of machine vision identify method for separating.
Fig. 2 is that the present invention is based on the integral layout structural schematic diagrams that the components of machine vision identify separation system.
Fig. 3 is that information of the invention judges decision flow diagram.
Fig. 4 is the operation schematic diagram differentiated to small components.
Fig. 5 is the operation schematic diagram differentiated to big components.
Fig. 6 is the Local map of No. two manipulators.
In figure: passing classification push sensor 1, classification push photoelectric sensor 2, conveyer belt 3, components recycling and processing device
4, big components sort pass band 5, robot movement regional scope 6, annular light source 7, No.1 manipulator 8, sensor 9, zero
10, No. two manipulators 11 of part, magnetic chuck 11.1,11.2, No. two industrial cameras 11.3 of vacuum chuck, small components classification area
12, main control computer 13.
Specific embodiment
The present invention is described in further detail in the following with reference to the drawings and specific embodiments:
The components based on machine vision identify method for separating as shown in the figure, comprising the following steps:
S1 collecting sample part diagram picture carries out feature extraction to sample part diagram picture and establishes components template library；
S2 judges whether components to be detected reach detection zone, and judges the attribute of components, specifically includes big
It is small, magnetic and non magnetic；
S3 carries out Image Acquisition and feature extraction to components to be detected: big components directly carry out Image Acquisition and spy
Sign is extracted；Magnetic small components carry out Image Acquisition and feature extraction after drawing using magnetic chuck；Non magnetic small components are adopted
Image Acquisition and feature extraction are carried out after being drawn with vacuum chuck again；Wherein, Image Acquisition uses the machinery equipped with industrial camera
Hand is realized；The absorption of small components is realized using another manipulator equipped with magnetic chuck, vacuum chuck and industrial camera.It is small by zero
It is the components that 200mm circle is formed by within area that component, which is defined as size in diameter, i.e. components are revolved around its geometric center
It circles and is formed by area of the round area less than the circle that diameter is 200mm, weight is less than 1000N.Remaining is defined as big by zero
Component.To guarantee shooting quality, Image Acquisition with install annular light source additional on manipulator.
S4 matches part diagram picture to be detected with template library image, confirms components type and quality；Using base
In the method for the template matching of binary image technology, a standard form Ti, figure to be identified are established to each template library image
As being X, in template image and images to be recognized, the foreground target pixel value after binaryzation is set as 1, and background pixel value is set as 0,
The size of template image and images to be recognized is M × N, by images to be recognized one by one with template matching, finds out its similarity Si,
The number that different images to be recognized X and standard form image Ti upper value are the point of " 1 " simultaneously is different,
So ratio Si is different.Rejection threshold value λ is set, if Si < λ, determinesIf Si >=λ, X ∈ Ti is determined.
If having Si < λ for all templates, determine that the quality of the components is problematic；If having for multiple template
Si >=λ then takes matching degree highest as output, completes components type and differentiate.
S5 will confirm that the components of type and quality are delivered to corresponding storing region, complete the classification of components.Specifically
Are as follows: qualified big components reach big components specification area with conveyer belt, and are passed by corresponding big components sort pass band
It is defeated；Qualified small components are directly placed by another manipulator；Defect ware reaches end with conveyer belt and is recycled.
Wherein, the Image Acquisition condition and feature of the sample components in above step S1 and S3 and components to be detected mention
Take mode all the same.The shooting environmental condition of template library image, illumination, shooting height are identical as input picture, small components root
Different poses is chosen according to its difference for taking a crane shot face and is shot one by one according to its absorption face, big components, to make template library
It is as detailed as possible, and there are the surface informations of thin portion difference to be grabbed by camera for the close components of guarantee.The seating surface of components
It is limited, can place on a moving belt is also limited for the position for shooting or drawing, so the absorption for small components
Face, the high angle shot face of big components are all shot according to different poses one by one when preparing template library, as detailed as possible.Industry
The edge detection method that camera carries the positioning of components using camera software kit, detects the rough profile side of components
After edge, it is known that its placement status is drawn or shot according to its placed side.
In the present invention, the image in template library uses the method as input picture to carry out image preprocessing to extract figure
As feature.Image preprocessing mainly includes image enhancement, filtering and noise reduction, image segmentation and edge detection, by image and background point
It separates out and, carry out characteristic parameter extraction for target object.Components type is completed using the method for template matching to differentiate.
