CN106897994A - A kind of pcb board defect detecting system and method based on layered image - Google Patents
A kind of pcb board defect detecting system and method based on layered image Download PDFInfo
- Publication number
- CN106897994A CN106897994A CN201710045720.3A CN201710045720A CN106897994A CN 106897994 A CN106897994 A CN 106897994A CN 201710045720 A CN201710045720 A CN 201710045720A CN 106897994 A CN106897994 A CN 106897994A
- Authority
- CN
- China
- Prior art keywords
- image
- defect
- pcb
- mapping
- module
- 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
- 230000007547 defect Effects 0.000 title claims abstract description 131
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000013507 mapping Methods 0.000 claims abstract description 77
- 238000012360 testing method Methods 0.000 claims abstract description 43
- 238000003709 image segmentation Methods 0.000 claims abstract description 21
- 238000007689 inspection Methods 0.000 claims abstract description 15
- 238000007781 pre-processing Methods 0.000 claims abstract description 15
- 238000003384 imaging method Methods 0.000 claims abstract description 8
- 238000004458 analytical method Methods 0.000 claims abstract description 4
- 239000000463 material Substances 0.000 claims description 21
- 230000009466 transformation Effects 0.000 claims description 21
- 238000006243 chemical reaction Methods 0.000 claims description 14
- 235000000391 Lepidium draba Nutrition 0.000 claims description 10
- 230000002708 enhancing effect Effects 0.000 claims description 9
- 238000001914 filtration Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 230000011218 segmentation Effects 0.000 claims description 8
- 230000000877 morphologic effect Effects 0.000 claims description 7
- 230000002950 deficient Effects 0.000 claims description 4
- 239000000284 extract Substances 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 4
- 241000276425 Xiphophorus maculatus Species 0.000 claims description 3
- 239000012634 fragment Substances 0.000 claims description 3
- 238000002203 pretreatment Methods 0.000 claims 1
- 238000001514 detection method Methods 0.000 description 15
- 238000005516 engineering process Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 235000013399 edible fruits Nutrition 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000004587 chromatography analysis Methods 0.000 description 1
- 238000000280 densification Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000002594 fluoroscopy Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000012372 quality testing Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 238000002601 radiography Methods 0.000 description 1
- 238000004064 recycling Methods 0.000 description 1
- 230000007847 structural defect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/003—Reconstruction from projections, e.g. tomography
- G06T11/006—Inverse problem, transformation from projection-space into object-space, e.g. transform methods, back-projection, algebraic methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10116—X-ray image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30141—Printed circuit board [PCB]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Pure & Applied Mathematics (AREA)
- Algebra (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of pcb board defect detecting system based on layered image and method, the pcb board defect detecting system based on layered image, including:Image capture module, image procossing rebuild module and PCB defect recognition modules;Image capture module includes X-ray image acquisition device, for gathering the projected image of PCB on-gauge plates and the projected image of PCB boards under test;Image procossing rebuilds module to be used to for the projected image of the projected image of PCB on-gauge plates and PCB boards under test to carry out Stratified Imaging, is obtained standard drawing and is treated mapping;PCB defect recognition modules include:Image pre-processing module, image registration module, image segmentation module and defect recognition module;For Treatment Analysis standard drawing and treating mapping, so as to carry out defect recognition to PCB boards under test, and testing result is exported.The invention also discloses a kind of pcb board defect inspection method of the above-mentioned pcb board defect detecting system based on layered image of application.
Description
Technical field
The present invention relates to the quality testing field in multilayer board production process, more particularly to one kind is applied many
Defects detection based on chromatographic technique and recognition methods in layer printed circuit board production process.
Background technology
Printed circuit board (PCB) (hereinafter referred to as pcb board) is the critical component of various electronic products, the direct shadow of quality of its performance
Ring the service life of electronic product.It is of future generation as electronic system product is to multi-functional, miniaturization, the development in lightweight direction
Electronic product be more densification by outstanding behaviours to pcb board requirement, become more meticulous and micro hole, this is caused to pcb board
Quality inspection it is more and more challenging.
Initial pcb board detection relies primarily on artificial vision, but artificial vision's detection uncertain factor is big, efficiency is low, after
Image processing techniques is applied to pcb board detection field, automatic optics inspection is generated, gathers to be checked by CCD camera
Pcb board image is surveyed, the image feeding image pick-up card of collection is digitized treatment afterwards, such as image carried out afterwards and is located in advance
Reason, image segmentation, template matches etc. judge that pcb board whether there is defect.
The method for being presently used for pcb board detection mainly has artificial vision-based detection, automatic optics inspection AOI (Automatic
Optic Inspection), automatic X-ray detection AXI (Automatic X-ray Inspection).For multi-layer PCB board
It is imitated with detection, AOI technologies are only capable of detecting the external behavior of product, and the production that one layer is often completed in process of production needs
Detected once.The X-ray DR (Digital Radiography) of relatively early application is although detection can provide fine definition
Fluoroscopy images, but because the information overlap on depth direction cannot also be detected to pcb board internal structural defects.
With the method Computer that x-ray ct technology will be similar in the appearance of medical domain, researchers
Laminography (CL, computer X-ray Stratified Imaging technology) is applied to the detection of industrial circle component defect.
The content of the invention
For defect present in prior art, it is an object of the invention to provide a kind of pcb board based on layered image
Defect detecting system and method, are directed to a kind of intellectuality of multi-layered high-density pcb board offer, are efficiently based on layered image
Pcb board defect detecting system and method, its defect location are accurately easy and effective.
To achieve the above objectives, the present invention is adopted the technical scheme that:
A kind of pcb board defect detecting system based on layered image, including:Image capture module, image procossing rebuild mould
Block and PCB defect recognition modules;
Described image acquisition module include X-ray image acquisition device, for gather PCB on-gauge plates projected image and
The projected image of PCB boards under test;
Described image processing reconstructed module is used to enter the projected image of the projected image of PCB on-gauge plates and PCB boards under test
Row Stratified Imaging, obtains standard drawing and treats mapping;
The PCB defect recognitions module includes:Image pre-processing module, image registration module, image segmentation module and lack
Fall into identification module;For Treatment Analysis standard drawing and treating mapping, so as to carry out defect recognition to PCB boards under test, and detection is exported
As a result.
On the basis of such scheme, described image pretreatment module is used for standard drawing and treats that mapping is filtered respectively
Noise processed and image enhancement processing.
On the basis of such scheme, described image registration module is used to extract positioning on PCB on-gauge plates and PCB boards under test
The position dimension feature in hole, is then zoomed in and out in standard drawing and the geometric position difference for treating on mapping by location hole, translated
And rotation process, realize treating that mapping is corresponded with the pixel of standard drawing.
