CN109540902A - A kind of polaroid Defect Detection system and its detection method - Google Patents
A kind of polaroid Defect Detection system and its detection method Download PDFInfo
- Publication number
- CN109540902A CN109540902A CN201811350658.XA CN201811350658A CN109540902A CN 109540902 A CN109540902 A CN 109540902A CN 201811350658 A CN201811350658 A CN 201811350658A CN 109540902 A CN109540902 A CN 109540902A
- Authority
- CN
- China
- Prior art keywords
- detection
- polaroid
- product
- processing
- camera
- 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
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- 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
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- 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)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The present invention passes through a kind of polaroid Defect Detection system and detection method, it mainly detected from Mark detection, bubble, beat the multiple directions such as trace detection, folding line detection to polaroid progress complete detection, process includes the detection triggering of product pan feeding, picture obtains, vision algorithm processing, as a result output and data storage, the system cooperate corresponding hardware that can quickly be deployed to production scene, adaptation and various various sizes of products, have a good application prospect.
Description
Technical field
The present invention relates to a kind of polaroid Defect Detection system and its detection methods, belong to intelligent industrial manufacturing field.
Background technique
In LCD industry, required to make display reach bright, clear etc., polaroid is as in display
Critical elements play the role of vital.Since polaroid can not be asked with visually picking out polar biased direction easily to solve this
Topic, it is common practice to: when manufacturing polaroid, can be stamped for identification partially in a certain region in the corner of polaroid release film
The multicolour pattern or text of extreme direction are referred to as arrow print.
Polaroid in process of production, needs to examine its defect (foreign matter, bubble, scratch etc.) by vision inspection apparatus
It looks into.Due to can not be domestic general visually to pick out deflection angle and the direction of polaroid, before polaroid is pasted glass surface
Time way be that production line personnel carry out artificial arrow print and automatic Mark.But the stickup program of polaroid is consistency operation, if
Be wherein have the polaroid of part be inverted the yields that if pan feeding anisotropy, can either reduce entire production output and
Improve manufacturing cost.
In addition, due to manufacture, will lead to polaroid in the production process of polaroid and various defects occur.As a result,
The Defect Detection of polaroid seems essential for the manufacturing process of polaroid.
Currently, external some big polaroid manufacturers, such as: South Korea's Samsung, SUMITOMO CHEMICAL have dedicated polarisation
Equipment is examined in piece flaw inspection, and equipment and technology is monopolized by Ji Jia foreign corporation, expensive.
From the point of view of domestic entire industry, equipment popularity rate is low, as domestic cost of labor increases year by year, the market demand compared with
It is high；Meanwhile similar equipment has very high technical threshold.Therefore, whether from market demand or from the art wall of equipment
From the point of view of building, researching and developing the system all has positive meaning.
Summary of the invention
Based on above-mentioned technical problem, the present invention provides a kind of polaroid Defect Detection system and its detection method, main to wrap
It includes to detect from product pan feeding and trigger, picture obtains, vision algorithm processing, the functions such as a result output and data storage, the system
Cooperate corresponding hardware that can quickly be deployed to production scene, adaptation and various various sizes of products, has and answer well
Use prospect.
In order to solve the above-mentioned technical problem, the technical scheme adopted by the invention is that:
A kind of polaroid Defect Detection system, it is characterised in that: including server；It is passed with the pan feeding of server electrical connection
Sensor, bubble measuring station, beats trace measuring station, folding line measuring station and PLC controller at Mark measuring station；The PLC controller with
The bubble measuring station beats trace measuring station, the electrical connection of folding line measuring station, is also connected with discharging sensing in the PLC controller
Device.
Preferably, the Mark measuring station sweeps camera imaging system and periphery Encoder1 including a PC, line, wherein institute
It states peripheral Encoder1 signal and line sweeps camera and is connected respectively with server；The bubble measuring station includes PC, a Liang Tai
Line sweeps camera imaging system and periphery Encoder2；The trace measuring station of beating includes a PC, more table top array camera imaging systems
With peripheral Encoder2；The folding line measuring station includes a PC, more table top array camera imaging systems and periphery Encoder2；Institute
Peripheral Encoder2 signal is stated to be connected with server and PLC controller respectively, wherein two lines sweep camera with it is described
Server is connected；More table top array camera imaging systems are electrically connected with PLC controller.
