CN110132982A - A kind of automatic optical detection method and equipment of high sensitivity - Google Patents
A kind of automatic optical detection method and equipment of high sensitivity Download PDFInfo
- Publication number
- CN110132982A CN110132982A CN201910448654.3A CN201910448654A CN110132982A CN 110132982 A CN110132982 A CN 110132982A CN 201910448654 A CN201910448654 A CN 201910448654A CN 110132982 A CN110132982 A CN 110132982A
- Authority
- CN
- China
- Prior art keywords
- imaging
- camera
- image
- variety
- flat samples
- 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
-
- 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
- G01N21/8806—Specially adapted optical and illumination features
Abstract
The invention discloses the automatic optical detection methods and equipment of a kind of high sensitivity, belong to technical field of optical detection.Highly sensitive automatic optical detection method of the invention, first, in accordance with imaging sequence continuous moving shooting in flat samples of a variety of imaging patterns of setting；Next, which steps be repeated alternatively until, has shot flat samples in the image of a variety of imaging patterns；Finally by algorithm by the image mosaic of identical imaging pattern at complete flat samples full figure, while testing result is obtained by defects detection and analysis tool.The method achieve the image for obtaining a variety of imaging patterns in single pass, camera can quickly shoot the image of a variety of imaging patterns under non-stop state, can get the sample image of different imaging patterns.In addition, realizing the automatic focusing of camera the present invention also provides highly sensitive automatic optical detection method, guarantee the image that high quality can be got in imaging process.
Description
Technical field
The present invention relates to flat samples defect detecting technique fields, are specifically related to a kind of automated optical inspection of high sensitivity
Survey method and apparatus.
Background technique
Automatic optics inspection (AOI) equipment is the key equipment for showing plate and semiconductor integrated circuit and chip manufacturing,
To diagnose and improve the product quality in process for making, manufacturing cost is reduced.It detection sensitivity, stability and can detect
To defect kind be automated optical detection equipment key technical index.With new technology application and technological level into
Step, requirement of the processing procedure to become increasingly complex to detection sensitivity and detection wide variety is also higher and higher, while to improve life
The detection time of yield, equipment is also important technical indicator.
Highly sensitive testing requirements detect imaging system image quality with higher, at this time in addition to requiring imaging system
Noise is lower outer, and inevitable requirement optical resolution is higher, the caused depth of field reduce then require system imaging process must it is steady,
Stablize.The diversification of detection type requires detection system that can obtain and analyzes various defect abnormal signals, such as basis: missing
Or the paramophias such as extra, dirty, scratch, material became uneven, irregular colour, the bright dark huge flaw indication of difference of background
Deng.The above several points all put forward new requirements whole equipment system: higher optical resolution, more stable imaging process,
Adapt to imaging pattern, more intelligent detection algorithm, the testing process of optimization and the control of number of drawbacks detection.
When carrying out defects detection to flat samples, the image information of the various modes obtained in flat samples can be answered
The defect type of high degree highlighted in flat samples is capable of in detection to various defect types.The prior art is to plate
The images of a variety of imaging patterns of sample obtains low efficiency, and camera needs to shoot different imaging moulds being fixed to setting shooting area
Formula can not achieve to flat samples in image in a variety of imaging pattern high-precisions, with the high speed detection for stablizing imaging effect, nothing
Method meets the detection demand of high-precision large scale flat samples.
Summary of the invention
The purpose of the invention is to overcome the prior art to complex process, film color difference is different, bright dark difference, polarization are poor
Deficiency in different equal complicated defect and highly sensitive defects detection, provide a kind of high sensitivity automatic optical detection method and
Equipment.Have the characteristics that high efficiency obtains the image of a variety of imaging patterns, high-precision, stabilization, high speed detection.
The present invention provides a kind of automatic optical detection method of high sensitivity, comprising the following steps:
According to imaging sequence continuous moving shooting in flat samples of a variety of imaging patterns of setting, a variety of imagings are acquired
The image of mode, and there is the region that partly overlaps between the image of two adjacent imaging patterns of the same race；
It steps be repeated alternatively until and all shot flat samples in the image of a variety of imaging patterns；
The image of a variety of imaging patterns of flat samples is obtained, and by the image mosaic of identical imaging pattern at complete flat
The sample full figure of plate sample.The detection of high efficient and flexible is made to defective data by defects detection and analysis tool simultaneously and is divided
Analysis.
Preferred embodiment: it according to imaging sequence continuous moving shooting in flat samples of a variety of imaging patterns of setting, adopts
Collect the image of a variety of imaging patterns, and there is the region that partly overlaps between the image of two adjacent imaging patterns of the same race, specifically
Are as follows:
Camera connects in flat samples according to preset motion profile according to the imaging sequence of a variety of imaging patterns of setting
Continuous follow shot, acquires the image of a variety of imaging patterns of flat samples, and the image of two adjacent imaging patterns of the same race it
Between have and partly overlap region, camera is driven by movement mechanism realize that camera is continuously moved relative to flat samples in shooting process
Dynamic shooting.
Preferred embodiment: it steps be repeated alternatively until and all shot flat samples at the image of a variety of imaging patterns, specifically
Are as follows:
Camera connects in flat samples according to preset motion profile according to the imaging sequence of a variety of imaging patterns of setting
Continuous follow shot, acquires the image of a variety of imaging patterns of flat samples, camera shot a variety of imaging patterns under it is a variety of
It repeats to recycle shooting according to the imaging sequence of a variety of imaging patterns after the image of imaging pattern, be put down until taking all to have shot
All images of the plate sample in a variety of imaging patterns.
Preferred embodiment: obtaining the image of a variety of imaging patterns of flat samples, and by the image mosaic of identical imaging pattern
At complete flat samples sample full figure, specifically:
Computer obtains flat samples in all images of a variety of imaging patterns, adjacent two figure of identical imaging pattern
Overlapping region is all had as between, flat samples are superimposed with each other in the overlapping region of the image of identical imaging pattern and have been spliced into
The sample full figure of whole flat samples.
Preferred embodiment: camera is driven by movement mechanism in shooting process realizes camera relative to flat samples continuous moving
Shooting, specifically:
Movement mechanism camera in shooting process, movement mechanism drives relative motion between camera or flat samples, phase
The every opposing plates sample of machine walks a length and just sends primary shooting signal to camera, and continuous moving is clapped under the conditions of non-stop
It takes the photograph.
Preferred embodiment: according to imaging sequence continuous moving shooting in flat samples of a variety of imaging patterns of setting, tool
Body are as follows:
According to a variety of imaging patterns and setting imaging sequence of setting, successively automatic cycle switches, and a variety of imaging patterns cut
It is consistent with the frequency that camera is shot to change frequency, shoots in flat samples according to preset motion profile continuous moving.
Preferred embodiment: where a variety of imaging patterns are as follows: bright-field imagery, dark-field imaging, grey field imaging, backlight imaging, polarization
One of imaging or a variety of combinations.
Preferred embodiment: where a variety of imaging patterns are as follows: red light source, green light source, yellow light sources, in white light source
One or more combinations.
Preferred embodiment: where a variety of imaging patterns are as follows: the light source of the same color of various levels of brightness illuminates.
