CN105301003A - Method for detecting bottle cap surface production date based on machine vision - Google Patents
Method for detecting bottle cap surface production date based on machine vision Download PDFInfo
- Publication number
- CN105301003A CN105301003A CN201510744035.0A CN201510744035A CN105301003A CN 105301003 A CN105301003 A CN 105301003A CN 201510744035 A CN201510744035 A CN 201510744035A CN 105301003 A CN105301003 A CN 105301003A
- Authority
- CN
- China
- Prior art keywords
- bottle cap
- image
- detection
- circle
- production date
- 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
Landscapes
- Image Analysis (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to a method for detecting a bottle cap surface production date based on machine vision. According to the method, an LED white ring light source illuminates the surface of a to-be-measured bottle cap. The method comprises the steps of collecting image information of a qualified bottle cap; collecting an image of a to-be-measured bottle cap; performing binarization processing on an image having undergone median filtering pre-processing; performing circle location, finding a bottle cap edge to serve as a round edge, and taking the circle center as an original point to establish a coordinate system; presetting the upper limit and the lower limit of the pixel point number of detection targets according to the pixel point number of sprayed production date codes on the qualified bottle cap, utilizing the round edge obtained in the step 4) and the coordinate system which has been established to set the inside of the circle as an region of interest (ROI), taking the sprayed production date codes as the detection targets, and performing bottle cap detection. A algorithm in use is simple, the steps are fewer, quick detection of the image is achieved, the detection efficiency is high, and the method is subjected to a small environmental influence and is high in accuracy.
Description
Technical field
The present invention relates to a kind of bottle cap surface production date online test method, belong to field of machine vision.
Background technology
In the industrial production line such as beverage, beer, mostly adopt the method for artificial visually examine to check that whether bottle is qualified.
During particularly bottle cap is produced, need to print the date of manufacture on bottle cap, have the situation of the fuzzy or drain spray of coding.
Because vision and limbs fatigue easily occur human eye, usually efficiency is very low, can cause undetected or flase drop, and cost of labor is also more and more higher.These substandard products come into the market the quality that consumer can be caused to distrust product, and impact is sold, and even cause the problem such as drink expired product health risk by mistake.
For the defect that above-mentioned manual detection exists, carry out the bottle cap detect delay based on machine vision, the detection system set up a set of optical imagery, image acquisition and Digital Image Processing and analyze as platform, not only there is theoretical foundation, and have very large economic worth, can guarantee that reducing and stop above-mentioned phenomenon in maximum magnitude occurs, and improves product quality.
Machine Vision Detection is in the existing application of a lot of detection field, also application precedent is had in bottle cap context of detection, but be all mostly (as patent CN102095733A to the Defect Detection of bottle cap Inner pad and outside surface, CN103884651A), these are all the detections for bottle cap presentation quality, and rarely have research for the detection of the coding information (as the date of manufacture) on bottle cap.In industrial production line, per secondly at least to detect more than 5 bottles, and need the existence detecting the coding of bottle cap date of manufacture, the feature such as integrality and clarity.This all has higher requirements to the real-time of detection system and software process.
Summary of the invention
The object of the invention is to, overcome the weak point of manual detection existence and the limitation of bottle cap detection, a kind of bottle cap surface production date online test method based on machine vision is provided.Technical scheme of the present invention is as follows:
Based on a bottle cap surface production date detection method for machine vision, the method LED white annular light source throw light on bottle cap to be measured surface, step is as follows:
1) gather the image information of qualified bottle cap, regulate camera and light source, the bottle cap background image making collection and the contrast of date of manufacture that it sprays reach maximum, so that follow-up process;
2) gather bottle cap image to be measured, medium filtering pre-service is carried out to gathered bottle cap image, detect target information to remove noise jamming and to strengthen;
3) adjust suitable threshold value according to actual environment, carry out binary conversion treatment to through the pretreated image of medium filtering, background and target are separated by image after binaryzation effectively;
4) to 3) image that obtains justifies location, and finding bottle cap edge as round edge, is that initial point sets up coordinate system with the center of circle, adopts the following step to carry out justifying location:
One, set an annular region, bottle cap edge is between inner ring and outer shroud.