On the basis of such scheme, described image segmentation module is used for by the standard after image registration resume module
Scheme and treat that mapping carries out contrast enhancing, the conversion of white top cap and iterative threshold segmentation respectively.
On the basis of such scheme, the defect recognition module is used for to be measured after image segmentation module is processed
Figure and standard drawing carry out differing from shadow computing, then carry out Morphological scale-space, finally carry out Classifcation of flaws.
A kind of pcb board defect inspection method based on layered image, examines using the above-mentioned pcb board defect based on layered image
Examining system, comprises the following steps:
The projection of S1, the projected image and PCB boards under test for gathering PCB on-gauge plates respectively using X-ray image acquisition device
Image;
S2, the projected image of the projected image of PCB standard editions and PCB boards under test is rebuild into module by image procossing respectively
In filtered back projection FDK (Feldkamp-Davis-Kress) restructing algorithm of platy structure carry out Stratified Imaging, respectively
To standard drawing and treating mapping;
S3, image preprocessing is carried out to standard drawing, then preserve standard drawing;PCB boards under test are detected, is opened to be measured
Figure, treating mapping carries out image preprocessing, then to standard drawing and treating that mapping carries out image registration;
After the completion of S4, image registration, standard drawing is isolated by image segmentation and background and line pad of mapping etc. is treated
Feature, then to standard drawing and treat that mapping makes the difference shadow computing;Then Morphological scale-space is carried out and filters the noise of non-defective part to obtain
Defect image, the contour feature finally according to defect is identified to defect type, obtains testing result.
On the basis of such scheme, described image pretreatment includes image denoising and image enhaucament;Described image denoising
Filtering noise reduction before image procossing is carried out using medium filtering, filter window uses 3 × 3 pixel sizes;
Described image enhancing becomes gradation of image interval of changing commanders and is divided into two sections or even multistage by linear gradation, then makees respectively
Linear transformation, each straightway corresponds to a linear transformation relation for part.
On the basis of such scheme, described image registration includes positioning loop truss and geometric transformation;Described image registration
Process be:
The setting circle of PCB on-gauge plates and the setting circle of PCB boards under test are detected using random Hough transformation, is extracted
The position dimension feature of PCB on-gauge plates setting circle and PCB board under test setting circles, then by the way that location hole is in standard drawing and treats mapping
On geometric transformation treat the size position deviation of mapping and be adjusted.
On the basis of such scheme, the adjustment is concretely comprised the following steps:
Assuming that A (x1, y1), B (x2, y2) it is two central coordinate of circle of setting circle of PCB on-gauge plates, A ' (x3, y3), B ' (x4, y4)
It is two central coordinate of circle of setting circle of PCB boards under test,
(1) the level angle θ for treating mapping positioning centre is calculatedABWith the level angle θ ' of standard drawing positioning centreAB, such as
Fruit θAB≠θ′AB, then treating mapping carries out rotation transformation, and the anglec of rotation is θAB-θ′AB(clockwise for just);
(2) calculate and treat the distance between two positioning centres of mapping LABThe distance between with two positioning centres of standard drawing
L′ABIf, LAB≠L′AB, then treat mapping and zoom in and out conversion;Scaling multiple is n=L 'AB/LAB;
(3) boundary position for treating mapping is calculated, and will treats that the target area of mapping extracts according to boundary position information,
That is extracted treats that mapping target area is with coordinate (x3-x1, y3-y1)、(x3-x1, y4-y2)、(x4-x2, y3-y1)、(x4-x2, y4-
y2) it is the rectangle of end points.
On the basis of such scheme, the detailed process of described image segmentation is:Contrast enhancing is first carried out, then is carried out white
Top cap is converted, and finally carries out image segmentation using iteration threshold method;The process that implements of the iteration threshold method is:
(1) maximum gradation value and minimum gradation value of image are obtained, g is designated as respectivelymaxAnd gmin, make initial threshold t0=
(gmax+gmin)2;
(2) according to threshold value tiIt is foreground and background to divide the image into, and the average gray value g of prospect is obtained respectivelyAWith background
Average gray value gB;
(3) new threshold value t is obtainedi+1=(gA+gB)/2;
(4) if ti=ti+1, gained is threshold value;Otherwise turn (2), iterate to calculate.
On the basis of such scheme, the identification process of the defect type is:
(1) contour feature first to defect image carries out extracting the boundary profile segments N for obtaining defect image1;
(2) and then according to the boundary profile segments N of defect image1Carry out first step identification;
If N1=4, then short circuit or open circuit are identified as, it is designated as 1 class defect;
If N1=2, then burr or defect are identified as, it is designated as 2 class defects;
If N1=1, then fifth wheel or cavity are identified as, it is designated as 3 class defects;
(3) it is that many material or scarce material determine defect type finally according to defect;
Then it is open circuit if lacking material if many material, are short circuit to 1 class defect;
Then it is defect if lacking material if many material, are burr to 2 class defects;
Then it is cavity if lacking material if many material, are fifth wheel to 3 class defects;
(4) in order to the defect on PCB boards under test more intuitively is presented into user, needed after classifying to defect type
Accurate position mark and type mark are carried out to defect.
On the basis of such scheme, the boundary profile segments N for obtaining defect image1Detailed process be:
(1) origin of coordinates is chosen, the upper left corner with defect image uses defect image 2 × 2 square structure as origin
Element expansion process, carries out contour feature extraction;
(2) each point coordinate value on the profile border of acquisition is stored among a two-dimensional array B successively;
(3) treat mapping and enter row threshold division, obtain binary map, be designated as I0;
(4) initial fragment number variable N=1 is set, I is returned according to the coordinate value in two-dimensional array B0In, write down at first point
Pixel value E1(0 or 1), then judges boundary coordinate in I successively0The pixel value of middle corresponding points, if i-th pixel value is Ei(i=
1,2,3 ... .n), when pixel value is converted once, N=N+1, until terminating point, writes down terminating point pixel value for En;
If N=1, N1=N=1;
If N ≠ 1 and En≠E1, then N1=N;
If N ≠ 1 and En=E1, then N1=N-1.
On the basis of such scheme, the mark of the defect asks the method for rectangular centre to enter using Minimum Enclosing Rectangle method
Line flag.
Brief description of the drawings
The present invention has drawings described below:
Fig. 1 is based on the pcb board defect detecting system structured flowchart of layered image;
Setting circle detects schematic diagram in Fig. 2 image registration modules;
Diagram to be measured in Fig. 3 image registration modules after geometric transformation is intended to;
Fig. 4 testing result schematic diagrames.
Specific embodiment
The present invention is described in further detail below in conjunction with accompanying drawing.