Preferably, the server includes database and the query interface that connect with the database.
A kind of detection method using polaroid Defect Detection system, characterized by the following steps:
(1) product is lined up, and for managing the product in current system, is lined up after product pan feeding into product, is produced after sorting
Product are lined up, it then follows the principle of first in, first out；
(2) tracking and triggering of polaroid position using sensor plus encoder assembles and assists corresponding algorithm inclined to track
Mating plate detection system position and trigger the camera at corresponding function station and take pictures and sort；
(3) Defect Detection successively carries out Mark detection, bubble inspection to the tracking and triggering of polaroid position according in step (2)
It surveys, beat trace detection and total four Detection tasks of folding line detection, specific implementation are as follows:
(a) Mark is detected, and for detecting automatic Mark and figure number Mark, including a PC, line sweep camera imaging system and periphery
Whether Encoder1, Encoder1 tracking have product to flow into our station, if it find that there is product inflow, then PC triggers camera and takes pictures simultaneously
Image algorithm processing is executed, server is finally write the result into；
(b) bubble detects, and for detecting polaroid bubble artifacts, including a PC, two lines sweep camera imaging system and periphery
Whether Encoder2, Encoder2 tracking have product to flow into our station, if it find that there is product inflow, then PC triggers camera and takes pictures simultaneously
Image algorithm processing is executed, server is finally write the result into；
(c) trace detection is played, trace flaw is beaten for detecting polaroid, predominantly detects the concave-convex defect inside polaroid, including one
Whether PC, more table top array camera imaging systems and periphery Encoder2, Encoder2 tracking have product to flow into our station, if it find that
There is product inflow, then PC triggers camera and takes pictures and execute image algorithm processing, finally writes the result into server；
(d) folding line detects, for detecting polaroid folding line flaw, including a PC, more table top array camera imaging systems and periphery
Whether Encoder2, Encoder2 tracking have product to flow into our station, if it find that there is product inflow, then PC triggers camera and takes pictures simultaneously
Image algorithm processing is executed, server is finally write the result into；
(4) distributed computing, the distributed computing platform based on WCF technology building simultaneously are supported to extend, by four inspections in step (3)
The result of survey station is aggregated on corresponding product, and corresponding data are stored in database.
Preferably, image processing algorithm employed in four Detection tasks uses GPU to accelerate skill in the step (3)
Art.
Preferably, add encoder tracking polaroid in the position of detection system for aiding sensors in the step (2)
And the camera for triggering corresponding function station is taken pictures, algorithmic procedure includes: to measure encoder single pulse in belt transmission process
Transmission distance is u, and pan feeding inductor is L with a distance from measuring station, then is converted into umber of pulse d=L/u of encoder, is based on this
Principle, the process of product tracking are as follows: when pan feeding inductor senses product, record the position d0 of current encoder, be denoted as
Product enters material position, opens the position d1 that thread reads encoder in real time, when | d1- d0-L/u | when < Δ, pass through IO card output one
The pulse of a fixed pulse width, triggering camera is taken pictures, while opening the Detection task at current detection station, and wherein Δ is according to system reality
Situation is set as 0.2 or 0.1 or other smaller numerical value.
Preferably, PC triggering camera is taken pictures when Mark is detected in the step (3) and the image algorithm processing executed includes
Pretreatment, main processing block and post-processing；It wherein pre-processes, including input picture, extracts polarizer region, evades partial region
And filtering；The main processing block includes that the judgement of ultraphotic open country, automatic Symbolic judgement and the judgement of OCV template matching, above three are sentenced
As long as disconnected be only that the result of affirmative just can enter the OK of post-processing as a result, the result for having any one to be negated in the process
Enter the NG process of post-processing.
Preferably, PC triggering camera is taken pictures when bubble detects in the step (3) and the image algorithm processing executed includes
Pretreatment, main processing block and post-processing, wherein pretreatment includes input picture, extracts polarizer region, evades partial region
It is calculated with polarizing film mean value/variance, the main processing block includes non-edge ring detection zone and edge ring detection zone, then to non-side
Edge ring detection zone and edge ring detection zone carry out algorithm process respectively and obtain corresponding doubtful bubble feature, and the post-processing includes
The doubtful bubble feature of acquirement is merged, and postsearch screening is carried out by algorithmic rule and finally determines bubble artifacts position.