Preferred embodiment: camera is driven by movement mechanism in shooting process realizes camera relative to flat samples continuous moving
Shooting, specifically:
The X-axis movement mechanism of camera movement mechanism in shooting process drives camera or the every walking certain length of flat samples
Camera will be triggered to be shot, camera or the flat samples continuous moving in the case where the X-axis movement mechanism of movement mechanism drives are shot.
Preferred embodiment: the invention also includes following steps based on the above technical solution:
When shooting to first piece of flat samples according to the imaging sequence continuous moving of a variety of imaging patterns of setting, camera is recorded
The most clear camera heights when being imaged every time in flat samples；
The camera is carried out to most clear camera heights guarantor when planar obit simulation and each imaging when continuous moving shooting
Save as optimum trajectory information；
When continuing flat samples after sensing, optimum trajectory information is called, controls camera relative to flat samples in horizontal plane
On along the optimum trajectory information movement when so that camera is shot with optimum height.
Preferred embodiment: described that first piece of flat samples are clapped according to the imaging sequence continuous moving of a variety of imaging patterns of setting
When taking the photograph, camera most clear camera heights when being imaged every time in flat samples are recorded, specifically:
Controller control Three-axis drive mechanism according to a variety of imaging patterns of setting imaging sequence driving camera relative to
First piece of flat samples carry out continuous moving shooting in each detection zone, and it is each in first piece of flat samples that computer records camera
Plane running track when imaging and the most clear camera axial height when being imaged in each detection zone.
Preferred embodiment: the controller control Three-axis drive mechanism drives according to the imaging sequence of a variety of imaging patterns of setting
Dynamic camera carries out continuous moving shooting in each detection zone relative to first piece of flat samples, specifically:
The X-axis driving unit of controller control Three-axis drive mechanism drives flat samples to be moved to camera along the x axis
The movement forward or backwards of total travel length is done under imaging line, while controller control camera takes pictures interval to plate according to setting
Each detection zone of sample carries out Image Acquisition；
The Y-axis driving unit of controller control Three-axis drive mechanism drives camera to translate setting width along the y axis, then
The detection content to repeat the above steps；
The Z axis driving unit that controller controls Three-axis drive mechanism drives camera to adjust each detection zone along Z-direction
Camera heights when upper imaging, the detection content to repeat the above steps are finished until all detection zone Image Acquisition.
Preferred embodiment: when the continuous flat samples after sensing, calling optimum trajectory information, controls camera relative to plate
Sample in the horizontal plane along the optimum trajectory information movement when so that camera is shot with optimum height, specifically:
When continuing flat samples after sensing, controller call camera when each detection zone is imaged clearest camera it is flat
It face running track and most clear camera heights is imaged in each detection zone forms cam curve；
Camera, when moving according to plane running track, while being moved relative to flat samples according to cam curve, is made
Stable focusing is remained between camera and flat samples, realizes that focusing progress image automatically according to the face type of flat samples adopts
Collection.
Preferred embodiment: the invention also includes following steps based on the above technical solution:
The initial imperfection obtained by the image detection of each imaging pattern is divided into the initial of corresponding each imaging pattern to lack
It falls into sublist and stores；
Spacing distance in the initial imperfection sublist of the image of each imaging pattern is met preset first judgement respectively to mark
Quasi- defect point merges, and obtains the final defect point sublist of the image of each imaging pattern and stores；
Spacing distance between the final defect point sublist of the image of each imaging pattern is met into preset second judgment criteria
Defect point merge, obtain final defect point table and store, each defect point in final defect point table include merge formed should
The image information of the defect point of the final defect point sublist of defect point；
It, will be in final defect point table according to the combination of the image of single imaging pattern or the image of multiple imaging patterns
Defect point is shown.
Another aspect of the present invention additionally provides a kind of automated optical detection equipment of high sensitivity, comprising:
Imaging mechanism comprising for obtaining the phase of a variety of imaging modality images of each detection zone in flat samples
Machine, camera lens and light source, and the light source is configured as that the light of different imaging patterns can be issued, and has the imaging mechanism more
Kind imaging pattern；
Movement mechanism is used to drive the three-axis moving mechanism of the relatively described flat samples movement of the imaging mechanism and holds
Hold in the palm the air bearing support platform of flat samples；
Controller is used to control the movement mechanism and imaging mechanism, makes the imaging mechanism according to a variety of of setting
The imaging sequence of imaging pattern continuous moving in flat samples is shot, and between the image of two adjacent imaging patterns of the same race
With the region that partly overlaps；
Computer, is used to obtain the image of a variety of imaging patterns of flat samples, and by the image of identical imaging pattern
It is spliced into the sample full figure of complete flat samples.Efficient spirit is made to defective data by defects detection and analysis tool simultaneously
Detection and analysis living.
Preferred embodiment: the computer prestores optimum trajectory information, the optimum trajectory information include the camera into
Planar obit simulation and camera heights most clear when being imaged every time when row continuous moving is shot；
The controller is also used to call the optimum trajectory information, controls the camera relative to flat samples in level
When on face along the optimum trajectory information movement, so that camera is shot with optimum height.
Preferred embodiment: the Three-axis drive mechanism includes Z axis driving unit, Y-axis driving unit and X-axis driving unit, institute
It states Z axis driving unit to be arranged on Y-axis driving unit, the Y-axis driving unit is arranged on the rack, and the bottom of rack is equipped with institute
Air bearing support platform is stated, the X-axis driving unit is arranged in air bearing support platform, and the camera and light source setting are driven in Z axis
On moving cell.
Preferred embodiment: the Z axis driving unit is Z axis straight line mould group, and the camera and light source are fixed at Z axis straight line
On the sliding block of mould group, the camera and light source follow the sliding block direction vertical up-or-down movement of Z axis straight line mould group, the Z axis straight line
Mould group is used to adjust the best focusing height of camera.
Preferred embodiment: the Y-axis driving unit is Y-axis straight line mould group, and the Z axis driving unit is fixed on Y-axis straight line mould
On the sliding block of group, the Y-axis straight line mould group is for driving camera and light source to move along a straight line along the y axis, the camera and light source
Image Acquisition is carried out to detection zone each in flat samples along the y axis.
Preferred embodiment: the X-axis driving unit is X-axis straight line mould group, and the flat samples are connected to X by vacuum chuck
In axis straight line mould group, the X-axis straight line mould group moves along a straight line back and forth along the x axis for drawing flat samples, the camera and
Light source carries out Image Acquisition to detection zone each in flat samples along the x axis relative to flat samples.
Preferred embodiment: the controller and imaging mechanism, movement mechanism and calculating mechatronics, the controller pass through control
Timing processed issues trigger signal to imaging mechanism, and control imaging mechanism obtains the plate of a variety of imaging patterns in a variety of imaging patterns
Sample image, the controller call the optimum trajectory information control movement mechanism of computer storage to transport by optimum trajectory information
It is dynamic.
Based on the above technical solution, compared with prior art, advantages of the present invention is as follows:
On the one hand a kind of automatic optical detection method of high sensitivity of the invention, this method are realized in single pass and are obtained
Take the image of a variety of imaging patterns, according to the imaging sequence of a variety of imaging patterns of setting when use, camera connects flat samples
Continuous follow shot；And there is the region that partly overlaps between the image of two adjacent imaging patterns of the same race, camera is in shooting process
Middle movement mechanism driving camera is shot relative to flat samples continuous moving, and camera can quickly be shot under non-stop state
The image of a variety of imaging patterns greatly improves the information obtaining ability of detection device and the image taking efficiency of camera.