A kind of pcb board defect detecting system based on layered image, including:Image capture module, image procossing rebuild mould
Block and PCB defect recognition modules;
Described image acquisition module include X-ray image acquisition device, for gather PCB on-gauge plates projected image and
The projected image of PCB boards under test;
Described image processing reconstructed module is used to enter the projected image of the projected image of PCB on-gauge plates and PCB boards under test
Row Stratified Imaging, obtains standard drawing and treats mapping;
The PCB defect recognitions module includes:Image pre-processing module, image registration module, image segmentation module and lack
Fall into identification module;For Treatment Analysis standard drawing and treating mapping, so as to carry out defect recognition to PCB boards under test, and detection is exported
As a result.
On the basis of such scheme, described image pretreatment module is used for standard drawing and treats that mapping is filtered respectively
Noise processed and image enhancement processing.
On the basis of such scheme, described image registration module is used to extract positioning on PCB on-gauge plates and PCB boards under test
The position dimension feature in hole, is then zoomed in and out in standard drawing and the geometric position difference for treating on mapping by location hole, translated
And rotation process, realize treating that mapping is corresponded with the pixel of standard drawing.
On the basis of such scheme, described image segmentation module is used for by the standard after image registration resume module
Scheme and treat that mapping carries out contrast enhancing, the conversion of white top cap and iterative threshold segmentation respectively.
On the basis of such scheme, the defect recognition module is used for to be measured after image segmentation module is processed
Figure and standard drawing carry out differing from shadow computing, then carry out Morphological scale-space, finally carry out Classifcation of flaws.
Embodiment 1, the pcb board defect detecting system based on layered image and method
Fig. 1 shows a kind of pcb board defect detecting system structured flowchart based on layered image that the present embodiment is provided, should
Detecting system includes:Image capture module, image procossing rebuild module and PCB defect recognition modules, the specific works of each module
Step is as follows:
S1, the projected image for gathering PCB on-gauge plates respectively using the X-ray image acquisition device in image capture module and
The projected image of PCB boards under test;
S2, the projected image of the projected image of PCB on-gauge plates and PCB boards under test is rebuild into module by image procossing respectively
In filtered back projection FDK (Feldkamp-Davis-Kress) restructing algorithm of platy structure carry out Stratified Imaging, respectively
Layered image (hereinafter referred to as standard drawing) and the layered image of PCB boards under test to PCB on-gauge plates is (hereinafter referred to as to be measured
Figure);
S3, in PCB defect recognition modules, image preprocessing is carried out to standard drawing using image pre-processing module, then
Preserve standard drawing;PCB boards under test are detected, opening treats mapping, and treating mapping using image pre-processing module carries out image
Pretreatment, then by the way that the positioning loop truss in image registration module and geometric transformation are to standard drawing and treat that mapping carries out image and matches somebody with somebody
It is accurate;
After the completion of S4, image registration, standard drawing is separated by image segmentation module and the background and line pad of mapping is treated
Etc. feature, then by the way that defect recognition module is to standard drawing and treats that mapping makes the difference shadow computing;Then carry out Morphological scale-space filter it is non-
The noise of defect part obtains defect image, and the contour feature finally according to defect is identified to defect type, is detected
As a result.
Can not completely accomplish that pixel is corresponded and Threshold segmentation obtains imperfect during due to image registration, poor shadow will be caused
Binary map after conversion there may be non-defective part.If these parts can not be removed, flase drop can be caused, so also needing to
Morphological scale-space is carried out to filter noise.Last then to separating defect image carries out Classifcation of flaws.
1st, image preprocessing includes image denoising and image enhaucament.
Image denoising carries out the filtering noise reduction before image procossing using medium filtering, and filter window is big using 3 × 3 pixels
It is small, by the noise filtering in image.
Image enhaucament becomes gradation of image interval of changing commanders and is divided into two sections or even multistage by linear gradation, then makees linear respectively
Conversion, each straightway corresponds to a linear transformation relation for part.
2nd, image registration includes positioning loop truss and geometric transformation.
Feature point extraction is a crucial step in image registration, treats that mapping is generally positioned with matching for standard drawing using setting
Mark is solved, and designer often adds location hole on pcb board and conveniently produces and tested during pcb board is designed
Positioning in journey.
Standard drawing that step S2 is obtained and mapping is treated, image preprocessing is carried out respectively, carried out after the completion of image preprocessing
Image registration, as shown in Figure 2.During image registration, the position dimension of PCB on-gauge plates and PCB board under test upper Positioning holes is extracted respectively
Feature, then zoomed in and out in standard drawing and the geometric position difference for treating on mapping by location hole, translate and rotation process come
Realization treats that mapping is corresponded with the pixel of standard drawing.
The setting circle in PCB on-gauge plates and PCB boards under test is detected from random Hough transformation in the present invention.With
The process of machine Hough transform, is pre-processed by carrying out gray processing, denoising, rim detection and morphology operations etc. to image,
Image space randomly selects three not conllinear points and maps it onto a point of parameter space, constitutes more to one mapping,
Hough transform is carried out instead of cycle calculations by using Multidimensional numerical, random Hough transformation compared with traditional Hough transform,
Reduce memory requirements and avoid quantization parameter space, detection speed and accuracy rate can be greatly improved.To a certain layering of pcb board
The standard drawing of image and treat that mapping carries out the testing result of location hole and is illustrated in fig. 2 shown below.As can be seen from Figure 2 by random
Hough transform can be good at detecting the setting circle on pcb board.Existed by figure after the position dimension feature for extracting setting circle
Geometric transformation on standard drawing and target chart board is realized treating that mapping is corresponded with the pixel of standard drawing.Assuming that A (x1, y1), B
(x2, y2) it is two central coordinate of circle of setting circle of PCB on-gauge plates, A ' (x3, y3), B ' (x4, y4) for two of PCB boards under test it is fixed
The central coordinate of circle of circle of position.There is setting circle on PCB on-gauge plates and PCB boards under test, therefore treat that the size position deviation of mapping can be with
It is adjusted by geometric transformation, adjustment is concretely comprised the following steps:
(1) the level angle θ for treating mapping positioning centre is calculatedABWith the level angle θ ' of standard drawing positioning centreAB.Such as
Fruit θAB≠θ′AB, then treating mapping carries out rotation transformation, and the anglec of rotation is θAB-θ′AB(clockwise for just);
(2) calculate and treat the distance between two positioning centres of mapping LABThe distance between with two positioning centres of standard drawing
L′ABIf, LAB≠L′AB, then treat mapping and zoom in and out conversion;Scaling multiple is n=L 'AB/LAB;
(3) boundary position for treating mapping is calculated, and will treats that the target area of mapping extracts according to positional information;Carried
What is taken treats that mapping target area is with coordinate (x3-x1, y3-y1)、(x3-x1, y4-y2)、(x4-x2, y3-y1)、(x4-x2, y4-y2)
It is the rectangle of end points, mapping is as shown in Figure 3 for treating after geometric transformation.