Preferably, PC triggering camera is taken pictures when playing trace detection in the step (3) and the image algorithm processing executed includes
Original graph is subjected to FFT transform, chooses suitable filter filtering, FFT inverse transformation, is highlighted finally by algorithm process and beats trace.
Preferably, PC triggering camera is taken pictures when folding line detects in the step (3) and the image algorithm processing executed includes
Pretreatment, main processing block and post-processing, wherein pretreatment includes extracting polarizer region, evading partial region and mean value filter
Wave, main processing block include informal voucher line algorithm process, black streaking algorithm process and marginal band algorithm process, above three judgement
As long as be only in the process affirmative result just can enter the OK of post-processing as a result, have any one negated result i.e.
Into the NG process of post-processing.
The beneficial effects of the present invention are: mainly being examined from Mark by a kind of polaroid Defect Detection system and detection method
Survey, bubble detection, beat trace detection, folding line detection etc. multiple directions to polaroid carry out complete detection, including from product pan feeding inspection
Triggering is surveyed, picture obtains, vision algorithm processing, and as a result output and data storage, the system cooperate corresponding hardware can be fast
Speed is deployed to production scene, adaptation and various various sizes of products, has a good application prospect.
Detailed description of the invention
Fig. 1 is a kind of system block diagram of polaroid Defect Detection system of the present invention；
Image processing algorithm flow chart when Fig. 2 is a kind of Mark detection of polaroid flaw detection method of the present invention；
Image processing algorithm flow chart when Fig. 3 is a kind of bubble detection of polaroid flaw detection method of the present invention；
Fig. 4 is the image processing algorithm flow chart of a kind of polaroid flaw detection method of the present invention beaten when trace detects；
Image processing algorithm flow chart when Fig. 5 is a kind of folding line detection of polaroid flaw detection method of the present invention.
Specific embodiment
Below in conjunction with Figure of description, the present invention is described further.
As shown in Figure 1, a kind of polaroid Defect Detection system, enters including server, with what the server was electrically connected
Material sensor, bubble measuring station, plays trace measuring station, folding line measuring station and PLC controller, the PLC control at Mark measuring station
Device and the bubble measuring station beat trace measuring station, the electrical connection of folding line measuring station, are also connected with discharging in the PLC controller
Sensor.The detection system main contents include electrical control, upper computer software and Machine Vision Detection algorithm, main process packet
It includes to detect from product pan feeding and trigger, picture obtains, vision algorithm processing, as a result output and data storage.The inspection software is matched
Production scene, adaptation and various various sizes of products can be quickly deployed to by closing corresponding hardware, have application well
Prospect.
The Mark measuring station includes a PC, line sweeps camera imaging system and periphery Encoder1 signal accesses, wherein
The periphery Encoder1 signal and line sweep camera and are connected respectively with server；The bubble measuring station includes PC, two
Platform line sweeps camera imaging system and the access of periphery Encoder2 signal；The trace measuring station of beating includes a PC, more table top battle array phases
Machine imaging system and periphery Encoder2 signal access；The folding line measuring station includes a PC, more table top array cameras imaging system
System and the access of periphery Encoder2 signal；The periphery Encoder2 signal is connected with server and PLC controller respectively,
Described in two lines sweep camera and be connected to the server and connect, more table top array camera imaging systems are electrical with PLC controller
Connection.It is respectively used to realize the Mark detection to polaroid, bubble detection, plays trace detection, folding line detection, realize to polaroid
The Omnibearing automatic of each performance indicator detects.
The server includes database and the query interface that connect with the database, facilitates client query.
