CN107894209A - A kind of dip roll Coating thickness machine vision detection method - Google Patents
A kind of dip roll Coating thickness machine vision detection method Download PDFInfo
- Publication number
- CN107894209A CN107894209A CN201710942080.6A CN201710942080A CN107894209A CN 107894209 A CN107894209 A CN 107894209A CN 201710942080 A CN201710942080 A CN 201710942080A CN 107894209 A CN107894209 A CN 107894209A
- Authority
- CN
- China
- Prior art keywords
- dip roll
- profile
- machine vision
- coating thickness
- detection method
- 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
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
- G01B11/06—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness for measuring thickness ; e.g. of sheet material
- G01B11/0616—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness for measuring thickness ; e.g. of sheet material of coating
Abstract
The embodiment of the invention discloses a kind of dip roll Coating thickness machine vision detection method, comprise the following steps：Gather side image during dip roll inactive state；It is area-of-interest to select the dip roll run up on direction of the side image and ullage subregion, and extracts the profile of the dip roll of ullage；When dip roll is run, the viscous extract body edge contour of the area-of-interest is obtained；Merge the profile of the dip roll and the viscous extract body edge contour, demarcate measurement point, make the measurement line with the central coordinate of circle of the profile of the dip roll, calculate the distance between the profile of the dip roll and described viscous extract body edge contour.Using the present invention, utilize mechanical vision inspection technology and image processing techniques, it is the defects of overcoming the method for offline inspection to feed back coating layer thickness information in real time, more accurate compared to conventional offline Thickness sensitivity, more efficient by Coating thickness machine vision on-line checking.
Description
Technical field
The present invention relates to coating field of compounding and field of machine vision, more particularly to a kind of dip roll Coating thickness machine
Device visible detection method.
Background technology
Print roll coating is a kind of technology that solvent-free coatable masking liquid is overlying on to substrate surface, is solvent-free compound key step
Suddenly, it is solvent-free compound because being gained the name in whole recombination process without using volatile organic solvent, it is one internationally recognized green
Color combination process, it is a kind of important processing method of package compound film.This patent is with domestic the most frequently used leaching roller at present
Coating machine is object, and Coating thickness is detected by the method for machine vision.
Coating thickness is that an important indicator of coating process is also the final embodiment of coating quality.Dip roll is coated with
Coat thickness detection is domestic at present to use offline inspection such as always：Weight method, Electronic Speculum detection etc..The method of offline inspection can not be real
When feedback coating layer thickness information.
The content of the invention
Technical problem to be solved of the embodiment of the present invention is, there is provided a kind of dip roll Coating thickness machine vision
Detection method.Using mechanical vision inspection technology and image processing techniques, the on-line automatic of coating layer thickness have effectively achieved
Detection.
In order to solve the above-mentioned technical problem, the embodiments of the invention provide a kind of dip roll Coating thickness machine vision
Detection method, comprise the following steps：
S1：Gather side image during dip roll inactive state；
S2：It is interested to select the dip roll run up on direction of the side image and ullage subregion
Region, and extract the profile of the dip roll of ullage；
S3：When dip roll is run, the viscous extract body edge contour of the area-of-interest is obtained；
S4：Merge the profile of the dip roll and the viscous extract body edge contour,
S5：Measurement point is demarcated, the measurement line with the central coordinate of circle of the profile of the dip roll is made, calculates the dip roll
Profile and the distance between the viscous extract body edge contour.
Further, the step of also being demarcated before step S1 including the use of Zhang Zhengyou scaling methods to camera.
Further, the method that step S2, step S3 extracts profile including the use of Canny detective operators.
Further, step S2, step S3 is also included the method for adjacent detection contour linkage.
Further, step S2 also includes the round method that carried out curve fitting to profile.
Implement the embodiment of the present invention, have the advantages that：The present invention is utilized at mechanical vision inspection technology and image
Reason technology, by Coating thickness machine vision on-line checking, coating can not be fed back in real time by overcoming the method for offline inspection
It is the defects of thickness information, more accurate compared to conventional offline Thickness sensitivity, more efficient.
Brief description of the drawings
Fig. 1 is the image when present invention collects dip roll inactive state；
Fig. 2 is the schematic diagram in the region interested to the present invention；
Fig. 3 is the schematic diagram of the profile of obtained dip roll；
Fig. 4 is the schematic diagram of the viscous extract body edge contour of obtained dip roll；
Fig. 5 is the schematic diagram that the profile of dip roll merges with viscous extract body edge contour；
Fig. 6 is the schematic diagram for measuring extracting liq thickness.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, the present invention is made into one below in conjunction with accompanying drawing
It is described in detail on step ground.