3rd, image segmentation module
Image segmentation is carried out standard drawing and respectively after mapping carries out registration.Image segmentation process first carries out contrast increasing
By force, then white top cap conversion is carried out, image is split using the method for iteration threshold finally.
Before view data after to treatment carries out defect recognition, image segmentation is one of most important step, it be by
Image procossing is transitioned into the key of target identification, and its main target is to divide an image into target area and non-target area.When one
Width image contains more background, and background it is uneven when, foreground target can not preferably be split using global threshold.Such as
Fruit is converted by top cap and first removes background, and row threshold division is entered again after obtaining more uniform foreground target, can so be obtained more
Preferable segmentation result.White top cap conversion (White Top-Hat, WTH) of image is defined as original image f and its opening operation figure
The difference of picture γ (f), i.e.,:
WTH (f)=f- γ (f) (1)
Image is first carried out into contrast enhancing, then carries out white top cap conversion, finally using the method for iteration threshold to image
Split.Iteration threshold algorithm implementation process is as follows:
(1) maximum gradation value and minimum gradation value of image are obtained, g is designated as respectivelymaxAnd gmin, make initial threshold t0=
(gmax+gmin)/2;
(2) according to threshold value tiIt is foreground and background to divide the image into, and the average gray value g of prospect is obtained respectivelyAWith background
Average gray value gB;
(3) new threshold value t is obtainedi+1=(gA+gB)/2;
(4) if ti=ti+1, gained is threshold value;Otherwise turn (2), iterate to calculate.
It is more uniform by the image background after contrast enhancing and the conversion of white top cap, by the region of black artifact effects
This it appears that influence of the artifact to wire is substantially eliminated.Therefore white top cap can be utilized in the image segmentation to pcb board
Uneven image in the method elimination background of conversion, recycling the method for Threshold segmentation carries out image segmentation.
4th, defects detection
Defect recognition module is, to the image after segmentation, to find out the difference for treating mapping and standard drawing, maximally effective side
Method is exactly to treat mapping and standard drawing to carry out differing from shadow computing, and the image to difference movie queen carries out Morphological scale-space to eliminate non-defective
Partial noise obtains defect image, and the contour feature finally according to defect is judged defect type.First to defect map
The contour feature of picture carries out extracting the boundary profile segments N for obtaining defect image1.The origin of coordinates is chosen, the present invention is with defect
The upper left corner of image is origin, to defect image using 2 × 2 square structure element expansion process, carries out contour feature extraction,
The each point coordinate value on the profile border of acquisition is stored among a two-dimensional array B successively.Treat mapping and enter row threshold division,
Binary map is obtained, I is designated as0;Setting initial fragment number variable N=1, I is returned according to the coordinate value in array B0In, write down first
Point pixel value E1(0 or 1), judges boundary coordinate in I successively0The pixel value of middle corresponding points, if i-th pixel value is Ei(i=1,
2,3 ... .n), when pixel value is converted once, N=N+1, until terminating point, writes down terminating point pixel value for En.If N=1, N1
=N=1;If N ≠ 1 and En≠E1, then N1=N;If N ≠ 1 and En=E1, then N1=N-1.Boundary profile point according to defect image
Hop count N1, complete first step identification.If N1=4, then short circuit or open circuit are identified as, it is designated as 1 class defect;If N1=2, then it is identified as
Burr or defect, are designated as 2 class defects;If N1=1, then fifth wheel or cavity are identified as, it is designated as 3 class defects.Finally according to defect
It is that many material or scarce material determine defect type.To 1 class defect, if many material, are short circuit, on the contrary it is breaking;To 2 class defects, if many material,
It is then burr, on the contrary defect;To 3 class defects, if many material, are fifth wheel, on the contrary it is empty.In order to more intuitively by pcb board
Defect be presented to user, need to carry out accurate position and type mark to it after classifying defect.At present for table
The method of planar defect mark asks Minimum Enclosing Rectangle method, connected component labeling method etc., the present invention to pass through Minimum Enclosing Rectangle method
Rectangular centre is sought, then defect is marked, testing result is as shown in Figure 4.
The content not being described in detail in this specification belongs to prior art known to professional and technical personnel in the field.
Claims (13)
1. a kind of pcb board defect detecting system based on layered image, it is characterised in that including:At image capture module, image
Reason rebuilds module and PCB defect recognition modules;
Described image acquisition module includes X-ray image acquisition device, and projected image and PCB for gathering PCB on-gauge plates are treated
The projected image of drafting board;
Described image processing reconstructed module is used to be divided the projected image of the projected image of PCB on-gauge plates and PCB boards under test
Layer imaging, obtains standard drawing and treats mapping;
The PCB defect recognitions module includes:Image pre-processing module, image registration module, image segmentation module and defect are known
Other module;For Treatment Analysis standard drawing and treating mapping, so as to carry out defect recognition to PCB boards under test, and testing result is exported.
2. the pcb board defect detecting system of layered image is based on as claimed in claim 1, it is characterised in that described image is pre-
Processing module is used for standard drawing and treats that mapping carries out filtering noise processed and image enhancement processing respectively.
3. the pcb board defect detecting system of layered image is based on as claimed in claim 1, it is characterised in that described image is matched somebody with somebody
Quasi-mode block is used to extract the position dimension feature of PCB on-gauge plates and PCB board under test upper Positioning holes, then by location hole in standard
Scheme and treat that the geometric position difference on mapping is zoomed in and out, translated and rotation process, the pixel one of mapping and standard drawing is treated in realization
One correspondence.
4. the pcb board defect detecting system of layered image is based on as claimed in claim 1, it is characterised in that described image point
Cut module for by the standard drawing after image registration resume module and treat mapping carry out respectively contrast enhancing, white top cap become
Change and iterative threshold segmentation.
5. the pcb board defect detecting system of layered image is based on as claimed in claim 1, it is characterised in that the defect is known
Then other module carries out morphology for mapping and standard drawing to carry out differing from shadow computing to treating after image segmentation module is processed
Treatment, finally carries out Classifcation of flaws.
6. a kind of pcb board defect inspection method based on layered image, using the base described in claim 1-5 any claims
In the pcb board defect detecting system of layered image, it is characterised in that comprise the following steps:
The projected image of S1, the projected image for gathering PCB on-gauge plates respectively using X-ray image acquisition device and PCB boards under test;
S2, by the projected image of the projected image of PCB standard editions and PCB boards under test respectively by image procossing rebuild module in
The filtered back projection FDK restructing algorithms of platy structure carry out Stratified Imaging, respectively obtain standard drawing and treat mapping;
S3, image preprocessing is carried out to standard drawing, then preserve standard drawing;PCB boards under test are detected, opening treats mapping,
Treating mapping carries out image preprocessing, then to standard drawing and treating that mapping carries out image registration;
After the completion of S4, image registration, standard drawing is isolated by image segmentation and the background and line pad feature of mapping is treated, then
To standard drawing and treating that mapping makes the difference shadow computing;Then Morphological scale-space is carried out and filters the noise of non-defective part to obtain defect map
Picture, the contour feature finally according to defect is identified to defect type, obtains testing result.
