CN108986086A - The detection of typographical display panel inkjet printing picture element flaw and classification method and its device - Google Patents
The detection of typographical display panel inkjet printing picture element flaw and classification method and its device Download PDFInfo
- Publication number
- CN108986086A CN108986086A CN201810727496.0A CN201810727496A CN108986086A CN 108986086 A CN108986086 A CN 108986086A CN 201810727496 A CN201810727496 A CN 201810727496A CN 108986086 A CN108986086 A CN 108986086A
- Authority
- CN
- China
- Prior art keywords
- image
- classification
- display panel
- original image
- typographical display
- 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
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000001514 detection method Methods 0.000 title claims abstract description 44
- 238000007641 inkjet printing Methods 0.000 title claims abstract description 34
- 230000007547 defect Effects 0.000 claims abstract description 44
- 238000012360 testing method Methods 0.000 claims abstract description 29
- 230000011218 segmentation Effects 0.000 claims abstract description 15
- 230000002950 deficient Effects 0.000 claims abstract description 13
- 238000010801 machine learning Methods 0.000 claims abstract description 10
- 238000007619 statistical method Methods 0.000 claims abstract description 10
- 238000004458 analytical method Methods 0.000 claims description 8
- 239000000463 material Substances 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 7
- 230000000295 complement effect Effects 0.000 claims description 4
- 238000003062 neural network model Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 3
- 238000003064 k means clustering Methods 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims description 2
- 238000003709 image segmentation Methods 0.000 claims 1
- 238000007639 printing Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 4
- 230000004927 fusion Effects 0.000 description 2
- 239000002096 quantum dot Substances 0.000 description 2
- 239000007921 spray Substances 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000006854 communication Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 230000003760 hair shine Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000003754 machining Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000007740 vapor deposition Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30141—Printed circuit board [PCB]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
Abstract
The present invention relates to a kind of detections of typographical display panel inkjet printing picture element flaw and classification method and its device.This method comprises the following steps: the piecemeal of measured panel image obtains；When receiving original image, the original image is pre-processed；When receiving pretreated original image, all block images of same panel are spliced；When receiving spliced original image, image is split；When receive segmentation after original image, classified with the classifier in machine learning to image, subsequent technique entered if without defect, if there is if provide prompt；When receiving the original image for prompting defective pixel after classification, positioning and disaggregated classification are carried out to described image defect pixel, and to its classification and quantity solution for statistical analysis for indicating associated disadvantages.Present invention seek to address that artificial detection and classify existing time length, low efficiency, testing result are influenced vulnerable to the subjective factor of people, the problem of consistency difference.
Description
Technical field
The present invention relates to typographical display panel detection devices, are a kind of typographical display panel inkjet printing pictures exactly
Plain defects detection and classification method and its device.
Background technique
Organic Light Emitting Diode (Organic-Emitting Diode, OLED) display, light emitting diode with quantum dots
(Quantum Dot Light Emitting Diodes, QLED) display and printing electrowetting (Electrowetting
Display, EWD) the typographical displays panel such as display is since its active shines, low-power consumption, wide viewing angle, low cost, is easily achieved
The advantages that Flexible Displays, is more and more paid close attention to by display circle, represents the developing direction of the following display technology.But due to print
Brush display panel technique very complicated, so that various defects inevitably occurs in product, to limit printing precision and work
Skill stability has eventually led to its extremely low product yield, has limited the scale of its industrial production.Therefore to inkjet printing processing procedure
The defects detection of rear panel can effectively control the reduction and cost that defect panel flows into product yield caused by subsequent handling
Loss.
For a long time, OLED, QLED and printing electrowetting such as show at the typographical displays panel inkjet printing processing procedure defect pixel
Detection be all to be carried out by the way of artificial.But print pixel defect is usually less than 0.001, and its gray scale and background phase
Closely, contrast is extremely low, artificially checks and such defect of classifying means that OLED, QLED and printing electrowetting shows etc. that printings are shown
The quality and measurement standard for showing panel cannot all be guaranteed.
Summary of the invention
The purpose of the present invention is to provide a kind of typographical display panel inkjet printing picture element flaw detection with classification method and
Its device, it is intended to solve that artificial detection and classification nozzle spray direction deviations, satellite ink droplet, positioning is inaccurate, ink droplet is bigger than normal, forms a film
Uniformity difference time length of this few quasi-representative defect, low efficiency, testing result is influenced vulnerable to the subjective factor of people, consistency is poor
Problem.
