CN100347843C - Apparatus for automatically distinguishing and aligning wafer of wafer cutter and method thereof - Google Patents
Apparatus for automatically distinguishing and aligning wafer of wafer cutter and method thereof Download PDFInfo
- Publication number
- CN100347843C CN100347843C CNB2005102006361A CN200510200636A CN100347843C CN 100347843 C CN100347843 C CN 100347843C CN B2005102006361 A CNB2005102006361 A CN B2005102006361A CN 200510200636 A CN200510200636 A CN 200510200636A CN 100347843 C CN100347843 C CN 100347843C
- Authority
- CN
- China
- Prior art keywords
- wafer
- feature pattern
- computer
- motion control
- image
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Dicing (AREA)
- Container, Conveyance, Adherence, Positioning, Of Wafer (AREA)
Abstract
The present invention relates to a device for automatically distinguishing and aligning wafers of a wafer scriber and a method thereof. The wafer scriber comprises a digital CCD video collecting machine, a digital image collecting card, a computer, a motion control card, a servo driving mechanism and a working table, wherein wafer feature pattern video signals which are collected by the digital CCD video collecting machine are input in the digital image collecting card; the computer receives the video signals of the digital image collecting card, standard wafer feature pattern template data stored in advance and the program of an image analysis system are input in the computer; the system compares the similarity between the received wafer feature pattern template data and standard template gray data stored in advance; the data of the image analysis result of the computer is input in the motion control card; pulse signals and direction signals of the motion control card are input in the servo driving mechanism; the working table can be movably adjusted in an X, Y two-dimension direction. The present invention greatly enhances the automation degree of the wafer scriber, the processing efficiency and the aligning precision.
Description
Technical field
The present invention relates to a kind of machine vision alignment system, particularly a kind of scribing machine is discerned the technique of alignment field automatically.
Background technology
Along with the continuous expansion and the development of IC industry and IT industry, the application of semiconductor chip is more and more wider, and scribing machine is the key equipment on the packaging technology link of back.Scribing machine is mainly used in silicon integrated circuit, light-emitting diode, the cutting-up processing of materials such as lithium niobate, piezoelectric ceramic, quartz, GaAs, sapphire, aluminium oxide, iron oxide, glass.Scribing machine is the key equipment on the semiconductor package wiring, and its function is that the very little wafer of 4-6 is cut into unit component.Wafer refers to the monocrystalline silicon disk more, refines by ordinary silicon salad system to form, and be the most frequently used semi-conducting material, be divided into specifications such as 4 inches, 5 inches, 6 inches, 8 inches by its diameter.Recently develop and 12 inches even bigger specification.Wafer is big more, and producible IC is just many on the same disk, can reduce cost; But require material technology and production technology higher.The situation of analyzing semiconductor market development, Chinese semiconductor industry will enter the period of a fast development.Estimate in following 5 years, with the growth rate that keeps about 20%.Accordingly, the demand of semiconductor production equipment also will have nearly 20% growth rate, and this research and development and production for scribing machine equipment is the new opportunity of a development.The wafer attenuate is on special equipment, grinds from wafer rear, wafer is thinned to the degree that is fit to encapsulation.Because the size increasing (from 4 inches, 5 inches, 6 inches, developing into 8 inches even 12 inches) of wafer in order to increase the mechanical strength of wafer, prevents that wafer from deforming, ftractureing in the course of processing, the thickness of wafer is also increasing always.But along with system develops towards compact direction, the thickness of module thins down after the Chip Packaging, therefore, before encapsulation, the reduced thickness of wafer must be arrived the acceptable degree, to satisfy the requirement of chip assembling.As 6 inches wafers, thickness is about 675 microns, is generally 150 microns behind the attenuate.In the operation of wafer attenuate, stressed uniformity will be crucial, otherwise wafer is easy to distortion, cracking.Behind the wafer attenuate, can carry out scribing.Scribing machine in the past is manually operated, and different processing wafers have different graphics chips, and each processing all needs to pass through operation manually later on by microscope imaging, and efficient is low, makes the automaticity of entire equipment be subjected to great restriction.