7. the pcb board defect inspection method of layered image is based on as claimed in claim 6, it is characterised in that described image is pre-
Treatment includes image denoising and image enhaucament;Described image denoising carries out the filtering noise reduction before image procossing using medium filtering,
Filter window uses 3 × 3 pixel sizes;
Described image enhancing becomes gradation of image interval of changing commanders and is divided into two sections or even multistage by linear gradation, then makees linear respectively
Conversion, each straightway corresponds to a linear transformation relation for part.
8. the pcb board defect inspection method of layered image is based on as claimed in claim 6, it is characterised in that described image is matched somebody with somebody
Standard includes positioning loop truss and geometric transformation;Described image registration process be:
The setting circle of PCB on-gauge plates and the setting circle of PCB boards under test are detected using random Hough transformation, extracts PCB marks
The position dimension feature of quasi- plate setting circle and PCB board under test setting circles, then by the way that location hole is in standard drawing and treats on mapping
The size position deviation that mapping is treated in geometric transformation is adjusted.
9. the pcb board defect inspection method of layered image is based on as claimed in claim 8, it is characterised in that the adjustment
Concretely comprise the following steps:
Assuming that A (x1, y1), B (x2, y2) it is two central coordinate of circle of setting circle of PCB on-gauge plates, A ' (x3, y3), B ' (x4, y4) be
Two central coordinate of circle of setting circle of PCB boards under test,
(1) the level angle θ for treating mapping positioning centre is calculatedABWith the level angle θ ' of standard drawing positioning centreABIf, θAB
≠θ′AB, then treating mapping carries out rotation transformation, and the anglec of rotation is θAB-θ′AB;
(2) calculate and treat the distance between two positioning centres of mapping LABWith the distance between two positioning centres of standard drawing L 'AB,
If LAB≠L′AB, then treat mapping and zoom in and out conversion;Scaling multiple is n=L 'AB/LAB;
(3) boundary position for treating mapping is calculated, and will treats that the target area of mapping extracts according to boundary position information, carried
What is taken treats that mapping target area is with coordinate (x3-x1, y3-y1)、(x3-x1, y4-y2)、(x4-x2, y3-y1)、(x4-x2, y4-y2)
It is the rectangle of end points.
10. the pcb board defect inspection method of layered image is based on as claimed in claim 6, it is characterised in that described image point
The detailed process cut is:Contrast enhancing is first carried out, then carries out white top cap conversion, finally carry out image using iteration threshold method
Segmentation;The process that implements of the iteration threshold method is:
(1) maximum gradation value and minimum gradation value of image are obtained, g is designated as respectivelymaxAnd gmin, make initial threshold t0=(gmax+
gmin)/2;
(2) according to threshold value tiIt is foreground and background to divide the image into, and the average gray value g of prospect is obtained respectivelyAIt is average with background
Gray value gB;
(3) new threshold value t is obtainedi+1=(gA+gB)/2;
(4) if ti=ti+1, gained is threshold value;Otherwise turn (2), iterate to calculate.
The 11. pcb board defect inspection methods based on layered image as claimed in claim 6, it is characterised in that the defect class
The identification process of type is:
(1) contour feature first to defect image carries out extracting the boundary profile segments N for obtaining defect image1;
(2) and then according to the boundary profile segments N of defect image1Carry out first step identification;
If N1=4, then short circuit or open circuit are identified as, it is designated as 1 class defect;
If N1=2, then burr or defect are identified as, it is designated as 2 class defects;
If N1=1, then fifth wheel or cavity are identified as, it is designated as 3 class defects;
(3) it is that many material or scarce material determine defect type finally according to defect;
Then it is open circuit if lacking material if many material, are short circuit to 1 class defect;
Then it is defect if lacking material if many material, are burr to 2 class defects;
Then it is cavity if lacking material if many material, are fifth wheel to 3 class defects;
(4) in order to the defect on PCB boards under test more intuitively is presented into user, needed after classifying to defect type to lacking
It is trapped into going accurate position mark and type mark.
The 12. pcb board defect inspection methods based on layered image as claimed in claim 11, it is characterised in that the acquisition
The boundary profile segments N of defect image1Detailed process be:
(1) origin of coordinates is chosen, the upper left corner with defect image uses defect image 2 × 2 square structure element as origin
Expansion process, carries out contour feature extraction;
(2) each point coordinate value on the profile border of acquisition is stored among a two-dimensional array B successively;
(3) treat mapping and enter row threshold division, obtain binary map, be designated as I0;
(4) initial fragment number variable N=1 is set, I is returned according to the coordinate value in two-dimensional array B0In, write down first pixel value
E1, then judge boundary coordinate in I successively0The pixel value of middle corresponding points, if i-th pixel value is Ei, when pixel value converts one
It is secondary, N=N+1, until terminating point, writes down terminating point pixel value for En;
If N=1, N1=N=1;
If N ≠ 1 and En≠E1, then N1=N;
If N ≠ 1 and En=E1, then N1=N-1.
