CN109507209A - A kind of film printing defect detecting system and method - Google Patents
A kind of film printing defect detecting system and method Download PDFInfo
- Publication number
- CN109507209A CN109507209A CN201910056248.2A CN201910056248A CN109507209A CN 109507209 A CN109507209 A CN 109507209A CN 201910056248 A CN201910056248 A CN 201910056248A CN 109507209 A CN109507209 A CN 109507209A
- Authority
- CN
- China
- Prior art keywords
- image
- module
- film printing
- detection
- template
- 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
- G01N21/956—Inspecting patterns on the surface of objects
- G01N21/95607—Inspecting patterns on the surface of objects using a comparative method
Landscapes
- 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 Processing (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of film printing defect detecting system and methods, system includes power supply module and the PLC synchronization control module and image capture module that connect respectively with power supply module, the PLC synchronization control module is successively connect with image capture module, image preprocessing and color space transformation module, image multi-pass decoding locating module and image template matching module, and described image template matching module is connect with BLOB Cluster Analysis module and off-line modeling module respectively.Compared with prior art, the positive effect of the present invention is: firstly, the present invention improves the accuracy of film printing defects detection by secondary positioning, can be applied to the film printing defects detection of complex pattern, texture, character and color；Secondly, can during film printing real-time monitoring film production situation；Again, it solves the mode for needing artificial detection in traditional detection method, greatly reduces cost of labor.
Description
Technical field
The present invention relates to image procossings and computer vision field, and in particular to a kind of film printing defect detecting system and
Method can detect film type flexible material printing quality defect automatically.
Background technique
With the continuous fast development of domestic and international printing technology, printing machine performance in terms of have it is very big
It is promoted, but for present case, quality problems still can be inevitably generated in printing process.How to subtract
The production consumption as caused by product quality problem generated in low printing process is one put in face of each printing enterprise
Pressing issues.
There are many kinds of class, such as ink dots to scratch, bite, chromatography deviation, surface blot, rise for the mass defect of film printing product
Wrinkle, misalignment quality, traditional detection method relies on human eye detection, but the eyes of people are checking printed matter defect by subjectivity
Factor is affected, and low efficiency is fed back the features such as slow, is not well positioned to meet the demand of Modern Press enterprise mass production.
To print product using NI Vision Builder for Automated Inspection real-time detection, human eye is replaced with camera, with computer hardware and figure
As processing software replaces human brain to carry out image analysis processing, the feedback of defect recognition and information, printing quality is substantially increased
Detection efficiency and detection quality, and the main flow direction of printing enterprise's quality control at present.
China Patent Publication No. CN201810189016, denomination of invention: plastic film printing character defect detecting device and
In method, device includes that module, delivery module, lighting module and acquisition module occur for electrostatic；Electrostatic occurs module and is installed on biography
Send module；Detection zone, installation lighting module and acquisition module above detection zone are provided on delivery module.The present invention can
Plastic film product is set to utilize CCD area array cameras timing according to the smooth transmission of predetermined speed by adjusting illumination brightness and angle
Complete acquisition image data, and print character defect is accurately and effectively determined using progressive backoff algorithm and two-way difference shadow method
Whether there is or not and its type, while to defect information carry out automatic alarm and reject operate.Whole system is participated in without personnel, is realized
The intelligence of whole system avoids manually participating at high cost, speed, it can be achieved that the automatic detection of character defect, classification and reject
Slowly, the problems such as discrimination is low accurately and efficiently realizes the detection to print character defect.But this method only realizes that character lacks
Sunken automatic detection is easy to be limited by Thinfilm pattern, texture, therefore the film that can not be applied to complex pattern and texture lacks
It falls into detection.
Summary of the invention
The shortcomings that in order to overcome the prior art, the present invention provides a kind of film printing defect detecting system and method, purports
In the film defects detection for solving the problems, such as complex pattern, character, texture and color, pass through secondary positioning and template matching algorithm
Programming count average value and relative error have the characteristics that reliable, accuracy is high, error is small, can be verified simultaneously with human eye.
The technical scheme adopted by the invention is that: a kind of film printing defect detecting system, including power supply module and
The PLC synchronization control module and image capture module connecting respectively with power supply module, the PLC synchronization control module is successively
With image capture module, image preprocessing and color space transformation module, image multi-pass decoding locating module and image template
It is connected with module, described image template matching module is connect with BLOB Cluster Analysis module and off-line modeling module respectively.
