CN109360794A - A kind of crystal silicon photovoltaic electrode of solar battery secondary printing precision visible detection method and device - Google Patents
A kind of crystal silicon photovoltaic electrode of solar battery secondary printing precision visible detection method and device Download PDFInfo
- Publication number
- CN109360794A CN109360794A CN201811181356.4A CN201811181356A CN109360794A CN 109360794 A CN109360794 A CN 109360794A CN 201811181356 A CN201811181356 A CN 201811181356A CN 109360794 A CN109360794 A CN 109360794A
- Authority
- CN
- China
- Prior art keywords
- camera
- printing
- grid line
- image
- silicon wafer
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L22/00—Testing or measuring during manufacture or treatment; Reliability measurements, i.e. testing of parts without further processing to modify the parts as such; Structural arrangements therefor
- H01L22/20—Sequence of activities consisting of a plurality of measurements, corrections, marking or sorting steps
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L22/00—Testing or measuring during manufacture or treatment; Reliability measurements, i.e. testing of parts without further processing to modify the parts as such; Structural arrangements therefor
- H01L22/10—Measuring as part of the manufacturing process
- H01L22/12—Measuring as part of the manufacturing process for structural parameters, e.g. thickness, line width, refractive index, temperature, warp, bond strength, defects, optical inspection, electrical measurement of structural dimensions, metallurgic measurement of diffusions
Landscapes
- Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Computer Hardware Design (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Power Engineering (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of crystal silicon solar-energy photovoltaic battery secondary printing precision visible detection methods, realize that device required for this method mainly includes array camera, camera fixed frame, rotation positioning platform, rack, bearing ink pad and control system, the control system is used to control rotation positioning platform and controls the acquisition and algorithm process work of silicon chip image；The method of the present invention installs board first and demarcates to camera, then it is taken pictures using the camera demarcated to the silicon wafer after secondary printing, then by machine machine vision processing algorithm to image zooming-out silicon wafer grid line edge of taking pictures, it detects the width of grid line on silicon wafer, and identifies and position the position of mark point in silicon wafer；Each camera calculates the offset between secondary printing grid line and one-step print grid line by the control system grid line width detected and the location information of mark point in exhibition by combination method camera；The features such as inspection method and system of the invention has recognition speed fast, and stability is good.
Description
Technical field
The present invention relates to Computer Vision Detection Technique, in particular to a kind of secondary print of crystal silicon photovoltaic electrode of solar battery
Brush precision visible detection method and device.
Background technique
Crystal silicon solar-energy photovoltaic battery printing equipment is a kind of automatically to realize silk screen and stock based on machine vision
Print system.In order to adapt to high efficiency and high-precision crystal silicon solar batteries production requirement, Appliance computer vision is to printing
The silicon wafer of equipment production carries out fast, accurately detection and has become necessity.
For the conventional solar cells piece of large area, according to single camera as detection device, then need to be equipped with non-
The camera of normal high pixel, this not only will increase the cost of equipment, and considerably increase the operand of image procossing.In fact, right
In the detection of cell piece grid line printing precision, entire cell piece is not needed all to be taken by camera, and uses distribution array
The different position of low pixel camera shooting cell piece of formula, and by the calibration of camera system, then do a series of image procossing,
Whether the precision that press printing is measured in feature identification and calculating meets the requirements.
Summary of the invention
It is an object of the invention to overcome shortcoming and deficiency in the prior art, a kind of crystal silicon photovoltaic solar battery is provided
Electrode secondary printing precision visible detection method and device are identified based on image procossing, feature, accurately detect the crystal silicon sun
Whether energy battery grid line prints accurately, and the offset that printing is not punctual.
In order to achieve the above object, the present invention adopts the following technical scheme that:
A kind of crystal silicon photovoltaic electrode of solar battery secondary printing precision visible detection method, includes the following steps:
(1) mounted array camera and parameter is set, is completed at the same time calibration of the optical calibrating plate to array camera；
(2) by silk-screen printing it is primary after, the silicon wafer after printing is rotated by 90 ° through rotation positioning platform is sent to array phase
Taking pictures on station below machine, while triggering array camera and taking pictures, the silicon chip image taken pictures then is transmitted to central processing unit；
(3) central processing unit extracts silicon chip edge by machine vision Processing Algorithm, detects grid line on silicon wafer
Width, and identify and position the position of mark point in silicon wafer when printing twice；
(4) position after the one-step print detected according to step (3) with grid line width and mark point after secondary printing
Confidence breath, calculates the offset between secondary printing grid line and one-step print grid line, secondary printing accuracy detection can be completed.