Summary of the invention
The purpose of this invention is to provide a kind of Automatic Wafer Identification Alignment method of Dicing Machine, solve the self-aligning problem of wafer; And solution further improves alignment speed, precision and consistency problem.
Technical scheme of the present invention:
This Automatic Wafer Identification Alignment method of Dicing Machine is characterized in that comprising following six parts:
A digital CCD image collection machine of gathering the wafer characteristic image;
A digital image acquisition card of pegging graft with main frame, it is imported the wafer feature pattern vision signal that above-mentioned digital CCD image collection machine is gathered, and this digital image acquisition card carries out denoising slicing, binaryzation, feature pattern extraction, carries out the matched and searched preliminary treatment according to feature pattern image;
A computer that receives above-mentioned digital image acquisition card vision signal, it is imported the standard form data of storing the wafer signature in advance, and the program of input picture analytical system, this system compares wafer circle feature pattern video signal data that receives and the similarity that prestores between the standard form gray feature flag data, according to analysis result planning movement locus, movement locus is changed into the command language of motion control card;
A motion control card of pegging graft with main frame is to its input aforementioned calculation machine graphical analysis result's command language data-signal;
A servo-actuating device is imported the pulse signal and the direction signal of above-mentioned motion control card output to it;
One at the equal workbench of removable adjustment of X, Y two-dimensional directional, and this workbench is connected with the output shaft of above-mentioned servo-actuating device.
This Automatic Wafer Identification Alignment method of Dicing Machine is characterized in that may further comprise the steps:
(1) digital image acquisition: to image card, image card carries out denoising slicing, binaryzation, feature pattern to the picture signal of input to extract, carries out the matched and searched preliminary treatment according to the feature pattern of wafer digital CCD image collection machine with the video signal transmission of wafer;
(2) extract the signature data of the standard form that prestores, carry out matched and searched, adjust workbench, search successfully until the feature pattern identification of wafer;
(3) result who obtains is delivered to the shared drive data segment, uses for motion control card, and motion control card drives servo-actuating device according to lookup result, adjusts workbench, until satisfying alignment request.
This Automatic Wafer Identification Alignment method of Dicing Machine step also can be described as:
A, in the memory of computer, store one group of wafer image standard form in advance based on gray scale;
B, wafer is placed under the camera lens of CCD video image acquisition machine;
C, the wafer greyscale video signal of gathering is passed to the digital image acquisition card, image is carried out denoising slicing, binaryzation, feature pattern extract, carry out the matched and searched preliminary treatment according to feature pattern;
D, through pretreated signal input computer, by the image analysis system of computer this signal data and the similarity that prestores between the standard form gray feature flag data are compared;
E, by computer registration signal is transferred to motion control card, workbench is adjusted in the motion control card transmission signals high speed and precision motion of giving servo-actuating device Control work platform again, forms a closed loop scribing machine wafer and discerns alignment device automatically;
F, circulation repeatedly, up to finding the wafer coordinate position of specifying the gray scale mark value to be complementary with template image.
The scribing machine wafer is discerned the alignment function method automatically, it is characterized in that may further comprise the steps:
Step 1: according to the difference rotary table of the angle that finds feature pattern and standard feature figure;
Step 2: the workbench X coordinate with this moment is a starting point, searches corresponding first feature pattern;
Step 3: travelling table directions X 1/4-3/4 wafer radius size, search corresponding second feature pattern;
Step 4:, adjust table mechanism according to the result who finds first and second feature pattern;
Step 5: the workbench X coordinate with this moment is a starting point, searches corresponding first feature pattern;
Step 6: 3/4-1 wafer radius size of travelling table directions X, search corresponding second feature pattern;
Step 7:, adjust table mechanism according to the result who finds first and second feature pattern;
Step 8: extract characteristic pattern and carry out accuracy checking, meet the demands to aim at and finish, do not satisfy precision, repeating step 1～7 is till precision satisfies condition.