The 13. pcb board defect inspection methods based on layered image as claimed in claim 11, it is characterised in that the defect
Mark ask the method for rectangular centre to be marked using Minimum Enclosing Rectangle method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710045720.3A CN106897994A (en) | 2017-01-20 | 2017-01-20 | A kind of pcb board defect detecting system and method based on layered image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710045720.3A CN106897994A (en) | 2017-01-20 | 2017-01-20 | A kind of pcb board defect detecting system and method based on layered image |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106897994A true CN106897994A (en) | 2017-06-27 |
Family
ID=59198935
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710045720.3A Pending CN106897994A (en) | 2017-01-20 | 2017-01-20 | A kind of pcb board defect detecting system and method based on layered image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106897994A (en) |
Cited By (41)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107966448A (en) * | 2017-11-17 | 2018-04-27 | 福建工程学院 | A kind of 2 dimension detection methods for PCB paste solder printing quality |
CN108279241A (en) * | 2017-10-20 | 2018-07-13 | 同济大学 | A kind of workpiece configurations detection method based on machine vision |
CN108389184A (en) * | 2017-10-20 | 2018-08-10 | 同济大学 | A kind of workpiece drilling number detection method based on machine vision |
CN108648168A (en) * | 2018-03-15 | 2018-10-12 | 北京京仪仪器仪表研究总院有限公司 | IC wafer surface defects detection methods |
CN109087286A (en) * | 2018-07-17 | 2018-12-25 | 江西财经大学 | A kind of detection method and application based on Computer Image Processing and pattern-recognition |
CN109242853A (en) * | 2018-11-05 | 2019-01-18 | 南京信息工程大学 | A kind of PCB defect intelligent detecting method based on image procossing |
CN109584215A (en) * | 2018-11-10 | 2019-04-05 | 东莞理工学院 | A kind of online vision detection system of circuit board |
CN109632815A (en) * | 2017-10-09 | 2019-04-16 | 上海明匠智能系统有限公司 | Surface defect on-line detecting system and method |
CN109785294A (en) * | 2018-12-24 | 2019-05-21 | 苏州江奥光电科技有限公司 | A kind of pcb board defective locations detection system and method |
CN109813727A (en) * | 2018-12-25 | 2019-05-28 | 苏州江奥光电科技有限公司 | A kind of pcb board weld defects detection method based on depth information |
CN109829886A (en) * | 2018-12-25 | 2019-05-31 | 苏州江奥光电科技有限公司 | A kind of pcb board defect inspection method based on depth information |
CN109859192A (en) * | 2019-02-01 | 2019-06-07 | 东莞中科蓝海智能视觉科技有限公司 | The detection recognition method of round high reflective workpiece |
CN109871895A (en) * | 2019-02-22 | 2019-06-11 | 北京百度网讯科技有限公司 | The defect inspection method and device of circuit board |
CN109961464A (en) * | 2019-03-22 | 2019-07-02 | 武汉艾维视智能技术有限公司 | A kind of high-precision screen printing net plate scan image method for registering |
CN110335233A (en) * | 2019-04-24 | 2019-10-15 | 武汉理工大学 | Express-way guard-rail plates defect detecting system and method based on image processing techniques |
CN110346704A (en) * | 2019-07-02 | 2019-10-18 | 广州视源电子科技股份有限公司 | Determination method, apparatus, equipment and the storage medium of test file in board test |
CN110351951A (en) * | 2019-07-05 | 2019-10-18 | 信泰电子(西安)有限公司 | The selecting system and method for location hole when PCB circuit board sharp processing |
CN110443777A (en) * | 2018-10-09 | 2019-11-12 | 广东仙童智能机器人科技有限公司 | A kind of PCB image alignment method, device and computer storage medium |
CN110473184A (en) * | 2019-08-06 | 2019-11-19 | 福建工程学院 | A kind of pcb board defect inspection method |
WO2019227762A1 (en) * | 2018-05-31 | 2019-12-05 | 歌尔股份有限公司 | Method, device and apparatus for detecting pixel defect of optical module |
CN110579479A (en) * | 2019-08-09 | 2019-12-17 | 康代影像科技(苏州)有限公司 | PCB maintenance system and maintenance method based on false point defect detection |
CN110599470A (en) * | 2019-08-30 | 2019-12-20 | 武汉科技大学 | Magnetic shoe surface defect detection system and method |
CN110838135A (en) * | 2019-10-31 | 2020-02-25 | 上海工程技术大学 | PCB bare board image registration method |
CN111080615A (en) * | 2019-12-12 | 2020-04-28 | 创新奇智(重庆)科技有限公司 | PCB defect detection system and detection method based on convolutional neural network |
CN111812545A (en) * | 2020-07-07 | 2020-10-23 | 苏州精濑光电有限公司 | Line defect detection method, device, equipment and medium |
CN111929239A (en) * | 2020-07-20 | 2020-11-13 | 浙江四点灵机器人股份有限公司 | AOI detection device and detection method for PCB part defects |
WO2021000524A1 (en) * | 2019-07-03 | 2021-01-07 | 研祥智能科技股份有限公司 | Hole protection cap detection method and apparatus, computer device and storage medium |
CN112200798A (en) * | 2020-10-29 | 2021-01-08 | 佛山市南海区广工大数控装备协同创新研究院 | Printed circuit board detection method based on X-ray layering technology |
CN112288782A (en) * | 2020-10-29 | 2021-01-29 | 佛山市南海区广工大数控装备协同创新研究院 | PCB defect detection method based on boundary segment number value |
CN113096119A (en) * | 2021-04-30 | 2021-07-09 | 上海众壹云计算科技有限公司 | Method and device for classifying wafer defects, electronic equipment and storage medium |
CN113129368A (en) * | 2020-01-15 | 2021-07-16 | 惠州市成泰自动化科技有限公司 | PCB direction identification method |
CN113192027A (en) * | 2021-04-29 | 2021-07-30 | 华南理工大学 | Detection method and application of high-power LED module packaging defects |
CN113379726A (en) * | 2021-06-30 | 2021-09-10 | 深圳中科飞测科技股份有限公司 | Line detection method, device, equipment and computer readable storage medium |
CN113628176A (en) * | 2021-07-28 | 2021-11-09 | 生益电子股份有限公司 | Conductive medium detection method, device, equipment and storage medium of PCB |
CN113781419A (en) * | 2021-08-31 | 2021-12-10 | 广州大学 | Defect detection method, visual system, device and medium for flexible PCB |
CN114998314A (en) * | 2022-07-18 | 2022-09-02 | 启东旺晟电子科技有限公司 | PCB (printed Circuit Board) defect detection method based on computer vision |
CN115141563A (en) * | 2022-08-15 | 2022-10-04 | 常州博疆新材料科技有限公司 | Full-biodegradable adhesive tape and preparation method thereof |
CN116168040A (en) * | 2023-04-26 | 2023-05-26 | 四川元智谷科技有限公司 | Component direction detection method and device, electronic equipment and readable storage medium |
CN116168021A (en) * | 2023-04-21 | 2023-05-26 | 中江立江电子有限公司 | Fault part identification system and method |