As a preferred technical solution, in the step (1), by customizing optical calibrating plate, in conjunction with calibration algorithm to phase
Camera in machine battle array is demarcated；Detailed process is as follows:
The optical calibrating plate of customization is fixed on the bearing ink pad for station of taking pictures, triggering array camera is carried out to optical calibrating
Plate is taken pictures, and the image that then will take pictures is transmitted to central processing unit by image pick-up card；In the central processing unit utilizes
Portion's benchmark calibration module handles optical calibrating plate image, thus calibrate each camera pixel equivalent and each camera it
Between relative positional relationship, so that camera fields of view when its normal photographing is unified into a coordinate system.
As a preferred technical solution, in the step (3), the central processing unit passes through machine machine vision processing algorithm
Silicon chip edge is extracted, detects the width of grid line on silicon wafer, specifically include the following steps:
(311) silicon chip image is filtered first, then obtains silicon chip edge feature using arithmetic operators
Position of the pixel under camera image coordinate system；
(312) centered on the pixel of silicon chip edge feature, and along the tangential direction and gradient of the edge feature of the pixel
Direction carries out the interpolation processing of selected range, obtains interpolation image；
(313) tangential direction to interpolation image along silicon chip edge feature seeks mean value, to obtain one-dimensional image battle array
Column；
(314) edge feature positioning is carried out using the edge Locating operator of sub-pix to one-dimensional pattern matrix, obtains silicon
Coordinate of the piece edge feature under interpolation image coordinate system, and the coordinate is converted by homogeneous transformation to camera image coordinate
System；
(315) according to the silicon chip edge feature obtained after step (311)~step (314) processing in camera image coordinate system
Under position, coordinate of the silicon wafer grid line feature under camera image coordinate system is calculated, to calculate silicon wafer grid line width.
As a preferred technical solution, in the step (3), the position of mark point in silicon wafer when printing twice is identified and positioned
It sets, the specific steps are as follows:
(321) it after each camera gets silicon chip image, is scanned in image, if detecting " crescent moon " in the picture
The pattern of shape then enters in next step；Otherwise it is judged as correct printing, exits the detection method；
(322) outer edge and inward flange that crescent moon pattern is identified by matching algorithm, wherein outer peripheral radian is greater than
The radian of inward flange；
(323) two edges of crescent moon are fitted respectively, fit two circles, and determine the position in the two centers of circle
It sets；
(324) parameter for combining known printing screen, determines that two center locations are belonging respectively to one-step print or two
The mark point position of secondary printing.
It is described that two edges of crescent moon are fitted respectively as a preferred technical solution, in step (323), specifically
To carry out multi-point sampling on the outer edge of crescent moon and inward flange respectively, and circular fitting is carried out with least square method.
It is described to calculate between secondary printing grid line and one-step print grid line as a preferred technical solution, in step (4)
Offset, specifically: according to step (3) grid line width data obtained and locating mark points data, comprehensive first and second print
The grid line width parameter of brush filament net is equipped with weight to the grid line width data and locating mark points data respectively, calculates one
Offset between secondary printing and secondary printing.