The present invention also provides a kind of film printing defect inspection methods, include the following steps:
(1) using a fixed width of acquisition and the film graphics off-line modeling of height, template image and corresponding is obtained
Detection parameters；
(2) gaussian filtering process is carried out to image to be detected that image capture module acquires in real time；Template size is 3X3；
(3) gaussian filtering image and template image that (2) step obtains are transformed into Lab space respectively；
(4) according to the positioning core region in off-line modeling to L spatial image coarse positioning, then at image block multithreading
Reason, every block of image randomly select the fine positioning again of the biggish region of variance on the basis of coarse positioning；
(5) model of detection image and off-line modeling is carried out template matching to score on the basis of (4) step positions
Then the Lab of detection image and modeled images layer is carried out product single layer and integrated color difference calculates, and summarizes to obtain residual error by analysis
BLOB；
(6) BLOB clustering: first according to the preset detection parameters of off-line modeling, compare the area and energy of cluster result
Figureofmerit determines whether misprint or noise spot, then according to the shape of BLOB to the residual point for exceeding template setting value
Sorted out；
(7) image is shown and defect information exports.
Compared with prior art, the positive effect of the present invention is:
Firstly, the present invention improves the accuracy of film printing defects detection by secondary positioning, complicated figure can be applied to
The film printing defects detection of case, texture, character and color；Secondly, can during film printing real-time monitoring film production
Situation；Again, it solves the mode for needing artificial detection in traditional detection method, greatly reduces cost of labor.
Detailed description of the invention
Examples of the present invention will be described by way of reference to the accompanying drawings, in which:
Fig. 1 is the structure chart of film printing defect detecting system of the present invention；
Fig. 2 is the data flow diagram of film printing defect detecting system of the present invention；
Fig. 3 is the off-line modeling flow chart of film printing defect detecting system of the present invention.
Specific embodiment
A kind of film printing defect detecting system, as shown in Figure 1, including acquiring image data in the detection zone of setting
Image capture module, image preprocessing and color space transformation module, off-line modeling and on-line study module, image multistage search
Rope locating module, image template matching module, BLOB Cluster Analysis module, the acquisition of control two-way ccd image and Serial Port Information are same
The PLC synchronization control module of step provides the power supply module of independent current source for system.Entire detection system is relevant to software
Data communication network includes two gigabit Ethernets, the two networks transmit image data and testing result data respectively, from soft
Guarantee that real-time detection result can quickly summarize from each measuring station on hardware to carry out statistics and show.System is connected using ethernet bus
Mode is connect, the extension of system detection unit is facilitated.Each measuring station is hung on two networks, after the completion of detection operation,
It will test result and be sent to image display, the images such as image needed for defect image data and modeling with data private local area network
Data are sent to modeling module by image transmitting private network, and modeling module mainly instructs modeling, and by the template built up according to
Each measuring station needs to be sent to corresponding measuring station.The data flow of whole system is as shown in Figure 2.
(1) image capture module realize image data acquisition, image capture module by two-way linear array colored CCD, apply it is resistance to
Moral camera lens, highlighted white LED line light source and its peripheral circuit composition, image data are captured by CCD, are transmitted by call back function
It is analyzed and processed to preprocessing module.
(2) color camera is clearer due to being imaged, and the time for exposure is short, and there are certain high-frequency noises, leads to product image
Brightness be not be straight line, there are random fluctuations, so if directly carry out processing model the threshold range that learns out compared with
Greatly, model curve is not smooth enough.Image is filtered for the above reason, image is smoothly pre-processed, by with
The threshold range of upper step study is obviously lower.
Color space transformation uses CIELab standard colorimetric system, and CIELab includes the identifiable entire color model of naked eyes
It encloses, there are many colour gamut ratio CMYK and RGB color field width of CIELab, and (in the space L*a*b*, L* indicates brightness, and a* indicates red and subtracts green, b*
Indicate green and subtract indigo plant), and they are included.Its calculation formula is:
A*=500 × (f (X/Xn)-f (Y/Yn))
B*=200 × (f (Y/Yn)-f (Z/Zn))
In this color space, the Euclidean distance of two kinds of colors is more nearly, it is easier to examine with the colour-difference that people is perceived
Fine color difference is measured, and generates uneven deviation not as RGB color model.
The factors such as in view of the complexity of product pattern color, colour mixture field color difference is small, and pattern color is shallow, we are quasi-
It is scheduled on CIELab and spatially carries out acetes chinensis, acetes chinensis is exactly that product to be detected and standard sample are carried out color difference comparison.
Simultaneously for quantitative analysis product color difference difference, single layer and integrated color difference analysis method have been used.As sample L*a*b* is
L1*, a1*, b1*, product L*a*b* to be detected is L2*, a2*, b2*.Then single layer color difference and integrated color difference detection calculate following fixed
Justice:
Δ L=L1*-L2*
Δ a=a1*-a2*
Δ b=b1*-b2*
Δc2=Δ L2+Δa2+Δb2
(3) image multi-pass decoding locating module is to allow product printing process to have one to guarantee full width face detection accuracy
Determine chromatography deviation, it is therefore desirable to which fine multi-pass decoding positioning is carried out to picture portion domain.Image multi-pass decoding locating module includes
Two step Search positioning.Since there are deviations for machine mechanical running part, so that there are spatial deviations for film graphics.Meanwhile
Since printing zone is there are technological fluctuation, spatial position of these characteristic areas in image greatly is caused to there is fluctuation.Therefore, exist
Before carrying out image detection, it is necessary to carry out an image greatly and the positioning of characteristic area Two step Search.
The positioning of image generally uses the matching location algorithm based on characteristic point, and first selection has certain special in reference map
Then the characteristic area of point matches location feature region using certain similarity criterion, according to characteristic area in realtime graphic
Position in reference map and the matching position in realtime graphic, so that it may be directed at two images.
Correlation function algorithm is the search localization method being widely used, it is grown up on the basis of poor sum of squares approach
's.If its core concept is template T and subgraph Si,jContent it is similar or consistent, then template T and subgraph Si,jPhase relation
Number R (i, j) is larger.Estimating for the similarity measure of correlation function can be derived by poor sum of squares approach, then be had according to above formula:
Wherein, M × N is Prototype drawing size, and it is a constant that the right Section 3, which indicates the gross energy of template, in formula, and is searched
Rope position (i, j) is unrelated；First item is the energy of subimage block under template map combining, it slowly changes with the position (i, j)；The
Binomial is the cross-correlation of subgraph and template image, is changed with (i, j), as Prototype drawing T and subgraph Si,jIt is taken when Similarity matching
Value is maximum.Therefore, the similarity measure of correlation function algorithm can be defined as follows:
Or it is normalized to:
It can be seen that the size, the size of region of search and search of Prototype drawing from similar topology degree above and formula
Location algorithm determines the speed of image alignment, and the determination of Prototype drawing size and region of search size is dependent on designer's
Experience, while also relying on the property of specific image.Prototype drawing and region of search are smaller, and search locating speed is faster, however not
Certainty factor is higher, matches the error of positioning with regard to big, therefore should determine Prototype drawing according to picture characteristics and experimental result
With the size of region of search.
(4) image template study module is in process of production since mass colour changes, and detection system needs automatic on-line
Habit or off-line learning product mass colour avoid generating wrong report.On-line study module is in normal detection process, by operator
Specified condition carries out the process of on-line study, and the template library of modification is the template data copy on this detection work station.From
The process of line modeling is as shown in Figure 3.
It (4.1) is entire image by two-way camera acquisition image mosaic；
(4.2) wherein 1 piece of individual zone-texture is selected in entire image center and mark positioning core region as benchmark；
(4.3) each area grade is simply provided in individual region, orientation factor, individual defect area including positioning core
With energy setting etc.；
(4.4) setting of individual region is applied to full width face；
(4.5) template is sent to measuring station and 10 rice product of on-line study；
(4.6) modeling is completed, and can normally be detected；
(5) template matching algorithm module is by previously selected training set, and system establishes color model.Production in real time
When, current image to be detected is compared with the color model established before, realizes quality complete detection.Image template matching
Module is to will test image on the basis of image and characteristic area Two step Search position greatly and template image compares, and is obtained residual
Poor BLOB.
(6) after BLOB cluster algorithm module is realtime graphic and corresponding template contrasting detection, to defective region into
Row clustering.Abandoned tender standard is sentenced according to modeling is preset, is compared the indexs such as area and the energy of cluster result, is realized product defects
It is classified and sentences useless.BLOB Cluster Analysis module be by carrying out graphic feature analysis to residual error Blob unit, can will be simple
Pattern-information is quickly converted to the shape information of pattern, such as figure centroid, graphics area, figure perimeter, the external minimum square of figure
Shape and other graphical informations, so that real defect and false defect be come out according to graphic feature different instructions.Typically
Blob feature has: the ratio of Blob area and boundary rectangle area, minimum external elliptical long axis angle etc..BLOB clustering master
It to include that following four detects.
A. strong ties point detects:
Residual point progress secondary splitting (being divided again with larger threshold value) being connected in zonule is counted afterwards, is connected
Residual points ConnectNum, is then determined compared with the threshold value of setting.
B. bulk strength detects:
Main method is the sum of all residual intensity in zoning, is then determined compared with the threshold value of setting,
Mainly detect weakness and the not residual point set of strong ties.
C. block energy detects:
The sum of the energy of all residual points, is then determined compared with the threshold value of setting in zoning, it is also inspection
Survey that large area color is weak sells of one's property the residual point of raw block.
D. block cluster detection
Whether so-called piece of cluster mainly has similitude using position adjacent area or has correlation under certain conditions,
Zonule have larger residual points, or have with integral strength in adjacent area it is strong, or have residual point close with adjacent area
Density etc. can be clustered, make weak piece of residual point by being clustered into the residual point of strong block.
(7) PLC synchronization control module be by encoder A phase signals control two-way linear array color CCD image acquisition and
The synchronization of RS485 Serial Port Information.
(8) power supply module provides independent current source for system.Power supply module includes 5V DC power supply, 12V DC
Power supply, 24V DC power supply, power system are encoder, PLC, line array CCD and the power supply of White LED linear light source.
The present invention also provides a kind of film printing defect inspection methods, acquire picture number in the detection zone of setting
Further include following steps (1) to step (8) according to, the film printing defect inspection method:
(1) film graphics of a fixed width and height are acquired；
(2) two-way camera acquisition image mosaic is first entire image by off-line modeling；Then it is selected in entire image center
Out wherein 1 piece of individual zone-texture as benchmark；And positioning core region is marked, while each region being simply provided in individual region
Detection level；Orientation factor, individual defect area and energy setting including positioning core；And it is applied to full width face；Obtain template
Image and corresponding detection parameters.
(3) image preprocessing carries out gaussian filtering process, template size 3X3 to entire image；
(4) gaussian filtering image and template image in (3) are transformed into Lab space by color space transformation respectively, convenient for into
Row product single layer and integrated color difference analysis；
(5) image multi-pass decoding locating module；It is slightly fixed to L spatial image according to the positioning core region in off-line modeling first
Position, then image block multiple threads, every block of image randomly selects the biggish region of variance again on the basis of coarse positioning
Fine positioning；Guarantee system will not generate wrong report because of orientation problem；
(6) image template matching algorithm；Detection image is built offline with before on the basis of multi-pass decoding positions first
The model of mould carries out template matching comparative analysis, then the Lab of detection image and modeled images layer progress product single layer and comprehensive
Color difference analysis is closed, and summarizes to obtain residual error BLOB；
(7) BLOB cluster algorithm；Blob analysis is primarily to carry out feature extraction and classifying, head to target image
Abandoned tender standard first is sentenced according to off-line modeling is preset, compares the indexs such as area and the energy of cluster result, it is wrong to determine whether printing
Then mistake or noise spot are sorted out the residual point beyond template setting value according to the shape of BLOB (such as length-width ratio)；
(8) image is shown and defect information exports.
After tested, with good stability and practicability of the invention.
Claims (10)
1. a kind of film printing defect detecting system, it is characterised in that: supply mould including power supply module and respectively with power supply
Block connection PLC synchronization control module and image capture module, the PLC synchronization control module successively with image capture module,
Image preprocessing and color space transformation module, image multi-pass decoding locating module are connected with image template matching module, described
Image template matching module is connect with BLOB Cluster Analysis module and off-line modeling module respectively.
2. a kind of film printing defect detecting system according to claim 1, it is characterised in that: described image acquisition module
It is made of two-way linear array colored CCD, Schneider camera lens, highlighted white LED line light source and its peripheral circuit.
3. a kind of film printing defect detecting system according to claim 1, it is characterised in that: described image pretreatment and
Color space transformation module include to the image mosaic of the two-way camera of image capture module acquisition at entire image carry out it is high
This filter preprocessing；And the pretreated image of gaussian filtering and template image are transformed into Lab space.
4. a kind of film printing defect detecting system according to claim 1, it is characterised in that: described image multi-pass decoding
Locating module includes that an image and characteristic area Two step Search greatly positions.
5. a kind of film printing defect detecting system according to claim 1, it is characterised in that: the off-line modeling module
Process it is as follows:
It (1) is entire image by two-way camera acquisition image mosaic；
(2) wherein 1 piece of individual zone-texture is selected in entire image center and mark positioning core region as benchmark；
(3) each area grade is simply provided in individual region, orientation factor, individual defect area and energy including positioning core
Setting；
(4) setting of individual region is applied to full width face；
(5) template is sent to measuring station and 10 rice product of on-line study；
(6) modeling is completed, and obtains color model.
6. a kind of film printing defect detecting system according to claim 1, it is characterised in that: described image template matching
Module compares image to be detected and template image on the basis of image and characteristic area Two step Search position greatly, obtains
Residual error BLOB.
7. a kind of film printing defect detecting system according to claim 1, it is characterised in that: the BLOB clustering
Module includes the detection of strong ties point, bulk strength detection, block energy detection and block cluster detection.
8. a kind of film printing defect inspection method, characterized by the following steps:
(1) using a fixed width of acquisition and the film graphics off-line modeling of height, template image and corresponding inspection are obtained
Survey parameter；
(2) gaussian filtering process is carried out to image to be detected that image capture module acquires in real time；Template size is 3X3；
(3) gaussian filtering image and template image that (2) step obtains are transformed into Lab space respectively；
(4) according to the positioning core region in off-line modeling to L spatial image coarse positioning, then image block multiple threads,
Every block of image randomly selects the fine positioning again of the biggish region of variance on the basis of coarse positioning；
(5) model of detection image and off-line modeling is carried out template matching comparative analysis on the basis of (4) step positions, so
The Lab of detection image and modeled images layer is carried out product single layer afterwards and integrated color difference calculates, and summarizes to obtain residual error BLOB；
(6) BLOB clustering: first according to the preset detection parameters of off-line modeling, the area and energy for comparing cluster result refer to
Mark, determines whether misprint or noise spot, is then carried out according to the shape of BLOB to the residual point beyond template setting value
Sort out；
(7) image is shown and defect information exports.
9. a kind of film printing defect inspection method according to claim 8, it is characterised in that: will figure described in (3) step
Calculation formula as being transformed into Lab space use are as follows:
A*=500 × (f (X/Xn)-f (Y/Yn))
B*=200 × (f (Y/Yn)-f (Z/Zn))
In formula, L* indicates brightness, and a* indicates red and subtracts green, and b* indicates green to subtract indigo plant.
10. a kind of film printing defect inspection method according to claim 8, it is characterised in that: carried out described in (5) step
The formula that product single layer and integrated color difference calculate are as follows:
Δ L=L1*-L2*
Δ a=a1*-a2*
Δ b=b1*-b2*
Δc2=Δ L2+Δa2+Δb2
In formula, L1*, a1*, b1* are the space L*a*b* of sample, and L2*, a2*, b2* are the space L*a*b* of product to be detected.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910056248.2A CN109507209A (en) | 2019-01-22 | 2019-01-22 | A kind of film printing defect detecting system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910056248.2A CN109507209A (en) | 2019-01-22 | 2019-01-22 | A kind of film printing defect detecting system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109507209A true CN109507209A (en) | 2019-03-22 |
Family
ID=65758217
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910056248.2A Pending CN109507209A (en) | 2019-01-22 | 2019-01-22 | A kind of film printing defect detecting system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109507209A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110632094A (en) * | 2019-07-24 | 2019-12-31 | 北京中科慧眼科技有限公司 | Pattern quality detection method, device and system based on point-by-point comparison analysis |
| CN110940670A (en) * | 2019-11-25 | 2020-03-31 | 佛山缔乐视觉科技有限公司 | Flexible printing label printing head draft detection system based on machine vision and implementation method thereof |
| CN111242896A (en) * | 2019-12-31 | 2020-06-05 | 电子科技大学 | Color printing label defect detection and quality rating method |
| CN111260625A (en) * | 2020-01-15 | 2020-06-09 | 征图新视(江苏)科技股份有限公司 | Automatic extraction method for offset printing large image detection area |
| CN112557399A (en) * | 2020-11-30 | 2021-03-26 | 河北白沙烟草有限责任公司 | Point inspection method and device for smoke machine equipment quality detection system |
| CN112730251A (en) * | 2020-12-30 | 2021-04-30 | 凌云光技术股份有限公司 | Device and method for detecting color defects of screen |
| CN112767306A (en) * | 2020-12-24 | 2021-05-07 | 凌云光技术股份有限公司 | Printed matter quality detection and receiving method and system |
| CN114295561A (en) * | 2022-03-03 | 2022-04-08 | 北京中科慧眼科技有限公司 | Color difference measuring method and system based on device-independent color space |
| CN114757954A (en) * | 2022-06-16 | 2022-07-15 | 南通人民彩印有限公司 | Carton printing color difference defect detection method based on artificial intelligence system |
| CN116482104A (en) * | 2023-02-10 | 2023-07-25 | 中恒永创(北京)科技有限公司 | Thermal transfer film detection method |
| CN118566248A (en) * | 2023-03-23 | 2024-08-30 | 深圳市振华兴智能技术有限公司 | Chip patch precision visual detection method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101799434A (en) * | 2010-03-15 | 2010-08-11 | 深圳市中钞科信金融科技有限公司 | Printing image defect detection method |
| CN105335692A (en) * | 2014-08-15 | 2016-02-17 | 软控股份有限公司 | Tire X-ray image detection and identification method and system |
| KR20180061528A (en) * | 2016-11-29 | 2018-06-08 | 충북대학교 산학협력단 | System and method for detecting foreign substance using lab color space |
| CN108267455A (en) * | 2018-03-08 | 2018-07-10 | 陕西科技大学 | plastic film printing character defect detecting device and method |
| CN108605097A (en) * | 2016-11-03 | 2018-09-28 | 华为技术有限公司 | Optical imaging method and its device |
| CN108918527A (en) * | 2018-05-15 | 2018-11-30 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of printed matter defect inspection method based on deep learning |
-
2019
- 2019-01-22 CN CN201910056248.2A patent/CN109507209A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101799434A (en) * | 2010-03-15 | 2010-08-11 | 深圳市中钞科信金融科技有限公司 | Printing image defect detection method |
| CN105335692A (en) * | 2014-08-15 | 2016-02-17 | 软控股份有限公司 | Tire X-ray image detection and identification method and system |
| CN108605097A (en) * | 2016-11-03 | 2018-09-28 | 华为技术有限公司 | Optical imaging method and its device |
| KR20180061528A (en) * | 2016-11-29 | 2018-06-08 | 충북대학교 산학협력단 | System and method for detecting foreign substance using lab color space |
| CN108267455A (en) * | 2018-03-08 | 2018-07-10 | 陕西科技大学 | plastic film printing character defect detecting device and method |
| CN108918527A (en) * | 2018-05-15 | 2018-11-30 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of printed matter defect inspection method based on deep learning |
Non-Patent Citations (2)
| Title |
|---|
| XINGGUANG QI等: "Research of paper surface defects detection system based on Blob algorithm", 《PROCEEDING OF THE IEEE INTERNATIONAL CONFERENCE ON INFORMATION AND AUTOMATION》 * |
| 张冬娟: "基于CCD的烫印区域识别检测技术的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110632094B (en) * | 2019-07-24 | 2022-04-19 | 北京中科慧眼科技有限公司 | Pattern quality detection method, device and system based on point-by-point comparison analysis |
| CN110632094A (en) * | 2019-07-24 | 2019-12-31 | 北京中科慧眼科技有限公司 | Pattern quality detection method, device and system based on point-by-point comparison analysis |
| CN110940670B (en) * | 2019-11-25 | 2023-04-28 | 佛山缔乐视觉科技有限公司 | Machine vision-based flexographic printing label printing first manuscript detection system and implementation method thereof |
| CN110940670A (en) * | 2019-11-25 | 2020-03-31 | 佛山缔乐视觉科技有限公司 | Flexible printing label printing head draft detection system based on machine vision and implementation method thereof |
| CN111242896A (en) * | 2019-12-31 | 2020-06-05 | 电子科技大学 | Color printing label defect detection and quality rating method |
| CN111260625A (en) * | 2020-01-15 | 2020-06-09 | 征图新视(江苏)科技股份有限公司 | Automatic extraction method for offset printing large image detection area |
| CN112557399A (en) * | 2020-11-30 | 2021-03-26 | 河北白沙烟草有限责任公司 | Point inspection method and device for smoke machine equipment quality detection system |
| CN112557399B (en) * | 2020-11-30 | 2022-11-22 | 河北白沙烟草有限责任公司 | Point inspection method and device for smoke machine equipment quality detection system |
| CN112767306A (en) * | 2020-12-24 | 2021-05-07 | 凌云光技术股份有限公司 | Printed matter quality detection and receiving method and system |
| CN112730251A (en) * | 2020-12-30 | 2021-04-30 | 凌云光技术股份有限公司 | Device and method for detecting color defects of screen |
| CN112730251B (en) * | 2020-12-30 | 2023-09-26 | 凌云光技术股份有限公司 | Device and method for detecting screen color defects |
| CN114295561B (en) * | 2022-03-03 | 2022-09-20 | 北京中科慧眼科技有限公司 | Color difference measuring method and system based on device-independent color space |
| CN114295561A (en) * | 2022-03-03 | 2022-04-08 | 北京中科慧眼科技有限公司 | Color difference measuring method and system based on device-independent color space |
| CN114757954A (en) * | 2022-06-16 | 2022-07-15 | 南通人民彩印有限公司 | Carton printing color difference defect detection method based on artificial intelligence system |
| CN116482104A (en) * | 2023-02-10 | 2023-07-25 | 中恒永创(北京)科技有限公司 | Thermal transfer film detection method |
| CN116482104B (en) * | 2023-02-10 | 2023-12-05 | 中恒永创(北京)科技有限公司 | Thermal transfer film detection method |
| CN118566248A (en) * | 2023-03-23 | 2024-08-30 | 深圳市振华兴智能技术有限公司 | Chip patch precision visual detection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109507209A (en) | A kind of film printing defect detecting system and method | |
| CN101799434B (en) | Printing image defect detection method | |
| CN104441526B (en) | Mould in-service monitoring protection system and method based on appearance profile coupling | |
| CN108109137A (en) | The Machine Vision Inspecting System and method of vehicle part | |
| CN118196068B (en) | Textile printing and dyeing quality monitoring system based on artificial intelligence | |
| CN106056597B (en) | Object visual detection method and device | |
| CN104766049B (en) | A kind of object color recognition methods and system | |
| KR102168724B1 (en) | Method And Apparatus for Discriminating Normal and Abnormal by using Vision Inspection | |
| CN113109348B (en) | Paddle image transfer printing defect identification method based on machine vision | |
| CN102262093A (en) | Machine vision-based on-line detection method for printing machine | |
| CN106341682A (en) | Method, system and device for detecting aging function | |
| MX2011003977A (en) | Method and system for item identification. | |
| CN109693140A (en) | A kind of intelligent flexible production line and its working method | |
| CN107248154A (en) | A kind of cloth aberration real-time on-line detecting method | |
| CN110348380A (en) | A kind of visualization grounding switch state video recognition system and method | |
| CN102608130A (en) | Smart card stain detecting system based on image feature matching technology and detection and detecting method | |
| CN105241883A (en) | Printer pattern online detection and color division method | |
| CN108971025A (en) | A kind of visible detection method and vision detection system | |
| CN105136811B (en) | OVMI optics gradual change magnetic ink printing product on-line detecting system and method | |
| CN216525503U (en) | Carbon fiber prepreg surface defect on-line measuring device based on machine vision | |
| CN111766248A (en) | A system and method for on-line detection of steel stamps based on color CCD | |
| CN114577805A (en) | MiniLED backlight panel defect detection method and device | |
| CN116188376A (en) | Digital printing stepping channel detection method based on twin neural network | |
| CN111784666A (en) | Learning and memory-based LED lamp bead defect detection method | |
| CN108491759A (en) | A kind of process detection device and its process detection method based on deep learning |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190322 |