A kind of dip roll Coating thickness machine vision detection method of the embodiment of the present invention, is entered by following steps
OK.
1st, camera calibration is carried out first, camera is demarcated using Zhang Zhengyou scaling methods, and this method template construct holds
Easily, easy to use, cost is low, and robustness is good, and accuracy rate is high.Post-equalization camera distortion is demarcated, while obtains inside and outside demarcation ginseng
Pixel distance, physical distance can be converted into by inside and outside parameter by number.
2nd, image is gathered when roller remains static, as shown in Figure 1.
3rd, in view of the refraction problems of light under liquid level, it is area-of-interest to manually select ullage partial roller region,
And the sub- profile of liquid level upper rolls is extracted with Canny detective operators, as shown in Figure 2.
4th, because Canny detective operators, the contour line of detection is many discrete curves, so needing adjacent detection wheel
Exterior feature connection.
First using relatively long profile as reference contours, the recurrence line segment of two profiles is fitted, another returns line end
To on the tropic of reference contours, reference contours end points also projects on the tropic spot projection, and the distance between two projections are just
It is the absolute distance of another profile.One threshold distance is set when merging profile, cannot be merged more than threshold value.
5th, due to extracting edge and the out-of-flatness of profile, and many small projections and breakpoint be present, if directly detecting meeting
Cause flase drop.So need the contour curve fitting circle according to extraction.The design uses principle of least square fitting circle.Fitting
Specific method is as follows：
With RiFor measuring point radius, θiIt is the phase angle of point position.Now, the rectangular co-ordinate (X of measuring pointi,Yi) be：
Xi=Ricosθi；Yi=Risinθi.Fitting circle schematic diagram assumes that the preferable center of circle is (X0, Y0), radius of circle R, builds
Founding deviation object function is：
So as to obtain：
It can show that two multiply and estimate calculated value and be：
The circle is roll profile, while obtains fitting circle central coordinate of circle (XR,YR) and radius information R, as shown in Figure 3.
6th, controlled motor moves roller, gathers picture now.
7th, it is area-of-interest equally to manually select ullage, and Canny detective operators extracting liqs are used in region
Edge contour.
8th, equally needed because the contour line of Canny detective operators detection is many discrete curves adjacent inspection
Survey contour linkage.Connection method is identical with step 4, and connection rear profile is as shown in Figure 4.
9th, two curves of above-mentioned fitting are drawn, and are filled with the grey of pixel value 128.
10th, look for a suitable measurement point manually in the right of figure below right side graph, and record the coordinate of the point.With fitting
Central coordinate of circle (XR,YR) 2 points of generation straight lines, using the straight line as measuring line.Operator, interpolation method are measured using Halcon
Neighbour's sampling is selected, edge is formed edge pair by simultaneous selection, you can obtains the intermarginal pixel distance D in both sides.
The inside and outside parameter obtained when the 11st, according to camera calibration, the pixel distance D of measurement is converted into physical distance D ＇.
The above disclosed power for being only a kind of preferred embodiment of the present invention, the present invention can not being limited with this certainly
Sharp scope, therefore the equivalent variations made according to the claims in the present invention, still belong to the scope that the present invention is covered.
Claims (5)
1. a kind of dip roll Coating thickness machine vision detection method, it is characterised in that comprise the following steps：
S1：Gather side image during dip roll inactive state；
S2：It is region of interest to select the dip roll run up on direction of the side image and ullage subregion
Domain, and extract the profile of the dip roll of ullage；
S3：When dip roll is run, the viscous extract body edge contour of the area-of-interest is obtained；
S4：Merge the profile of the dip roll and the viscous extract body edge contour,
S5：Measurement point is demarcated, the measurement line with the central coordinate of circle of the profile of the dip roll is made, calculates the wheel of the dip roll
The distance between wide and described viscous extract body edge contour.
2. dip roll Coating thickness machine vision detection method according to claim 1, it is characterised in that step S1
The step of also being demarcated before including the use of Zhang Zhengyou scaling methods to camera.
3. dip roll Coating thickness machine vision detection method according to claim 2, it is characterised in that step
The method of S2, step S3 including the use of Canny detective operators extraction profile.
4. dip roll Coating thickness machine vision detection method according to claim 3, it is characterised in that step
S2, step S3 are also included the method for adjacent detection contour linkage.