The offset calculated between secondary printing grid line and one-step print grid line as a preferred technical solution,
The algorithm of following steps should be used:
(411) grid line for passing through secondary printing in camera fields of view is sampled first and is taken a little, utilize the width of setting
Threshold value judges that the width of each sampled point, the serial number and width value to more each sampled point can be divided into sampled point
Three classes: 1. sequence number value is smaller and width value is greater than the threshold value set；2. width value is less than the threshold value of setting；3. sequence number value compared with
Big and width value is greater than the threshold value of setting；
(412) point by first kind sampled point in top edge fits first straight line L with least square method1；First will be removed
The point of the top edge of sampled point after class fits second straight line L with least square method2；By the lower edge of third class sampled point
Point fits third straight line L using least square method3；The point of lower edge other than third class sampled point is used into least square
Method fits the 4th straight line L4；If the number of sampled point only has two or three in same class, then directly in above classification
The straight line fitting for giving up such calculates, and directly enables L1=L3Or L2And L4；Then L is calculated1And L3Angle average value, and and L2
And L4Angle average value make it is poor, can be obtained this grid line first and second print between angle deviation；Again with same
Step and method calculate other grid lines, the deviation in other camera fields of view, are averaged, finally obtain with grid line width data
The printing deviation being calculated；
(413) it after obtaining the printing deviation being calculated with grid line width data, is calculated by calculating the shape of crescent moon
Printing deviation, specifically: this two o'clock was calculated by the magazine inward flange central coordinate of circle of any two of array camera
First heart straight line l1, can similarly obtain the second company heart straight line l2, third connect heart straight line l3, third heart straight line l4, similarly
Method calculates the corresponding l in the outer edge center of circle1’、l2’、l3’、l4'；Finally calculate separately l1With l1’、l2With l2’、l3With l3’、l4With
l4' between angle, seek the average value of this four angles as the printing deviation value calculated by crescent shape；
It is equipped with weight according to the printing deviation that printing drift angle obtains above-mentioned steps (412) and step (413) two methods,
Finally using the weighted average of the two as final printing deviation and output.
The detection device of crystal silicon photovoltaic electrode of solar battery secondary printing precision visible detection method, including control system
System, rack, the side of being mounted on the rack rotation positioning platform and be mounted on rack side and be used for fixed array camera phase
Machine fixed frame；Four bearing ink pads of outer edge equal distribution on the rotation positioning platform, can be made often by rotation positioning platform
A bearing ink pad is successively rotated to the lower section of array camera, and the array camera includes four cameras, and camera position can be transferred, and make
Array camera can arranged orthogonal in the top of bearing ink pad；
The array camera is for taking pictures to the optical calibrating plate or silicon wafer that are placed on bearing ink pad, the optics mark
Fixed board is used to carry out location position to array camera；The control system is used to control the rotation and control of rotation positioning platform
Array camera acquires image and image procossing.
The control system includes central processing unit and connects respectively with central processing unit as a preferred technical solution,
Camera controller, image pick-up card, register control, image processor, the memory, man-machine interface connect；
The register control connection rotation positioning platform reaches bearing ink pad specified to control rotation positioning platform rotation
Position；
The camera controller connects array camera with image pick-up card, controls triggering array camera by camera controller
It takes pictures, obtains camera and take pictures and be converted into digital signal input central processing unit through image pick-up card after image, pass through central processing
Device and image processor handle the digital picture of silicon wafer, complete the secondary printing accuracy detection of target silicon wafer.
The optical calibrating plate or silicon wafer are fixed on bearing ink pad by way of vacuum suction as a preferred technical solution,
On.
The present invention has the following advantages compared with the existing technology and effect:
(1) present system use low resolution multiple industrial camera distributed arrangement modes, have compared with reduce
Grid line testing cost；
(2) grid line width visible detection method used by the method for the present invention has the anti-noise ability relatively opened and higher fortune
Calculate efficiency；
(3) data for comprehensively considering grid line width data Yu printing mark point that the method for the present invention uses, to calculate one, two
The method of offset between secondary printing, robustness with higher.
Detailed description of the invention
Fig. 1 is crystal silicon photovoltaic electrode of solar battery secondary printing precision visible detection method flow chart of the invention；
Fig. 2 is crystal silicon photovoltaic electrode of solar battery secondary printing precision vision inspection apparatus schematic diagram of the invention；
Fig. 3 is crystal silicon photovoltaic electrode of solar battery secondary printing precision vision inspection apparatus control system structure of the invention
At schematic diagram；
Fig. 4 (a), Fig. 4 (b) are simple first and second printing grid line and mark point schematic shapes in present example respectively；
Fig. 5 is the figure in present example in the case of correct secondary printing；
Fig. 6 is the schematic diagram in present example in the case of incorrect secondary printing；
Fig. 7 is grid line width in present example, crescent shape and its corresponding center of circle detection schematic diagram；
Calculate the schematic diagram of printing deviation when Fig. 8 in the embodiment of the present invention by grid line width data；
The printing deviation value calculated in the embodiment of the present invention by crescent shape when Fig. 9；
It is shown in Fig. 2, Fig. 3: 1-array camera；2-array camera framves；3-optical calibrating plates；4,8,9,10-four
Bearing ink pad；5-rotation positioning platforms；6-racks；7-control system figures.