The concrete workflow that this automatic identification alignment system is searched feature pattern is: the computer image analysis treatment system is carried out preliminary treatment with the digital picture that image pick-up card is transferred to computer, gets rid of the influence of other disturbance factors to real image; Next system carries out the extraction of feature pattern according to concrete wafer, and as the mark of aiming at, computer is searched feature pattern in the realtime graphic that the image card transmission comes; Leave in the shared data section of appointment with regard to the director data that carries out changing into behind the trajectory planning motion control card according to the result that obtains if find feature pattern, motion control card is adjusted workbench according to the data-driven servo-actuating device in the shared data section; Image pick-up card will be reorientated wafer image on the aftertable again and be gathered again and be transferred to computer, system can analyze, feature pattern is searched image once more, and the result who searches judged that alignment procedures just completes successfully if satisfy the alignment precision requirement.If feature pattern does not find, the abnormality processing data are used in the motion control meeting, table mechanism are adjusted, up to finding feature pattern.
Because there are many meetings factors that generation is disturbed to wafer mapper such as water in the processing environment of scribing machine, just must the image of image pick-up card collection be handled, and just can accurately analyze image.System reaches the purpose of aligning according to the actual motion that obtains the feature pattern result who finds being changed into behind the result of feature pattern workbench.
Because be subjected to the influence of processing environment, the image pick-up card images acquired needs to comprise preliminary treatment such as denoising, binaryzation after being transferred to computer.Because the requirement of alignment precision, workbench directions X need progressive accuracy and the last precision of aiming at that certain distance guarantees search procedure that move.
The present invention adopts digital CCD images acquired, image card is analyzed images, and the recognition software system analysing digital image is according to analysis result planning movement locus, movement locus is changed into the command language of motion control card, and the motion control card control is finally finished alignment function.Make the workbench of carrying wafer can under the control of computer, finish the alignment function of wafer automatically by actual needs, changed in the past and can only rely on manual operations, efficient is low, the situation that automaticity is low, effectively raise the working (machining) efficiency of equipment, alleviated labour intensity, further improved wafer working (machining) efficiency, alignment speed and precision, have the strong characteristics of consistency, improved automation degree of equipment greatly.
Description of drawings
Fig. 1 is a structured flowchart of the present invention.
Fig. 2 is the flow chart of alignment methods of the present invention.
Fig. 3 is the operational flowchart of alignment methods of the present invention.
Fig. 4 is the FB(flow block) of alignment methods step of the present invention.
Fig. 5 is the structure composition frame chart of the embodiment of the invention.
Fig. 6 is the alignment methods graphical analysis view of the embodiment of the invention.
Fig. 7 is that the alignment methods step images of the embodiment of the invention is analyzed view.
Embodiment
Embodiment is referring to Fig. 1,5, and the scribing machine wafer that the present invention describes is discerned alignment system automatically and comprised that following six soft or hard assemblies constitute: JAICV50 numeral CCD, CORECOBanditII CV digital image acquisition card, computer image analysis treatment system, PMAC2-PC﹠amp; LITE eight axis motion control cards, by X, Y, Z, the θ X-Y workbench of totally 4 servo-actuating devices that constitute of dimension motors, carrying wafer to be processed.
Referring to Fig. 2, the present invention discerns alignment system automatically and searches the concrete workflow of feature pattern and be: the computer image analysis treatment system is carried out preliminary treatment with the digital picture that image pick-up card is transferred to computer, gets rid of the influence of other disturbance factors to real image; Next system carries out the extraction of feature pattern according to concrete wafer, and as the mark of aiming at, computer is searched feature pattern in the realtime graphic that the image card transmission comes; Leave in the shared data section of appointment with regard to the director data that carries out changing into behind the trajectory planning motion control card according to the result that obtains if find feature pattern, motion control card is adjusted workbench according to the data-driven servo-actuating device in the shared data section; Image pick-up card will be reorientated wafer image on the aftertable again and be gathered again and be transferred to computer, system can analyze, feature pattern is searched image once more, and the result who searches judged that alignment procedures just completes successfully if satisfy the alignment precision requirement.If feature pattern does not find, the abnormality processing data are used in the motion control meeting, table mechanism are adjusted, up to finding feature pattern.