The present invention also provides a kind of polaroid flaw detection method, include the following steps: that (1) product is lined up, for managing
The product in current system is managed, product is carried out after product pan feeding and lines up, and sub-elect product and line up, it then follows the original of first in, first out
Then；(2) tracking and triggering of polaroid position using sensor plus encoder assembles and assists corresponding algorithm and tracks polarisation
Piece detection system position and trigger the camera at corresponding function station and take pictures and sort；(3) Defect Detection, according to step (2)
In Mark detection, bubble detection successively carried out to the tracking of polaroid position and triggering, play trace detection and folding line detection amounts to four
A Detection task, specific implementation are as follows:
(a) Mark is detected, and for detecting automatic Mark and figure number Mark, including a PC, line sweep camera imaging system and periphery
The access of Encoder1 signal, whether Encoder1 tracking has product to flow into our station, if it find that there is product inflow, then PC triggers phase
Machine takes pictures and executes image algorithm processing, finally writes the result into server；Wherein, image algorithm processing is as shown in Fig. 2, packet
Pretreatment, main processing block and post-processing are included, wherein pre-process, including input picture, polarizer region is extracted, evades part area
Domain and filtering, the main processing block include the judgement of ultraphotic open country, automatic Symbolic judgement and the judgement of OCV template matching, above three
As long as be only in deterministic process affirmative result just can enter the OK of post-processing as a result, have any one negated knot
Fruit is the NG process for entering post-processing.
(b) bubble detects, for detecting polaroid bubble artifacts, including a PC, two lines sweep camera imaging system and
Peripheral Encoder2 signal access, whether Encoder2 tracking has product to flow into our station, if it find that there is product inflow, then PC is touched
Hair camera takes pictures and executes image algorithm processing, finally writes the result into server；Wherein, image algorithm processing is such as Fig. 3 institute
Show, including pretreatment, main processing block and post-processing, wherein pretreatment includes input picture, extracts polarizer region, evades portion
Subregion and polarizing film mean value/variance calculate, and the main processing block includes non-edge ring detection zone and edge ring detection zone, then
Algorithm process is carried out to non-edge ring detection zone and edge ring detection zone respectively and obtains corresponding doubtful bubble feature, is located after described
Reason includes that the doubtful bubble feature that will be obtained merges, and carries out postsearch screening by algorithmic rule and finally determine bubble artifacts place
Position.
(c) trace detection is played, trace flaw is beaten for detecting polaroid, predominantly detects the concave-convex defect inside polaroid, including
Whether one PC, more table top array camera imaging systems and the access of periphery Encoder2 signal, Encoder2 tracking have product inflow
Our station, if it find that there is product inflow, then PC triggers camera and takes pictures and execute image algorithm processing, finally writes the result into service
Device；Wherein, image algorithm processing as shown in figure 4, include original graph is carried out FFT transform, choose suitable filter filtering,
FFT inverse transformation highlights finally by algorithm process and beats trace.
(d) folding line detects, for detecting polaroid folding line flaw, including a PC, more table top array camera imaging systems and
Peripheral Encoder2 signal access, whether Encoder2 tracking has product to flow into our station, if it find that there is product inflow, then PC is touched
Hair camera takes pictures and executes image algorithm processing, finally writes the result into server；Image algorithm processing is as shown in figure 5, packet
Pretreatment, main processing block and post-processing are included, wherein pretreatment includes extracting polarizer region, evading partial region and mean value filter
Wave, main processing block include informal voucher line algorithm process, black streaking algorithm process and marginal band algorithm process, above three judgement
As long as be only in the process affirmative result just can enter the OK of post-processing as a result, have any one negated result i.e.
Into the NG process of post-processing.
(4) distributed computing, the distributed computing platform based on WCF technology building simultaneously are supported to extend, will be four in step (3)
The result of a measuring station is aggregated on corresponding product, and corresponding data are stored in database.Point based on WCF technology building
Cloth computation model simultaneously is supported to extend.Since each station is distributed in different places and in order to accelerate arithmetic speed, each station task
An at least industrial personal computer, and need the result of every station task being aggregated into corresponding product.
During image procossing, using corresponding vision algorithm, different templates data are established according to different products
And parameter, the product to differ greatly can also be handled, there is very big suitability and stability.