CN116433666A (en) * | 2023-06-14 | 2023-07-14 | 江西萤火虫微电子科技有限公司 | Board card line defect online identification method, system, electronic equipment and storage medium |
CN117474977A (en) * | 2023-12-27 | 2024-01-30 | 山东旭美尚诺装饰材料有限公司 | Quick detection method and system for European pine plate pits based on machine vision |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN201034949Y (en) * | 2007-02-05 | 2008-03-12 | 华南理工大学 | X-ray detection device of printed circuit board |
CN102288858A (en) * | 2011-05-19 | 2011-12-21 | 昆山鑫立电子科技有限公司 | Detecting device for printed circuit board |
CN102721695A (en) * | 2012-05-18 | 2012-10-10 | 深圳大学 | Method for detecting printed circuit board defect |
CN202854066U (en) * | 2012-10-19 | 2013-04-03 | 东莞凯佳智芯电子科技有限公司 | Device for detecting multilayer PCB (Printed Circuit Board) by using X-ray technology |
CN104764712A (en) * | 2015-04-29 | 2015-07-08 | 浙江工业大学 | Method for detecting quality of inner wall of via hole of PCB |
-
2017
- 2017-01-20 CN CN201710045720.3A patent/CN106897994A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN201034949Y (en) * | 2007-02-05 | 2008-03-12 | 华南理工大学 | X-ray detection device of printed circuit board |
CN102288858A (en) * | 2011-05-19 | 2011-12-21 | 昆山鑫立电子科技有限公司 | Detecting device for printed circuit board |
CN102721695A (en) * | 2012-05-18 | 2012-10-10 | 深圳大学 | Method for detecting printed circuit board defect |
CN202854066U (en) * | 2012-10-19 | 2013-04-03 | 东莞凯佳智芯电子科技有限公司 | Device for detecting multilayer PCB (Printed Circuit Board) by using X-ray technology |
CN104764712A (en) * | 2015-04-29 | 2015-07-08 | 浙江工业大学 | Method for detecting quality of inner wall of via hole of PCB |
Non-Patent Citations (6)
Title |
---|
刘西锋 等: "基于层析成像技术的PCB板缺陷检测系统", 《机械工程与自动化》 * |
刘阳: "基于图像处理的PCB焊接缺陷检测技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
杨庆华 等: "基于机械视觉的PCB裸板缺陷自动检测方法", 《中国机械工程》 * |
柳林: "《数字图像处理实验指导教程》", 31 January 2014, 浙江大学出版社 * |
段大高 王建勇: "《图像处理与应用》", 31 January 2013, 北京邮电大学出版社 * |
胡秀亮 等: "基于视觉图像处理技术的PCB缺陷检测自动检测系统", 《微型机与应用》 * |
Cited By (59)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109632815A (en) * | 2017-10-09 | 2019-04-16 | 上海明匠智能系统有限公司 | Surface defect on-line detecting system and method |
CN108279241A (en) * | 2017-10-20 | 2018-07-13 | 同济大学 | A kind of workpiece configurations detection method based on machine vision |
CN108389184A (en) * | 2017-10-20 | 2018-08-10 | 同济大学 | A kind of workpiece drilling number detection method based on machine vision |
CN107966448A (en) * | 2017-11-17 | 2018-04-27 | 福建工程学院 | A kind of 2 dimension detection methods for PCB paste solder printing quality |
CN108648168A (en) * | 2018-03-15 | 2018-10-12 | 北京京仪仪器仪表研究总院有限公司 | IC wafer surface defects detection methods |
US11587219B2 (en) | 2018-05-31 | 2023-02-21 | Goertek Inc. | Method and apparatus for detecting pixel defect of optical module, and device |
WO2019227762A1 (en) * | 2018-05-31 | 2019-12-05 | 歌尔股份有限公司 | Method, device and apparatus for detecting pixel defect of optical module |
CN109087286A (en) * | 2018-07-17 | 2018-12-25 | 江西财经大学 | A kind of detection method and application based on Computer Image Processing and pattern-recognition |
CN110443777B (en) * | 2018-10-09 | 2023-03-10 | 广东仙童智能机器人科技有限公司 | PCB image alignment method and device and computer storage medium |
CN110443777A (en) * | 2018-10-09 | 2019-11-12 | 广东仙童智能机器人科技有限公司 | A kind of PCB image alignment method, device and computer storage medium |
CN109242853A (en) * | 2018-11-05 | 2019-01-18 | 南京信息工程大学 | A kind of PCB defect intelligent detecting method based on image procossing |
CN109242853B (en) * | 2018-11-05 | 2021-10-19 | 南京信息工程大学 | PCB defect intelligent detection method based on image processing |
CN109584215A (en) * | 2018-11-10 | 2019-04-05 | 东莞理工学院 | A kind of online vision detection system of circuit board |
CN109785294A (en) * | 2018-12-24 | 2019-05-21 | 苏州江奥光电科技有限公司 | A kind of pcb board defective locations detection system and method |
CN109829886A (en) * | 2018-12-25 | 2019-05-31 | 苏州江奥光电科技有限公司 | A kind of pcb board defect inspection method based on depth information |
CN109813727B (en) * | 2018-12-25 | 2021-08-03 | 苏州江奥光电科技有限公司 | PCB welding defect detection method based on depth information |
CN109813727A (en) * | 2018-12-25 | 2019-05-28 | 苏州江奥光电科技有限公司 | A kind of pcb board weld defects detection method based on depth information |
CN109859192A (en) * | 2019-02-01 | 2019-06-07 | 东莞中科蓝海智能视觉科技有限公司 | The detection recognition method of round high reflective workpiece |
CN109871895B (en) * | 2019-02-22 | 2021-03-16 | 北京百度网讯科技有限公司 | Method and device for detecting defects of circuit board |
CN109871895A (en) * | 2019-02-22 | 2019-06-11 | 北京百度网讯科技有限公司 | The defect inspection method and device of circuit board |
CN109961464A (en) * | 2019-03-22 | 2019-07-02 | 武汉艾维视智能技术有限公司 | A kind of high-precision screen printing net plate scan image method for registering |
CN110335233A (en) * | 2019-04-24 | 2019-10-15 | 武汉理工大学 | Express-way guard-rail plates defect detecting system and method based on image processing techniques |
CN110335233B (en) * | 2019-04-24 | 2023-06-30 | 武汉理工大学 | Highway guardrail plate defect detection system and method based on image processing technology |
CN110346704B (en) * | 2019-07-02 | 2022-05-20 | 广州视源电子科技股份有限公司 | Method, device and equipment for determining test file in board test and storage medium |
CN110346704A (en) * | 2019-07-02 | 2019-10-18 | 广州视源电子科技股份有限公司 | Determination method, apparatus, equipment and the storage medium of test file in board test |
WO2021000524A1 (en) * | 2019-07-03 | 2021-01-07 | 研祥智能科技股份有限公司 | Hole protection cap detection method and apparatus, computer device and storage medium |
CN110351951A (en) * | 2019-07-05 | 2019-10-18 | 信泰电子(西安)有限公司 | The selecting system and method for location hole when PCB circuit board sharp processing |
CN110351951B (en) * | 2019-07-05 | 2021-02-05 | 信泰电子(西安)有限公司 | System and method for selecting positioning hole during PCB contour machining |
CN110473184A (en) * | 2019-08-06 | 2019-11-19 | 福建工程学院 | A kind of pcb board defect inspection method |
CN110579479A (en) * | 2019-08-09 | 2019-12-17 | 康代影像科技(苏州)有限公司 | PCB maintenance system and maintenance method based on false point defect detection |
WO2021027184A1 (en) * | 2019-08-09 | 2021-02-18 | 苏州康代智能科技股份有限公司 | Pcb maintenance system and maintenance method based on false point defect detection |
CN110599470B (en) * | 2019-08-30 | 2022-04-08 | 武汉科技大学 | Magnetic shoe surface defect detection system and method |
