CN209148563U - A kind of double imaging type transparent component defect detecting devices - Google Patents
A kind of double imaging type transparent component defect detecting devices Download PDFInfo
- Publication number
- CN209148563U CN209148563U CN201821738995.1U CN201821738995U CN209148563U CN 209148563 U CN209148563 U CN 209148563U CN 201821738995 U CN201821738995 U CN 201821738995U CN 209148563 U CN209148563 U CN 209148563U
- Authority
- CN
- China
- Prior art keywords
- module
- transparent component
- detection
- workbench
- ultrasound
- 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.)
- Active
Links
Abstract
The utility model discloses a kind of double imaging type transparent component defect detecting devices, described device includes ultrasound detection module, optical detecting module, motion-control module, data fusion module, multichannel deep learning module and auxiliary manipulator, ultrasound detection module, optical detecting module and auxiliary manipulator pass through Ethernet and motion-control module communication connection, communication connection, motion-control module also pass through bus and display alarm module communication connection two-by-two by Ethernet for data fusion module, multichannel deep learning module and motion-control module.The utility model detects more accurate, automatic operation, promotes detection efficiency using ultrasound and the double image checkings of optics.
Description
Technical field
The utility model relates to 3C industry transparent component product defects detection device and method, specifically one kind is based on
The 3C transparent component defect intelligent detection device of ultrasound and the double imagings of optics.
Background technique
In China's 3C industry, transparent component constantly increases in application, in particular with quickly propelling for 5G technology, to glass
The component of material is widely used in terminal communication products.It is saturating for 3C but due to the continuous promotion of assembling quality
The defects detection of bright component is also increasingly stringenter.Currently, most of also rest on by the stage for manually visualizing detection identification, people
Work method labor intensity is very big, and due to needing to be detected by secondary light source, has injury larger the eyesight of worker.
In addition, with the continuous improvement of 3C transparent component examination criteria, it is single by the risk that manually there is missing inspection.
Currently, transparent purchase part defects detection on the market is most of based on artificial, a small amount of detection for having automation is set
Standby, based on single optical detection means, first camera collection image is initial data, at image recognition algorithm
Reason, and analyze and judge whether there is defect.It is simple by conventional but since 3C transparent component size is small, defect is unobvious
Optical principle is detected, and recognition accuracy has to be hoisted.Thus, there is an urgent need to a kind of new detection methods, by melting
A variety of detection means are closed, so that fine defects are exposed, so that Tiny Mass defect existing for identification product is easier, for enterprise
Industry promotes product qualification rate, reduces manufacturing cost.
Utility model content
In order to solve the above technical problems, the utility model provides a kind of double imaging type transparent component defects detection dresses
It sets, improves detection accuracy.
In order to solve the above-mentioned technical problem, the utility model takes following technical scheme:
A kind of double imaging type transparent component defect detecting devices, described device include ultrasound detection module, optical detection mould
Block, motion-control module, data fusion module, multichannel deep learning module and auxiliary manipulator, ultrasound detection module, optics
Detection module and auxiliary manipulator pass through Ethernet and motion-control module communication connection, data fusion module, multichannel depth
Study module and motion-control module pass through Ethernet communication connection two-by-two.
The motion-control module also passes through bus and display alarm module communication connection.
The ultrasound detection module includes variable frequency ultrasonic transmitter, focused ultrasonic transducer, C sweeps workbench, ultrasound is examined
Micrometer controller and ultrasound detection workbench are equipped with X to motor and X to screw rod in ultrasound detection workbench side, X to screw rod with
X is connected to motor, and X has Y-direction plate by connecting swivel nut attaching on screw rod, which is equipped with Y-direction motor and Y-direction screw rod, Y
It is connect to screw rod with Y-direction motor, C sweeps workbench and is mounted on Y-direction screw rod by nut, variable frequency ultrasonic transmitter and focusing
Ultrasonic probe is swept workbench with C by support rod and is connect, and X is equipped with X to screw end and returns to zero sensor, Y-direction screw rod end to positioning
End is equipped with Y-direction positioning and returns to zero sensor, X to motor, Y-direction motor, variable frequency ultrasonic transmitter, focused ultrasonic transducer respectively with
Ultrasound detection microcontroller communication connection.
The ultrasound detection module further includes sound-absorbing frame, which is located on ultrasound detection workbench for enclosing
Firmly 3C transparent component to be detected.