To achieve the above object, the technical scheme is that a kind of typographical display panel inkjet printing picture element flaw is examined
Survey and classification method, include the following steps:
Step S1, the typographical display panel after inkjet printing is placed in detection platform, piecemeal obtains typographical display face to be detected
The original image of plate；
Step S2, the piecemeal original image that step S1 is received is pre-processed；
Step S3, to after pretreatment and all piecemeal original images of same typographical display panel splice；
Step S4, spliced original image is split, obtains detection target；
Step S5, to the original image after segmentation, two classification are carried out to image using the classifier in machine learning, if zero defect
Pixel then carries out subsequent technique, if thening follow the steps S6；
Step S6, to the original image for prompting defective pixel after classification, positioning and the disaggregated classification of defect pixel are carried out, and to it
Classification and the quantity solution for statistical analysis for indicating associated disadvantages.
In an embodiment of the present invention, pretreatment mode is to carry out intensity histogram to piecemeal original image in the step S2
Figure equalizes to improve brightness of image, enhances details.
In an embodiment of the present invention, in the step S3 connecting method include Image Feature Point Matching, coordinate transform and
Image co-registration, described image Feature Points Matching are to carry out Gradient Features extraction and pairing to pretreated piecemeal original image,
The coordinate transform is the relationship between the coordinate system according to two piecemeal original images to be spliced, by matrix multiplication by two
The coordinate system of person 11%, described image fusion is spatially to be matched the image data of the same object of separate sources
Then all message complementary senses of each image are effectively combined by algorithm, are spliced into piece image, facilitate subsequent to entire by standard
The classification and statistics of typographical display panel defect.
In an embodiment of the present invention, the specific implementation process of the step S4 are as follows: calculated using improved K- mean cluster
Method is split spliced original image, finally obtains luminescent material print zone and the bank background area of image.
In an embodiment of the present invention, the specific implementation process of the step S5 are as follows: to the original image after segmentation, use
SVM classifier and histograms of oriented gradients feature in machine learning carry out two classification to image, enter if defect-free pixel
Subsequent technique, if thening follow the steps S6.
1. in an embodiment of the present invention, the specific implementation process of the step S6 are as follows: to prompting defective picture after classification
The original image of element, using the physical location of coordinate transform locating defective pixels, using probabilistic neural network model to image line
The disaggregated classification that defect pixel is realized in feature training is managed, and to its solution for statistical analysis for finally indicating associated disadvantages.
The present invention also provides a kind of detection of typographical display panel inkjet printing picture element flaw and sorters, comprising:
Ink-jet printer in the bank for luminescent material etc. to be printed to typographical display panel, provides for defect pixel detection
Test object；
Removable detection platform for fixing the typographical display panel after inkjet printing, and can be printed according to preset step-length is mobile
Brush display panel；
Camera is set to above the typographical display panel, acquires testing image for piecemeal；
Computer is connect with the camera, for controlling the camera piecemeal acquisition testing image, and is waited for the piecemeal received
Altimetric image is analyzed and exports analysis result.
In an embodiment of the present invention, the computer analyzes the testing image and exports the tool of analysis result
Body realizes that steps are as follows:
Step S01, the piecemeal testing image received is pre-processed；
Step S02, to after pretreatment and all piecemeal testing images of same typographical display panel splice；
Step S03, spliced testing image is split, obtains detection target；
Step S04, to the testing image after segmentation, two classification are carried out to image using the classifier in machine learning, if intact
It is sunken then carry out subsequent technique, if thening follow the steps S05；
Step S05, to the testing image for prompting defective pixel after classification, positioning and the disaggregated classification of defect pixel are carried out, and right
Its classification and the quantity solution for statistical analysis for indicating associated disadvantages.
Compared to the prior art, the invention has the following advantages: the method for the present invention can it is automatic, easily to ink-jet
Printing panel defect carries out detection classification, overcome it is traditional directly using print zone area be finely divided class rotational invariance and
Scale invariability difference and the poor robustness that using least square method and interpolation algorithm textural characteristics are carried out with parameter fitting classification
Disadvantage solves the problems, such as that artificial detection time length, low efficiency, testing result are poor vulnerable to the subjective factor influence of people, consistency.?