5. dip roll Coating thickness machine vision detection method according to claim 4, it is characterised in that step S2
Also include the round method that carried out curve fitting to profile.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710942080.6A CN107894209A (en) | 2017-10-11 | 2017-10-11 | A kind of dip roll Coating thickness machine vision detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710942080.6A CN107894209A (en) | 2017-10-11 | 2017-10-11 | A kind of dip roll Coating thickness machine vision detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107894209A true CN107894209A (en) | 2018-04-10 |
Family
ID=61803114
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710942080.6A Pending CN107894209A (en) | 2017-10-11 | 2017-10-11 | A kind of dip roll Coating thickness machine vision detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107894209A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3330961A (en) * | 1964-04-15 | 1967-07-11 | Eastman Kodak Co | Photoelectric skip detector for use with a viscous layer applicator |
| JPH04233405A (en) * | 1990-08-31 | 1992-08-21 | Windmoeller & Hoelscher | Method and apparatus for measuring thickness of paint film |
| US20040194699A1 (en) * | 2002-07-08 | 2004-10-07 | Bruce Nesbitt | Apparatus for simultaneously coating and measuring parts |
| CN102388288A (en) * | 2009-04-16 | 2012-03-21 | 利乐拉瓦尔集团及财务有限公司 | System and method of measuring the thickness of a layer of coating |
| CN104942401A (en) * | 2015-06-15 | 2015-09-30 | 中国地质大学（武汉） | Tube blank cold centering method based on binocular stereoscopic vision and tube blank cold centering device |
| CN107218895A (en) * | 2017-06-07 | 2017-09-29 | 深圳市华星光电技术有限公司 | A kind of optical thickness monitor device and monitoring method |
-
2017
- 2017-10-11 CN CN201710942080.6A patent/CN107894209A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3330961A (en) * | 1964-04-15 | 1967-07-11 | Eastman Kodak Co | Photoelectric skip detector for use with a viscous layer applicator |
| JPH04233405A (en) * | 1990-08-31 | 1992-08-21 | Windmoeller & Hoelscher | Method and apparatus for measuring thickness of paint film |
| US20040194699A1 (en) * | 2002-07-08 | 2004-10-07 | Bruce Nesbitt | Apparatus for simultaneously coating and measuring parts |
| CN102388288A (en) * | 2009-04-16 | 2012-03-21 | 利乐拉瓦尔集团及财务有限公司 | System and method of measuring the thickness of a layer of coating |
| CN104942401A (en) * | 2015-06-15 | 2015-09-30 | 中国地质大学（武汉） | Tube blank cold centering method based on binocular stereoscopic vision and tube blank cold centering device |
| CN107218895A (en) * | 2017-06-07 | 2017-09-29 | 深圳市华星光电技术有限公司 | A kind of optical thickness monitor device and monitoring method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104359403B (en) | Planar part dimension measurement method based on sub-pixel edge algorithm | |
| CN106651828B (en) | Method for measuring sub-pixel of product size under industrial small-scale motion blur imaging condition | |
| JP6620477B2 (en) | Method and program for detecting cracks in concrete | |
| CN102799867B (en) | Meter pointer angle identification method based on image processing | |
| CN104574393B (en) | A kind of three-dimensional pavement crack pattern picture generates system and method | |
| CN107945155B (en) | Toothpaste tube shoulder defect detection method based on Gabor filter | |
| CN102175700A (en) | Method for detecting welding seam segmentation and defects of digital X-ray images | |
| CN106996748A (en) | A kind of wheel footpath measuring method based on binocular vision | |
| CN104050446A (en) | Meter pointer image identification method based on pointer width character | |
| CN106500619B (en) | The camera internal imaging sensor installation error separation method that view-based access control model measures | |
| CN106569220A (en) | Laser ranging correction method and device applying same | |
| CN103279956A (en) | Method for detecting positional accuracy of components of chip mounter | |
| CN109489566A (en) | Lithium battery diaphragm material cuts checking method for width, detection system and device | |
| CN104331695B (en) | A kind of circle marker symbol shape quality detection method of robust | |
| CN106600592A (en) | Track long chord measurement method based on the splicing of continuous frame images | |
| CN105839505A (en) | Three-dimensional visual detection method and detection device for road surface damage information | |
| CN104517298A (en) | Moving object detector | |
| CN103606149A (en) | Method and apparatus for calibration of binocular camera and binocular camera | |
| CN113554667B (en) | Three-dimensional displacement detection method and device based on image recognition | |
| CN107421438A (en) | Film size detection control system and its method based on machine vision | |
| CN108871185B (en) | Method, device and equipment for detecting parts and computer readable storage medium | |
| CN107449485A (en) | Liquid level image identification method | |
| CN101494740B (en) | Calibration element for calibrating the magnification ratio of a camera, and a calibration method | |
| CN107367245B (en) | Invalid point detection and elimination method in optical three-dimensional profile measurement | |
| TW201931308A (en) | Video type water ruler scale detecting and identifying system and method |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180410 |