The optical detecting module includes optical detection workbench, LED line light source, angle adjustable reflection concave mirror, fixture, back
Scape plate, detection video camera and optical detection microcontroller, LED line light source, angle adjustable reflection concave mirror, fixture and background board setting
On optical detection workbench, angle adjustable reflection concave mirror is connect with stepper motor, and 3C transparent component to be detected is placed on fixture
On, optical detection workbench be equipped with column, detection video camera by support rod connect with column so that detect position for video camera in
Above 3C transparent component to be detected, detection video camera, LED line light source, stepper motor are communicated with optical detection microcontroller respectively
Connection.
The LED line light source and angle adjustable reflection concave mirror are respectively equipped with two groups, and it is transparent to be respectively provided at 3C to be detected
The region at component both ends carries out independent camera shooting and obtains photo.
The utility model has the following beneficial effects:
1) theoretical based on Multi-source Information Fusion, the ultrasound image in multiple channels and optical imagery are subjected to fusion detection, gram
Take the defect of single detection technique；
2) three kinds of specifications: 2.25MHz, 4.5MHz, 9MHZ are configured using ultrasonic transmitters, generates three kinds of supersonic guide-waves respectively
Ultrasound image under frequency covers different type defects；
3) using can the intensity of light source and the adjustable LED line light source of irradiating angle, by imaging of repeatedly taking pictures, reduction is to be checked
Survey the optical interference at 3C transparent component entrance port；
4) convolutional neural networks for using multichannel, after sample off-line training, can it is on-line automatic to ultrasound image and
Optical imagery is merged, and the feature of defect sample is automatically extracted, and substitution is artificial, completes the inspection that 3C transparent component is online, accurate
It surveys.
Detailed description of the invention
Attached drawing 1 is utility model device catenation principle schematic diagram；
Attached drawing 2 is the theory of mechanics schematic diagram of utility model device；
Attached drawing 3 is the catenation principle schematic diagram of ultrasound detection module in the utility model；
Attached drawing 4 is the theory of mechanics schematic diagram of ultrasound detection module in the utility model；
Attached drawing 5 is the catenation principle schematic diagram of optical detecting module in the utility model；
Attached drawing 6 is the theory of mechanics schematic diagram of optical detecting module in the utility model；
Attached drawing 7 is the structural schematic diagram of sound-absorbing frame in the utility model；
Attached drawing 8 is the utility model ultrasonic imaging schematic illustration；
Attached drawing 9 is the schematic network structure that the utility model multichannel deep learning module uses.
Specific embodiment
For the feature, technological means and specific purposes achieved, function that can further appreciate that the utility model, below
The utility model is described in further detail in conjunction with attached drawing and specific embodiment.
As shown in figure 1 and 2, the utility model discloses a kind of double imaging type transparent component defect detecting devices, described
Device includes ultrasound detection module 5, optical detecting module 2, motion-control module, data fusion module, multichannel deep learning
Module and auxiliary manipulator 4, ultrasound detection module, optical detecting module and auxiliary manipulator pass through Ethernet and motion control mould
Block communication connection, data fusion module, multichannel deep learning module and motion-control module pass through the Ethernet company of communicating two-by-two
It connects, motion-control module also passes through bus and display alarm module communication connection.Pass through movement after ultrasonic imaging and optical imagery
Control module is forwarded to after data fusion module handled, and is forwarded to multichannel deep learning mould by data fusion module again
Block is recognized, and recognition result is sent to motion-control module；Motion-control module passes through CAN bus and display alarm module phase
Connection sends final detection data in display alarm module；Auxiliary manipulator realizes feeding, transhipment and its work of blanking,
Realize the carrying 3C transparent component to be detected of automation.
Auxiliary manipulator 4 is mounted in manipulator base 13, is arranged in entrance 11, ultrasound detection module 5, optical detection mould
Among block 2 and outlet 12,3C transparent component 10 to be detected by manually or automatically conveyer belt (this field routine techniques, in figure not
Mark) it send to after going out to entrance 11, the ultrasound detection work being transported to by auxiliary manipulator 4 from entrance 11 in ultrasound detection module 5
Make on platform 3, after ultrasound detection module 5 completes detection, is transported on the optical detection workbench 1 in optical detecting module 2, etc.