Image processing algorithm employed in four Detection tasks uses GPU acceleration technique in the step (3), and described
Add encoder tracking polaroid in the position of detection system for aiding sensors in step (2) and triggers the phase at corresponding function station
Machine is taken pictures, and algorithmic procedure includes: in belt transmission process, and measuring encoder single pulse transmission distance is u, pan feeding induction
Device is L with a distance from measuring station, then is converted into umber of pulse d=L/u of encoder, is based on this principle, the process of product tracking
Are as follows: when pan feeding inductor senses product, the position d0 of current encoder is recorded, product is denoted as and enters material position, open thread
The position d1 for reading encoder in real time, when | d1- d0-L/u | < Δ meets, and the arteries and veins of a fixed pulse width is exported by IO card
Punching, triggering camera are taken pictures, while opening the Detection task at current detection station, and wherein Δ is the offset of very little, and theoretically 0, but
It is the stability and reliability in order to improve system, this value can be set to 0.1 or 0.2 or other smaller numbers, class
Like fuzzy control method, this value is adjusted according to the actual conditions of system.
It is obvious to a person skilled in the art that invention is not limited to the details of the above exemplary embodiments, Er Qie
In the case where without departing substantially from spirit or essential attributes of the invention, the present invention can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power
Benefit requires rather than above description limits, it is intended that by fall in claim with all in the meaning and scope of important document
Variation is included within the present invention.Any reference signs in the claims should not be construed as limiting the involved claims.
In addition, it should be understood that although this specification is described in terms of embodiments, but not each embodiment is only wrapped
Containing an independent technical solution, this description of the specification is merely for the sake of clarity, and those skilled in the art should
It considers the specification as a whole, the technical solutions in the various embodiments may also be suitably combined, forms those skilled in the art
The other embodiments being understood that.
Claims (10)
1. a kind of polaroid Defect Detection system, it is characterised in that: the pan feeding including server, with server electrical connection
Sensor, bubble measuring station, beats trace measuring station, folding line measuring station and PLC controller at Mark measuring station；The PLC controller
With the bubble measuring station, beat trace measuring station, folding line measuring station is electrically connected, discharging is also connected in the PLC controller and is passed
Sensor.
2. a kind of polaroid Defect Detection system according to claim 1, it is characterised in that: the Mark measuring station includes
One PC, line sweep camera imaging system and periphery Encoder1, wherein the periphery Encoder1 signal and line sweep camera difference
It is connected with server；The bubble measuring station sweeps camera imaging system and periphery Encoder2 including a PC, two lines；Institute
Stating and beating trace measuring station includes a PC, more table top array camera imaging systems and periphery Encoder2；The folding line measuring station includes
One PC, more table top array camera imaging systems and periphery Encoder2；The periphery Encoder2 signal respectively with server and
PLC controller is connected, wherein two lines are swept camera and are connected to the server and connect；More table top array camera imaging systems
It is electrically connected with PLC controller.
3. a kind of polaroid Defect Detection system according to claim 1, it is characterised in that: the server includes data
Library and the query interface being connect with the database.
4. a kind of detection method using the polaroid Defect Detection system as described in any one of claim 1-3, special
Sign is: including the following steps:
(1) product is lined up, and for managing the product in current system, is lined up after product pan feeding into product, is produced after sorting
Product are lined up, it then follows the principle of first in, first out；
(2) tracking and triggering of polaroid position using sensor plus encoder assembles and assists corresponding algorithm inclined to track
Mating plate detection system position and trigger the camera at corresponding function station and take pictures and sort；
(3) Defect Detection successively carries out Mark detection, bubble inspection to the tracking and triggering of polaroid position according in step (2)
It surveys, beat trace detection and total four Detection tasks of folding line detection, specific implementation are as follows:
(a) Mark is detected, and for detecting automatic Mark and figure number Mark, including a PC, line sweep camera imaging system and periphery
Whether Encoder1, Encoder1 tracking have product to flow into our station, if it find that there is product inflow, then PC triggers camera and takes pictures simultaneously
Image algorithm processing is executed, server is finally write the result into；
(b) bubble detects, and for detecting polaroid bubble artifacts, including a PC, two lines sweep camera imaging system and periphery
Whether Encoder2, Encoder2 tracking have product to flow into our station, if it find that there is product inflow, then PC triggers camera and takes pictures simultaneously
Image algorithm processing is executed, server is finally write the result into；
(c) trace detection is played, trace flaw is beaten for detecting polaroid, predominantly detects the concave-convex defect inside polaroid, including one
Whether PC, more table top array camera imaging systems and periphery Encoder2, Encoder2 tracking have product to flow into our station, if it find that
There is product inflow, then PC triggers camera and takes pictures and execute image algorithm processing, finally writes the result into server；
(d) folding line detects, for detecting polaroid folding line flaw, including a PC, more table top array camera imaging systems and periphery
Whether Encoder2, Encoder2 tracking have product to flow into our station, if it find that there is product inflow, then PC triggers camera and takes pictures simultaneously
Image algorithm processing is executed, server is finally write the result into；
(4) distributed computing, the distributed computing platform based on WCF technology building simultaneously are supported to extend, by four inspections in step (3)
The result of survey station is aggregated on corresponding product, and corresponding data are stored in database.
