CN110261401A - A kind of industrial vision detection system - Google Patents
A kind of industrial vision detection system Download PDFInfo
- Publication number
- CN110261401A CN110261401A CN201910684117.9A CN201910684117A CN110261401A CN 110261401 A CN110261401 A CN 110261401A CN 201910684117 A CN201910684117 A CN 201910684117A CN 110261401 A CN110261401 A CN 110261401A
- Authority
- CN
- China
- Prior art keywords
- image
- camera
- product
- algorithm
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/89—Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles
- G01N21/892—Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles characterised by the flaw, defect or object feature examined
Landscapes
- Engineering & Computer Science (AREA)
- Textile Engineering (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of industrial vision detection system, including industrial camera module, image processing unit module and human-computer interface module, the industrial camera module is interconnected by image processing unit module and the human-computer interface module；Present system component is simplified, and draft normization is high, can be replicated rapidly；The influence to producing line technique is reduced, without pause, positioning product, dragging camera；It is few to reduce cost of labor, Material Cost, occupied space；Model calibration is easy to operate, low to operation technological requirement.
Description
Technical field
The present invention relates to industrial production line vision detection technologies, and in particular to a kind of assembly line product equipment defect knowledge
Other artificial intelligence technology.
Background technique
With the continuous promotion of production automation integrated level, vision-based detection is generally applied to product production line production, dress
In matching or packing；Vision detection technology, which refers to using technological means, replaces human eye to measure come the product produced to production line
And judgement.
Industrial vision detection device on the market at present, is generally required and is stopped using mechanical device to product, positioned,
And many light source equipment auxiliary lightings are installed, it then drives camera to each point by mechanical electric system, gradually acquires product figure
Picture recycles traditional images algorithm to be detected, and entire conceptual design is at a workstation form；There are following for such form
Disadvantage:
(1) intrusive influence is caused on producing line, is embodied in the interference to original technique beat, and the occupancy to producing line space；
In order to adapt to multiple product detection demand and increase camera quantity simply can bring it is too fat to move on device structure.
(2) light, Mechatronic Systems integrated complex, design, manufacture debugging cost are high, and high failure rate；Increase machinery positioning dress
It sets and generally requires change line sheet, considerably increase project construction difficulty and cost；
(3) system operatio difficulty is big, changes the line of production and generally requires manual setting calibration detection target when remodeling, in addition such work station is set
Standby flexible poor, transplantability is poor, is difficult to take into account the different product of shape.
Summary of the invention
In order to solve the above technical problems, it is an object of that present invention to provide a kind of industrial vision detection systems.
In order to achieve the above objectives, technical scheme is as follows: a kind of industrial vision detection system, including industrial camera
Module, image processing unit module and human-computer interface module, the industrial camera module by image processing unit module with
The human-computer interface module interactive connection.
Further, described image processing unit module handles video flowing, and comprehensive each frame result exports robust
Property preferably finally identification report, specifically contain following steps:
S1. the acquisition of training data sample: above-mentioned detection target is marked in the position of workpiece image and is cut out；Screening has certain
The sample image of difference is stored, while sending it to server, using as used in training cloud recognizer, data are adopted
Collection will embody enough diversity, such as different shadow environment, target morphology, background characteristics, surface curvature and shooting angle
Deng；
S2. it post-processes logical algorithm: AI algorithm being merged or spliced in the recognition result of multiframe, complete product ruler is obtained
The final recognition result for spending range, to export Finalise report；
S3. the intelligence of demarcation flow: demarcating for the template of same line product different model, for each type product,
Worker only needs to set calibration mode for system by man-machine interface, and then standard sample is placed in producing line from viewing field of camera
In flow through, system just learns model target type to be detected and point automatically, and is stored in model template；
S4. product triggering and model switching: when detecting, the triggering of each product can be using 2D and photoelectricity trigger switch
In conjunction with scheme, the scheme that combines of 2D camera image trigger method or 3D and 2D camera；And model is switched online, then it can be with
It is realized using characteristics of image triggering or barcode scanning, then calls corresponding model template, to realize that flexibility detects.
Further, the 2D camera image trigger method is that product feature is judged using algorithm to realize triggering meter
Number, the product feature include the verification of 2D characteristics of image, the verification of pipeline background, light stream Stren gsth test etc..
Further, 3D the and 2D camera association schemes are to pass through the life of structured light device using 3D structured light technique
At tens of thousands of a light points object and environment are scanned, and be imaged, then by implanting cloud PCL algorithm and a range of triangle
After algorithm calculates, the 3D depth image of accurate object and environment is obtained.
The beneficial effects of the present invention are: (1) system unit is simplified, draft normization is high, can replicate rapidly；
(2) influence to producing line technique is reduced, without pause, positioning product, dragging camera；
(3) it is few that cost of labor, Material Cost, occupied space are reduced.
(4) model calibration is easy to operate, low to operation technological requirement.
Detailed description of the invention
In figure: Fig. 1 is the work flow diagram of present system.
Fig. 2 is 2D camera image trigger method algorithm flow of the present invention.
Fig. 3 is 3D and 2D camera association schemes algorithm flow of the present invention.
Specific embodiment
Below in conjunction with the embodiment of the present invention, technical solution in the embodiment of the present invention understands, completely into flexible hand row
Ground description.
In conjunction with shown in Fig. 1 to Fig. 3: a kind of industrial vision detection system, including industrial camera module, image processing unit mould
Block and human-computer interface module, which is characterized in that the industrial camera module by image processing unit module with it is described man-machine
Interface module interactive connection.
Described image processing unit module handles video flowing, and comprehensive each frame result exports robustness preferably most
Identification report eventually, specifically contains following steps:
S1. the acquisition of training data sample: above-mentioned detection target is marked in the position of workpiece image and is cut out；Screening has certain
The sample image of difference is stored, while sending it to server, using as used in training cloud recognizer, data are adopted
Collection will embody enough diversity, such as different shadow environment, target morphology, background characteristics, surface curvature and shooting angle
Deng；
S2. it post-processes logical algorithm: AI algorithm being merged or spliced in the recognition result of multiframe, complete product ruler is obtained
The final recognition result for spending range, to export Finalise report；
S3. the intelligence of demarcation flow: demarcating for the template of same line product different model, for each type product,
Worker only needs to set calibration mode for system by man-machine interface, and then standard sample is placed in producing line from viewing field of camera
In flow through, system just learns model target type to be detected and point automatically, and is stored in model template；
S4. product triggering and model switching: when detecting, the triggering of each product can be using 2D and photoelectricity trigger switch
In conjunction with scheme, the scheme that combines of 2D camera image trigger method or 3D and 2D camera；And model is switched online, then it can be with
It is realized using characteristics of image triggering or barcode scanning, then calls corresponding model template, to realize that flexibility detects.
The 2D camera image trigger method, which is available with, judges product feature to realize flip-flop number；Software algorithm
Come first by feature extraction (the 2D characteristics of image of product verifies, pipeline background verification, light stream Stren gsth test etc.) algorithm
Judge whether product enters viewing field of camera, work and count to trigger vision system, 2D characteristics of image algorithm is available but not
It is limited to edge extracting, angle point, straight line, circular arc, specific template matching, SIFT/SURF feature；Pipeline background verification be judge it is defeated
Whether some background characteristics (texture, profile, color) of line sending are blocked；Light stream Stren gsth test is each picture calculated in visual field
Element is along the velocity vector in pipeline direction, and by three, (2D characteristics of image verifies the decision of final image triggering algorithm, conveying
The verification of line background, light stream Stren gsth test) it is obtained jointly based on certain weight distribution, this scheme can save trigger sensor,
It relates to system more to simplify, to the more demanding of image algorithm.
3D the and 2D camera association schemes are that height measuring and calculating, locations of contours parting, flip-flop number etc. can be achieved at the same time often
See functional requirement, and cost is relatively low；Its technical characterstic is, using 3D structured light technique, the number generated by structured light device
Ten thousand light points are scanned object and environment, and are imaged, then by implanting a cloud PCL algorithm and range of triangle algorithm meter
After calculation, the 3D depth image of accurate object and environment is obtained；The scheme that 2D+3D sensor combines needs to calculate height very much
Degree, locations of contours parting, flip-flop number application have apparent performance advantage and cost advantage.
The beneficial effects of the present invention are: (1) system unit is simplified, draft normization is high, can replicate rapidly；
(5) influence to producing line technique is reduced, without pause, positioning product, dragging camera；
(6) it is few that cost of labor, Material Cost, occupied space are reduced.
Model calibration is easy to operate, low to operation technological requirement.
It although an embodiment of the present invention has been shown and described, for the ordinary skill in the art, can be with
A variety of variations, modification, replacement can be carried out to these embodiments without departing from the principles and spirit of the present invention by understanding
And deformation, the scope of the present invention by.
Claims (4)
1. a kind of industrial vision detection system, including industrial camera module, image processing unit module and human-computer interface module,
It is characterized in that, the industrial camera module is interconnected by image processing unit module and the human-computer interface module.