Optical detecting module 2 complete detection after, be transported to outlet 12 go out after, by manually or automatically conveyer belt (this field routine techniques,
Do not indicated in figure) it is pushed to next process；Further, due to ultrasonic module detection unit and optical detection unit working time
It is longer than the auxiliary manipulator time, therefore, by rhythm control, it can be achieved that ultrasound detection module and optical detecting module it is parallel
Work.
As shown in figures 3 and 4, the ultrasound detection module includes variable frequency ultrasonic transmitter, focused ultrasonic transducer 7, C
Workbench 9, ultrasound detection microcontroller and ultrasound detection workbench 5 are swept, is equipped with X to motor in 5 side of ultrasound detection workbench
14 and X is connected to screw rod 15, X to screw rod and X to motor, and X has Y-direction plate 16 by connecting swivel nut attaching on screw rod, the Y-direction plate
16 are equipped with Y-direction motor 17 and Y-direction screw rod 18, and Y-direction screw rod is connect with Y-direction motor, and C sweeps workbench and is mounted on Y-direction by nut
On screw rod, variable frequency ultrasonic transmitter and focused ultrasonic transducer are swept workbench with C by support rod and are connect, and X is last to screw rod 15
End is equipped with X and returns to zero sensor 19 to positioning, and 18 end of Y-direction screw rod is equipped with Y-direction positioning and returns to zero sensor 20, and X is electric to motor, Y-direction
Machine, variable frequency ultrasonic transmitter, focused ultrasonic transducer respectively with ultrasound detection microcontroller communication connection.Ultrasound detection module
It further include sound-absorbing frame 8, which is located on ultrasound detection workbench for enclosing 3C transparent component to be detected.It is such as attached
Shown in Fig. 7, which has corresponding electromagnet A, electromagnet B convenient for connecting with ultrasound detection microcontroller, by sound-absorbing
Material 81, flexible sound-absorbing material 82 are alternatively formed.
Ultrasonic transmitter and focused ultrasonic transducer carry out 3C transparent component to be detected flat under the drive that C sweeps workbench
Face C is swept, and ultrasound detection microcontroller realizes the motion control that workbench is swept to C by CAN bus.It is electric to motor and Y-direction using X
Machine drives C to sweep workbench in the stroke movement in the direction X-Y, so as to carry out ultrasonic imaging in specified region.Ultrasound detection
Microcontroller controls electromagnet by the isolating amplifier circuit of I/O and is acted, and changes the acoustic characteristic of sound-absorbing frame, to subtract
Interference of the ultrasonic echo of weak 3C transparent component circumferential side frame to testing result.
X drives X to move to positioning screw rod to motor, realizes that C sweeps the X-direction positioning of workbench；Y-direction motor drives Y-direction silk
The movement of bar positioning element realizes that C sweeps the Y-direction positioning of workbench；Sensor is returned to zero using positioning, it is transparent to every 3C to be detected
After the completion of component detection, zero adjustment is returned again；C, which is swept, to be installed variable frequency ultrasonic transmitter by support rod on workbench and gathers
Burnt ultrasonic probe, variable frequency ultrasonic transmitter and focused ultrasonic transducer are arranged with 3C transparent component to be detected at vertical direction,
It is contacted with 3C transparent component to be detected surface.The 3C transparent component to be detected for being coated with ultrasonic coupling agent is placed on ultrasound detection work
Make on platform, component surrounding to be detected installs sound-absorbing frame, reduces noise jamming.
As depicted in figures 5 and 6, the optical detecting module includes optical detection workbench 1, LED line light source 23, adjustable angle
Degree reflection concave mirror 24, fixture 21, background board 22, detection video camera 25 and optical detection microcontroller, LED line light source, adjustable angle
Degree reflection concave mirror, fixture and background board are arranged on optical detection workbench, and angle adjustable reflection concave mirror is connect with stepper motor,
3C transparent component to be detected is placed on fixture, optical detection workbench be equipped with column, detection video camera by support rod with
Column connects so that video camera, LED line light source, stepper motor are detected above 3C transparent component to be detected in detection position for video camera
Respectively with optical detection microcontroller communication connection.LED line light source and angle adjustable reflection concave mirror are respectively equipped with two groups, and divide
The region for not being located at 3C transparent component to be detected both ends carries out independent camera shooting and obtains photo.Optical detection microcontroller passes through CAN
Bus realizes that the angle to the brightness of LED line light source and its angle adjustable reflection concave mirror controls, and controls detection video camera
Take pictures；Further, LED line light source and angle adjustable reflection concave mirror configure two sets, complete the both ends of 3C transparent component to be detected
Individual photos realize clamped one time positioning, repeatedly take pictures.