5. a kind of polaroid flaw detection method according to claim 4, it is characterised in that: four in the step (3)
Image processing algorithm employed in Detection task uses GPU acceleration technique.
6. a kind of polaroid flaw detection method according to claim 4, it is characterised in that: be used in the step (2)
Aiding sensors add encoder tracking polaroid in the position of detection system and trigger the camera at corresponding function station and take pictures, algorithm
Process includes: in belt transmission process, and measuring encoder single pulse transmission distance is u, and pan feeding inductor is from measuring station
Distance is L, then is converted into umber of pulse d=L/u of encoder, is based on this principle, the process of product tracking are as follows: when pan feeding incudes
When device senses product, the position d0 of current encoder is recorded, product is denoted as and enters material position, thread is opened and reads encoder in real time
Position d1, as | d1- d0-L/u | when < Δ, the pulse of a fixed pulse width is exported by IO card, triggering camera is taken pictures, together
The Detection task at Shi Kaiqi current detection station, wherein Δ is set as 0.2 or 0.1 or other smaller numbers according to system actual conditions
Value.
7. a kind of polaroid flaw detection method according to claim 4, it is characterised in that: Mark in the step (3)
PC triggering camera is taken pictures when detection and the image algorithm processing executed includes pretreatment, main processing block and post-processing；It is wherein pre-
Processing, including input picture, extract polarizer region, evade partial region and filtering；The main processing block includes ultraphotic open country
Judgement, automatic Symbolic judgement and the judgement of OCV template matching, the result in above three deterministic process being only affirmative could be into
As long as entering the OK of post-processing as a result, the result for having any one to be negated enters the NG process of post-processing.
8. a kind of polaroid flaw detection method according to claim 4, it is characterised in that: bubble in the step (3)
PC triggering camera is taken pictures when detection and the image algorithm processing executed includes pretreatment, main processing block and post-processing, wherein in advance
Processing includes input picture, extracts polarizer region, evades partial region and polarizing film mean value/variance calculating, the main process task
Module includes non-edge ring detection zone and edge ring detection zone, then is carried out respectively to non-edge ring detection zone and edge ring detection zone
Algorithm process obtains corresponding doubtful bubble feature, and the post-processing includes that the doubtful bubble feature that will be obtained merges, and passes through
Algorithmic rule carries out postsearch screening and finally determines bubble artifacts position.
9. a kind of polaroid flaw detection method according to claim 4, it is characterised in that: beat trace in the step (3)
PC triggering camera is taken pictures when detection and the image algorithm processing executed includes that original graph is carried out to FFT transform, chooses suitable filter
The filtering of wave device, FFT inverse transformation, highlight finally by algorithm process and beat trace.
10. a kind of polaroid flaw detection method according to claim 4, it is characterised in that: folding line in the step (3)
PC triggering camera is taken pictures when detection and the image algorithm processing executed includes pretreatment, main processing block and post-processing, wherein in advance
Processing includes extracting polarizer region, evading partial region and mean filter, and main processing block includes informal voucher line algorithm process, black
Striped algorithm process and marginal band algorithm process, be only in above three deterministic process affirmative result just can enter after
As long as the OK of processing is as a result, the result for having any one to be negated enters the NG process of post-processing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811350658.XA CN109540902A (en) | 2018-11-14 | 2018-11-14 | A kind of polaroid Defect Detection system and its detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811350658.XA CN109540902A (en) | 2018-11-14 | 2018-11-14 | A kind of polaroid Defect Detection system and its detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109540902A true CN109540902A (en) | 2019-03-29 |
Family
ID=65847073
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811350658.XA Pending CN109540902A (en) | 2018-11-14 | 2018-11-14 | A kind of polaroid Defect Detection system and its detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109540902A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110090808A (en) * | 2019-05-17 | 2019-08-06 | 智翼博智能科技(苏州)有限公司 | Polaroid full-automatic charging appearance detecting device and detection method |
| CN110153020A (en) * | 2019-05-17 | 2019-08-23 | 智翼博智能科技(苏州)有限公司 | Polaroid full-automatic charging Mark detection device and detection method |
| CN111036578A (en) * | 2019-12-27 | 2020-04-21 | 无锡动视宫原科技有限公司 | Detection method and detection system for fine concave-convex points on surface of polarizer |
| CN111539952A (en) * | 2020-05-15 | 2020-08-14 | 南京工程学院 | Scratch detection Internet of things system and scratch detection and result sharing method |
| CN113030095A (en) * | 2021-02-24 | 2021-06-25 | 杭州微纳智感光电科技有限公司 | Polaroid appearance defect detecting system |
| CN115330693A (en) * | 2022-07-18 | 2022-11-11 | 广州超音速自动化科技股份有限公司 | Multi-system labeling signal combination control method, system and platform |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1576828A (en) * | 2003-07-28 | 2005-02-09 | 日东电工株式会社 | Inspection method and system for sheet products |
| CN101086483A (en) * | 2006-06-09 | 2007-12-12 | 富士胶片株式会社 | Apparatus and method for inspecting film defect |
| CN102590221A (en) * | 2012-02-24 | 2012-07-18 | 深圳大学 | Apparent defect detecting system and detecting method of polarizer |
| CN103048333A (en) * | 2012-12-07 | 2013-04-17 | 北京优纳科技有限公司 | Appearance detection equipment and method |
| CN103837545A (en) * | 2014-03-14 | 2014-06-04 | 华中科技大学 | Lens imaging device and method |
| CN206192914U (en) * | 2016-11-05 | 2017-05-24 | 陕西东方航空仪表有限责任公司 | Magnetic ring part visual detection control system with pulse location |
| CN106872487A (en) * | 2017-04-21 | 2017-06-20 | 佛山市南海区广工大数控装备协同创新研究院 | The surface flaw detecting method and device of a kind of view-based access control model |
| CN107315017A (en) * | 2017-08-02 | 2017-11-03 | 苏州市朗电机器人有限公司 | Optical film Visual intelligent detection device |
| CN207158348U (en) * | 2017-08-14 | 2018-03-30 | 翰博高新材料(合肥)股份有限公司 | A kind of polaroid cleans automatic putting device |
| CN108535253A (en) * | 2017-03-03 | 2018-09-14 | 住友化学株式会社 | Annotation equipment, defect inspecting system and film manufacturing method |
| CN207992081U (en) * | 2018-02-02 | 2018-10-19 | 嵊州市东浩电子科技有限公司 | A kind of optical film defect detection equipment |
-
2018
- 2018-11-14 CN CN201811350658.XA patent/CN109540902A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1576828A (en) * | 2003-07-28 | 2005-02-09 | 日东电工株式会社 | Inspection method and system for sheet products |
| CN101086483A (en) * | 2006-06-09 | 2007-12-12 | 富士胶片株式会社 | Apparatus and method for inspecting film defect |
| CN102590221A (en) * | 2012-02-24 | 2012-07-18 | 深圳大学 | Apparent defect detecting system and detecting method of polarizer |
| CN103048333A (en) * | 2012-12-07 | 2013-04-17 | 北京优纳科技有限公司 | Appearance detection equipment and method |
| CN103837545A (en) * | 2014-03-14 | 2014-06-04 | 华中科技大学 | Lens imaging device and method |
| CN206192914U (en) * | 2016-11-05 | 2017-05-24 | 陕西东方航空仪表有限责任公司 | Magnetic ring part visual detection control system with pulse location |
| CN108535253A (en) * | 2017-03-03 | 2018-09-14 | 住友化学株式会社 | Annotation equipment, defect inspecting system and film manufacturing method |
| CN106872487A (en) * | 2017-04-21 | 2017-06-20 | 佛山市南海区广工大数控装备协同创新研究院 | The surface flaw detecting method and device of a kind of view-based access control model |
| CN107315017A (en) * | 2017-08-02 | 2017-11-03 | 苏州市朗电机器人有限公司 | Optical film Visual intelligent detection device |
| CN207158348U (en) * | 2017-08-14 | 2018-03-30 | 翰博高新材料(合肥)股份有限公司 | A kind of polaroid cleans automatic putting device |
| CN207992081U (en) * | 2018-02-02 | 2018-10-19 | 嵊州市东浩电子科技有限公司 | A kind of optical film defect detection equipment |
Non-Patent Citations (1)
| Title |
|---|
| 赖文威: "偏光片外观缺陷成像机理与检测技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110090808A (en) * | 2019-05-17 | 2019-08-06 | 智翼博智能科技(苏州)有限公司 | Polaroid full-automatic charging appearance detecting device and detection method |
| CN110153020A (en) * | 2019-05-17 | 2019-08-23 | 智翼博智能科技(苏州)有限公司 | Polaroid full-automatic charging Mark detection device and detection method |
| CN111036578A (en) * | 2019-12-27 | 2020-04-21 | 无锡动视宫原科技有限公司 | Detection method and detection system for fine concave-convex points on surface of polarizer |
| CN111539952A (en) * | 2020-05-15 | 2020-08-14 | 南京工程学院 | Scratch detection Internet of things system and scratch detection and result sharing method |
| CN113030095A (en) * | 2021-02-24 | 2021-06-25 | 杭州微纳智感光电科技有限公司 | Polaroid appearance defect detecting system |
| CN115330693A (en) * | 2022-07-18 | 2022-11-11 | 广州超音速自动化科技股份有限公司 | Multi-system labeling signal combination control method, system and platform |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109540902A (en) | A kind of polaroid Defect Detection system and its detection method | |
| JP3132565B2 (en) | Defect inspection method and apparatus | |
| Li et al. | Automatic industry PCB board DIP process defect detection system based on deep ensemble self-adaption method | |
| CN102374993B (en) | Unit and method for detecting appearance of three dimensional pin of integrated circuit package chip | |
| CN103839255A (en) | Method and device for detecting video image matting tampering | |
| CN113256570A (en) | Visual information processing method, device, equipment and medium based on artificial intelligence | |
| CN109283182A (en) | A kind of detection method of battery welding point defect, apparatus and system | |
| CN116109777A (en) | Complementary pseudo-multi-mode feature system and method | |
| CN107561736B (en) | LCD defect detection method based on Fourier transform and Hough transform | |
| CN109165603A (en) | A kind of Ship Detection and device | |
| You | PCB defect detection based on generative adversarial network | |
| CN117825393A (en) | Online monitoring system and method for automatic production line of polaroid | |
| Ghamisi et al. | Anomaly detection in automated fibre placement: Learning with data limitations | |
| CN111738963B (en) | Tobacco bale package appearance detection method based on deep self-learning | |
| CN108921642A (en) | One kind intelligent fruits and vegetables checkout apparatus based on computer vision and method | |
| CN103984967B (en) | A kind of automatic checkout system being applied to Commercial goods labels detection and automatic testing method | |
| CN117030612A (en) | Household film element surface defect detection system and detection method based on machine vision | |
| CN115947066B (en) | Belt tearing detection method, device and system | |
| Niskanen et al. | Experiments with SOM based inspection of wood | |
| CN110472085A (en) | 3-D image searching method, system, computer equipment and storage medium | |
| CN115601686A (en) | Method, device and system for confirming delivery of articles | |
| Xiong | A design of bare printed circuit board defect detection system based on YOLOv8 | |
| JP7460391B2 (en) | LEARNING DEVICE, IMAGE INSPECTION DEVICE, PROGRAM, AND LEARNING METHOD | |
| CN113624759A (en) | Apple nondestructive testing method based on machine learning | |
| Jaffery et al. | Real time visual inspection system (RTVIS) for calibration of mechanical gauges |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190329 |