Specific embodiment
In order to which the purpose of the present invention, technical solution and advantage is more clearly understood, with reference to the accompanying drawings and embodiments,
The present invention is further described in detail.It should be appreciated that described herein, the specific embodiments are only for explaining the present invention,
It is not limited to the present invention.
Embodiment
As shown in Figure 1, a kind of crystal silicon photovoltaic electrode of solar battery secondary printing precision visible detection method of the invention,
Include the following steps:
(1) mounted array camera and parameter is set, by customizing optical calibrating plate, in conjunction with calibration algorithm in camera battle array
Camera is demarcated, and after the completion of calibration, takes out optical calibrating plate；Printing machine starts to carry out the printing of silicon wafer grid line；
(2) after the every printing of printing machine is primary, the silicon wafer after printing is rotated by 90 ° through rotation positioning platform is sent to array phase
Taking pictures on station below machine, while triggering array camera and taking pictures, the silicon chip image taken pictures then is transmitted to central processing unit；
(3) central processing unit extracts silicon chip edge by machine vision Processing Algorithm, detects grid line on silicon wafer
Width, and identify and position the position of mark point in silicon wafer when printing twice；
(4) position after the one-step print detected according to step (3) with grid line width and mark point after secondary printing
Confidence breath, calculates the offset between secondary printing grid line and one-step print grid line, secondary printing accuracy detection can be completed.
The crystal silicon photovoltaic electrode of solar battery secondary printing precision visible detection method of the present embodiment is now described in detail；
In step (1), complete machine is installed, and complete the calibration of array camera；
The optical calibrating plate of customization is fixed on the bearing ink pad for station of taking pictures, triggering array camera is carried out to optical calibrating
Plate is taken pictures, and the image that then will take pictures is transmitted to central processing unit by image pick-up card；In the central processing unit utilizes
Portion's benchmark calibration module handles optical calibrating plate image, thus calibrate each camera pixel equivalent and each camera it
Between relative positional relationship, so that camera fields of view when its normal photographing is unified into a coordinate system.
In step (2), if Fig. 4 (a), Fig. 4 (b) are to print by primary, secondary grid line and mark point, every time after printing
Silicon wafer be rotated by 90 ° through rotation positioning platform and be sent to taking pictures on station below array camera, while triggering array camera bat
According to the silicon chip image taken pictures then is transmitted to central processing unit；
In step (3), silicon wafer grid line width detection includes the following steps:
(311) silicon chip image is filtered first, then obtains silicon chip edge feature using arithmetic operators
Position of the pixel under camera image coordinate system；
Specifically, digital noise processing is removed by silicon wafer digital picture of the image processor to input, and will gone
Noise image carries out Edge Gradient Feature processing by arithmetic operators, obtains the pixel of silicon chip edge feature in image coordinate
Coordinate under system；
In the present embodiment 1, described image coordinate system is the X using the image upper left corner of array camera acquisition as coordinate origin
Axis positive direction is directed toward image front-right along coordinate origin, and Y-axis positive direction is directed toward immediately below image along coordinate origin；
(312) centered on the pixel of silicon chip edge feature, and along the tangential direction and gradient of the edge feature of the pixel
Direction carries out the interpolation processing of selected range；The tangential direction and gradient direction of the edge feature along the pixel, refer to
The tangential direction and coordinate system direction of the gradient direction as interpolation image；
Specifically, central processing unit by image processor obtain coordinate of the silicon chip edge feature under image coordinate system with
And the tangential nature and Gradient Features of silicon chip edge, and according to the tangential nature and Gradient Features of silicon chip edge, establish interpolation graphs
The linear interpolation processing that selected range is carried out as coordinate system and with this coordinate system, obtains interpolation image；
(313) tangential direction to the interpolation image of acquisition along silicon chip edge feature seeks mean value, to obtain one-dimensional
Pattern matrix；
Specifically, according to the direction of interpolation image coordinate system and the tangential direction of silicon chip edge feature, to interpolation image
It carries out gray value and seeks mean value, export one-dimensional pattern matrix；
(314) to one-dimensional pattern matrix using obtaining edge feature sub-pix position after the edge Locating operator of sub-pix
It sets, and sub-pixel location is converted by homogeneous transformation to camera image coordinate system；
Specifically, edge is carried out using the edge Locating operator of sub-pix inside the control system to one-dimensional pattern matrix
Feature is accurately positioned, and obtains coordinate of the edge feature under interpolation image coordinate system, and by homogeneous transformation that silicon chip edge is special
It converts to camera image coordinate system the position of sign；
(315) it after according to the position of the silicon chip edge feature obtained after step (311)~step (314) processing, calculates
To coordinate of the silicon wafer grid line feature under image coordinate system, to calculate the developed width of silicon wafer grid line.