One end of support rod is mounted on column, height-adjustable, and column is connected through a screw thread and optical detection workbench
It is rigidly connected, camera placements is detected on the other end of support rod, with 3C transparent component vertical direction interval to be detected
40~60cm.Background board and fixture are placed on optical detection workbench by rigid connection, and 3C transparent component to be detected is by auxiliary
Help machinery to be manually placed on fixture, workpiece to be detected in place after, fixture Automatic-clamping, fixture is carried out by optical detection microcontroller
Driving.Angle adjustable reflection concave mirror is placed on the both ends of 3C transparent component to be detected, drives each LED line light source self-movement,
To realize the fine tuning of detection light source incident angle.
The motion-control module realizes the parallel control to ultrasound detection module and optical detecting module, issues detection
Instruction, and receive ultrasound image and optical imagery；Further, by beat realize ultrasound detection module and optical detecting module and
Row work promotes detection efficiency；Auxiliary manipulator realizes automation 3C to be detected is saturating under the control of motion-control module
Bright component is transported in ultrasound detection module or optical detecting module.
For the data fusion module by software realization, carrier is independent data fusion dsp chip, by ultrasound detection mould
The ultrasound image of block and the light image of optical detecting module carry out the pretreatment of image；To under three kinds of different supersonic guide-wave frequencies
Ultrasound image carry out Boolean calculation, and generate new image；Concave mirror lighting source is reflected to the angle adjustable under different angle,
Merging forms new optical imagery, reduces interference.
The multichannel deep learning module uses multichannel convolutive neural network, by independent convolutional neural networks
Dsp chip is realized, can pass through 10 channels, wherein 6 channel ultrasounds, 4 channel opticals, input the same transparent structure of 3C to be detected parallel
The ultrasound image and optical imagery of part.
The auxiliary manipulator realizes the automatically working of feeding, transhipment and its blanking.
The display alarm module realizes the display of testing result, for defective 3C transparent component to be detected, energy
Operator is prompted by sound-light alarm.
3C transparent component to be measured is transported to detection position by auxiliary manipulator, is examined using ultrasound detection module and optics
It surveys module and ultrasonic imaging and optical imagery is carried out to 3C transparent component to be measured respectively, image data is transmitted to motion control mould
Block；
Image data is forwarded to the warm module of data and pre-processed by motion-control module, again by data fusion module
It is forwarded to multichannel deep learning module to be recognized, recognition result is re-send into motion-control module, motion-control module
It will test data to be sent in display alarm module.
The method is when carrying out ultrasonic imaging, using C-scan imaging method, by configure 2.25MHz, 4.5MHz,
Tri- kinds of 9MHZ different ultrasonic frequencies generate corresponding ultrasound image, cover different type defects, super to pressing for detection
The signal strength size that sound reflecting is returned generates 255 grades of gray level images, and by normalized, the gray scale of maximum intensity is 0 pair
Black is answered, the smallest gray scale of intensity is 1 corresponding white.
It is described that image data is pre-processed by data warm module, specifically: to ultrasound detection module
Ultrasound image under tri- kinds of 2.25MHz, 4.5MHz, 9MHZ different supersonic guide-wave frequencies carries out Boolean XOR operation two-by-two, and raw
Three images of Cheng Xin amount to six ultrasound detection images；To optical detecting module detection when, to obtained under different angle to
Few two optical imagery images, form individual new optical imagery by image interfusion method, then adjust separately 3C transparent component
Working as degree of illuminating and taking pictures for both ends, amounts to and obtains four optical detection images.
The multichannel deep learning module realizes that it is more that image data is passed to this using convolutional neural networks dsp chip
Before channel depth study module, by data fusion module to the six ultrasound detection images and four optical detection figures of imaging
Picture is divided into the image group of 128 × 128 sizes, and ultrasound detection image is identical with optical detection picture size, identical bits when segmentation
The image set is classified as one group；Image after segmentation in groups, by batch input multichannel deep learning module, every batch of image is 60 secondary
The image of individual 128 × 128 size, after carrying out convolution sum pondization operation twice, then the neural network connected entirely twice, it is defeated
Out in 512 dimensional vectors, then by soft return 512 dimensional vectors are exported as vector, then will test result and pass through Industrial Ethernet
Bus transfer reminds operator to motion-control module, finally by display alarm module.