In typographical display panel inkjet printing processing procedure just by nozzle spray direction deviation, satellite ink droplet, positioning is inaccurate, ink droplet is bigger than normal, at
This few class defects detection of film uniformity difference are positioned and are solved, and the printing panel for preventing defect excessive flows into the techniques rings such as subsequent vapor deposition
Section improves product quality and yield, reduces subsequent machining cost.
Detailed description of the invention
Fig. 1 is the implementation process of typographical display panel inkjet printing picture element flaw of the present invention detection and classification method
Figure.
Fig. 2 is the realization stream of typographical display panel inkjet printing picture element flaw of the present invention detection and the device classified
Cheng Tu.
Specific embodiment
With reference to the accompanying drawing, technical solution of the present invention is specifically described.
The present invention provides a kind of detection of typographical display panel inkjet printing picture element flaw and classification methods, including walk as follows
It is rapid:
Step S1, the typographical display panel after inkjet printing is placed in detection platform, piecemeal obtains typographical display face to be detected
The original image of plate；
Step S2, the piecemeal original image that step S1 is received is pre-processed；
Step S3, to after pretreatment and all piecemeal original images of same typographical display panel splice；
Step S4, spliced original image is split, obtains detection target；
Step S5, to the original image after segmentation, two classification are carried out to image using the SVM classifier in machine learning, if nothing
Defect pixel then carries out subsequent technique, if thening follow the steps S6；
Step S6, to the original image for prompting defective pixel after classification, positioning and the disaggregated classification of defect pixel are carried out, and to it
Classification and the quantity solution for statistical analysis for indicating associated disadvantages.
Pretreatment mode is to carry out gray-level histogram equalization to piecemeal original image to improve image in the step S2
Brightness enhances details.
Connecting method includes Image Feature Point Matching, coordinate transform and image co-registration in the step S3, and described image is special
It is according to two that sign point matching, which is to the progress Gradient Features extraction of pretreated piecemeal original image and pairing, the coordinate transform,
Relationship between the coordinate system of piecemeal original image to be spliced, by matrix multiplication by the coordinate system 11% of the two
A, described image fusion is spatially to be registrated the image data of the same object of separate sources, then will by algorithm
Each all message complementary senses of image effectively combine, are spliced into piece image, facilitate subsequent to entire typographical display panel defect
Classification and statistics.
The specific implementation process of the step S4 are as follows: using improved K- means clustering algorithm to spliced original image
It is split, the final luminescent material print zone for obtaining image and bank background area.
The specific implementation process of the step S5 are as follows: to the original image after segmentation, using the support in machine learning to
Amount machine bonding position histogram of gradients feature carries out two classification to image, subsequent technique is entered if zero defect, if executing
Step S6.
The specific implementation process of the step S6 are as follows: to defective original image is prompted after classification, utilize coordinate transform
Image deflects pixel physical location is positioned, defect pixel is realized to image texture characteristic training using probabilistic neural network model
Disaggregated classification, and to its solution for statistical analysis for finally indicating associated disadvantages.
The present invention also provides a kind of detection of typographical display panel inkjet printing picture element flaw and sorters, comprising:
Ink-jet printer in the bank for luminescent material etc. to be printed to typographical display panel, provides for defect pixel detection
Test object；
Removable detection platform for fixing the typographical display panel after inkjet printing, and can be printed according to preset step-length is mobile
Brush display panel；
Camera is set to above the typographical display panel, acquires testing image for piecemeal；
Computer is connect with the camera, for controlling the camera piecemeal acquisition testing image, and is waited for the piecemeal received
Altimetric image is analyzed and exports analysis result.
The computer testing image is analyzed and export analysis result the specific implementation steps are as follows:
Step S01, the piecemeal testing image received is pre-processed；
Step S02, to after pretreatment and all piecemeal testing images of same typographical display panel splice；
Step S03, spliced testing image is split, obtains detection target；
Step S04, to the testing image after segmentation, two classification are carried out to image using the SVM classifier in machine learning, if nothing
Defect pixel then carries out subsequent technique, if thening follow the steps S05；
Step S05, to the testing image for prompting defective pixel after classification, positioning and the disaggregated classification of defect pixel are carried out, and right
Its classification and the quantity solution for statistical analysis for indicating associated disadvantages.