Referring to Fig. 3, scribing machine wafer of the present invention is discerned the method for operation that the feature pattern result who finds is changed into behind the result that alignment system obtains feature pattern the workbench actual motion automatically and is: step 1 is according to the difference rotary table that finds the angle of feature pattern and standard feature figure; Step 2 is a starting point with the workbench X coordinate of this moment, searches corresponding first feature pattern; Step 3 travelling table directions X 1/2 wafer radius size is searched corresponding second feature pattern; Step 4 is adjusted table mechanism according to the result who finds first and second feature pattern; Step 5 is a starting point with the workbench X coordinate of this moment, searches corresponding first feature pattern; Step 6 travelling table directions X is 1 wafer radius size, searches corresponding second feature pattern; Step 7 is adjusted table mechanism according to the result who finds first and second feature pattern; Step 8 is extracted characteristic pattern and is carried out accuracy checking, meets the demands to aim to finish, and does not satisfy precision, and repeating step 1～7 is till precision satisfies condition.
Referring to Fig. 4,6,7, scribing machine wafer of the present invention is discerned alignment system identification automatically and aimed at the algorithm that obtains feature pattern and be: (1) digitized video collection: image card carries out denoising slicing, binaryzation, defined feature figure to realtime graphic source image and extracts, carries out according to feature pattern preliminary treatment such as matched and searched; (2) extract standard form,, the defined feature mark of realtime graphic source images is searched for, mates, analyzes, searched, adjust workbench, search successfully until identification with the feature pattern comparison of realtime graphic source images; (3) result who obtains is delivered to the shared drive data segment, uses for sports cards, and sports cards is according to lookup result, and drive motors is adjusted workbench, adjusts the back until work and satisfies alignment request.
Claims (2)
1. Automatic Wafer Identification Alignment method of Dicing Machine is characterized in that using a kind of scribing machine wafer of following six parts that comprises and discerns alignment device automatically:
A digital CCD image collection machine of gathering the wafer characteristic image;
A digital image acquisition card of pegging graft with main frame, it is imported the wafer feature pattern vision signal that above-mentioned digital CCD image collection machine is gathered, and this digital image acquisition card carries out denoising slicing, binaryzation, feature pattern extraction, carries out the matched and searched preliminary treatment according to feature pattern image;
A computer that receives above-mentioned digital image acquisition card vision signal, it is imported the standard form data of storing the wafer signature in advance, and the program of input picture analytical system, this system compares wafer circle feature pattern video signal data that receives and the similarity that prestores between the standard form gray feature flag data, according to analysis result planning movement locus, movement locus is changed into the command language of motion control card;
A motion control card of pegging graft with main frame is to its input aforementioned calculation machine graphical analysis result's command language data-signal;
A servo-actuating device is imported the pulse signal and the direction signal of above-mentioned motion control card output to it;
One at the equal workbench of removable adjustment of X, Y two-dimensional directional, and this workbench is connected with the output shaft of above-mentioned servo-actuating device;
Above-mentioned Automatic Wafer Identification Alignment method of Dicing Machine:
A, in the memory of computer, store one group of wafer image standard form in advance based on gray scale;
B, wafer is placed under the camera lens of CCD video image acquisition machine;
C, the wafer greyscale video signal of gathering is passed to the digital image acquisition card, image is carried out denoising slicing, binaryzation, feature pattern extract, carry out the matched and searched preliminary treatment according to feature pattern;
D, through pretreated signal input computer, by the image analysis system of computer this signal data and the similarity that prestores between the standard form gray feature flag data are compared;
E, by computer registration signal is transferred to motion control card, motion control card transmission signals is again given the precise motion of servo-actuating device Control work platform, adjusts workbench, forms a closed loop scribing machine wafer and discerns alignment device automatically;
F, circulation repeatedly, up to finding the wafer coordinate position of specifying the gray scale mark value to be complementary with template image.
2. Automatic Wafer Identification Alignment method of Dicing Machine according to claim 1 is characterized in that may further comprise the steps:
Step 1: according to the difference rotary table of the angle that finds feature pattern and standard feature figure;
Step 2: the workbench X coordinate with this moment is a starting point, searches corresponding first feature pattern;
Step 3: travelling table directions X 1/4-3/4 wafer radius size, search corresponding second feature pattern;
Step 4:, adjust table mechanism according to the result who finds first and second feature pattern;
Step 5: the workbench X coordinate with this moment is a starting point, searches corresponding first feature pattern;
Step 6: 3/4-1 wafer radius size of travelling table directions X, search corresponding second feature pattern;
Step 7:, adjust table mechanism according to the result who finds first and second feature pattern;
Step 8: extract characteristic pattern and carry out accuracy checking, meet the demands to aim at and finish, do not satisfy precision, repeating step 1～7 is till precision satisfies condition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005102006361A CN100347843C (en) | 2005-10-24 | 2005-10-24 | Apparatus for automatically distinguishing and aligning wafer of wafer cutter and method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005102006361A CN100347843C (en) | 2005-10-24 | 2005-10-24 | Apparatus for automatically distinguishing and aligning wafer of wafer cutter and method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1787201A CN1787201A (en) | 2006-06-14 |
| CN100347843C true CN100347843C (en) | 2007-11-07 |
Family
ID=36784588
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005102006361A Expired - Fee Related CN100347843C (en) | 2005-10-24 | 2005-10-24 | Apparatus for automatically distinguishing and aligning wafer of wafer cutter and method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100347843C (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8278595B2 (en) * | 2007-03-16 | 2012-10-02 | Electro Scientific Industries, Inc. | Use of predictive pulse triggering to improve accuracy in link processing |
| CN101996398B (en) * | 2009-08-12 | 2012-07-04 | 睿励科学仪器（上海）有限公司 | Image matching method and equipment for wafer alignment |
| CN102528950B (en) * | 2010-12-08 | 2014-12-10 | 北京中电科电子装备有限公司 | Debugging device of scribing machine material transmission system |
| CN103972121B (en) * | 2013-01-30 | 2017-03-01 | 正恩科技有限公司 | The disconnected GTG detection method of wafer |
| JP6224350B2 (en) * | 2013-05-17 | 2017-11-01 | 株式会社ディスコ | Processing equipment |
| CN105304541A (en) * | 2015-09-30 | 2016-02-03 | 秦皇岛视听机械研究所 | Automatic semiconductor wafer positioning device |
| CN107219782A (en) * | 2017-05-12 | 2017-09-29 | 芜湖乐佳自动化机械有限公司 | A kind of dust-proof control system of high-low pressure transformation cabinet precision |
| CN111477575B (en) * | 2020-04-10 | 2023-08-25 | 江苏京创先进电子科技有限公司 | Alignment method of dicing saw with multiple cutting channels |
| CN112730248A (en) * | 2020-12-29 | 2021-04-30 | 无锡圆方半导体测试有限公司 | Method and system for preventing chip test pattern from deviating |
| CN113172778A (en) * | 2021-04-28 | 2021-07-27 | 华虹半导体(无锡)有限公司 | Taizhou ring removing method and positioning device for Taizhou ring removing |
| CN114326597B (en) * | 2021-12-30 | 2022-07-22 | 沈阳和研科技有限公司 | Automatic tool capable of realizing movement of each shaft of dicing saw and using method thereof |
| CN114420619B (en) * | 2022-03-30 | 2022-07-01 | 晶芯成(北京)科技有限公司 | Wafer alignment device and method, and wafer testing method |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH08166534A (en) * | 1994-12-14 | 1996-06-25 | Toshiba Corp | Automatic focusing method for image processor |
-
2005
- 2005-10-24 CN CNB2005102006361A patent/CN100347843C/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH08166534A (en) * | 1994-12-14 | 1996-06-25 | Toshiba Corp | Automatic focusing method for image processor |