As shown in Fig. 7, ultrasound detection microcontroller controls electromagnet A and electromagnet B by the isolating amplifier circuit of I/O
It is acted respectively, thus the acoustic characteristic of compressed flexible sound-absorbing material；It is super for tri- kinds of differences of 2.25MHz, 4.5MHz, 9MHZ
Guided Waves, electromagnet A and electromagnet B motion flow are executed by following rule:
(a) home position is in for the supersonic guide-wave of 2.25MHz, electromagnet A and electromagnet B, compressed flexible is not inhaled
The supersonic guide-wave of sound material, 2.25MHz can be decayed by three layers of sound-absorbing material and two sheets of flexible sound-absorbing material, reduce boundary
The interference of back wave.
(b) for the supersonic guide-wave of 4.5MHz, electromagnet A is in home position, not compressed flexible sound-absorbing material, but electromagnetism
Iron B is moved be in the right, compressed flexible sound-absorbing material, and the supersonic guide-wave of 4.5MHz can be soft by two layers of sound-absorbing material and one layer
Property sound-absorbing material is decayed, and the interference of boundary echo is reduced.
(c) supersonic guide-wave of 9MHz, electromagnet A and electromagnet B are moved to the right, all compressed flexible sound absorber
Material, the supersonic guide-wave of 9MHz can be decayed by one layer of sound-absorbing material, reduce the interference of boundary echo.
As shown in Fig. 8, when carrying out the detection of 3C transparent component, supersonic guide-wave enters 3C transparent component by couplant, such as
3C transparent component to be detected does not have defect (surface and inside), and guided wave can travel to always the bottom of 3C transparent component to be detected
Face.Since 3C transparent component to be detected bottom surface is smooth and is parallel to test surface, it is based on ultrasonic reflection principle, supersonic guide-wave is the bottom of by
Face is reflected and returns to aggregation ultrasonic probe, fluctuation of the supersonic guide-wave in 3C transparent component material to be detected on the direction of propagation
Equation are as follows:
pi(y, t)=∑ pi0ej(ωc-kx) (1)
In formula (1): Pi0For supersonic guide-wave amplitude；K is supersonic guide-wave wave number, and k=w/c；ω is angular frequency；C is ultrasound
Spread speed of the guided wave in 3C transparent component material.Interface echo of the supersonic guide-wave in test surface are as follows:
prf(y,t)|Y=0=∑ prf0ejωt (2)
The guided wave back wave that supersonic guide-wave is returned from bottom reflection are as follows:
Since the defect impedance of 3C transparent component has differences with 3C transparent component material impedance, supersonic guide-wave encounters defect
(in 3C transparent component material internal any point yiPlace) have reflection signal prd(y,t),
By (4) formula it is found that defect reflection Bobbi test surface back wave lags yi/ c, it is (y more advanced than bottorm echoT-yi)/
c.If with y=yiFor section, then available sound wave is in the material along the incidence wave at direction of propagation any point.Aggregation ultrasound is visited
The ultrasonic pulse of return is changed into electric pulse by head, after A/D is converted, by every bit collected ultrasonic echo characteristic quantity
It is showed in the form of gray scale, then just obtains scan image.
As shown in Fig. 9, multichannel deep learning module is realized using convolutional neural networks dsp chip, in incoming processing
Before, 128 × 128 are divided into the detection image of 10 channels (6 channel ultrasounds, 4 channel opticals) by data fusion module (4)
The image group (10 layers, 1 channel corresponding 1 layer) of size, ultrasound detection image is identical with optical detection picture size, and when segmentation is identical
The image of position is classified as one group；Image after segmentation in groups, by batch (image of individual secondary 128 × 128 size of every batch of 60) input
In multichannel deep learning module.