The specific implementation process of above-mentioned steps S01 to step S05 sees above the description in method, and details are not described herein again.
The following are specific implementation processes of the invention.
As shown in Figure 1, for a kind of typographical display panel inkjet printing picture element flaw detection provided in an embodiment of the present invention with
The implementation flow chart of classification method, is implemented as follows:
In step s101, typographical display panel image to be detected is obtained with high resolution industrial CCD camera subregion；
In the present embodiment, the typographical display panel after computer control starting CCD camera subregion acquisition inkjet printing processing procedure, and
Image is passed in computer.
In step s 102, when receiving original image, described image is pre-processed；
In the present embodiment due to uneven illumination, picture contrast is low and image communication process in noise jamming will affect acquisition
The quality of original image using gray-level histogram equalization, improve brightness of image, enhancing is thin therefore when being pre-processed
Section, the gradation data for accumulating overstocked image somewhere is evenly distributed in entire domain, reaches Image entropy maximum.
In step s 103, the original image is spliced when original image after receiving pretreatment；
In the present embodiment, the Panorama Mosaic algorithm based on Feature Points Matching is preferably used, two width is extracted first and waits spelling
The Gradient Features of the image connect, constitutive characteristic describe son and are matched, then according between the coordinate system of two images to be spliced
Relationship, by matrix multiplication by coordinate system 11% of the two, thus by the picture number of the same object of separate sources
According to being spatially registrated, finally by all message complementary senses of each image, effectively combines, be spliced into piece image.
In step S104, when receiving spliced original image, the original image is split；
In the present embodiment, the spliced original image, which is split, is realized using improved K- means clustering algorithm,
Image after the splicing is specifically transformed into Lab color space from RGB color, b value is clustered, and passes through b value
Probability distribution histogram peak, determine initial cluster center, then by minimum Eustachian distance method constantly to cluster centre into
Row iteration updates, until reaching the preset condition of convergence, segmentation is completed, the final cluster for obtaining inkjet printing ink marks and bank back
The cluster of scape.
In step s105, original image after receiving segmentation carries out rough sort to the original image；
In the present embodiment, it is that first classification acquisition prints defective picture enough that the original image after described pair of segmentation, which carries out two classification,
The sample image of element and defect-free pixel, the Hog feature that image is extracted in classification are trained, and generate corresponding SVM classifier, so
Classify after extracting Hog feature to the original image after the segmentation using this classifier afterwards.If without defect pixel
Into subsequent technique, if there is then providing prompt.
In step s 106, when the panel image for receiving prompt defect pixel, the original image is finely divided
Class simultaneously counts positioning；
It in the present embodiment, is to pass through meter to prompting the inkjet printing panel of defective pixel to be finely divided class after two classification
Energy, entropy, consistency and the correlation of the gray level co-occurrence matrixes of segmentation object are calculated to describe its texture features, utilizes probabilistic neural
Network model realizes the disaggregated classification of defect to image texture characteristic training, is that ink droplet is excessive, nozzle offset by defect exhaustive division
It is poor with uniformity, while it is statisticallyd analyze.
In the present embodiment, to the actual location of the defect pixel of the disaggregated classification using progress camera mark first
Fixed, internal reference matrix, outer ginseng matrix and distortion factor, are then become image pixel coordinates by multiplying relevant parameter matrix needed for obtaining
Image physical coordinates system is changed to, then arrives camera coordinates system, is finally transformed into world coordinate system, realizes that the reality of defect pixel is fixed
Position, indicates corresponding defect solution.
As shown in Fig. 2, for a kind of typographical display panel inkjet printing picture element flaw detection provided in an embodiment of the present invention with
The implementation flow chart of sorter, is implemented as follows:
Ink-jet printer 201 in the bank for luminescent material etc. to be printed to display panel, provides inspection for defect pixel detection
Survey object；
Mechanical mobile device 202, for according to give stepping mobile printing panel, facilitate the subsequent classification to whole base board defect
And statistics；
High resolution industrial CCD camera 203 is arranged above movable fixture, and it can with respect to the distance of typographical display panel
It adjusts, the acquisition for image information；
Computer 204 is electrically connected with the camera 203, reads the image information of the camera 203 by video card, and to being acquired
Image carries out mathematical analysis and exports analysis result.