A kind of automatic optical detection method of high sensitivity of the invention, this method is in the image for obtaining a variety of imaging patterns
Shi Shixian is focused automatically, and this method is shot each detection zone of first piece of flat samples by preset order, records camera
Plane running track and camera heights when most understanding are imaged in each detection zone；Then by the plane of the camera of record
It running track and most clear camera heights is imaged in each detection zone saves as optimum trajectory information；Finally after sensing
When continuous flat samples, controller calls optimum trajectory information, and control camera is relative to flat samples in the horizontal plane along described flat
The movement of face running track, and shot in each detection zone with optimal focal height.Acquire out the best clear of flat samples
Image, improve the stability to flat samples Surface testing, ensure that detection sensitivity.
A kind of automatic optical detection method of high sensitivity of the invention used a set of high efficiency smart detection instrument and
Method can analyze and show optimal imaging pattern assembled scheme and scheme is directly applied to defect inspection method, so that should
The use of method and apparatus is easier, efficient, and detection effect is more outstanding.
A kind of automated optical detection equipment of high sensitivity of the invention, the detection device pass through imaging mechanism, fitness machine
Structure, computer and the work of controller coordinated, realize the quick obtaining and automatic focusing function of various modes image.Realization pair
Flat samples meet the large scale plate of high-precision, complex process in a variety of imaging pattern high-precisions, stable and high speed detection
The highly sensitive detection demand of sample.
Detailed description of the invention
Fig. 1 is the overall structure diagram of the embodiment of the present invention；
Fig. 2 is the movement mechanism structural schematic diagram of the embodiment of the present invention；
Fig. 3 is the imaging mechanism structural schematic diagram of the embodiment of the present invention；
Fig. 4 is the triggering timing diagram of the embodiment of the present invention；
Fig. 5 is the shooting schematic diagram of the embodiment of the present invention；
Fig. 6 is the focusing structure figure of the embodiment of the present invention；
Fig. 7 is the focusing principle figure of the embodiment of the present invention.
Appended drawing reference: 101- rack, 102- movement mechanism, 103- imaging mechanism, 104- computer, 105- controller,
201- flat samples, 202- pedestal, 203- board, 204- circuit box, 205-Z axis driving unit, 206-X axis driving unit,
207-Y axis driving unit, 208- air bearing support platform, 209- gas circuit case, 301- camera, 302- camera lens, 303- dark-field imaging,
304- bright-field imagery, the imaging of 305- backlight, the imaging of 306- ash field, 307- red light source, 308- green light source, 309- blue light
Source, 310- white light source, the first signal of 401-, 402- second signal, 403- third signal, 404- fourth signal, 411- the 5th
Signal, the 6th signal of 412-, the 7th signal of 413-, 500- horizontal direction, the first image-region of 501-, the second image district of 502-
Domain, 503- third image-region, the 4th image-region of 504-, the 5th image-region of 505-, the 6th image-region of 506-, 511-
One moving distance, the second moving distance of 512-, 513- third moving distance, the 4th moving distance of 514-, 515- the 5th it is mobile away from
From, 602- vertical direction, the first position 604-, the second position 605-, 606- the third place, the 4th position 607-, 608- the 5th
Position.
Specific embodiment
With reference to the accompanying drawing and specific embodiment the present invention is described in further detail.
Referring to shown in Fig. 1, Fig. 2, Fig. 4 and Fig. 5, first aspect of the embodiment of the present invention provides a kind of the automatic of high sensitivity
Optical detecting method comprising following steps:
Step 1: according to the imaging sequence of a variety of imaging patterns of setting, camera 301 is in flat samples 201 according to default
Motion profile continuous moving shooting, acquire flat samples 201 a variety of imaging patterns image, the motion profile is according to reality
It needs to set with the concrete shape of flat samples 201；And there is part between the image of two adjacent imaging patterns of the same race
Overlapping region also has overlapping region in certain single pass, does not need opposite direction again between the image of a variety of imaging patterns
Flat samples 201 are shot.
Camera 301 in shooting process movement mechanism 102 X-axis movement mechanism 206 drive flat samples 201 it is every walking one
Measured length will trigger camera 301 and be shot, naturally it is also possible to drive phase using the X-axis movement mechanism 206 of movement mechanism 102
The every walking certain length of machine 301 will trigger camera 301 and be shot.Flat samples 201 are moved in the X-axis of movement mechanism 102
Under mechanism 206 drives, camera 301 carries out continuous moving shooting to it.Movement mechanism 201 camera 301 in shooting process, fortune
Motivation structure 201 drives flat samples 201 often to walk a length and sends primary shooting signal to camera 301, and camera 301 is not stopping
Under conditions of continuous moving shoot.
Step 2: according to the imaging sequence of a variety of imaging patterns set in step 1, camera 301 presses flat samples 201
According to the running track continuous moving shooting of setting, the image of a variety of imaging patterns of flat samples 201 is acquired, camera 301 is shot
It repeats to recycle according to the imaging sequence of a variety of imaging patterns after the image of a variety of imaging patterns under complete a variety of imaging patterns and clap
It takes the photograph, until having shot flat samples 201 in all images of a variety of imaging patterns.
Step 3: computer 104 obtain flat samples 201 a variety of imaging patterns all images, identical imaging pattern
Overlapping region is all had between two adjacent images, by flat samples 201 in the overlapping region of the image of identical imaging pattern
It is superimposed with each other and is spliced into the sample full figure of complete flat samples, the quantity of the product full figure of flat samples sample is a variety of imaging patterns
Particular number.It is to obtain completely using overlapping region is all had between adjacent two image of identical imaging pattern
To the image of flat samples 201, leakage bat and missing inspection are prevented.
Specifically, in the present embodiment, the present invention is respectively with three kinds of imaging modality images combinations and two kinds of imaging pattern figures
As combination is illustrated respectively.It should be known that imaging modality images integrated mode of the invention is far above the number of example.
When being imaged using three kinds of imaging patterns, as shown in a in Fig. 4.The sequence controller of controller 105 receives outer
The first signal of portion 401, meeting is according to Sequential logic output second signal 402,403 and of third signal after receiving the first signal 401
Fourth signal 404, that is, output second signal 402 is used to trigger the first after receiving first rising edge of the first signal 401
Imaging pattern adopts figure, exports third signal 403 after receiving second rising edge of the first signal 401 to trigger second of imaging
Mode adopts figure, exports fourth signal 404 after receiving the third rising edge of the first signal 401 to trigger the third imaging pattern
Adopt figure.It circuits sequentially, until completing all Image Acquisition tasks.It is achievable such as using the combined imaging of three kinds of imaging patterns: to make
The bright field combination lighting carried out with red light source 307, green light source 308, blue-light source 309 is imaged to realize to color sensitivity
The detection of different defects, detection of defect different to film thickness etc..Use the bright field of green light source 308, white light source 310
The detection etc. that the combination lighting that backlight, the dark field of white light source 310 carry out is imaged to realize the defect to different shape type.