Further, the blending image for 128 × 128 × 10, the A1 convolutional layer being sent into convolutional network, using 5 × 5 windows
After mouth convolution operation, 120 width images of 124 × 124 pixels are generated, then compression processing is carried out by the pond the A2 layer in module, it is raw
At 120 width images of 62 × 62 pixels, later, second of convolution operation is carried out, the A3 convolutional layer being sent into convolutional network, again
After the operation of 3 × 3 window convolutions, 240 width images of 60 × 60 pixels are generated, then carried out by the pond the A4 layer in convolutional network
Compression processing generates 240 width images of 30 × 30 pixels, later, handles by the full articulamentum of the A5 of convolutional network, output 4096
The vector of dimension further handles by the full articulamentum of A6 in convolutional network, exports the vector of 128 dimensions, finally, again by
The soft recurrence layer of A7 in convolutional network exports 4 dimensional vectors, indicate 3C transparent component to be detected belong to 4 classes (normal, spot defect,
Scratch defects, crack defect) probability density distribution, to recognize 3C transparent component to be checked with the presence or absence of defect, finally, then will
Testing result informs operator by Industrial Ethernet bus transfer to motion-control module, finally by display alarm module
Member.
Further, the off-line training sample database of multichannel depth convolutional neural networks can increase sample size.Thus,
The precision of detection can be promoted with the increase of sample size；Multichannel depth convolutional neural networks can also be made by user
With update is trained in the process, it also can choose and regularly updated by device manufacturer.
It should be noted that it is practical new to be not limited to this above is only the preferred embodiment of the utility model
Type, although the utility model is described in detail referring to embodiment, for those skilled in the art, still
It can modify to technical solution documented by previous embodiment or equivalent replacement of some of the technical features,
But within the spirit and principle of the utility model, any modification, equivalent replacement, improvement and so on should be included in
Within the protection scope of the utility model.
Claims (6)
1. a kind of double imaging type transparent component defect detecting devices, which is characterized in that described device includes ultrasound detection module, light
Learn detection module, motion-control module, data fusion module, multichannel deep learning module and auxiliary manipulator, ultrasound detection
Module, optical detecting module and auxiliary manipulator are data fusion module, more by Ethernet and motion-control module communication connection
Channel depth study module and motion-control module pass through Ethernet communication connection two-by-two.
2. double imaging type transparent component defect detecting devices according to claim 1, which is characterized in that the motion control
Module also passes through bus and display alarm module communication connection.
3. double imaging type transparent component defect detecting devices according to claim 2, which is characterized in that the ultrasound detection
Module includes that variable frequency ultrasonic transmitter, focused ultrasonic transducer, C sweep workbench, ultrasound detection microcontroller and ultrasound detection
Workbench is equipped with X to motor and X to screw rod in ultrasound detection workbench side, and X is connected to screw rod and X to motor, and X is to screw rod
Upper to have Y-direction plate by connecting swivel nut attaching, which is equipped with Y-direction motor and Y-direction screw rod, and Y-direction screw rod is connect with Y-direction motor,
C sweeps workbench and is mounted on Y-direction screw rod by nut, variable frequency ultrasonic transmitter and focused ultrasonic transducer by support rod with
C sweeps workbench connection, and X is equipped with X to screw end and returns to zero sensor to positioning, and Y-direction screw end is equipped with Y-direction positioning and returns to zero sensing
Device, X are communicated with ultrasound detection microcontroller respectively to motor, Y-direction motor, variable frequency ultrasonic transmitter, focused ultrasonic transducer
Connection.
4. double imaging type transparent component defect detecting devices according to claim 3, which is characterized in that the ultrasound detection
Module further includes sound-absorbing frame, which is located on ultrasound detection workbench for enclosing 3C transparent component to be detected.
5. double imaging type transparent component defect detecting devices according to claim 4, which is characterized in that the optical detection
Module includes optical detection workbench, LED line light source, angle adjustable reflection concave mirror, fixture, background board, detection video camera and light
Detection microcontroller is learned, LED line light source, angle adjustable reflection concave mirror, fixture and background board are arranged on optical detection workbench,
Angle adjustable reflection concave mirror is connect with stepper motor, and 3C transparent component to be detected is placed on fixture, on optical detection workbench
Equipped with column, detects video camera and connect with column by support rod so that detecting position for video camera on 3C transparent component to be detected
Side, detection video camera, LED line light source, stepper motor respectively with optical detection microcontroller communication connection.
