CN106896115A - Varnished glass Defect Detection device based on area array cameras parallel connection acquisition system - Google Patents
Varnished glass Defect Detection device based on area array cameras parallel connection acquisition system Download PDFInfo
- Publication number
- CN106896115A CN106896115A CN201710094355.5A CN201710094355A CN106896115A CN 106896115 A CN106896115 A CN 106896115A CN 201710094355 A CN201710094355 A CN 201710094355A CN 106896115 A CN106896115 A CN 106896115A
- Authority
- CN
- China
- Prior art keywords
- glass
- image
- area array
- program
- defect detection
- 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
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
- G01N21/958—Inspecting transparent materials or objects, e.g. windscreens
Abstract
The present invention commonly uses artificial mode for present glass production producer carries out the situation of Defect Detection to varnished glass, devises a kind of varnished glass Defect Detection system based on CCD area array cameras.By moving stage, 5 CCD area array cameras with camera lens, 1 bar shaped feux rouges line clears off source, netting twine and interchanger, host computer, and corresponding fixation means composition, it is broadly divided into two modules, one is glass image collection module, and two is glass image processing module, is attached by network between two modules.Processing routine judges glassy zone using side operator is sought, and extracts statistics flaw information.
Description
Technical field
The present invention relates to online glass detection field.Artificial mode is commonly used to applying for present glass production producer
The situation that paint glass carries out Defect Detection devises a kind of vision on-line detecting system based on area array cameras.Painting can be detected
Paint the flaw number of glass, flaw size and flaw species is identified, while the requirement of General Production Lines can be met, detect
High precision, testing result is reliable and stable.
Background technology
The Defect Detection to glass is all that offline method is detected with special testing equipment for a long time.And it is online
Some external mechanisms of detection technique have had suitable level, and domestic level relatively falls behind.
The LASOR companies of Germany are a companies for being provided exclusively in line detecting system.The 2F1 detections system that the said firm releases
System instead of original laser measuring technology using the technology for detection glass flaws of Digital photographic.There are many enterprises in current China
The on-line checking of float glass, such as Luoyang Glass Works, Qinhuangdao credit China, SHENZHEN NANBO etc. are all carried out using the equipment.Germany
Innomence production FCO on-line glass defect inspection systems be that flaw is detected using the principle of grating interference, but
The accuracy of detection of this equipment depends on the graduating accuracy of grating, therefore involves great expense, so using comparatively fewer.
The country also has a lot of enterprises to carry out the development of online detection instrument, but its product falls behind, and automatization level is not high,
Function is excessively single.Wuhan University of Technology devises a set of glass flaws on-line detecting system based on ARM and DSP.The system
The technical sophistication being related to, implements relatively difficult, is unfavorable for the actual production of input.
If developing, a set of stability, accuracy, applicability, reliability are all fine, and white paint with low cost
Glass flaws online detection instrument, especially develops the testing equipment with independent intellectual property right significant.It is also comprehensive
Close national power take part in international competition power lifting an important symbol.
The content of the invention
It is an object of the invention to provide a kind of on-line measuring device of white paint glass, the various flaws on glass are capable of detecting when
Defect simultaneously preserves result, and adapts to the detection of larger amplitude colourless paint glass.
To achieve the above object, the invention provides a kind of based on area array CCD camera parallel connection image capturing system
Automatic detection device.The requirement of glass production line can be met, the information of image can be effectively extracted, the place of image is carried out
Manage and then export testing result and store.
Whole detection system of the invention includes hardware system and software systems.Hardware system can be divided into IMAQ part
And image processing section, data transmission module and mobile platform.Image capture module includes camera, camera lens, light source.Data transfer
Module is transmitted through interchanger multiple signals is transmitted into host computer and processed using Ethernet.Mobile platform includes camera branch
Frame, motor, proximity switch, PLC;Software systems are higher using programming efficiency, the more preferable LabVIEW softwares of man-machine interaction.