2. a kind of industrial vision detection system according to claim 1, which is characterized in that described image processing unit module
Video flowing is handled, and comprehensive each frame result output robustness preferably finally identification report, specifically contains following step
It is rapid:
S1. the acquisition of training data sample: above-mentioned detection target is marked in the position of workpiece image and is cut out；Screening has certain
The sample image of difference is stored, while sending it to server, using as used in training cloud recognizer, data are adopted
Collection will embody enough diversity, such as different shadow environment, target morphology, background characteristics, surface curvature and shooting angle
Deng；
S2. it post-processes logical algorithm: AI algorithm being merged or spliced in the recognition result of multiframe, complete product ruler is obtained
The final recognition result for spending range, to export Finalise report；
S3. the intelligence of demarcation flow: demarcating for the template of same line product different model, for each type product,
Worker only needs to set calibration mode for system by man-machine interface, and then standard sample is placed in producing line from viewing field of camera
In flow through, system just learns model target type to be detected and point automatically, and is stored in model template；
S4. product triggering and model switching: when detecting, the triggering of each product can be using 2D and photoelectricity trigger switch
In conjunction with scheme, the scheme that combines of 2D camera image trigger method or 3D and 2D camera；And model is switched online, then it can be with
It is realized using characteristics of image triggering or barcode scanning, then calls corresponding model template, to realize that flexibility detects.
3. a kind of industrial vision detection system according to claim 2, which is characterized in that the 2D camera image triggering
Scheme is to judge product feature using algorithm to realize flip-flop number, and the product feature includes the verification of 2D characteristics of image, conveying
The verification of line background, light stream Stren gsth test etc..
4. a kind of industrial vision detection system according to claim 2, which is characterized in that 3D and 2D camera combination side
Case is scanned by tens of thousands of a light points that structured light device generates to object and environment using 3D structured light technique,
And it is imaged, then after being calculated by implanting a cloud PCL algorithm and range of triangle algorithm, the 3D for obtaining accurate object and environment is deep
Spend image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910684117.9A CN110261401A (en) | 2019-07-26 | 2019-07-26 | A kind of industrial vision detection system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910684117.9A CN110261401A (en) | 2019-07-26 | 2019-07-26 | A kind of industrial vision detection system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110261401A true CN110261401A (en) | 2019-09-20 |
Family
ID=67912105
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910684117.9A Pending CN110261401A (en) | 2019-07-26 | 2019-07-26 | A kind of industrial vision detection system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110261401A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111629205A (en) * | 2020-07-28 | 2020-09-04 | 天津美腾科技股份有限公司 | System and method applied to industrial camera simulation test |
| CN113252665A (en) * | 2021-04-27 | 2021-08-13 | 深圳市安仕新能源科技有限公司 | Product testing method and device, electronic equipment and storage medium |
| CN113703722A (en) * | 2021-07-20 | 2021-11-26 | 南京邮电大学 | Graphical 3D camera management system and working method thereof |
| AT524101A1 (en) * | 2020-07-27 | 2022-02-15 | Tgw Logistics Group Gmbh | System and method for registering a number of goods in a load carrier |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201819669U (en) * | 2010-09-17 | 2011-05-04 | 佛山市迈迅机电有限公司 | Vision inspection system for ceramic tiles |
| CN102305950A (en) * | 2011-04-29 | 2012-01-04 | 无锡众望四维科技有限公司 | Method for automatically detecting missed accessory during automatic assembly of injector by using machine visual system |
| CN103091332A (en) * | 2013-01-16 | 2013-05-08 | 浙江科技学院 | Detection method and detection system of U-shaped powder pipe based on machine vision |
| CN103286081A (en) * | 2013-05-07 | 2013-09-11 | 浙江工业大学 | Monocular multi-perspective machine vision-based online automatic sorting device for steel ball surface defect |
| CN105486684A (en) * | 2016-01-13 | 2016-04-13 | 广东产品质量监督检验研究院 | Non-destructive testing method for mahogany furniture wood varieties based on network platform |
| CN106770349A (en) * | 2016-11-24 | 2017-05-31 | 四川吉盛印铁有限公司 | A kind of tinplate dual sided quality detection method |
| CN106872487A (en) * | 2017-04-21 | 2017-06-20 | 佛山市南海区广工大数控装备协同创新研究院 | The surface flaw detecting method and device of a kind of view-based access control model |
| CN107607540A (en) * | 2017-07-26 | 2018-01-19 | 广东工业大学 | A kind of T-shirt on-line checking and method for separating based on machine vision |
| CN109934809A (en) * | 2019-03-08 | 2019-06-25 | 深慧视(深圳)科技有限公司 | A kind of paper labels character defect inspection method |
-
2019
- 2019-07-26 CN CN201910684117.9A patent/CN110261401A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201819669U (en) * | 2010-09-17 | 2011-05-04 | 佛山市迈迅机电有限公司 | Vision inspection system for ceramic tiles |
| CN102305950A (en) * | 2011-04-29 | 2012-01-04 | 无锡众望四维科技有限公司 | Method for automatically detecting missed accessory during automatic assembly of injector by using machine visual system |
| CN103091332A (en) * | 2013-01-16 | 2013-05-08 | 浙江科技学院 | Detection method and detection system of U-shaped powder pipe based on machine vision |
| CN103286081A (en) * | 2013-05-07 | 2013-09-11 | 浙江工业大学 | Monocular multi-perspective machine vision-based online automatic sorting device for steel ball surface defect |
| CN105486684A (en) * | 2016-01-13 | 2016-04-13 | 广东产品质量监督检验研究院 | Non-destructive testing method for mahogany furniture wood varieties based on network platform |
| CN106770349A (en) * | 2016-11-24 | 2017-05-31 | 四川吉盛印铁有限公司 | A kind of tinplate dual sided quality detection method |
| CN106872487A (en) * | 2017-04-21 | 2017-06-20 | 佛山市南海区广工大数控装备协同创新研究院 | The surface flaw detecting method and device of a kind of view-based access control model |
| CN107607540A (en) * | 2017-07-26 | 2018-01-19 | 广东工业大学 | A kind of T-shirt on-line checking and method for separating based on machine vision |
| CN109934809A (en) * | 2019-03-08 | 2019-06-25 | 深慧视(深圳)科技有限公司 | A kind of paper labels character defect inspection method |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AT524101A1 (en) * | 2020-07-27 | 2022-02-15 | Tgw Logistics Group Gmbh | System and method for registering a number of goods in a load carrier |
| CN111629205A (en) * | 2020-07-28 | 2020-09-04 | 天津美腾科技股份有限公司 | System and method applied to industrial camera simulation test |
| CN113252665A (en) * | 2021-04-27 | 2021-08-13 | 深圳市安仕新能源科技有限公司 | Product testing method and device, electronic equipment and storage medium |
| CN113252665B (en) * | 2021-04-27 | 2023-02-10 | 深圳市安仕新能源科技有限公司 | Product testing method and device, electronic equipment and storage medium |
| CN113703722A (en) * | 2021-07-20 | 2021-11-26 | 南京邮电大学 | Graphical 3D camera management system and working method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110261401A (en) | A kind of industrial vision detection system | |
| US20210286339A1 (en) | Automated 360-degree dense point object inspection | |
| EP3963414A2 (en) | Automated 360-degree dense point object inspection | |
| CN107084992B (en) | Capsule detection method and system based on machine vision | |
| JP5671281B2 (en) | Position / orientation measuring apparatus, control method and program for position / orientation measuring apparatus | |
| CN110214120B (en) | Belt inspection system and method | |
| CN106770321B (en) | A kind of plastic part defect inspection method based on multi-threshold section | |
| CN102529019B (en) | Method for mould detection and protection as well as part detection and picking | |
| CN107470170A (en) | PCB detection sorting systems and method based on machine vision | |
| CN102441581A (en) | Machine vision-based device and method for online detection of structural steel section size | |
| CN109712114A (en) | A kind of system and its detection method applied to zipper defects detection | |
| CN111539927A (en) | Detection process and algorithm of automobile plastic assembly fastening buckle lack-assembly detection device | |
| CN115144399A (en) | Assembly quality detection method and device based on machine vision | |
| Liguori et al. | An on-line stereo-vision system for dimensional measurements of rubber extrusions | |
| CN110763137A (en) | Method and device for determining product posture | |
| CN208555097U (en) | A kind of workpiece shapes based on computer vision detection and sorting device | |
| CN109701890A (en) | Magnetic tile surface defect detection and method for sorting | |
| Schreiber et al. | Online visual quality inspection for weld seams | |
| US10228239B2 (en) | Measuring apparatus, measuring method, and article manufacturing method | |
| CN102529020B (en) | Visual-motor integrated device for mould detection and protection as well as part detection and picking | |
| Koodtalang et al. | A design of automated inspections of both shape and height simultaneously based on stereo vision and plc | |
| CN115608643A (en) | Industrial detection method and system of industrial Internet | |
| US11748838B2 (en) | System and method for three-dimensional scan of moving objects longer than the field of view | |
| US11132579B2 (en) | Contour recognition device, contour recognition system and contour recognition method | |
| CN109063688A (en) | LED status recognition methods based on intelligent inspection robot |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190920 |
|
| WD01 | Invention patent application deemed withdrawn after publication |