When being imaged using two kinds of imaging patterns, as shown in b in Fig. 4.The sequence controller of controller 105 receives outer
The 5th signal 411 of portion, meeting is according to the 6th signal 412 of Sequential logic output, the 7th signal 413 after receiving the 5th signal 411, i.e.,
It receives the 6th signal 412 of output after first rising edge of the 5th signal 411 and is used to trigger the first imaging pattern and adopt figure, connect
It receives and exports the 7th signal 413 after second rising edge of the 5th signal 411 and adopt figure to trigger second of imaging pattern.Successively follow
Ring, until completing all Image Acquisition tasks.It is achievable such as using the combined imaging of two kinds of imaging patterns: to use green light source
The bright field combination lighting that the last 308 light, 308 dim light of green light source carry out is imaged to realize the detection to the bright dark huge defect of difference
Deng.
The present embodiment illustrates the actual imaging process of flat samples 201 with the scheme of a in Fig. 4, as shown in Figure 5.With t0
Moment, at the t0 moment, movement mechanism 102 issued the first signal 401, and the processing of via controller 105 issues second as starting explanation
Signal 402, imaging mechanism 103 get be immediately switched to after second signal 402 the first imaging pattern and using the first at
As pattern acquiring to the first image-region 501.Hereafter imaging mechanism 103 500 moves first movement distance 511 in the horizontal direction,
Arrival time t1 at this time, movement mechanism 102 issue the first signal 401, and the processing of via controller 105 issues third signal 403, at
Camera structure 103 is immediately switched to second of imaging pattern after getting third signal 403 and gets the second image-region 502.
Hereafter imaging mechanism 103 500 moves the second moving distance 512 in the horizontal direction, at this time arrival time t2, and movement mechanism 102 is sent out
First signal 401 out, the processing of via controller 105, issues fourth signal 404, after imaging mechanism 103 gets fourth signal 404
It is immediately switched to the third imaging pattern and gets third image-region 503.
Hereafter imaging mechanism 103 500 moves third moving distance 513 in the horizontal direction, at this time arrival time t3, fitness machine
Structure 102 issues the first signal 401, and the processing of via controller 105 issues second signal 402, and imaging mechanism 103 gets the second letter
It is immediately switched to the first imaging pattern again after numbers 402 and gets the 4th image-region 504.Hereafter imaging mechanism 103 is along water
Square to 500 mobile 4th moving distances 514, arrival time t4 at this time, movement mechanism 102 issues the first signal 401, through controlling
The processing of device 105, issues third signal 403, and imaging mechanism 103 is immediately switched to second one-tenth after getting third signal 403 again
As mode and get the 5th image-region 505.Hereafter imaging mechanism 103 500 moves the 5th moving distance in the horizontal direction
515, arrival time t5,102 part of movement mechanism issue the first signal 401 at this time, and the processing of via controller 105 issues the 4th letter
Numbers 404, imaging mechanism 103 gets after fourth signal 404 and is immediately switched to the third imaging pattern and gets the 6th figure
As region 506.So on up to all Image Acquisition of completion.
Wherein the 4th acquired in the first image-region 501 and the first imaging pattern acquired in the first imaging pattern
Image-region 504 connects just, the second image-region 502 and second of imaging pattern acquired in second of imaging pattern
The 5th acquired image-region 505 connects just, third image-region 503 acquired in the third imaging pattern and
6th image-region 506 acquired in three kinds of imaging patterns connects just.First image-region 501 and the 4th image-region
504 overlap, and the second image-region 502 overlaps with the 5th image-region 505, third image-region 503 and
Six image-regions 506 overlap, and overlapping region accounts for the 2% left of entire camera 301 pixel number in 500 directions in the horizontal direction
It is right.
In this way, each imaging pattern can get the figure of complete flat samples 201 during single pass
Picture.First movement distance 511, the second moving distance 512, third moving distance 513, the 4th moving distance the 514, the 5th it is mobile away from
The one third ratio of 301 pixel number of camera from 515 substantially in the horizontal direction 500, even there is 3000 pictures in this direction
Element, the distance are then about the length in 1000 pixels corresponding 201.Specific length needs to skim the number of overlaid pixel,
First movement distance 511 (the 4th moving distance 514), the second moving distance 512 (the 5th moving distance 515), third are moved simultaneously
The distance of dynamic distance 513 is because the difference of imaging pattern can be not completely equivalent.During actual scanning, the present embodiment can be used
Moving movement mechanism 201 can also be used to complete Image Acquisition in mobile imaging mechanism 103 in description.
Preferred embodiment scheme: according to the imaging sequence of a variety of imaging patterns of setting in the step 1, camera 301 exists
It is shot in flat samples 201 according to preset motion profile continuous moving, specifically:
According to a variety of imaging patterns and setting imaging sequence of setting, successively automatic cycle switches, and a variety of imaging patterns cut
It is consistent with the frequency that camera 301 is shot to change frequency, shoots in flat samples 201 according to preset motion profile continuous moving.
Preferred embodiment scheme: a variety of imaging patterns in the step 1 are as follows: bright-field imagery 304, dark-field imaging 303, Hui Chang
One of imaging 306, backlight imaging 305 or a variety of combinations.Or red light source 307, green light source 308, yellow light sources
309, one of white light source 310 or a variety of combinations.Or the light source illumination of the same color of various levels of brightness.
Referring to shown in Fig. 1, Fig. 2, Fig. 6 and Fig. 7, the second aspect of the present invention, on the basis of first aspect embodiment this
Invention additionally provides a kind of automatic optical detection method of high sensitivity, method includes the following steps:
Step 1: controller 105 controls Three-axis drive mechanism and drives phase according to the imaging sequence of a variety of imaging patterns of setting
Machine 301 carries out continuous moving shooting in each detection zone relative to first piece of flat samples 201, and computer 104 records camera 301
Plane running track when being imaged every time in first piece of flat samples 201 and the most clear phase when being imaged in each detection zone
Machine axial height；
Step 2: camera 301 is carried out to most clear camera axis when planar obit simulation and each imaging when continuous moving shooting
To highly saving as optimum trajectory information；
Step 3: when continuing flat samples 1 after sensing, calling optimum trajectory information, control camera 301 relative to plate sample
(i.e. camera 301 drives along 207 traffic direction of Y-axis driving unit and X-axis when product 201 are moved along plane running track in the horizontal plane
The movement of 206 traffic direction of moving cell) so that camera 301 is shot in each detection zone with optimum height.
Preferred embodiment scheme: controller 105 in above-mentioned steps 1 control Three-axis drive mechanism according to setting it is a variety of at
As the imaging sequence driving camera 301 of mode carries out continuous moving bat in each detection zone relative to first piece of flat samples 201
It takes the photograph, specifically:
Step 11: the X-axis driving unit 206 that controller 105 controls Three-axis drive mechanism drives flat samples 201 along X-axis
Direction is moved under the imaging line of camera 301 movement forward or backwards for doing total travel length, while controller 105 controls camera
301 carry out Image Acquisition to each detection zone of flat samples 201 by interval of taking pictures is set；
Step 12: the Y-axis driving unit 207 that controller 105 controls Three-axis drive mechanism drives camera 301 along the y axis
Translation setting width, then repeat the above steps 11 detection content；
Step 13: the Z axis driving unit 205 that controller 105 controls Three-axis drive mechanism drives camera 301 along Z-direction
Adjust the camera heights being imaged when most understanding in each detection zone, repeat the above steps 12 detection content until all detections
Area image acquisition finishes.