Two, seek area is set in annular region, and search direction is from inner ring to outer shroud;
Three, with the inner ring center of circle for initial point, the scounting line of many fixed intervals is set between inner ring and outer shroud;
Four, when certain point-polarity on scounting line changes from gray-scale value 1 to gray-scale value 0, looked for frontier point is thought of as;
Five, the frontier point composition round edge found on each bar scounting line is utilized;
5) bound of the pixel number detecting target is preset according to the pixel number of the date of manufacture coding on qualified bottle cap, utilize step 4) round edge that obtains and the coordinate system set up, region of interest ROI is set as by circle, using date of manufacture coding in ROI as detection target, according to actual conditions, set the pixel size of suitable detection target, the impact of the irrelevant tiny pixel stain caused because of light intensity in filtering image, if it is qualified that the pixel number detecting target is in the bound of regulation, otherwise be judged to be substandard product.The invention has the beneficial effects as follows:
First, the method algorithm used is simple, and step is few, and achieve the quick detection of image, detection efficiency is high; Secondly, the method is affected by environment less, and accuracy rate is high; Use photoelectricity to trigger to take pictures, real-time is good; This detection method is novel, applied widely, all applicable for proper Operation of Industrial bottled Product checking; The method can be attached on existing production line, realizes the on-line checkingi that limit produces frontier inspection survey.The present invention is applicable to bottle cap surface production date high speed detection in automated production.
Accompanying drawing explanation
Fig. 1 is used for the device schematic diagram detected the bottle cap surface production date based on machine vision of the present invention.
Fig. 2 of the present invention based on machine vision to bottle cap surface production date detection method process flow diagram.
Fig. 3 (a) the inventive method original bottle cap image to be measured, the bottle cap image that (b) step 1 collects.
The qualified bottle cap result of determination figure that Fig. 4 (a) the inventive method detects, (b) defective bottle cap result of determination figure.
Embodiment
Below in conjunction with accompanying drawing and instantiation, the present invention will be further described.
See Fig. 1, a kind of bottle cap surface production date detection system device based on machine vision, functionally divides 5 modules, i.e. travelling belt 1, image capture module 2, optical trigger 3, lighting module 4, image procossing and identification 5.Above-mentioned image capture module 2 uses 1,200,000 pixel CCD camera, and lighting module 4 uses LED annular light source, and image procossing and identification module 5 use industrial computer.Bottle to be measured is transported to station of taking pictures by travelling belt 1, and whether described photoelectric trigger device 3 arrives picture-taking position for detecting bottle, to produce pulse signal being transported to industrial computer 5, industrial computer sends signal triggering camera 2 and takes pictures; Annular light source 4 is for providing illumination to improve picture quality when camera is taken pictures, CCD camera is positioned at directly over travelling belt 1, and takes pictures by the middle of annular light source, and the image of shooting is transported in industrial computer and processes.
See Fig. 2, of the present invention based on the process flow diagram of machine vision to bottle cap surface production date detection method.Step is as follows:
(1) industrial camera is placed into the station to be measured directly over production line workpiece conveyor, position to be measured is placed the product of qualified bottle cap, regulate the exposure compensating of camera own and annular light source brightness to make acquisition bottle cap background image and date of manufacture coding contrast reach maximum.
(2) start travelling belt, bottle to be measured arrives camera site with travelling belt, triggers camera and obtains bottle cap image to be measured, be sent to industrial computer image processing system through passage.
(3) first pre-service being carried out to image, through repeatedly testing, using medium filtering effect best, image object information enhancement after process.
(4) to image binaryzation after step (4) process, according to image grey level histogram, set suitable threshold value, improve image recognition rate.
(5) find bottle cap edge as round edge in image after binarization, and be that initial point sets up coordinate system with the center of circle.Method is as follows: the first step, take bottle cap as benchmark, sets an annular region, bottle cap edge is between inner ring and outer shroud.Second step, using annular region as find circle edge region, search direction is from inner ring to outer shroud; 3rd step, with the inner ring center of circle for starting point arranges many rays, between inner ring and outer shroud, intercept adjacent spaces is the scounting line that 10px line segment is done.4th step, when certain point-polarity on line segment changes from gray-scale value 1 to gray-scale value 0, is thought of as looked for frontier point.5th step, setting minimum boundary intensity is 20, and scounting line width is 3px, and rounded edge pixel is 3px, can obtain according to above condition with round edge scounting line put into frontier point composition.
(6) round edge will be found inner as processing region, and date of manufacture coding is as detection target.According to actual conditions, set the pixel size of suitable detection target, when detect number of targets be judged as in the threshold value of setting qualified, otherwise defective.
Shown in Fig. 3, left image is camera shooting original image, and right figure is the image gathered after adjusting camera exposure rate and light source, and it is simpler that the image after regulating like this makes subsequent treatment judge.
See Fig. 4, upper figure is the judged result of qualified bottle cap, can see that bottle outlet edge is the round edge found in figure, only detects the coding in circle when detecting target.Figure below is the testing result when date of manufacture is imperfect on bottle cap, has defect when the date of manufacture, all can be judged as defective time unintelligible.
Claims (1)
1. based on a bottle cap surface production date detection method for machine vision, the method LED white annular light source throw light on bottle cap to be measured surface, step is as follows:
1) gather the image information of qualified bottle cap, regulate camera and light source, the bottle cap background image making collection and the contrast of date of manufacture that it sprays reach maximum, so that follow-up process;
2) gather bottle cap image to be measured, medium filtering pre-service is carried out to gathered bottle cap image, detect target information to remove noise jamming and to strengthen;
3) adjust suitable threshold value according to actual environment, carry out binary conversion treatment to through the pretreated image of medium filtering, background and target are separated by image after binaryzation effectively;
4) to 3) image that obtains justifies location, and finding bottle cap edge as round edge, is that initial point sets up coordinate system with the center of circle, adopts the following step to carry out justifying location:
One, set an annular region, bottle cap edge is between inner ring and outer shroud.
Two, seek area is set in annular region, and search direction is from inner ring to outer shroud;
Three, with the inner ring center of circle for initial point, the scounting line of many fixed intervals is set between inner ring and outer shroud;
Four, when certain point-polarity on scounting line changes from gray-scale value 1 to gray-scale value 0, looked for frontier point is thought of as;
Five, the frontier point composition round edge found on each bar scounting line is utilized.
5) bound of the pixel number detecting target is preset according to the pixel number of the date of manufacture coding on qualified bottle cap, utilize step 4) round edge that obtains and the coordinate system set up, region of interest ROI is set as by circle, using date of manufacture coding in ROI as detection target, according to actual conditions, set the pixel size of suitable detection target, the impact of the irrelevant tiny pixel stain caused because of light intensity in filtering image, if it is qualified that the pixel number detecting target is in the bound of regulation, otherwise be judged to be substandard product.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510744035.0A CN105301003A (en) | 2015-11-05 | 2015-11-05 | Method for detecting bottle cap surface production date based on machine vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510744035.0A CN105301003A (en) | 2015-11-05 | 2015-11-05 | Method for detecting bottle cap surface production date based on machine vision |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105301003A true CN105301003A (en) | 2016-02-03 |
Family
ID=55198511
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510744035.0A Pending CN105301003A (en) | 2015-11-05 | 2015-11-05 | Method for detecting bottle cap surface production date based on machine vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105301003A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105865570A (en) * | 2016-06-15 | 2016-08-17 | 天津大学 | Machine vision-based glass bottle liquid level detection method |
CN105865569A (en) * | 2016-05-12 | 2016-08-17 | 天津大学 | Glass bottle liquid level detection method based on visible light communication |
CN106018412A (en) * | 2016-05-12 | 2016-10-12 | 天津大学 | Bottle cap surface production date detection method based on visible light communication |
CN106052793A (en) * | 2016-06-15 | 2016-10-26 | 天津大学 | Machine vision based liquid level sub-quality product marking method |
CN110108712A (en) * | 2019-04-20 | 2019-08-09 | 东莞中科蓝海智能视觉科技有限公司 | Multifunctional visual sense defect detecting system |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0070666A2 (en) * | 1981-07-17 | 1983-01-26 | W.R. Grace & Co. | Optical inspection system |
WO2003024859A1 (en) * | 2001-09-20 | 2003-03-27 | Krones Ag | Method for inspecting closures on containers |
CN102095733A (en) * | 2011-03-02 | 2011-06-15 | 上海大学 | Machine vision-based intelligent detection method for surface defect of bottle cap |
CN102519972A (en) * | 2011-12-10 | 2012-06-27 | 山东明佳包装检测科技有限公司 | Detection method of PET bottle cap and liquid level |
CN103604806A (en) * | 2013-12-04 | 2014-02-26 | 天津普达软件技术有限公司 | Method for detecting seal ring defect |
CN103606169A (en) * | 2013-12-04 | 2014-02-26 | 天津普达软件技术有限公司 | Method for detecting defects of bottle cap |
CN103604808A (en) * | 2013-12-04 | 2014-02-26 | 天津普达软件技术有限公司 | Visual detection method for bottle cap defects |
CN203512149U (en) * | 2013-08-23 | 2014-04-02 | 天津森茂科技有限公司 | Medicine package three-phase code image detecting system |
-
2015
- 2015-11-05 CN CN201510744035.0A patent/CN105301003A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0070666A2 (en) * | 1981-07-17 | 1983-01-26 | W.R. Grace & Co. | Optical inspection system |
WO2003024859A1 (en) * | 2001-09-20 | 2003-03-27 | Krones Ag | Method for inspecting closures on containers |
CN102095733A (en) * | 2011-03-02 | 2011-06-15 | 上海大学 | Machine vision-based intelligent detection method for surface defect of bottle cap |
CN102519972A (en) * | 2011-12-10 | 2012-06-27 | 山东明佳包装检测科技有限公司 | Detection method of PET bottle cap and liquid level |
CN203512149U (en) * | 2013-08-23 | 2014-04-02 | 天津森茂科技有限公司 | Medicine package three-phase code image detecting system |
CN103604806A (en) * | 2013-12-04 | 2014-02-26 | 天津普达软件技术有限公司 | Method for detecting seal ring defect |
CN103606169A (en) * | 2013-12-04 | 2014-02-26 | 天津普达软件技术有限公司 | Method for detecting defects of bottle cap |
CN103604808A (en) * | 2013-12-04 | 2014-02-26 | 天津普达软件技术有限公司 | Visual detection method for bottle cap defects |
Non-Patent Citations (1)
Title |
---|
李喆等: "基于机器视觉的瓶盖表面检测技术的研究", 《仪表技术》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105865569A (en) * | 2016-05-12 | 2016-08-17 | 天津大学 | Glass bottle liquid level detection method based on visible light communication |
CN106018412A (en) * | 2016-05-12 | 2016-10-12 | 天津大学 | Bottle cap surface production date detection method based on visible light communication |
CN105865570A (en) * | 2016-06-15 | 2016-08-17 | 天津大学 | Machine vision-based glass bottle liquid level detection method |
CN106052793A (en) * | 2016-06-15 | 2016-10-26 | 天津大学 | Machine vision based liquid level sub-quality product marking method |
CN110108712A (en) * | 2019-04-20 | 2019-08-09 | 东莞中科蓝海智能视觉科技有限公司 | Multifunctional visual sense defect detecting system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105424722A (en) | Bottle cap unqualified product marking method based on machine vision | |
CN105301003A (en) | Method for detecting bottle cap surface production date based on machine vision | |
CN105865570A (en) | Machine vision-based glass bottle liquid level detection method | |
US7684034B2 (en) | Apparatus and methods for container inspection | |
US8014586B2 (en) | Apparatus and methods for container inspection | |
CN106052793A (en) | Machine vision based liquid level sub-quality product marking method | |
CN101566582B (en) | Medicine bottle label information online detection system in powder injection production based on mechanical vision | |
US7667836B2 (en) | Apparatus and methods for container inspection | |
CN104237252A (en) | Machine-vision-based method and device for intelligently detecting surface micro-defects of product | |
CN201707305U (en) | Inspection system for inspecting quality of container caps | |
CN103604808A (en) | Visual detection method for bottle cap defects | |
CN103454285A (en) | Transmission chain quality detection system based on machine vision | |
CN105046700A (en) | Brightness correction and color classification-based fruit surface defect detection method and system | |
CN103884720A (en) | FPGA (Field Programmable Gate Array)-based real-time PET (Polyethylene Terephthalate) bottle cover defect detection system and method | |
CN109461156B (en) | Threaded sealing plug assembly detection method based on vision | |
CA2786889A1 (en) | Method for monitoring the quality of the primer layer applied to a motor-vehicle body prior to painting | |
CN106824816A (en) | A kind of PE based on machine vision bottles of detection and method for sorting | |
Abdelhedi et al. | Design of automatic vision-based inspection system for monitoring in an olive oil bottling line | |
CN104118609A (en) | Labeling quality detecting method and device | |
CN106093057A (en) | A kind of online gluing detection method | |
CN101565109A (en) | Aluminum cap packaging online detection system in medical powder injection production based on mechanical vision | |
CN108776143A (en) | A kind of online vision inspection apparatus and method of the small stain of egg eggshell surface | |
CN111665251A (en) | Visual detection method for surface defects | |
CN105510339A (en) | Beverage bottle fullness detection method based on machine vision | |
CN106841229A (en) | A kind of online PE bottles of detection method of bottle sealing defect based on machine vision |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into 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: 20160203 |