Non-Patent Citations (3)
| Title |
|---|
| 全自动划片机自动识别对准技术的研究 杨云龙，刘金荣.电子工业专用设备，第110期 2004 * |
| 高精度机器视觉对准***的研究与设计 汪宏昇，史铁林.光学技术，第30卷第2期 2004 * |
| 高精度自动贴片机视觉对准***及其图像处理 夏奇，周明才，汪宏昇，史铁林.光学技术，第30卷第2期 2004 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1787201A (en) | 2006-06-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100347843C (en) | Apparatus for automatically distinguishing and aligning wafer of wafer cutter and method thereof | |
| CN101057256A (en) | Face feature collating device, face feature collating method, and program | |
| CN111951237A (en) | Visual appearance detection method | |
| CN110660104A (en) | Industrial robot visual identification positioning grabbing method, computer device and computer readable storage medium | |
| CN108920992A (en) | A kind of positioning and recognition methods of the medical label bar code based on deep learning | |
| CN101033958A (en) | Mechanical vision locating method | |
| CN1777915A (en) | Face image candidate area search method, face image candidate area search system, and face image candidate area search program | |
| CN111179233A (en) | Self-adaptive deviation rectifying method based on laser cutting of two-dimensional parts | |
| CN113554757A (en) | Three-dimensional reconstruction method and system for workpiece track based on digital twinning | |
| CN1225502A (en) | Method of separating chips from semiconductor wafer | |
| CN114549599A (en) | Wafer rapid pre-alignment method and device, electronic equipment and storage medium | |
| CN114004814A (en) | Coal gangue identification method and system based on deep learning and gray scale third moment analysis | |
| CN1693037A (en) | Method and equipment for deep information extraction for micro-operation tool based-on microscopic image processing | |
| US11373366B2 (en) | Method for improving modeling speed of digital slide scanner | |
| CN110942063A (en) | Certificate text information acquisition method and device and electronic equipment | |
| CN107579028B (en) | Method and device for determining edge of incomplete wafer and scribing device | |
| CN113705487B (en) | Precision workpiece identification and technological parameter association system and identification method | |
| US20050105790A1 (en) | Wafer observation position designating apparatus and wafer display position designating method | |
| JP2000097677A (en) | Image recognizing method | |
| JPH08330390A (en) | Picking-up device and method | |
| CN112419349B (en) | Artificial intelligent object fragment image splicing method | |
| CN115272751A (en) | Fruit grading sorting method based on deep learning and disordered grabbing | |
| US20240062988A1 (en) | Machine vision-based automatic focusing and automatic centering method and system | |
| CN101037186A (en) | MEMS automatic leading interlinking equipment based on zoom microscope location | |
| Yap et al. | Implementation of vision-assisted path planning system for bulk die sorting in semiconductor industry |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Assignee: CETC Beijing Electronic Equipment Co., Ltd. Assignor: No.45 Inst., China Electronic science and Technology Group Corp. Contract fulfillment period: 2007.3.1 to 2014.12.29 contract change Contract record no.: 2009990000523 Denomination of invention: Apparatus for automatically distinguishing and aligning wafer of wafer cutter and method thereof Granted publication date: 20071107 License type: Exclusive license Record date: 2009.5.18 |
|
| LIC | Patent licence contract for exploitation submitted for record |
Free format text: EXCLUSIVE LICENSE; TIME LIMIT OF IMPLEMENTING CONTACT: 2007.3.1 TO 2014.12.29; CHANGE OF CONTRACT Name of requester: BEIJING ZHONGDIAN SCIENCE ELECTRONIC EQUIPMENT CO. Effective date: 20090518 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20071107 Termination date: 20141024 |
|
| EXPY | Termination of patent right or utility model |