Preferred embodiment scheme: in above-mentioned steps 3 after sensing continue flat samples 1 when, call optimum trajectory information, control
When camera 301 processed is moved along plane running track in the horizontal plane relative to flat samples 201, so that camera 301 is in each inspection
It surveys on region and is shot with optimum height, specifically:
When continuing flat samples 201 after sensing, the optimum trajectory information that controller 105 calls computer 104 to store, control
Device 105 controls the plane running track of camera 301 and in each detection zone when each detection zone is imaged clearest of camera 301
Most clear camera heights are imaged on domain and form cam curve；
Relative to flat samples 201, when moving according to plane running track, (camera 301 drives list along Y-axis to camera 301
First 207 traffic directions and 206 traffic direction of X-axis driving unit), while being moved according to cam curve, make camera 301 and plate
Stable focusing is remained between sample 201, realizes and Image Acquisition is carried out according to the face type auto-focusing of flat samples 201.
Specifically, automatic focusing principle is as shown in Figure 6.Camera 301 carries camera lens 302, and 602 movements can along the vertical direction
Make flat samples 201 realized on imaging mechanism 103 be imaged on it is clear with it is fuzzy between change.Camera 301 and camera lens 302 are
When three positions 606, it is ensured that the detection zone of flat samples 201 is in coke-like state.Obtain about 606 the third place totally 5
The image of the flat samples 201 of position, i.e. first position 604 on 602, the second position 605, the third place along the vertical direction
606, the image of the 4th position 607, the 5th position 608.By to the first position 604, the second position 605, the third place
606, the clarity result of the image of the 4th position 607, the 5th position 608 is fitted, and the optimal image of clarity can be obtained will
Location, the as best focus position of the current point of flat samples 201 on vertical direction 602, approximating method and knot
Fruit is as shown in Figure 7.
According to this method, all images in entire flat samples 201 are got with image acquiring method shown in Fig. 5, are obtained
First position 604, the second position 605, the third place 606, fourth position 607, fiveth position of the position on vertical direction 602
608 clarity data in totally 5 position heights.It can be fitted with these data and inform against block flat samples 201 during the scanning process
Each imaging position optimum focusing Z axis height and position.Guaranteed by imaging mechanism 103 and movement mechanism 102, is replaced other
When flat samples 201, the entire surface type of other flat samples 201 can be remained unchanged, therefore above step only needs to complete once i.e.
It can.In detection process hereafter, by controller 105 control Z axis driving unit 205 come so that camera 301 in Z-direction
Near the upper total best focus position in flat samples 201, to guarantee to be imaged in the entire scanning processes of flat samples 201
Mechanism 103 can get the extraordinary image data of clarity.
It should be noted that using 5 location points as calculating in the present embodiment, but 5 positions not can only be selected, with energy
Fit best focus position be can, in the present embodiment recommend at least five location point, and between location point distance needs be more than
Field depth.It should be noted that the approximating method in the present embodiment uses Gauss curve fitting, but it is not limited to Gauss curve fitting one
Kind of mode, with can fit best focus position be can, recommend Gauss curve fitting method in the present embodiment.
The third aspect of the present invention, the present invention also provides a kind of high sensitivity on the basis of first aspect embodiment
Automatic optical detection method, method includes the following steps:
It is each a variety of to be divided into correspondence by step S101 for the initial imperfection point list that the image detection of a variety of imaging patterns obtains
The initial imperfection point sublist of the image of imaging pattern simultaneously stores.Specifically, the image of the imaging pattern of the embodiment of the present invention is three
Kind, the image of image, the second imaging pattern including the first imaging pattern and the image of third imaging pattern, initial imperfection point
It records in list there are three all defect point information that the image detection of imaging pattern arrives, defect point information includes the seat of defect point
It is marked with and the channel information of defect point, the channel information of defect point refers to the defect point by the image detection of which imaging pattern
It obtains.The initial imperfection point list that the image detection of three imaging patterns obtains is divided into the image of corresponding three imaging patterns
Initial imperfection point sublist, the initial imperfection point sublist of the image of three imaging patterns is stored in database.
Step S102, respectively by the initial imperfection point sublist of the image of each imaging pattern and spacing distance meet it is default
The first judgment criteria defect point merge, obtain the final defect point sublist of the image of each imaging pattern and store.Specifically
Ground calculates the distance between defect point in the initial imperfection point sublist of the image of each imaging pattern using coordinate, by three at
As the image of mode initial imperfection point sublist in and spacing distance meet preset first judgment criteria defect point merge, obtain
To the final defect point sublist of the image of three imaging patterns, the final defect point sublist of the image of three imaging patterns is stored
In the database, the coordinate of each defect point and the image information of defect point are recorded in each final defect point sublist.
Step S103, will be between the final defect point sublist of the image of each imaging pattern and spacing distance meets preset
The defect point of two judgment criterias merges, and obtains final defect point table and stores, each defect point in final defect point table includes
Merge the image information for forming the defect point of final defect point sublist of the defect point.Specifically, the image of three imaging patterns
Final defect point sublist in the coordinate of defect point can be understood as the same coordinate system calibration, the coordinate that uses calculates three
The distance between all defect point in the final defect point sublist of the image of imaging pattern, most by the image of three imaging patterns
Between whole defect point sublist and spacing distance meets the defect point merging of preset second judgment criteria, obtains a final defect point
Table, in the database by the storage of final defect point table, each defect point in final defect point table include to merge to form the defect
The image information of the defect point of the final defect point sublist of point, i.e., each defect point record in final defect point table have merging shape
At the defect point final defect point sublist defect point from the image of which or which imaging pattern.The present invention is real
It is identical with preset first judgment criteria to apply preset second judgment criteria in example.The embodiment of the present invention is remembered using a character string
Record identifies each defect point in final defect point table, and imaging pattern proposes defect point information in such a way that character string records
It takes and retrieves and is convenient, efficient.
Step S104 finally will most according to the combination of the image of single imaging pattern or the image of multiple imaging patterns
Excellent most stable of defect point testing result is shown.
Shown in referring to figure 1, figure 2 and figure 3, fourth aspect present invention additionally provides a kind of automatic optics inspection of high sensitivity
Equipment, the detection device include:
The bottom of rack 101, the rack 101 is equipped with pedestal 202, and pedestal 202 has been used to support the object of entire detection device
Structure is managed, plays carrying in the bottom of the detection device, which is fashioned iron steel using material.The bottom of the rack 101
Seat 202 is equipped with board 203, and board 203 is used to do benchmark for movement mechanism 102 and imaging mechanism 103, in the detection device
Benchmark is made reference, while playing carrying.Movement mechanism 102 and imaging mechanism 103 carry out adjustment on the basis of board 203
And operation.Marble or steelframe material can be used in the board 203, and for the precision for guaranteeing detection, which wants in index
Ask flatness within several silks.The rack 101 is equipped with outline border, and outline border is used to support supplementary structure and plays the isolation of board 203
Effect.Isolating door, sensor and other mechanical parts are housed on the outline border of the rack 101.The rack 101 is equipped with circuit box 204, electricity
Road case 204 for installing circuit devcie, circuit control device and circuit switching, circuit box 204 be used for for movement mechanism 102, at
Camera structure 102, computer 104 and controller 105 establish connection.The rack 101 is additionally provided with gas circuit case 209, and gas circuit case 209 is used for
Gas circuit device, gas circuit controller part and gas circuit switching are installed.