CN110599470A (en) * | 2019-08-30 | 2019-12-20 | 武汉科技大学 | Magnetic shoe surface defect detection system and method |
CN110838135A (en) * | 2019-10-31 | 2020-02-25 | 上海工程技术大学 | PCB bare board image registration method |
CN111080615B (en) * | 2019-12-12 | 2023-06-16 | 创新奇智(重庆)科技有限公司 | PCB defect detection system and detection method based on convolutional neural network |
CN111080615A (en) * | 2019-12-12 | 2020-04-28 | 创新奇智(重庆)科技有限公司 | PCB defect detection system and detection method based on convolutional neural network |
CN113129368A (en) * | 2020-01-15 | 2021-07-16 | 惠州市成泰自动化科技有限公司 | PCB direction identification method |
WO2021143295A1 (en) * | 2020-01-15 | 2021-07-22 | 惠州市成泰自动化科技有限公司 | Method for recognizing direction of pcb |
CN111812545A (en) * | 2020-07-07 | 2020-10-23 | 苏州精濑光电有限公司 | Line defect detection method, device, equipment and medium |
CN111929239A (en) * | 2020-07-20 | 2020-11-13 | 浙江四点灵机器人股份有限公司 | AOI detection device and detection method for PCB part defects |
CN112200798A (en) * | 2020-10-29 | 2021-01-08 | 佛山市南海区广工大数控装备协同创新研究院 | Printed circuit board detection method based on X-ray layering technology |
CN112200798B (en) * | 2020-10-29 | 2024-02-02 | 佛山市南海区广工大数控装备协同创新研究院 | Printed circuit board detection method based on X-ray layering technology |
CN112288782A (en) * | 2020-10-29 | 2021-01-29 | 佛山市南海区广工大数控装备协同创新研究院 | PCB defect detection method based on boundary segment number value |
CN113192027B (en) * | 2021-04-29 | 2023-03-24 | 华南理工大学 | Detection method and application of high-power LED module packaging defects |
CN113192027A (en) * | 2021-04-29 | 2021-07-30 | 华南理工大学 | Detection method and application of high-power LED module packaging defects |
CN113096119A (en) * | 2021-04-30 | 2021-07-09 | 上海众壹云计算科技有限公司 | Method and device for classifying wafer defects, electronic equipment and storage medium |
CN113096119B (en) * | 2021-04-30 | 2024-06-07 | 上海众壹云计算科技有限公司 | Method and device for classifying wafer defects, electronic equipment and storage medium |
CN113379726A (en) * | 2021-06-30 | 2021-09-10 | 深圳中科飞测科技股份有限公司 | Line detection method, device, equipment and computer readable storage medium |
CN113628176A (en) * | 2021-07-28 | 2021-11-09 | 生益电子股份有限公司 | Conductive medium detection method, device, equipment and storage medium of PCB |
CN113781419A (en) * | 2021-08-31 | 2021-12-10 | 广州大学 | Defect detection method, visual system, device and medium for flexible PCB |
CN114998314A (en) * | 2022-07-18 | 2022-09-02 | 启东旺晟电子科技有限公司 | PCB (printed Circuit Board) defect detection method based on computer vision |
CN115141563A (en) * | 2022-08-15 | 2022-10-04 | 常州博疆新材料科技有限公司 | Full-biodegradable adhesive tape and preparation method thereof |
CN116168021A (en) * | 2023-04-21 | 2023-05-26 | 中江立江电子有限公司 | Fault part identification system and method |
CN116168021B (en) * | 2023-04-21 | 2023-08-29 | 中江立江电子有限公司 | Fault part identification system and method |
CN116168040A (en) * | 2023-04-26 | 2023-05-26 | 四川元智谷科技有限公司 | Component direction detection method and device, electronic equipment and readable storage medium |
CN116433666B (en) * | 2023-06-14 | 2023-08-15 | 江西萤火虫微电子科技有限公司 | Board card line defect online identification method, system, electronic equipment and storage medium |
CN116433666A (en) * | 2023-06-14 | 2023-07-14 | 江西萤火虫微电子科技有限公司 | Board card line defect online identification method, system, electronic equipment and storage medium |
CN117474977A (en) * | 2023-12-27 | 2024-01-30 | 山东旭美尚诺装饰材料有限公司 | Quick detection method and system for European pine plate pits based on machine vision |
CN117474977B (en) * | 2023-12-27 | 2024-03-22 | 山东旭美尚诺装饰材料有限公司 | Quick detection method and system for European pine plate pits based on machine vision |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106897994A (en) | A kind of pcb board defect detecting system and method based on layered image | |
CN109961049B (en) | Cigarette brand identification method under complex scene | |
CN107389701A (en) | A kind of PCB visual defects automatic checkout system and method based on image | |
CN107966454A (en) | A kind of end plug defect detecting device and detection method based on FPGA | |
CN112686858A (en) | Visual defect detection method, device, medium and equipment for mobile phone charger | |
CN102567979A (en) | Vehicle-mounted infrared night vision system and multi-source images fusing method thereof | |
CN109087286A (en) | A kind of detection method and application based on Computer Image Processing and pattern-recognition | |
CN108564561A (en) | Pectoralis major region automatic testing method in a kind of molybdenum target image | |
CN109584215A (en) | A kind of online vision detection system of circuit board | |
CN110261391A (en) | A kind of LED chip appearance detection system and method | |
CN113240623B (en) | Pavement disease detection method and device | |
CN111695373B (en) | Zebra stripes positioning method, system, medium and equipment | |
CN111967288A (en) | Intelligent three-dimensional object identification and positioning system and method | |
CN113034474A (en) | Test method for wafer map of OLED display | |
CN113516619A (en) | Product surface flaw identification method based on image processing technology | |
CN107507130A (en) | A kind of quickly QFN chip pins image obtains and amplification method | |
US7231086B2 (en) | Knowledge-based hierarchical method for detecting regions of interest | |
CN108985294B (en) | Method, device and equipment for positioning tire mold picture and storage medium | |
CN110866917A (en) | Tablet type and arrangement mode identification method based on machine vision | |
CN110264463A (en) | A kind of material counting method based on matlab image procossing | |
CN116524315A (en) | Mask R-CNN-based lung cancer pathological tissue section identification and segmentation method | |
CN113989588A (en) | Self-learning-based intelligent evaluation system and method for pentagonal drawing test | |
JP2023535005A (en) | Method and system or apparatus for recognizing objects in electronic images | |
Junhong et al. | On-line inspection system for finished circuit board based on machine vision | |
CN117974601B (en) | Method and system for detecting surface defects of silicon wafer based on template matching |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170627 |
|
RJ01 | Rejection of invention patent application after publication |