The above are preferred embodiments of the present invention, all any changes made according to the technical solution of the present invention, and generated function is made
When with range without departing from technical solution of the present invention, all belong to the scope of protection of the present invention.
Claims (8)
1. a kind of typographical display panel inkjet printing picture element flaw detection and classification method, which comprises the steps of:
Step S1, the typographical display panel after inkjet printing is placed in detection platform, piecemeal obtains typographical display face to be detected
The original image of plate；
Step S2, the piecemeal original image that step S1 is received is pre-processed；
Step S3, to after pretreatment and all piecemeal original images of same typographical display panel splice；
Step S4, spliced original image is split, obtains detection target；
Step S5, to the original image after segmentation, two classification are carried out using the classifier air exercise print pixel image in machine learning,
Subsequent technique is carried out if zero defect, if thening follow the steps S6；
Step S6, to the original image for prompting defective pixel after classification, positioning and the disaggregated classification of defect pixel are carried out, and to scarce
Fall into classification and the quantity solution for statistical analysis for indicating associated disadvantages.
2. typographical display panel inkjet printing picture element flaw detection according to claim 1 and classification method, feature exist
In, pretreatment mode is to carry out gray-level histogram equalization to piecemeal original image to improve brightness of image in the step S2,
Enhance details.
3. typographical display panel inkjet printing picture element flaw detection according to claim 1 and classification method, feature exist
In connecting method includes Image Feature Point Matching, coordinate transform and image co-registration, described image characteristic point in the step S3
It is to wait spelling according to two width with being to the progress Gradient Features extraction of pretreated piecemeal original image and pairing, the coordinate transform
Relationship between the coordinate system of the piecemeal original image connect, it is described by matrix multiplication by coordinate system 11% of the two
Image co-registration is spatially to be registrated the image data of the same object of separate sources, then passes through algorithm for each image
All message complementary senses effectively combine, are spliced into piece image, facilitate the subsequent classification to whole typographical display panel defect and system
Meter.
4. typographical display panel inkjet printing picture element flaw detection according to claim 1 and classification method, feature exist
In the specific implementation process of the step S4 are as follows: carried out using improved K- means clustering algorithm to spliced original image
Segmentation, finally obtains luminescent material print zone and the bank background area of image.
5. typographical display panel inkjet printing picture element flaw detection according to claim 1 and classification method, feature exist
In the specific implementation process of the step S5 are as follows: to the original image after segmentation, using in machine learning SVM classifier and
Histograms of oriented gradients feature carries out two classification to image, subsequent technique is entered if defect-free pixel, if thening follow the steps
S6。
6. typographical display panel inkjet printing picture element flaw detection according to claim 1 and classification method, feature exist
In the specific implementation process of the step S6 are as follows: fixed using coordinate transform to the original image for prompting defective pixel after classification
The physical location of position defect pixel realizes the subdivision of defect pixel using probabilistic neural network model to image texture characteristic training
Class, and to its solution for statistical analysis for finally indicating associated disadvantages.
7. a kind of typographical display panel inkjet printing picture element flaw detection and sorter characterized by comprising
Ink-jet printer in the bank for luminescent material to be printed to typographical display panel, provides inspection for defect pixel detection
Survey object；
Removable detection platform for fixing the typographical display panel after inkjet printing, and can be printed according to preset step-length is mobile
Brush display panel；
Camera is set to above the typographical display panel, acquires testing image for piecemeal；
Computer is connect with the camera, for controlling the camera piecemeal acquisition testing image, and is waited for the piecemeal received
Altimetric image is analyzed and exports analysis result.
8. typographical display panel inkjet printing picture element flaw detection according to claim 7 and sorter, feature exist
The testing image is analyzed in, the computer and exports analysis result the specific implementation steps are as follows:
Step S01, the piecemeal testing image received is pre-processed；
Step S02, to after pretreatment and all piecemeal testing images of same typographical display panel splice；
Step S03, spliced testing image is split, obtains detection target；
Step S04, to the testing image after segmentation, two classification are carried out to pixel image using the classifier in machine learning, if
Zero defect then carries out subsequent technique, if thening follow the steps S05；
Step S05, to the testing image for prompting defective pixel after classification, positioning and the disaggregated classification of defect pixel are carried out, and right
Its classification and the quantity solution for statistical analysis for indicating associated disadvantages.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810727496.0A CN108986086A (en) | 2018-07-05 | 2018-07-05 | The detection of typographical display panel inkjet printing picture element flaw and classification method and its device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810727496.0A CN108986086A (en) | 2018-07-05 | 2018-07-05 | The detection of typographical display panel inkjet printing picture element flaw and classification method and its device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108986086A true CN108986086A (en) | 2018-12-11 |
Family
ID=64536946
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810727496.0A Pending CN108986086A (en) | 2018-07-05 | 2018-07-05 | The detection of typographical display panel inkjet printing picture element flaw and classification method and its device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108986086A (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109658403A (en) * | 2018-12-18 | 2019-04-19 | 福州大学 | A kind of colored electroweting display unit defect identification method and its device |
| CN109741347A (en) * | 2018-12-30 | 2019-05-10 | 北京工业大学 | A kind of image partition method of the iterative learning based on convolutional neural networks |
| CN109903287A (en) * | 2019-03-29 | 2019-06-18 | 北京百度网讯科技有限公司 | Quality determining method and device |
| CN110077126A (en) * | 2019-05-31 | 2019-08-02 | 西北工业大学 | Artificial intelligence assists printed electronics bootstrap to optimize method for improving |
| CN110428409A (en) * | 2019-07-31 | 2019-11-08 | 海南喜福家具制造有限公司 | Furniture quality detecting method and system |
| 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 |
| CN111044522A (en) * | 2019-12-14 | 2020-04-21 | 中国科学院深圳先进技术研究院 | Defect detection method and device and terminal equipment |
| CN111260621A (en) * | 2020-01-14 | 2020-06-09 | 湖南大学 | Method for positioning and identifying surface defects of printed circuit board |
| CN111376590A (en) * | 2018-12-29 | 2020-07-07 | Tcl集团股份有限公司 | Ink-jet printing equipment and ink-jet printing method and device thereof |
| CN111430577A (en) * | 2020-03-31 | 2020-07-17 | 深圳市华星光电半导体显示技术有限公司 | Uniformity calculation method, device, controller and storage medium |
| CN112083895A (en) * | 2019-06-14 | 2020-12-15 | 苏州美一杯信息科技有限公司 | Food customized printing system and control method thereof |
| CN112164042A (en) * | 2020-09-22 | 2021-01-01 | 华中科技大学 | Automatic defect identification method and system in three-dimensional sand mold printing and sanding process |
| CN113139507A (en) * | 2021-05-12 | 2021-07-20 | 保定金迪地下管线探测工程有限公司 | Automatic capturing method and system for drainage pipeline defect photos |
| CN114226262A (en) * | 2020-09-09 | 2022-03-25 | 宜谷京科技实业有限公司 | Flaw detection method, flaw classification method and flaw detection system |
| CN114332547A (en) * | 2022-03-17 | 2022-04-12 | 浙江太美医疗科技股份有限公司 | Medical object classification method and apparatus, electronic device, and storage medium |
| CN114998278A (en) * | 2022-06-16 | 2022-09-02 | 福州大学 | NLED/QLED/OLED pixel defect detection method and system based on feature pyramid network FPN |
| CN116643423A (en) * | 2023-07-24 | 2023-08-25 | 广东灿达股份有限公司 | Liquid crystal panel fault detection system and detection method |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102507592A (en) * | 2011-11-01 | 2012-06-20 | 河海大学常州校区 | Fly-simulation visual online detection device and method for surface defects |
| CN102854191A (en) * | 2012-07-18 | 2013-01-02 | 湖南大学 | Real-time visual detection and identification method for high speed rail surface defect |
| CN103733322A (en) * | 2011-06-17 | 2014-04-16 | 弗劳恩霍弗实用研究促进协会 | Method for providing a prediction model for crack detection and method for crack detection on a semiconductor structure |
| CN104392432A (en) * | 2014-11-03 | 2015-03-04 | 深圳市华星光电技术有限公司 | Histogram of oriented gradient-based display panel defect detection method |
| US20150221076A1 (en) * | 2014-02-06 | 2015-08-06 | Kla-Tencor Corporation | Defect detection and classification based on attributes determined from a standard reference image |
| CN105184793A (en) * | 2015-09-02 | 2015-12-23 | 广东电网有限责任公司汕尾供电局 | Electric energy meter sample appearance and PCB element detection method |
| CN107169953A (en) * | 2017-04-07 | 2017-09-15 | 西安电子科技大学 | Bridge concrete surface crack detection method based on HOG features |
| CN107967681A (en) * | 2017-11-24 | 2018-04-27 | 常熟理工学院 | Defect inspection method is hindered in a kind of elevator compensation chain punching based on machine vision |
-
2018
- 2018-07-05 CN CN201810727496.0A patent/CN108986086A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103733322A (en) * | 2011-06-17 | 2014-04-16 | 弗劳恩霍弗实用研究促进协会 | Method for providing a prediction model for crack detection and method for crack detection on a semiconductor structure |
| CN102507592A (en) * | 2011-11-01 | 2012-06-20 | 河海大学常州校区 | Fly-simulation visual online detection device and method for surface defects |
| CN102854191A (en) * | 2012-07-18 | 2013-01-02 | 湖南大学 | Real-time visual detection and identification method for high speed rail surface defect |
| US20150221076A1 (en) * | 2014-02-06 | 2015-08-06 | Kla-Tencor Corporation | Defect detection and classification based on attributes determined from a standard reference image |
| CN104392432A (en) * | 2014-11-03 | 2015-03-04 | 深圳市华星光电技术有限公司 | Histogram of oriented gradient-based display panel defect detection method |
| CN105184793A (en) * | 2015-09-02 | 2015-12-23 | 广东电网有限责任公司汕尾供电局 | Electric energy meter sample appearance and PCB element detection method |
| CN107169953A (en) * | 2017-04-07 | 2017-09-15 | 西安电子科技大学 | Bridge concrete surface crack detection method based on HOG features |
| CN107967681A (en) * | 2017-11-24 | 2018-04-27 | 常熟理工学院 | Defect inspection method is hindered in a kind of elevator compensation chain punching based on machine vision |
Non-Patent Citations (2)
| Title |
|---|
| 汪志亮等: "面向OLED屏像素缺陷检测的新方法", 《计算机工程与应用》 * |
| 郑钢丰: "《柱状杆结构中的逆散射理论与缺陷形状重构》", 30 September 2011 * |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109658403B (en) * | 2018-12-18 | 2022-05-17 | 福州大学 | Method and device for identifying defects of color electrowetting display unit |
| CN109658403A (en) * | 2018-12-18 | 2019-04-19 | 福州大学 | A kind of colored electroweting display unit defect identification method and its device |
| CN111376590A (en) * | 2018-12-29 | 2020-07-07 | Tcl集团股份有限公司 | Ink-jet printing equipment and ink-jet printing method and device thereof |
| CN111376590B (en) * | 2018-12-29 | 2021-07-27 | Tcl科技集团股份有限公司 | Ink-jet printing equipment and ink-jet printing method and device thereof |
| CN109741347B (en) * | 2018-12-30 | 2021-03-16 | 北京工业大学 | Iterative learning image segmentation method based on convolutional neural network |
| CN109741347A (en) * | 2018-12-30 | 2019-05-10 | 北京工业大学 | A kind of image partition method of the iterative learning based on convolutional neural networks |
| CN109903287A (en) * | 2019-03-29 | 2019-06-18 | 北京百度网讯科技有限公司 | Quality determining method and device |
| CN110077126A (en) * | 2019-05-31 | 2019-08-02 | 西北工业大学 | Artificial intelligence assists printed electronics bootstrap to optimize method for improving |
| CN112083895A (en) * | 2019-06-14 | 2020-12-15 | 苏州美一杯信息科技有限公司 | Food customized printing system and control method thereof |
| CN110428409B (en) * | 2019-07-31 | 2023-07-04 | 海南广营喜福科技有限公司 | Furniture quality inspection method and system |
| CN110428409A (en) * | 2019-07-31 | 2019-11-08 | 海南喜福家具制造有限公司 | Furniture quality detecting method and system |
| 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 |
| CN111044522B (en) * | 2019-12-14 | 2022-03-11 | 中国科学院深圳先进技术研究院 | Defect detection method and device and terminal equipment |
| CN111044522A (en) * | 2019-12-14 | 2020-04-21 | 中国科学院深圳先进技术研究院 | Defect detection method and device and terminal equipment |
| CN111260621A (en) * | 2020-01-14 | 2020-06-09 | 湖南大学 | Method for positioning and identifying surface defects of printed circuit board |
| CN111260621B (en) * | 2020-01-14 | 2023-05-09 | 湖南大学 | Positioning and identifying method for surface defects of printed circuit board |
| CN111430577A (en) * | 2020-03-31 | 2020-07-17 | 深圳市华星光电半导体显示技术有限公司 | Uniformity calculation method, device, controller and storage medium |
| CN111430577B (en) * | 2020-03-31 | 2022-05-31 | 深圳市华星光电半导体显示技术有限公司 | Uniformity calculation method, device, controller and storage medium |
| CN114226262A (en) * | 2020-09-09 | 2022-03-25 | 宜谷京科技实业有限公司 | Flaw detection method, flaw classification method and flaw detection system |
| CN112164042A (en) * | 2020-09-22 | 2021-01-01 | 华中科技大学 | Automatic defect identification method and system in three-dimensional sand mold printing and sanding process |
| CN113139507A (en) * | 2021-05-12 | 2021-07-20 | 保定金迪地下管线探测工程有限公司 | Automatic capturing method and system for drainage pipeline defect photos |
| CN113139507B (en) * | 2021-05-12 | 2022-06-17 | 保定金迪地下管线探测工程有限公司 | Automatic capturing method and system for drainage pipeline defect photos |
| CN114332547A (en) * | 2022-03-17 | 2022-04-12 | 浙江太美医疗科技股份有限公司 | Medical object classification method and apparatus, electronic device, and storage medium |
| CN114998278A (en) * | 2022-06-16 | 2022-09-02 | 福州大学 | NLED/QLED/OLED pixel defect detection method and system based on feature pyramid network FPN |
| CN116643423A (en) * | 2023-07-24 | 2023-08-25 | 广东灿达股份有限公司 | Liquid crystal panel fault detection system and detection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108986086A (en) | The detection of typographical display panel inkjet printing picture element flaw and classification method and its device | |
| CN108548820A (en) | Cosmetics paper labels defect inspection method | |
| CN109214420A (en) | The high texture image classification method and system of view-based access control model conspicuousness detection | |
| CN111242896A (en) | Color printing label defect detection and quality rating method | |
| CN111325203A (en) | American license plate recognition method and system based on image correction | |
| CN109117885B (en) | Stamp identification method based on deep learning | |
| US20050023354A1 (en) | Color calibration for color bar codes | |
| CN108304785A (en) | Road traffic sign detection based on self-built neural network and recognition methods | |
| CN104881665A (en) | Chip character identification and verification method and apparatus | |
| CN109145964B (en) | Method and system for realizing image color clustering | |
| CN110910372B (en) | Deep convolutional neural network-based uniform light plate defect detection method | |
| CN111310628A (en) | Paper currency forming mode inspection and identification method based on paper currency printing pattern characteristics | |
| CN113888536B (en) | Printed matter double image detection method and system based on computer vision | |
| CN107240112A (en) | Individual X Angular Point Extracting Methods under a kind of complex scene | |
| CN114119591A (en) | Display screen picture quality detection method | |
| CN115728309A (en) | Ink-jet printing line defect identification method and process regulation and control method | |
| CN114577805A (en) | MiniLED backlight panel defect detection method and device | |
| Suh et al. | Fusion of global-local features for image quality inspection of shipping label | |
| CN111179423B (en) | Three-dimensional infrared image generation method based on two-dimensional infrared image | |
| CN110400320B (en) | Segmentation method of electrowetting defective pixels | |
| Brumm et al. | A deep learning approach for the classification task of gravure printed patterns | |
| CN110298347B (en) | Method for identifying automobile exhaust analyzer screen based on GrayWorld and PCA-CNN | |
| CN108734158B (en) | Real-time train number identification method and device | |
| CN115311293B (en) | Rapid matching method for printed matter pattern | |
| CN114148103B (en) | Solder paste printing mark recognition method and system based on neural network 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: 20181211 |