6. double imaging type transparent component defect detecting devices according to claim 5, which is characterized in that the LED line light
The region that source and angle adjustable reflection concave mirror are respectively equipped with two groups, and are respectively provided at 3C transparent component to be detected both ends carries out only
Vertical camera shooting obtains photo.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201821738995.1U CN209148563U (en) | 2018-10-25 | 2018-10-25 | A kind of double imaging type transparent component defect detecting devices |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201821738995.1U CN209148563U (en) | 2018-10-25 | 2018-10-25 | A kind of double imaging type transparent component defect detecting devices |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN209148563U true CN209148563U (en) | 2019-07-23 |
Family
ID=67270835
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201821738995.1U Active CN209148563U (en) | 2018-10-25 | 2018-10-25 | A kind of double imaging type transparent component defect detecting devices |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN209148563U (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109115805A (en) * | 2018-10-25 | 2019-01-01 | 广东华中科技大学工业技术研究院 | Transparent component defect detecting device and method based on ultrasound and the double imagings of optics |
| CN110346004A (en) * | 2019-08-16 | 2019-10-18 | 杭州山科智能科技股份有限公司 | A kind of flow measuring data fusion method of two-channel ultrasonic time difference method |
| CN110827256A (en) * | 2019-10-31 | 2020-02-21 | 广东华中科技大学工业技术研究院 | Optical and thermal infrared multi-stage imaging detection method and device for defects of transparent component |
-
2018
- 2018-10-25 CN CN201821738995.1U patent/CN209148563U/en active Active
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109115805A (en) * | 2018-10-25 | 2019-01-01 | 广东华中科技大学工业技术研究院 | Transparent component defect detecting device and method based on ultrasound and the double imagings of optics |
| CN110346004A (en) * | 2019-08-16 | 2019-10-18 | 杭州山科智能科技股份有限公司 | A kind of flow measuring data fusion method of two-channel ultrasonic time difference method |
| CN110827256A (en) * | 2019-10-31 | 2020-02-21 | 广东华中科技大学工业技术研究院 | Optical and thermal infrared multi-stage imaging detection method and device for defects of transparent component |
| CN110827256B (en) * | 2019-10-31 | 2022-04-26 | 广东华中科技大学工业技术研究院 | Optical and thermal infrared multi-stage imaging detection method and device for defects of transparent component |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109115805A (en) | Transparent component defect detecting device and method based on ultrasound and the double imagings of optics | |
| CN209148563U (en) | A kind of double imaging type transparent component defect detecting devices | |
| CN104792793B (en) | Optical defect detection method and system | |
| CN103913468B (en) | Many defects of vision checkout equipment and the method for large-scale LCD glass substrate on production line | |
| CN108765416A (en) | PCB surface defect inspection method and device based on fast geometric alignment | |
| CN100371677C (en) | Method and device for analysing the surface of a substrate | |
| CN102192911B (en) | System for detecting quality of metal cap based on machine vision | |
| CN211070921U (en) | Instrument appearance detection device based on 3D scanning method | |
| CN108413873A (en) | A kind of online dimensional measurement of phone housing and surface defects detection system and method | |
| CN107957425A (en) | Transparent material defect detecting system and method | |
| CN110216080A (en) | A kind of Quality Monitoring Control System of the PCB processing producing line based on image comparison | |
| CN104792869B (en) | The Ultrasonic Nondestructive system of low-voltage electrical apparatus electrical contact brazing quality | |
| CN109127464A (en) | A kind of vision detects sorting equipment and its control method automatically | |
| CN110501347A (en) | A kind of rapid automatized Systems for optical inspection and method | |
| CN208207914U (en) | PCB surface defect detecting device based on fast geometric alignment | |
| CN110523657A (en) | A kind of instrument appearance delection device and its working method based on 3D scanning method | |
| CN211122594U (en) | Transparent plate defect detection device based on machine vision | |
| CN108267460A (en) | For the matrix form vision detection system and method for transparent material defects detection | |
| CN109342561B (en) | Curved surface weldment ultrasonic detection device and method | |
| CN208155258U (en) | A kind of online dimensional measurement of phone housing and surface defects detection system | |
| CN107238955A (en) | Liquid crystal module screen detecting system | |
| CN113607748B (en) | Optical coherence tomography detection system and method for transparent or translucent articles | |
| CN209927731U (en) | Visual identification equipment for weld joints of workpieces | |
| CN104655646A (en) | Glass substrate internal defect checking system and checking method for height position of internal defect | |
| CN206832700U (en) | The defects of one kind is based on infrared distance measuring sensor detection means |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant |