CN108418553A - Validity feature extracting method applied to the black angle detection of cell piece - Google Patents
Validity feature extracting method applied to the black angle detection of cell piece Download PDFInfo
- Publication number
- CN108418553A CN108418553A CN201810231844.5A CN201810231844A CN108418553A CN 108418553 A CN108418553 A CN 108418553A CN 201810231844 A CN201810231844 A CN 201810231844A CN 108418553 A CN108418553 A CN 108418553A
- Authority
- CN
- China
- Prior art keywords
- image
- black
- curve
- cell piece
- angle
- 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
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000001514 detection method Methods 0.000 title claims abstract description 17
- 230000007547 defect Effects 0.000 claims abstract description 22
- 238000000605 extraction Methods 0.000 claims abstract description 10
- 239000000284 extract Substances 0.000 claims abstract description 6
- 238000009826 distribution Methods 0.000 claims abstract description 5
- 238000012545 processing Methods 0.000 claims description 5
- 238000009499 grossing Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 13
- 238000003066 decision tree Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 239000000969 carrier Substances 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000003749 cleanliness Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 239000013078 crystal Substances 0.000 description 1
- 229910021419 crystalline silicon Inorganic materials 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 238000005401 electroluminescence Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000004020 luminiscence type Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 229910052710 silicon Inorganic materials 0.000 description 1
- 239000010703 silicon Substances 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000010792 warming Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02S—GENERATION OF ELECTRIC POWER BY CONVERSION OF INFRARED RADIATION, VISIBLE LIGHT OR ULTRAVIOLET LIGHT, e.g. USING PHOTOVOLTAIC [PV] MODULES
- H02S50/00—Monitoring or testing of PV systems, e.g. load balancing or fault identification
- H02S50/10—Testing of PV devices, e.g. of PV modules or single PV cells
- H02S50/15—Testing of PV devices, e.g. of PV modules or single PV cells using optical means, e.g. using electroluminescence
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/50—Photovoltaic [PV] energy
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The present invention is applied to the validity feature extracting method of the black angle detection of cell piece, by carrying out after rejecting other interference and method line projection, carries out indicatrix extraction；Indicatrix extraction step is as follows：F1）It is extracted average value, standard variance, greatest gradient, the features such as minimal gradient herein from curve；F2）The fitting that straight line is carried out to curve, extracts angle and two features of deviation of straight line；F3）Carry out the distribution displaying in average gray and Gradient Features space；F4）Feature is analyzed, judges whether black angular defect.
Description
Technical field
The present invention relates to solar battery sheet preparation processes, specifically, its displaying is a kind of to be applied to the black angle detection of cell piece
Validity feature extracting method.
Background technology
The continuous development of society, the demand to the energy are growing on and on, and non-renewable energy resources is promoted to peter out.Simultaneously because
The mankind cause environmental problem constantly to deteriorate the unreasonable use of the energy, and wherein global warming is particularly problematic,
Seriously threaten the survival and development of the mankind.Solar energy power generating due to cleanliness without any pollution, safe and reliable, easy for installation, and
It can be combined well with building, have become one of the important directions of development new energy at present.Convert solar energy into electric energy
Solar photovoltaic assembly is needed, photovoltaic module is made of numerous solar battery sheets, and solar battery sheet is in production process
In, defect is inevitably caused, this will seriously affect the luminous efficiency and service life of solar battery sheet, it is therefore necessary to
Defects detection is carried out to solar battery sheet, and rejects and contains defective cell piece.
According to the difference of solar battery sheet making material, monocrystalline, polycrystalline and non-crystalline silicon can be divided into.What is be discussed herein is
Cell piece is two kinds of front, i.e. single crystal battery piece and polycrystalline cell piece.The method for detecting internal flaw comparative maturity is EL
（Electroluminescence）.That is battery lighting defects detection.
The test philosophy of EL such as Fig. 1, the additional forward bias voltage of crystal-silicon solar cell, power supply are injected to solar cell
A large amount of nonequilibrium carriers, electroluminescent rely on a large amount of nonequilibrium carriers injected from diffusion region constantly recombination luminescence, put
Go out photon；It recycles CCD camera to capture these photons, is shown after being handled by computer, entire test
Journey is carried out in darkroom.
" black angle " defect be wherein there are one angle or more than one is apparent partially darker than other positions in four angles of cell piece,
And it slowly brightens from outside to inside.If Fig. 3 is a typical black angular defect, wherein upper left and upper right corner ratio other places are apparent partially
Secretly.The position at black angle can influence luminous efficiency, so must detected.
Black angle detection at this stage, is to pass through Blob mostly（Block）Method.Pass through direct binaryzation, dynamic two-value first
Change method extracts the block at black angle（blob）, feature extraction then is carried out to blob again, finally by decision tree or grader
Method distinguishes normal and black angle.The shortcomings that the method is that some black angles are not obvious, it is difficult to be partitioned into complete black corner
Point, to cause missing inspection or flase drop.
It solves the above problems applied to the validity feature extracting method of the black angle detection of cell piece therefore, it is necessary to provide.
Invention content
The object of the present invention is to provide a kind of validity feature extracting methods applied to the black angle detection of cell piece.
Technical solution is as follows：
A kind of validity feature extracting method applied to the black angle detection of cell piece rejects other interference and normal throwing by carrying out
Movie queen carries out indicatrix extraction；
Indicatrix extraction step is as follows：
F1）It is extracted average value, standard variance, greatest gradient, the features such as minimal gradient herein from curve；
F2）The fitting that straight line is carried out to curve, extracts angle and two features of deviation of straight line；
F3）Carry out the distribution displaying in average gray and Gradient Features space；
F4）Feature is analyzed, judges whether black angular defect.
Further, rejecting other interference implementation causes is：The defects of other classifications and black；If do not rejected,
It can cause to contain other defect information in the curve of projection, cause the signal at black angle that can be weakened；The specific steps are：
B1) using the window of a 15x15, mean value smoothing processing is carried out to image with mask；
B2) treated, and smoothed image is compared as background image, original image and background image pixels point；
B3) step B2）In, dark 35% pixel is other defect in original image, must not participate in annular projection；
B4）Last and mask images combine, and generate the image that may finally be projected.
Further, method line projection is as follows：
D1）Take the angle of end four；
D2）Each angle is projected along 45 degree or 135 degree to progress method direction inside image；
D3）Normalization：Each effective pixel in method direction is not fully identical, so need to use valid pixel number, it is right
Curve is normalized；
D4）Two curves of upper left and upper right are below, and slowly walk upwards；And the company curve of lower-left and bottom right is upper
Face, and without apparent rule；Reflect that the content of image can be reflected in method line projection.
Compared with prior art, the present invention is protected by carrying out indicatrix extraction after rejecting other interference and method line projection
Demonstrate,prove the accuracy of detection of the black angle detection of cell piece.
Description of the drawings
Fig. 1 is EL testing principle schematic diagrames；
Fig. 2 is the flow diagram of the present invention；
Fig. 3 is canny processing schematic diagrames；
Fig. 4 is edge point search schematic diagram；
Fig. 5 is fitting a straight line schematic diagram；
Fig. 6 is cover plate image schematic diagram；
Fig. 7 is original image and background image pixels point contrast schematic diagram；
Fig. 8 is the image schematic diagram that may finally be projected；
Fig. 9 is projection theory schematic diagram；
Figure 10 is projection step schematic diagram；
Figure 11 is drop shadow curve's schematic diagram；
Figure 12 is average gray and the distribution schematic diagram in Gradient Features space；
Figure 13 is decision tree schematic diagram.
Specific implementation mode
Embodiment：
Referring to Fig. 2, the present embodiment shows a kind of black angle detection method of cell piece, cell piece is positioned first, by battery
Piece center and picture centre overlap, convenient for projecting and covering mask（Cover plate）Image.In order to make the curve after projection that can more embody
The feature at black angle needs to mask other defect, i.e., does not participate in projection.Then started with the fixed point at angle, along 45 degree or 135 degree
Internally carry out method direction projection.Two-dimensional signal is changed into one-dimensional curve to analyze；Feature is carried out then for curve
Extraction.Ultimate analysis feature judges whether black angular defect.
It is as follows：
A kind of black angle detection method of cell piece, includes the following steps：
S1 it) is aligned：Cell piece is moved on into picture centre；
S2）Reject other interference；
S3）Method line projection；
S4）Indicatrix extracts；
S5）Classification；
Wherein：
S1）It is as follows：
A1 canny processing) is carried out to image, as a result such as Fig. 3；
A2) divide four sides internally to be searched for from periphery, find marginal point（That is non-zero points on Canny figures）, find and stop, as a result such as
Fig. 4；
A3 after) having looked for marginal point, fitting a straight line is carried out to four edges respectively, as a result such as Fig. 5；
A4 four intersection points of quadrangle) are found out；Four intersection points are averaging to get to the center of cell piece again；
A5 angle) is obtained by the mean value calculation of four straight lines；There are a center and angle, then by geometric transformation by cell piece
The center of image is moved on to, alignment is completed；
A6) generation includes the cover plate image of the intermediate cell piece with main grid, such as Fig. 6.
The defects of other classifications and black；If do not rejected, can cause to contain other defect in the curve of projection
Information causes the signal at black angle that can be weakened, therefore need to carry out S2）Step；
S2）It is as follows：
B1) using the window of a 15x15, mean value smoothing processing is carried out to image with mask；
B2) treated, and smoothed image is compared as background image, original image and background image pixels point；
B3) step B2）In, dark 35% pixel is other defect in original image, must not participate in annular projection, with reference to Fig. 7；
B4）Last and mask images combine, and generate the image that may finally be projected, such as Fig. 8.
S3）It is as follows：
D1）With reference to Fig. 9, Figure 10, the angle of end four is taken；
D2）Each angle is projected along 45 degree or 135 degree to progress method direction inside image；
D3）Normalization：Each effective pixel in method direction is not fully identical, so need to use valid pixel number, it is right
Curve is normalized；
D4）Drop shadow curve such as Figure 11, wherein the two of upper left and upper right curve is below, and slowly walks upwards；And lower-left and
The company curve of bottom right is above, and no apparent rule；Reflect that the content of image can be reflected in method line projection.
S4）It is as follows：
F1）It is extracted average value, standard variance, greatest gradient, the features such as minimal gradient herein from curve；
F2）The fitting that straight line is carried out to curve, extracts angle and two features of deviation of straight line；
F3）Carry out the distribution displaying in average gray and Gradient Features space, such as Figure 12, wherein saturate point is normal sample
This, the light point of color is defect sample；
F4）Feature is analyzed, judges whether black angular defect.
S5）Referring to Fig.1 3, classified using simplest decision tree.
Compared with prior art, the present invention can go out black angular defect by automatic decision：First image is aligned；Then
Normal direction projection is carried out, can really reflect the changing rule of black angle image；Then feature extraction is carried out to drop shadow curve；It is most laggard
Row classification.93% can be reached by being experimentally confirmed this method discrimination, and rate of false alarm can be controlled 1%.
Above-described is only some embodiments of the present invention.For those of ordinary skill in the art, not
Under the premise of being detached from the invention design, various modifications and improvements can be made, these belong to the protection model of the present invention
It encloses.
Claims (3)
1. a kind of validity feature extracting method applied to the black angle detection of cell piece, it is characterised in that：By carrying out rejecting other
After interference and method line projection, indicatrix extraction is carried out；
Indicatrix extraction step is as follows：
F1 average value, standard variance, greatest gradient, the features such as minimal gradient) are extracted herein from curve；
F2 the fitting that straight line) is carried out to curve, extracts angle and two features of deviation of straight line；
F3 the distribution displaying in average gray and Gradient Features space) is carried out；
F4 feature) is analyzed, judges whether black angular defect.
2. a kind of validity feature extracting method applied to the black angle detection of cell piece according to claim 1, feature exist
In：Rejecting other interference implementation causes is：The defects of other classifications and black；If do not rejected, the song of projection can be caused
Other defect information is contained in line, causes the signal at black angle that can be weakened；The specific steps are：
B1) using the window of a 15x15, mean value smoothing processing is carried out to image with mask；
B2) treated, and smoothed image is compared as background image, original image and background image pixels point；
B3) step B2) in, dark 35% pixel is other defect in original image, must not participate in annular projection；
B4) last and mask images combine, and generate the image that may finally be projected.
3. a kind of validity feature extracting method applied to the black angle detection of cell piece according to claim 2, feature exist
In：Method line projection is as follows：
D1 the angle of end four) is taken；
D2) each angle is projected along 45 degree or 135 degree to progress method direction inside image；
D3 it) normalizes：Each effective pixel in method direction is not fully identical, so need to use valid pixel number, it is right
Curve is normalized；
D4) two curves of upper left and upper right are below, and slowly walk upwards；And the company curve of lower-left and bottom right is upper
Face, and without apparent rule；Reflect that the content of image can be reflected in method line projection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810231844.5A CN108418553A (en) | 2018-03-20 | 2018-03-20 | Validity feature extracting method applied to the black angle detection of cell piece |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810231844.5A CN108418553A (en) | 2018-03-20 | 2018-03-20 | Validity feature extracting method applied to the black angle detection of cell piece |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108418553A true CN108418553A (en) | 2018-08-17 |
Family
ID=63132887
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810231844.5A Pending CN108418553A (en) | 2018-03-20 | 2018-03-20 | Validity feature extracting method applied to the black angle detection of cell piece |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108418553A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106780455A (en) * | 2016-12-07 | 2017-05-31 | 五邑大学 | A kind of product surface detection method based on the local neighborhood window for sliding |
| CN107561087A (en) * | 2017-08-31 | 2018-01-09 | 广东工业大学 | A kind of mouse logo positioning and defect inspection method based on machine vision |
-
2018
- 2018-03-20 CN CN201810231844.5A patent/CN108418553A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106780455A (en) * | 2016-12-07 | 2017-05-31 | 五邑大学 | A kind of product surface detection method based on the local neighborhood window for sliding |
| CN107561087A (en) * | 2017-08-31 | 2018-01-09 | 广东工业大学 | A kind of mouse logo positioning and defect inspection method based on machine vision |
Non-Patent Citations (1)
| Title |
|---|
| 韩芳芳: "表面缺陷视觉在线检测关键技术研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109084957B (en) | Defect detection and color sorting method and system for photovoltaic solar crystalline silicon cell | |
| Chen et al. | Accurate and robust crack detection using steerable evidence filtering in electroluminescence images of solar cells | |
| CN109584222A (en) | A kind of failure modes of the photovoltaic module image based on unmanned plane and discrimination method | |
| CN106408605A (en) | Method for judging dirtiness of photovoltaic cell panel based on color and texture identification technology | |
| CN113989241B (en) | Photovoltaic module EL defect detection method based on fusion of image processing and deep learning | |
| CN107481237A (en) | A kind of infrared array image hot spot detection method based on multiframe temperature characterisitic | |
| CN112669306A (en) | Solar cell defect detection method and system based on saliency map | |
| CN106057700B (en) | A kind of detection method on red of the side of solar battery sheet | |
| CN117036825A (en) | Solar cell panel detection method, medium and system | |
| CN111260617A (en) | Solar cell panel defect detection method based on deep learning | |
| Liu et al. | A CISG method for internal defect detection of solar cells in different production processes | |
| CN107742283A (en) | A kind of method of cell piece outward appearance grid line thickness inequality defects detection | |
| CN108418554A (en) | The black angle detection method of cell piece | |
| Rodriguez et al. | Segmentation and error detection of PV modules | |
| Zhang et al. | Research on image defect detection of silicon panel based on prewitt and canny operator | |
| CN111598851B (en) | Solar cell broken piece detection method based on morphological image processing | |
| CN108418553A (en) | Validity feature extracting method applied to the black angle detection of cell piece | |
| CN108447053A (en) | Normal projecting method applied to the black angle detection of cell piece | |
| Ahan et al. | AI-assisted cell-level fault detection and localization in solar PV electroluminescence images | |
| CN108418551A (en) | Cell piece center alignment methods | |
| CN108418552A (en) | Annulus projecting method applied to the detection of cell piece evil mind | |
| CN115330768B (en) | Quality grading method for solar cell panel | |
| CN108445018B (en) | Effective characteristic curve extraction method applied to battery piece black heart detection | |
| CN108807206A (en) | Cell piece evil mind detection method | |
| Wang et al. | Detection of Small Targets in Photovoltaic Cell Defect Polarization Imaging Based on Improved YOLOv7 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180817 |
|
| RJ01 | Rejection of invention patent application after publication |