The mobile platform also includes support, motor, driving-chain, bearing, bearing block, carrying roller and gear.The gear is consolidated
It is connected in the two ends of each carrying roller, all of carrying roller is connected by driving-chain, motor drives all of carrying roller to do to revolve by driving-chain
Transhipment is dynamic.
The camera support is arranged on the center of the mobile platform X-direction, the surface of Z-direction.
5 cameras are installed, the camera is arranged on camera holder and is arranged on phase by T-nut on the camera support
On machine support, convenient disassembly.
The proximity switch is arranged on the marginal position of the mobile platform, and phase will be opened when having glass to come
Machine and light source are detected to it.
The light source is located between 620nm-760nm using red strip source, the wavelength of the light of release, can be effective
Prevent picture quality reduces because of the change of optical source wavelength.
The interchanger and host computer are installed on the workbench of mobile platform side, are easy to operating personnel to check, whole
Body structure chart is as shown in Figure 1.
The inner surface spraying reflectorized material of the light shield, can so be reflected back the light that script is radiated at outside glass
Glass surface, to the dark place of glass surface, especially top, forms the effect of light filling.Measure the stability requirement for illumination
Highest, as long as illumination occurs the change of 10-20%, measurement result may deflect away from 1-2 pixel, cause image to change, this
It is not the problem of software, it is necessary to from the angle of system design, the interference of ambient light is excluded, while to ensure active illumination light source
Stability of photoluminescence, so as to improve the precision and stability of measurement.
The camera uses CCD area array cameras, and resolution ratio is 2592 × 1944, and detectability can reach:Scratch width
More than 0.1mm, raised, recessed, more than calculus diameter 0.1mm, 0.01 square millimeter of flaw area.If measured object is
Motion, then influence (fuzzy pixel=speed of moving body * camera exposure of the motion blur to the precision of images must be considered
Time).On the premise of certainty of measurement is ensured, the line speed that disclosure satisfy that is 9m/min to the present invention.
Other features of the invention will be described in detail in subsequent specific embodiment.
Brief description of the drawings
Accompanying drawing is, for providing further understanding of the invention, and to constitute the part of specification, with following tool
Body embodiment is used to explain the present invention together, is not construed as limiting the invention.In the accompanying drawings:
Fig. 1 is glass inspection systems structure chart;
Fig. 2 is the structure chart of carrying roller part;
Fig. 3 is self adaptation ROI exemplary plots;
Fig. 4 is host computer total system functional diagram.
Description of reference numerals:
1. support stand;2. light shield;3. camera;4. light source;5. glass to be measured;6. proximity switch;7. host computer;8. hand over
Change planes;9. motor;10. belt wheel;11. transmission belts;12. bearings;13. bearing blocks;14. carrying rollers;15. rubber sleeves
Specific embodiment
Below with reference to accompanying drawing, specific implementation of the invention is elaborated:
Accompanying drawing 1 is glass detection device overall structure figure, mainly includes that 1 support stand, 2 light shield, 3 camera, 4 light sources 5 are to be measured
The transmission belt of 10 belt wheel of glass 6 proximity switch, 7 host computer, 8 interchanger, 9 motor 11, patch rolls platform when setting 5 glass run to be measured
Side, and to run from bottom to up, after activation system, 9 electric motor startings simultaneously drive an active carrying roller to rotate, and are driven through 11 transmission belts
It is whole to roll platform motion.When 5 glass to be measured are by 6 proximity switch, 3 cameras and 4 light sources are automatically opened up, and 3 cameras are according to one
Framing rate is taken pictures, and photograph is transferred into 7 host computers is processed;Because the image that 3 cameras shoot is coloured image, and
The image of program treatment is gray level image, so carrying out planes of color spatial extraction to the image for photographing first, is changed into
Gray-scale map, extraction algorithm is weighted mean method, and formula is:Gray=0.30*R (i, j)+0.59*G (i, j)+0.11*B (i, j),
Wherein (i, j) is coordinate value.When seeking side operator and detecting the upper edge of glass, program is proceeded by image processing program
Flaw is counted.Now, program needs to judge glass region, it is to avoid by background error detection.The method for avoiding error detection is to lead to
Cross and seek when operator finds glass, image is divided into several regions by the straight line where side, if gray value japanning higher
Glass, black is set to by background, and now one piece of brightness highest is glassy zone in several regions of image;If grey
The relatively low varnished glass of angle value, white is set to by background, and a piece that now brightness is minimum in several regions of image is glass
Glass region.If not searching out the side of glass, that is, think that whole image is all glassy zone.The method for counting flaw still makes
With side operator is sought, the edge of gray scale mutation in glassy zone is found, and, that is, have found flaw by these edge extractings out.Close
The reason adjustment raceway speed of service, the frame per second and height of camera make the photograph of shooting to embody monolithic glass after splicing.To each
Photo carries out flaw and extracts and be added, when in image processing program after seeking side operator and detect the lower edge of glass, program
Stop statistics flaw, thus by the flaw Information Statistics on monolithic glass out.
The present invention is removed background area using self adaptation ROI methods, determines target area, it is to avoid the generation of error detection.
Such as the situation of accompanying drawing 3, that is, the glass image that there is top and the formation of the right deflection is in quadrangle in the visual field, with the feelings of accompanying drawing 3
The judgement of glassy zone is illustrated as a example by condition.If top is respectively y=a with the equation of straight line where the right1x+b1, y=a2x+b2Connection
Vertical x=0, can solve x1=0, y1=b1Obtain A point coordinates.And then x can be solved2=(b2-b1)/(a1-a2), y2=a1x2+b1
Obtain B point coordinates.Simultaneous formula y=n (n is the following ordinate in region), can solve x3=(n-b2)/a2y3=n, that is, obtain C points
Coordinate.And the coordinate of D points obviously for (0, n).When there is top in former ROI with the right and vertical the right in glass, the right
The equation of place straight line is changed into x=c.A and D point coordinates is constant, and simultaneous solution obtains x2=c, y2=a1x2+b1Obtain B point coordinates.Connection
Vertical y=n, solves x3=c, y3=n is to obtain C point coordinates.4 points of ABCD of connection, then draws the ROI of glass region.According to
The glassy zone ROI of other quadrangle situations can be tried to achieve according to same thinking, and when glass is all in image or in the absence of glass,
The judgement of average gray value can be carried out to region, higher than the judgement of given threshold to be all glass, now ROI is original image;It is low
It is judged as that it is zero now to set ROI areas in the absence of glass in given threshold.
If accompanying drawing 4 is invention software main function of system.First, after activation system, to ensure the security of system, need
Login password is set, and user input password can enter program main interface.There are system parameter setting, such as camera IP in main interface
Set, communication interface is set, the time for exposure is set, the parameter setting such as threshold value of each image processing module, so that convenient operation
Personnel debug.Parameter setting finish after by carry out on-line checking, if finding flaw on glass, system record, display
Flaw quantity and position are simultaneously alarmed;If without flaw, system continues to detect;It is initial firstly the need of carrying out after system operation
Change, parameters reset or reset, and subsequent picture acquisition module starts, the glass image passed through on collection raceway, by network
View data is transferred to image processing module.When image processing module detects the top of glass, detection program starts point
Analyse the flaw of glass surface.If it find that flaw, then system is recorded and alarmed, if not finding flaw, system proceeds
Detection, until complete glass image of collection.When image processing module detect glass it is following when, system stop Defect Detection,
Until detecting the next piece of top of glass.
Claims (4)
1. a kind of varnished glass Defect Detection equipment based on area array CCD camera parallel connection acquisition system, mainly includes:Strip light
Source, CCD area array cameras and camera lens, roll platform, work station and supporting detection program.
2. require that a kind of described varnished glass Defect Detection based on area array CCD camera parallel connection acquisition system sets according to right 1
It is standby, it is characterised in that:Instead of manual detection, high precision, speed is fast, and accuracy rate is high, simple structure, and modularization is good, is easy to safeguard
And maintenance, cost is relatively low.
3. require that a kind of described varnished glass Defect Detection based on area array CCD camera parallel connection acquisition system sets according to right 1
It is standby, it is characterised in that:Whole system is divided into image capture module and image processing module, image capture module be camera lens, camera,
Light source, fixed section bar and rolling platform;Image processing module is extracted for the color of image plane space of work station and its institute's band
Program, find glass edge program, judge glassy zone program, identification extraction flaw program and statistics program.Two modules
Between data be transmitted by network.
4. require that a kind of described varnished glass Defect Detection based on area array CCD camera parallel connection acquisition system sets according to right 1
It is standby, it is characterised in that handling process is following steps:The CCD area array cameras on crossbeam are fixed on according to certain frame per second to lower section
Shot, when glass passes through, seeking side operator and can detect the upper edge of glass in processing routine now detects that program is opened
Beginning work, it is necessary first to judge glass region, in case the flaw on background area is erroneously detected.Algorithm is:By seeking side
Operator finds the side of glass, and image is divided into several regions by the straight line where side, if gray value varnished glass higher, will
Background is set to black, and now one piece of brightness highest is glassy zone in several regions of image;If gray value compared with
Low varnished glass, white is set to by background, and a piece that now brightness is minimum in several regions of image is glassy zone.
If not searching out the side of glass, that is, think that whole image is all glassy zone.The method of flaw is counted still using seeking side
Operator, finds the edge of gray scale mutation in glassy zone, and, that is, have found flaw by these edge extractings out.At image
In reason program seek the lower edge that side operator detects glass after, program stopped statistics flaw.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710094355.5A CN106896115A (en) | 2017-02-21 | 2017-02-21 | Varnished glass Defect Detection device based on area array cameras parallel connection acquisition system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710094355.5A CN106896115A (en) | 2017-02-21 | 2017-02-21 | Varnished glass Defect Detection device based on area array cameras parallel connection acquisition system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106896115A true CN106896115A (en) | 2017-06-27 |
Family
ID=59185817
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710094355.5A Pending CN106896115A (en) | 2017-02-21 | 2017-02-21 | Varnished glass Defect Detection device based on area array cameras parallel connection acquisition system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106896115A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108621472A (en) * | 2018-03-30 | 2018-10-09 | 天津市天锻压力机有限公司 | A kind of automatic rotary positioning apparatus and method |
CN108982529A (en) * | 2018-09-30 | 2018-12-11 | 广州中国科学院沈阳自动化研究所分所 | A kind of power battery coating quality vision-based detection platform and detection method |
CN109752395A (en) * | 2017-11-01 | 2019-05-14 | 深圳市迪姆自动化有限公司 | A kind of glass surface Defect Detection equipment |
CN109883325A (en) * | 2019-03-13 | 2019-06-14 | 成都格拉斯曼科技有限公司 | A kind of real-time detection device of float glass width |
CN110596126A (en) * | 2018-05-25 | 2019-12-20 | 上海翌视信息技术有限公司 | Sheet glass edge flaw detection method based on image acquisition |
CN111912860A (en) * | 2020-08-17 | 2020-11-10 | 华辉玻璃(中国)有限公司 | Rapid detection system for glass flaws |
CN115356355A (en) * | 2022-10-21 | 2022-11-18 | 沃卡姆(山东)真空玻璃科技有限公司 | Automatic detection blanking conveying line and detection method for vacuum laminated glass |
CN115366526A (en) * | 2022-08-24 | 2022-11-22 | 广东华中科技大学工业技术研究院 | Thermal spraying positioning correction method and device for glass cover plate and glass cover plate fitting equipment |
-
2017
- 2017-02-21 CN CN201710094355.5A patent/CN106896115A/en active Pending