Imaging mechanism 103, the imaging mechanism 103 include for acquiring a variety of of each detection zone in flat samples 201
The camera 301 of mode image and the light source that a variety of imaging patterns are presented are equipped with camera lens 302 in the front end of camera 301.The imager
The camera 301 of structure 103 is mainly responsible for image acquisition, which is preferably face formation camera, and face formation camera is with higher
Pixel number and very high data transmission rate.In the present embodiment, which has up to even tens of million pixel numbers of number million,
The specific data transmission rate of camera 301 is up to 2GB/s.The camera lens 302 of the camera 103 belongs to high resolution imaging lens, at this
There is micrometer resolution in embodiment.The camera lens 302 has long reach, and color difference and distortion with very little.
Lighting method workable for the light source of the imaging mechanism 103 includes but are not limited to bright-field imagery 304, backlight imaging
305, dark-field imaging 303 or grey field imaging 306.Different light sources can be used in each lighting method of the imaging mechanism 103
Color includes but are not limited to red light source 307, green light source 308, blue-light source 309, white light source 310.The imaging mechanism
The different intensities of light source can be used in 103 each lighting method.The imaging pattern of the imaging mechanism 103 can be in lighting method
It is freely combined between light source colour, intensity, illustrates but is not limited only to following example: using red light source 307, green light source
308, the bright field combination lighting imaging that blue-light source 309 carries out；Use the bright-field imagery 304 of green light source 308, white light source
The combination lighting imaging that 310 backlight imaging 305, the dark-field imaging 303 of white light source 310 carry out；Use green light source 308
Qiang Guang, 304 combination lighting of bright-field imagery that the dim light of green light source 308 carries out are imaged.
Movement mechanism 102, the movement mechanism 102 include for driving imaging mechanism 103 in 201 upper direction of flat samples
Three-axis moving mechanism and support flat samples 201 air bearing support platform 208.The three-axis moving mechanism includes that Z axis driving is single
Member 205, Y-axis driving unit 207 and X-axis driving unit 206, Z axis driving unit 205 are arranged on Y-axis driving unit 207, Y-axis
Driving unit 207 is arranged in rack 101, and the bottom of rack 101 is equipped with air bearing support platform 208, and X-axis driving unit 206 is set
It sets in air bearing support platform 208, camera 301, camera lens 302 and light source are arranged on Z axis driving unit 205.
Z axis driving unit 205 is Z axis straight line mould group, and camera 301 and light source are fixed at the sliding block of Z axis straight line mould group
On, camera 301 and light source follow the sliding block direction vertical up-or-down movement of Z axis straight line mould group to each detection in flat samples 201
Region carries out Image Acquisition, and Z axis straight line mould group is used to adjust the best focusing height of camera 301.Y-axis driving unit 207 is Y-axis
Straight line mould group, Z axis driving unit 205 are fixed on the sliding block of Y-axis straight line mould group, and Y-axis straight line mould group is for driving 301 He of camera
Light source moves along a straight line along the y axis, and camera 301 and light source carry out figure to each detection zone in flat samples 201 along the y axis
As acquisition.X-axis driving unit 206 is X-axis straight line mould group, and flat samples 201 are connected to X-axis straight line mould group by vacuum chuck
On, X-axis straight line mould group is for drawing the round-trip linear motion along the x axis of flat samples 201, and camera 301 and light source are relative to flat
Plate sample 201 carries out Image Acquisition to detection zone each in flat samples 201 along the x axis.
Computer 104, computer 104 are used to store preset optimum trajectory information and obtain the various modes of camera shooting
Image information, optimum trajectory information include planar obit simulation when the camera carries out continuous moving shooting and every time imaging when it is most clear
The camera heights of Chu.And and analysis is done to various modes image information and handles to obtain defect information, detection various modes image
The defect of information simultaneously generates report；Wherein, preset optimum trajectory information is configured as: clapping camera 301 along the sequence of setting
Each detection zone is taken the photograph, and the detection zone is shot with corresponding Z axis height in each detection zone.
Controller 105, the controller 105 are electrically connected with imaging mechanism 103, movement mechanism 102 and computer 104, control
Device 105 issues trigger signal timing to imaging mechanism 103 by sequence controller, controls imaging mechanism 103 in a variety of imaging moulds
Formula shoots the image of a variety of imaging patterns, and the optimum trajectory information that controller 105 calls computer 104 to store controls movement mechanism
102 press optimum trajectory information movement, so that imaging mechanism 103 is shot with optimum height.
It is used to control movement mechanism 102 and imaging mechanism 103 with the controller 105, makes the imaging mechanism 103 according to setting
The imaging sequence of fixed a variety of imaging patterns continuous moving in flat samples 201 is shot, and two adjacent imaging moulds of the same race
There is the region that partly overlaps between the image of formula.The controller 105 is responsible for the timing control of signal triggering in this detection device,
According to the difference of imaging pattern, the electrical control division branch of the controller 105 issues different trigger signals, to guarantee camera
301 can shoot image in correct imaging pattern.The controller 105 is completed at the same time including by state control in this detection device
The other function that system, the movement of movement mechanism 102, the triggering of imaging mechanism 103 control, each function combination of message transmission are completed.The control
The process control section of device 105 processed is completed jointly by PC and electric control system, the main configuration including detection method and parameter,
Condition monitoring, performance management and test, interaction, abnormality processing, as the result is shown etc..
Those skilled in the art can carry out various modifications to the embodiment of the present invention and modification, if these modifications and change
For type within the scope of the claims in the present invention and its equivalent technologies, then these modifications and variations are also in protection scope of the present invention
Within.
The prior art that the content being not described in detail in specification is known to the skilled person.
Claims (22)
1. a kind of automatic optical detection method of high sensitivity, which comprises the following steps:
According to imaging sequence continuous moving shooting in flat samples of a variety of imaging patterns of setting, a variety of imaging patterns are acquired
Image, and have between the image of two adjacent imaging patterns of the same race and partly overlap region；
It steps be repeated alternatively until and all shot flat samples in the image of a variety of imaging patterns；
The image of a variety of imaging patterns of flat samples is obtained, and by the image mosaic of identical imaging pattern at complete plate sample
The sample full figure of product；Make the detection and analysis of high efficient and flexible to defective data by defects detection and analysis tool simultaneously.
2. according to the method described in claim 1, it is characterized by:
According to imaging sequence continuous moving shooting in flat samples of a variety of imaging patterns of setting, a variety of imaging patterns are acquired
Image, and have between the image of two adjacent imaging patterns of the same race and partly overlap region, specifically:
Camera is continuously moved in flat samples according to preset motion profile according to the imaging sequence of a variety of imaging patterns of setting
Dynamic shooting, acquires the image of a variety of imaging patterns of flat samples, and have between the image of two adjacent imaging patterns of the same race
Overlap region, and camera is driven by movement mechanism realize that camera is clapped relative to flat samples continuous moving in shooting process
It takes the photograph.