A kind of components identification separation system based on machine vision, including main control computer 13, conveyer belt 3, sensor
9,8, No. two manipulators 11 of No.1 manipulator, small components classification area 12, classification push sensor 1, classification push photoelectric sensing
Device 2, big components sort pass band 5 are set with area to be tested and big components processing region on conveyer belt 3, and sensor 9 is set
It sets in area to be tested, No.1 8 hands of machinery and No. two manipulators 11 are arranged in 3 two sides of conveyer belt, are respectively positioned on area to be tested, greatly
Components processing region is located at area to be tested downstream, and big components processing region includes classification push sensor 1, classification push
Photoelectric sensor 2 and big components sort pass band 5, classification push sensor 1, the classification push setting of photoelectric sensor 2 are passing
Send band 3 side, big components sort pass band 5 is located at 3 other side of conveyer belt, transmission corresponding with classification push photoelectric sensor 2
Band 3, sensor 9,8, No. two manipulators 11 of No.1 manipulator, classification push sensor 1 and classification push photoelectric sensor 2 by
Main control computer control.3 end of conveyer belt is additionally provided with components recycling and processing device 4.Equipped with industrial phase on No.1 manipulator 8
Machine, and annular light source 7 is furnished with after mechanical paw, a magnetic chuck, a vacuum chuck and are housed on two manipulators 11
A industrial camera.The model ZYE1-P100/40 that the circular electromagnetic chunk of No. two manipulators 11 is selected, disk major diameter are
100mm, path 42mm, power 15w, attraction 1200N are self-possessed for 1900N.The model that vacuum chuck is selected
ZP100HS, sucker diameter are 100mm, and suction cup type is heavy-load type, and material is silicon rubber.Small components are defined as size and exist
Diameter is the components that 200mm circle is formed by within area, i.e. components rotate a circle around its geometric center and are formed by circle
Area be less than diameter be 200mm circle area, weight be less than 1000N.Remaining is defined as big components.
Invention is further elaborated using specific embodiment below, the circular electric of No. two manipulators 11 shown in Fig. 2
The model ZYE1-P100/40 that magnetic-disc is selected, disk major diameter are 100mm, path 42mm, power 15w, attraction
For 1200N, it is self-possessed for 1900N.The model ZP100HS that vacuum chuck is selected, sucker diameter are 100mm, and suction cup type is
Heavy-load type, material are silicon rubber.
The working region of No.1 manipulator 8 and No. two manipulators 11 is the circle that radius is R, R=1m, as shown in Fig. 2, machine
Tool hands movement regional scope 6.The transmission speed of conveyer belt be 0~4m/s, favor speed 0.2m/s, width 400mm~
Between 800mm, length is between 3000mm~4000mm, with a thickness of 10mm.The feeding interval time of 3 one end of conveyer belt is t, t
It completes a components detection identification and classifies required total time.Feeding time interval is set, can avoid machine occur
Tool makees situations such as uncoordinated, missing inspection by hand.
As shown in Figures 2 and 3, on conveyer belt 3 close to manipulator end setting sensor 9 with judge components 10 size,
It is magnetic, non magnetic and whether reach area to be tested, and realize by industrial camera the accurate positioning and shooting of components.Specifically
Process is as follows: if being judged as big magnetic or non magnetic components, as shown in figure 5, directly being transmitted using No.1 manipulator 8
To its shooting, collecting image on band 3, but heavy parts shoot multiple splicings and are identified；If being judged as small magnetic parts,
As shown in figs. 4 and 6, then magnetic chuck 11.1 and No. two industrial cameras 11.3 on No. two manipulators 11 is used to position it
And it draws, then it is directed at shooting, collecting image with the industrial camera clamped on No.1 manipulator 8；If being judged as small non magnetic
Magnetic chuck 11.1 on No. two manipulators 11 is then changed to vacuum chuck 11.2 as shown in figs. 4 and 6 by components, then right
Components are positioned and are drawn, then the industrial camera clamped on this components and No.1 manipulator 8 is directed at shooting, collecting figure
Picture.
As shown in Figure 1, inputting master control meter for the industrial camera shooting, collecting image on No.1 manipulator 8 as input picture
Calculation machine 13.
Main control computer is to the image of acquisition from image enhancement, filtering and noise reduction, and image segmentation, three aspect of edge detection is successively
It is pre-processed, image and background separation is come out, carry out characteristic parameter extraction for target object.The feature master generally extracted
To include morphological feature, gray feature, textural characteristics, binary image technology can be used, i.e., set the gray value of target part
For maximum, and the gray value of background parts is set to minimum, is usually set to zero.
The shooting environmental condition of template library image, illumination, shooting height are identical as the input picture of components to be detected, small
Components are chosen different poses according to its difference for taking a crane shot face and are shot one by one according to its absorption face, big components, thus
Keep template library as detailed as possible, and there are the surface informations of thin portion difference to be grabbed by camera for the close components of guarantee.Template library
In image use the method as input picture carry out image preprocessing to extract characteristics of image.
Using the method for the template matching based on binary image technology, a master die is established to each template library image
Plate Ti, images to be recognized X, in template image and images to be recognized, the foreground target pixel value after binaryzation is set as 1, back
Scape pixel value is set as 0, their picture size size is M × N, by images to be recognized one by one with template matching, finds out its phase
Like degree Si,
The number that different images to be recognized X and standard form image Ti upper value are the point of " 1 " simultaneously is different,
So the ratio is different.Rejection threshold value λ is set, if Si < λ, determinesIf Si >=λ, X ∈ Ti is determined.If
There is Si < λ for all templates, then determines that the quality of the components is problematic, as illustrated in fig. 1 and 2, defective in quality zero
Component is transferred directly to transmission end of tape recovery processing；If having Si >=λ for multiple template, the highest conduct of matching degree is taken
Output is completed components type and is differentiated.As shown in Figures 2 and 3, large parts directly pushes photoelectricity by the classification of transmission end of tape
Sensor and classification driving means cooperation are classified, and are then delivered to by big components sort pass band and are put region.It is small-sized
Components are then rotated a certain angle by the pedestal of No. two manipulators to specified identification region, then will draw thereon small-sized zero
Part classification is put down.
The present invention uses No.1 manipulator and No. two manipulator cooperatings, fast and accurately can not only guide and position
Tested part, realizes flexible to various sizes auto parts and components open defect and precision recognition detection, while cooperating master control meter
Quick-witted energy analysis process system is calculated, intelligent decision and sorting can be carried out to the information of acquisition, and do not need because of element size
It is different and replace different manipulator fixtures, save cost.Small parts are drawn using sucking disc type mechanical hand simultaneously
Method, neither will cause imaging interference, component surface will not be caused crawl damage, and reduce standard machinery hand for
The difficulty of different components crawl programmings, strong operability.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily expect change or replacement,
It should be included within the scope of protection of the present invention.
Claims (7)
1. a kind of components based on machine vision identify method for separating, which comprises the following steps:
S1 collecting sample part diagram picture carries out feature extraction to sample part diagram picture and establishes components template library；
S2 judges whether components to be detected reach detection zone, and judges the attribute of components, specifically includes size, magnetism
It is and non magnetic, wherein small components refer to that size justifies the components being formed by within area, and weight in diameter for 200mm
Less than 1000N, remaining is big components；
S3 carries out Image Acquisition and feature extraction to components to be detected: big components directly carry out Image Acquisition and feature mentions
It takes；Magnetic small components carry out Image Acquisition and feature extraction after drawing using magnetic chuck；Non magnetic small components are using true
Suction disk carries out Image Acquisition and feature extraction after drawing again；Wherein, Image Acquisition uses the manipulator equipped with industrial camera real
It is existing；The absorption of small components is realized using another manipulator equipped with magnetic chuck, vacuum chuck and industrial camera；
S4 matches part diagram picture to be detected with template library image, confirms components type and quality；
S5 will confirm that the components of type and quality are delivered to corresponding storing region, complete the classification of components.
2. the components according to claim 1 based on machine vision identify method for separating, it is characterised in that: the step
The Image Acquisition condition and feature extraction mode of sample components and components to be detected in S1 and S3 are all the same.
3. the components according to claim 2 based on machine vision identify method for separating, it is characterised in that: the step
Annular light source is additionally provided on the manipulator of Image Acquisition in S3.
4. the components according to claim 3 based on machine vision identify method for separating, it is characterised in that: the step
The detailed process of S5 are as follows: qualified big components reach big components specification area with conveyer belt, and by corresponding big components
The transmission of sort pass band；Qualified small components are directly placed by another manipulator；Defect ware with conveyer belt reach end into
Row recycling.
5. a kind of components based on machine vision identify separation system, it is characterised in that: including main control computer, conveyer belt,
Sensor, No.1 manipulator, No. two manipulators, small components classification area, classification push sensor, classification push photoelectric sensing
Device, big components sort pass band are set with area to be tested and big components processing region, the sensing on the conveyer belt
Device setting is arranged in area to be tested, the No.1 manipulator and No. two manipulators in conveyor belt two sides, is respectively positioned on area to be detected
Domain, the No.1 manipulator is for acquiring components image information, and No. two manipulators are for drawing and putting small components；
The big components processing region is located at area to be tested downstream, and the big components processing region includes classification push sensing
Device, classification push photoelectric sensor and big components sort pass band, the classification push sensor, classification push photoelectric sensing
Device is arranged in conveyer belt side, and the big components sort pass band is located at the conveyer belt other side, pushes photoelectric sensing with classification
Device is corresponding, the conveyer belt, sensor, No.1 manipulator, No. two manipulators, classification push sensor and classification push photoelectric transfer
Sensor is controlled by main control computer.
6. the components according to claim 5 based on machine vision identify separation system, it is characterised in that: the transmission
Belt end is additionally provided with components recycling and processing device.
7. the components according to claim 5 based on machine vision identify separation system, it is characterised in that: the No.1
It is provided with industrial camera on manipulator, is provided with magnetic chuck, vacuum chuck and industrial camera on No. two manipulators.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710393224.7A CN107138431B (en) | 2017-05-27 | 2017-05-27 | A kind of components identification method for separating and system based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710393224.7A CN107138431B (en) | 2017-05-27 | 2017-05-27 | A kind of components identification method for separating and system based on machine vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107138431A CN107138431A (en) | 2017-09-08 |
| CN107138431B true CN107138431B (en) | 2019-11-08 |
Family
ID=59780854
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710393224.7A Active CN107138431B (en) | 2017-05-27 | 2017-05-27 | A kind of components identification method for separating and system based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107138431B (en) |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107886496A (en) * | 2017-09-30 | 2018-04-06 | 北京得华机器人技术研究院有限公司 | A kind of quality determining method of the auto-parts based on form fit |
| CN107570430B (en) * | 2017-10-26 | 2018-07-31 | 中国人民解放军国防科技大学 | Intelligent robot-based part sorting method in mechanical equipment maintenance process |
| CN107671008A (en) * | 2017-11-13 | 2018-02-09 | 中国科学院合肥物质科学研究院 | A kind of part stream waterline automatic sorting boxing apparatus of view-based access control model |
| CN107831176B (en) * | 2017-12-04 | 2023-10-27 | 苏州斯莱克精密设备股份有限公司 | Slender soft part detection device |
| CN108176607A (en) * | 2017-12-13 | 2018-06-19 | 全椒祥瑞塑胶有限公司 | A kind of injecting products quality efficient detection method |
| CN108480240A (en) * | 2018-03-19 | 2018-09-04 | 李晓 | A kind of pearl intelligent sorting system based on machine vision |
| CN109013375B (en) * | 2018-05-31 | 2021-01-01 | 郑州轻工业学院 | Detection device and detection method for angle steel of electric power iron tower |
| CN109746931A (en) * | 2019-03-13 | 2019-05-14 | 厦门中高智能电器科学研究院有限公司 | A kind of polymorphic type foreign matter removing robot towards GIS cavity |
| CN109927030A (en) * | 2019-03-28 | 2019-06-25 | 云南国土资源职业学院 | A kind of robot visual orientation method of appearance components detection |
| CN110308151A (en) * | 2019-07-22 | 2019-10-08 | 重庆大学 | A kind of surface of workpiece defect identification method and device based on machine vision |
| CN111366584B (en) * | 2020-04-09 | 2023-01-03 | 安徽科技学院 | Method and equipment for classifying damage of recyclable parts of scraped car based on visual identification |
| CN113510132A (en) * | 2020-04-10 | 2021-10-19 | 广东鸿山环境集团有限公司 | Intelligent garbage treatment and classification system |
| CN111553445B (en) * | 2020-05-20 | 2024-04-19 | 北京三一智造科技有限公司 | Component identification method and device, storage medium and electronic equipment |
| CN112699743A (en) * | 2020-12-15 | 2021-04-23 | 黄冈格罗夫氢能汽车有限公司 | System for identifying hydrogen energy fuel cell vehicle indicator lamp and button switch by one key |
| CN112707162A (en) * | 2020-12-17 | 2021-04-27 | 东莞市李群自动化技术有限公司 | Feeding method and multi-channel feeding machine |
| CN114455312A (en) * | 2021-12-27 | 2022-05-10 | 广州优尼冲压有限公司 | Replaceable visual discharging system |
| CN115945404B (en) * | 2023-02-15 | 2023-06-27 | 苏州天准科技股份有限公司 | Defect detection equipment and detection method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4678920A (en) * | 1985-06-17 | 1987-07-07 | General Motors Corporation | Machine vision method and apparatus |
| CN202516964U (en) * | 2011-12-22 | 2012-11-07 | 浙江金刚汽车有限公司 | Plate element procedure turning mechanical hand |
| CN102974551A (en) * | 2012-11-26 | 2013-03-20 | 华南理工大学 | Machine vision-based method for detecting and sorting polycrystalline silicon solar energy |
| CN103706575A (en) * | 2013-12-31 | 2014-04-09 | 江苏大学 | Device and method for grading and sorting lenses on line based on two-stage image acquisition |
| CN105953138A (en) * | 2015-12-31 | 2016-09-21 | 广东工业大学 | Machine vision light source device with light spot shape controllable and implementing method thereof |
| CN106216256A (en) * | 2016-07-04 | 2016-12-14 | 佛山科学技术学院 | A kind of air spring rod multiple position automatic checkout equipment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN205600756U (en) * | 2016-04-28 | 2016-09-28 | 扬州大学 | Optic platform snatchs manipulator |
-
2017
- 2017-05-27 CN CN201710393224.7A patent/CN107138431B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4678920A (en) * | 1985-06-17 | 1987-07-07 | General Motors Corporation | Machine vision method and apparatus |
| CN202516964U (en) * | 2011-12-22 | 2012-11-07 | 浙江金刚汽车有限公司 | Plate element procedure turning mechanical hand |
| CN102974551A (en) * | 2012-11-26 | 2013-03-20 | 华南理工大学 | Machine vision-based method for detecting and sorting polycrystalline silicon solar energy |
| CN103706575A (en) * | 2013-12-31 | 2014-04-09 | 江苏大学 | Device and method for grading and sorting lenses on line based on two-stage image acquisition |
| CN105953138A (en) * | 2015-12-31 | 2016-09-21 | 广东工业大学 | Machine vision light source device with light spot shape controllable and implementing method thereof |
| CN106216256A (en) * | 2016-07-04 | 2016-12-14 | 佛山科学技术学院 | A kind of air spring rod multiple position automatic checkout equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107138431A (en) | 2017-09-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107138431B (en) | A kind of components identification method for separating and system based on machine vision | |
| CN106053479B (en) | A kind of vision detection system of the workpiece appearance defects based on image procossing | |
| CN105891233B (en) | Lens surface defect intelligent checking system and its implementation based on machine vision | |
| CN109225915B (en) | A kind of rivet quality classification and Automated Sorting System and method | |
| CN103736672B (en) | A kind of eyeglass is classified sorting equipment online | |
| CN106204614A (en) | A kind of workpiece appearance defects detection method based on machine vision | |
| CN106872474A (en) | Exploded perspective vision testing machine and its detection method | |
| CN106000904A (en) | Automatic sorting system for household refuse | |
| Liu et al. | Automatic detection technology of surface defects on plastic products based on machine vision | |
| CN106290382A (en) | Bubble-cap tablet package defective vision detection device and method | |
| JP7055223B2 (en) | Mask defect inspection method | |
| CN105690393A (en) | Four-axle parallel robot sorting system based on machine vision and sorting method thereof | |
| CN109332211A (en) | A kind of part size detection system and method based on image processing techniques | |
| CN112630230A (en) | Online surface defect detection method based on photometric stereo method | |
| CN114758236A (en) | Non-specific shape object identification, positioning and manipulator grabbing system and method | |
| CN106680293A (en) | Machine vision based motor rotor gib defect detection system and method | |
| CN111487192A (en) | Machine vision surface defect detection device and method based on artificial intelligence | |
| CN114346821A (en) | Workpiece processing method and system and readable storage medium | |
| CN113744247A (en) | PCB welding spot defect identification method and system | |
| CN111330874A (en) | Detection device and detection method for pollution or impurity defects of bottom area of medicine bottle | |
| Moirogiorgou et al. | Intelligent robotic system for urban waste recycling | |
| CN109668897A (en) | The vision detection system and its detection method of cyclic annular precision component surface micro defect | |
| CN107246841A (en) | A kind of multiple views grouping system device of auto parts and components | |
| CN203069148U (en) | Finished magnetic ring image automatic detection system | |
| CN106364903A (en) | Monocular three-dimensional vision sorting method for stacked workpieces |
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 |