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109752395A (en) * | 2017-11-01 | 2019-05-14 | 深圳市迪姆自动化有限公司 | A kind of glass surface Defect Detection equipment |
CN108621472A (en) * | 2018-03-30 | 2018-10-09 | 天津市天锻压力机有限公司 | A kind of automatic rotary positioning apparatus and method |
CN110596126A (en) * | 2018-05-25 | 2019-12-20 | 上海翌视信息技术有限公司 | Sheet glass edge flaw detection method based on image acquisition |
CN110596127A (en) * | 2018-05-25 | 2019-12-20 | 上海翌视信息技术有限公司 | Sheet glass edge flaw detection system based on image acquisition |
CN110596127B (en) * | 2018-05-25 | 2022-07-08 | 翌视智能科技(上海)有限公司 | Sheet glass edge flaw detection system based on image acquisition |
CN108982529A (en) * | 2018-09-30 | 2018-12-11 | 广州中国科学院沈阳自动化研究所分所 | A kind of power battery coating quality vision-based detection platform and detection method |
CN109883325A (en) * | 2019-03-13 | 2019-06-14 | 成都格拉斯曼科技有限公司 | A kind of real-time detection device of float glass width |
CN111912860A (en) * | 2020-08-17 | 2020-11-10 | 华辉玻璃(中国)有限公司 | Rapid detection system for glass flaws |
CN111912860B (en) * | 2020-08-17 | 2023-05-09 | 华辉玻璃(中国)有限公司 | Rapid detection system for glass flaws |
CN115366526A (en) * | 2022-08-24 | 2022-11-22 | 广东华中科技大学工业技术研究院 | Thermal spraying positioning correction method and device for glass cover plate and glass cover plate fitting equipment |
CN115366526B (en) * | 2022-08-24 | 2023-07-28 | 广东华中科技大学工业技术研究院 | Thermal spraying positioning correction method and device for glass cover plate and glass cover plate laminating equipment |
CN115356355A (en) * | 2022-10-21 | 2022-11-18 | 沃卡姆(山东)真空玻璃科技有限公司 | Automatic detection blanking conveying line and detection method for vacuum laminated glass |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106896115A (en) | Varnished glass Defect Detection device based on area array cameras parallel connection acquisition system | |
CN110596134B (en) | Sheet glass edge flaw detection method based on image acquisition | |
CN100419410C (en) | Defect inspection data processing system | |
CN106226157B (en) | Concrete structure member crevices automatic detection device and method | |
CN101382502B (en) | Surface blot detecting system and detecting method thereof | |
CN102323272A (en) | Filter paper defect detecting system and detection method thereof based on machine vision technique | |
CN113418933B (en) | Flying shooting visual imaging detection system and method for detecting large-size object | |
CN110567976A (en) | mobile phone cover plate silk-screen defect detection device and detection method based on machine vision | |
CN208366871U (en) | Detection system | |
CN205427436U (en) | Display device's counterpoint check out test set and exposure process systems | |
CN102879404A (en) | System for automatically detecting medical capsule defects in industrial structure scene | |
CN111610190B (en) | Appearance inspection device for household appliances | |
CN202177587U (en) | Filter paper defect detecting system based on machine vision technology | |
CN108195847A (en) | A kind of relief pattern Online Video detection device and its detection method | |
CN105067629B (en) | A kind of chromatic circle resistance open defect and resistance value integrated testing method | |
CN103091332B (en) | Detection method and detection system of U-shaped powder pipe based on machine vision | |
CN105699386B (en) | A kind of automatic cloth inspection labeling method using contact-type image sensor | |
CN114910483B (en) | Device, method and system for detecting quality of super-crystal plate | |
CN116256366A (en) | Chip defect detection method, detection system and storage medium | |
Telljohann | Introduction to building a machine vision inspection | |
CN211825732U (en) | Non-contact film pressing backlight panel detection device and backlight panel automatic detection line | |
CN201218722Y (en) | Optical detector for printed circuit board | |
CN111330869A (en) | Visual detection method and system for on-line grading of lens | |
CN116256909B (en) | Real-time detection processing system and processing method for liquid crystal coating | |
CN219915416U (en) | Detection equipment for image sensing chip |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170627 |