3. according to the method described in claim 1, it is characterized by:
It steps be repeated alternatively until and all shot flat samples in the image of a variety of imaging patterns, specifically:
Camera is continuously moved in flat samples according to preset motion profile according to the imaging sequence of a variety of imaging patterns of setting
Dynamic shooting, acquires the image of a variety of imaging patterns of flat samples, camera shot a variety of imaging patterns under a variety of imagings
It repeats to recycle shooting according to the imaging sequence of a variety of imaging patterns after the image of mode, has all shot plate sample until taking
All images of the product in a variety of imaging patterns.
4. according to the method described in claim 1, it is characterized by:
The image of a variety of imaging patterns of flat samples is obtained, and by the image mosaic of identical imaging pattern at complete plate sample
Product sample full figure, specifically:
Computer obtains flat samples in all images of a variety of imaging patterns, adjacent two image of identical imaging pattern it
Between all have overlapping region, by flat samples the overlapping region of the image of identical imaging pattern be superimposed with each other be spliced into it is complete
The sample full figure of flat samples.
5. according to the method described in claim 2, it is characterized by:
Camera is driven by movement mechanism realize that camera is shot relative to flat samples continuous moving in shooting process, specifically:
Movement mechanism camera in shooting process, movement mechanism drives relative motion between camera or flat samples, and camera is every
Opposing plates sample walks a length and just sends primary shooting signal to camera, and continuous moving is shot under the conditions of non-stop.
6. according to the method described in claim 1, it is characterized by:
According to setting a variety of imaging patterns imaging sequence in flat samples continuous moving shoot, specifically:
A variety of imaging patterns and the successively automatic cycle switching of setting imaging sequence according to setting, the switching frequency of a variety of imaging patterns
Rate is consistent with the frequency that camera is shot, and shoots in flat samples according to preset motion profile continuous moving.
7. according to the method described in claim 1, it is characterized by:
Wherein, a variety of imaging patterns are as follows: one of bright-field imagery, dark-field imaging, grey field imaging, backlight imaging, polarization imaging
Or a variety of combination.
8. according to the method described in claim 1, it is characterized by:
Wherein, a variety of imaging patterns are as follows: one of red light source, green light source, yellow light sources, white light source or a variety of groups
It closes.
9. according to the method described in claim 1, it is characterized by:
Wherein, a variety of imaging patterns are as follows: the light source of the same color of various levels of brightness illuminates.
10. according to the method described in claim 2, it is characterized by:
Camera is driven by movement mechanism realize that camera is shot relative to flat samples continuous moving in shooting process, specifically:
The X-axis movement mechanism drive camera of movement mechanism or the every walking certain length of flat samples will in shooting process for camera
Triggering camera is shot, and camera or the flat samples continuous moving in the case where the X-axis movement mechanism of movement mechanism drives are shot.
11. -10 described in any item methods according to claim 1, which is characterized in that itself the following steps are included:
When shooting to first piece of flat samples according to the imaging sequence continuous moving of a variety of imaging patterns of setting, record camera is flat
Most clear camera heights when being imaged every time in plate sample；
The planar obit simulation when camera is carried out continuous moving shooting is saved as with camera heights most clear when being imaged every time
Optimum trajectory information；
When continuing flat samples after sensing, optimum trajectory information is called, control camera is relative to flat samples edge in the horizontal plane
When the optimum trajectory information movement, so that camera is shot with optimum height.
12. according to the method for claim 11, it is characterised in that:
It is described to first piece of flat samples according to a variety of imaging patterns of setting imaging sequence continuous moving shoot when, record camera
The most clear camera heights when being imaged every time in flat samples, specifically:
Controller controls Three-axis drive mechanism according to the imaging sequence driving camera of a variety of imaging patterns of setting relative to first piece
Flat samples carry out continuous moving shooting in each detection zone, and computer record camera is imaged every time in first piece of flat samples
When plane running track and the most clear camera axial height when being imaged in each detection zone.
13. according to the method for claim 12, it is characterised in that:
Controller control Three-axis drive mechanism according to a variety of imaging patterns of setting imaging sequence driving camera relative to
First piece of flat samples carry out continuous moving shooting in each detection zone, specifically:
The X-axis driving unit drive flat samples of controller control Three-axis drive mechanism are moved to the imaging of camera along the x axis
The movement forward or backwards of total travel length is done under line, while controller control camera takes pictures interval to flat samples according to setting
Each detection zone carry out Image Acquisition；
The Y-axis driving unit of controller control Three-axis drive mechanism drives camera to translate setting width along the y axis, then repeats
The detection content of above-mentioned steps；
The Z axis driving unit that controller controls Three-axis drive mechanism drive camera along Z-direction adjust in each detection zone at
As when camera heights, the detection content to repeat the above steps finishes until all detection zone Image Acquisition.
14. according to the method for claim 11, it is characterised in that:
When the continuous flat samples after sensing, optimum trajectory information is called, controls camera relative to flat samples in horizontal plane
On along the optimum trajectory information movement when so that camera is shot with optimum height, specifically:
When continuing flat samples after sensing, controller calls the plane fortune of camera camera when each detection zone is imaged clearest
It row track and most clear camera heights is imaged in each detection zone forms cam curve；
Camera, when moving according to plane running track, while being moved relative to flat samples according to cam curve, makes camera
Stable focusing is remained between flat samples, realizes and progress Image Acquisition is focused according to the face type of flat samples automatically.
15. -10 described in any item methods according to claim 1, which is characterized in that can be carried out by preset judgment criteria
The analysis of defect merges, to obtain optimal most stable of testing result comprising following steps:
The initial imperfection obtained by the image detection of each imaging pattern is divided into initial imperfection of corresponding each imaging pattern
Table simultaneously stores；
Spacing distance in the initial imperfection sublist of the image of each imaging pattern is met into preset first judgment criteria respectively
Defect point merges, and obtains the final defect point sublist of the image of each imaging pattern and stores；
Spacing distance between the final defect point sublist of the image of each imaging pattern is met into lacking for preset second judgment criteria
Trapping spot merges, and obtains final defect point table and stores, and each defect point in final defect point table includes to merge to form the defect
The image information of the defect point of the final defect point sublist of point；
According to the combination of the image of single imaging pattern or the image of multiple imaging patterns, by the defects of final defect point table
Point is shown.
16. a kind of automated optical detection equipment of high sensitivity characterized by comprising
Imaging mechanism comprising for obtaining camera, the mirror of a variety of imaging modality images of each detection zone in flat samples
Head and light source, and the light source is configured as that the light of different imaging patterns can be issued, make the imaging mechanism have it is a variety of at
As mode；
Movement mechanism is used to drive the three-axis moving mechanism of the relatively described flat samples movement of the imaging mechanism and support flat
The air bearing support platform of plate sample；
Controller is used to control the movement mechanism and imaging mechanism, makes the imaging mechanism according to a variety of imagings of setting
The imaging sequence of mode continuous moving in flat samples is shot, and is had between the image of two adjacent imaging patterns of the same race
Partly overlap region；
Computer, is used to obtain the image of a variety of imaging patterns of flat samples, and by the image mosaic of identical imaging pattern
At the sample full figure of complete flat samples.High efficient and flexible is made to defective data by defects detection and analysis tool simultaneously
Detection and analysis.
17. detection device according to claim 16, it is characterised in that:
The computer prestores optimum trajectory information, and the optimum trajectory information includes that the camera carries out continuous moving shooting
When planar obit simulation and every time imaging when most clear camera heights；
The controller is also used to call the optimum trajectory information, controls the camera relative to flat samples in the horizontal plane
When along the optimum trajectory information movement, so that camera is shot with optimum height.
18. detection device according to claim 16, it is characterised in that:
The Three-axis drive mechanism includes Z axis driving unit, Y-axis driving unit and X-axis driving unit, the Z axis driving unit
It is arranged on Y-axis driving unit, the Y-axis driving unit is arranged on the rack, and it is flat that the bottom of rack is equipped with air bearing support
Platform, the X-axis driving unit are arranged in air bearing support platform, and the camera and light source are arranged on Z axis driving unit.
19. detection device according to claim 18, it is characterised in that:
The Z axis driving unit is Z axis straight line mould group, and the camera and light source are fixed on the sliding block of Z axis straight line mould group,
The camera and light source follow the sliding block direction vertical up-or-down movement of Z axis straight line mould group, and the Z axis straight line mould group is for adjusting phase
The best focusing height of machine.
20. detection device according to claim 18, it is characterised in that:
The Y-axis driving unit is Y-axis straight line mould group, and the Z axis driving unit is fixed on the sliding block of Y-axis straight line mould group, institute
Y-axis straight line mould group is stated for driving camera and light source to move along a straight line along the y axis, the camera and light source are along the y axis to flat
Each detection zone carries out Image Acquisition in plate sample.
21. detection device according to claim 18, it is characterised in that:
The X-axis driving unit is X-axis straight line mould group, and the flat samples are connected in X-axis straight line mould group by vacuum chuck,
The X-axis straight line mould group moves along a straight line back and forth along the x axis for drawing flat samples, and the camera and light source are relative to plate
Sample carries out Image Acquisition to detection zone each in flat samples along the x axis.
22. detection device according to claim 16, it is characterised in that:
The controller and imaging mechanism, movement mechanism and calculating mechatronics, the controller pass through control sequential to imaging
Mechanism issues trigger signal, and control imaging mechanism obtains the flat samples image of a variety of imaging patterns, institute in a variety of imaging patterns
Stating controller calls the optimum trajectory information control movement mechanism of computer storage by optimum trajectory information movement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910448654.3A CN110132982A (en) | 2019-05-27 | 2019-05-27 | A kind of automatic optical detection method and equipment of high sensitivity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910448654.3A CN110132982A (en) | 2019-05-27 | 2019-05-27 | A kind of automatic optical detection method and equipment of high sensitivity |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110132982A true CN110132982A (en) | 2019-08-16 |
Family
ID=67582155
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910448654.3A Pending CN110132982A (en) | 2019-05-27 | 2019-05-27 | A kind of automatic optical detection method and equipment of high sensitivity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110132982A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110987959A (en) * | 2019-12-16 | 2020-04-10 | 广州量子激光智能装备有限公司 | Online burr detection method |
| CN111855681A (en) * | 2020-07-02 | 2020-10-30 | 快克智能装备股份有限公司 | Online non-stop AOI (automatic optical inspection) detection method |
| CN113459084A (en) * | 2021-05-21 | 2021-10-01 | 广东拓斯达科技股份有限公司 | Robot parameter calibration method, device, equipment and storage medium |
| CN114638824A (en) * | 2022-05-11 | 2022-06-17 | 成都数联云算科技有限公司 | Fusion method, device, equipment and medium for collecting images based on AOI equipment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101960295A (en) * | 2008-03-04 | 2011-01-26 | 有限会社共同设计企画 | Electronic component inspecting method and apparatus used in the method |
| CN105259189A (en) * | 2015-10-21 | 2016-01-20 | 凌云光技术集团有限责任公司 | Glass defect imaging system and method |
-
2019
- 2019-05-27 CN CN201910448654.3A patent/CN110132982A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101960295A (en) * | 2008-03-04 | 2011-01-26 | 有限会社共同设计企画 | Electronic component inspecting method and apparatus used in the method |
| CN105259189A (en) * | 2015-10-21 | 2016-01-20 | 凌云光技术集团有限责任公司 | Glass defect imaging system and method |
Non-Patent Citations (1)
| Title |
|---|
| 孙英: "基于优化滤波器的冷轧带钢表面缺陷分割与合并", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110987959A (en) * | 2019-12-16 | 2020-04-10 | 广州量子激光智能装备有限公司 | Online burr detection method |
| CN111855681A (en) * | 2020-07-02 | 2020-10-30 | 快克智能装备股份有限公司 | Online non-stop AOI (automatic optical inspection) detection method |
| CN113459084A (en) * | 2021-05-21 | 2021-10-01 | 广东拓斯达科技股份有限公司 | Robot parameter calibration method, device, equipment and storage medium |
| CN114638824A (en) * | 2022-05-11 | 2022-06-17 | 成都数联云算科技有限公司 | Fusion method, device, equipment and medium for collecting images based on AOI equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110132982A (en) | A kind of automatic optical detection method and equipment of high sensitivity | |
| CN110006905B (en) | Large-caliber ultra-clean smooth surface defect detection device combined with linear area array camera | |
| KR101246958B1 (en) | Specimen inspecting apparatus using multi-line senser camera and multi-light | |
| KR101949257B1 (en) | Apparatus for Testing Display device module and Method for Testing the same | |
| US20090147245A1 (en) | System and method for measuring optical resolution of lens | |
| CN101545870B (en) | High-speed automatic detector of microscopic pinhole defect in chromium plate film | |
| CN206772850U (en) | CCD lines for Intelligent Measurement surface defects of products sweep camera system | |
| JP2017181188A (en) | Circuit board inspection device | |
| CN109406527B (en) | System and method for detecting fine appearance defects of micro camera module lens | |
| CN108769454A (en) | It is a kind of intelligently to adjust weld seam negative plate digitization scanning device and its scanning drive method | |
| JP2006300913A (en) | Appearance inspection apparatus of light guide plate | |
| CN109827974B (en) | Resin optical filter film crack detection device and detection method | |
| CN108548825A (en) | A kind of transparent plate defect detecting device and method based on two-dimentional illumination | |
| JP2013250225A (en) | Visual inspection device and visual inspection method | |
| CN104062099A (en) | Testing device and method for LED light strips with side luminance | |
| IE80676B1 (en) | A measurement system | |
| WO2017157043A1 (en) | Measurement system based on point laser imaging | |
| CN209992407U (en) | Large-caliber ultra-clean smooth surface defect detection device combined with linear array camera | |
| CN207366467U (en) | A kind of surface defect real-time detection apparatus for increasing material manufacturing | |
| CN104330419A (en) | Method and device for detecting film | |
| CN106053485A (en) | Machine vision-based novel algorithm of intelligent circular inspection of steel ball surface defects | |
| CN210720188U (en) | Rapid automatic optical detection system | |
| CN108489830A (en) | Crack propagation measuring system | |
| CN112730442A (en) | Automatic online product surface defect detection device and system based on machine vision | |
| KR20120049826A (en) | Position alignment device, position alignment method, and computer readable recording medium having position alignment program recorded thereon |
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: 20190816 |