The identification and positioning of silicon wafer grid line mark point, include the following steps:
(321) it after each camera gets silicon chip image, is scanned in image by image processor, if in image
In detect the pattern of " crescent moon " shape as shown in FIG. 6, then enter in next step；Otherwise as shown in figure 5, being judged as correct print
Brush, exits this detection method；
(322) crescent moon pattern outer edge and inward flange are identified by matching algorithm, wherein defining the biggish edge of radian
For outer edge, on the contrary is inward flange, as shown in Figure 7；
(323) multi-point sampling is carried out on the outer edge of crescent moon and inward flange respectively, and carries out circular arc with least square method
Fitting fits two circles, and determines the position in the two centers of circle, as shown in Figure 7；
(324) parameter for combining known printing screen, determines that two center locations are belonging respectively to one-step print or two
The mark point position of secondary printing.
In step (4), collects every silicon chip image and pass through above-mentioned grid line checking method for width and grid line reference point identifying
With localization method, grid line width data obtained and locating mark points data；The grid line width of comprehensive first and second printing screen
Parameter is equipped with weight appropriate to grid line width data and locating mark points data respectively, calculates one-step print and secondary print
Secondary printing accuracy detection can be completed in offset between brush；Algorithm used in the step, specific as follows:
Grid line width data is first obtained, by taking a grid line in a camera fields of view as an example, for as shown in Figure 8: first giving up
The part of main grid is gone to, then the grid line for passing through secondary printing in camera fields of view is sampled in the method equidistantly sampled, takes n point
(n value is generally higher than 25) obtains corresponding top edge coordinate (x in each sampled pointup,yup), lower edge coordinate (xdown,
ydown) and width w；Sampled point can be divided into following three classes: 1. serial number by comparing the serial number and width value of each sampled point
It is worth the threshold value that smaller and width value is set greater than one；2. the threshold value that width value is set less than one；3. sequence number value is larger
And width value is greater than the threshold value of a setting.Wherein, above-mentioned " threshold value of setting " need to be according to selected in actual production
The difference of deflection angle is provided and is calculated when the difference of silk screen web plate grid line specification, silicon wafer are into plate.Classify after sampled point, by
Point ((x in such as Fig. 8 of a kind of sampled point in top edgeup1,yup1)……(xupi,yupi)) with least square method fit first
Straight line L1；It will be except the point ((x in such as Fig. 8 of the top edge of the sampled point after the first kindupi,yupi)……(xupn,yupn)) with most
Small square law fits second straight line L2；By the point ((x in such as Fig. 8 of the lower edge of third class sampled pointdowni+k,
ydowni+k)……(xdownn,ydownn)) with least square method fit third straight line L3；Under other than third class sampled point
The point ((x in such as Fig. 8 at edgedown1,ydown1)……(xdowni+k,ydowni+k)) with least square method fit the 4th straight line
L4.If the number of sampled point is less (such as two or three) in same class in above classification, then directly give up the straight of such
Line the Fitting Calculation, directly enables L1=L3Or L2And L4.Then L is calculated1And L3Angle average value, and and L2And L4Angle
It is poor that average value is made, and the deviation of angle between the printing of this grid line first and second can be obtained；Again with same step and method meter
Other grid lines, the deviation in other camera fields of view are calculated, is averaged, finally obtains the print being calculated with grid line width data
Brush deviation.
After obtaining the printing deviation being calculated with grid line width data, it is inclined that printing is calculated by calculating the shape of crescent moon
Difference；Detailed process are as follows: as shown in Figure 9: carrying out multi-point sampling to the crescent shaped inside and outside edge that step (322) detects, and
The corresponding circle in inside and outside edge and central coordinate of circle are acquired out with least square method；By inward flange central coordinate of circle in camera A with
Inward flange central coordinate of circle in camera B calculated the first heart straight line l of this two o'clock1；Second company's heart straight line can similarly be obtained
l2, third connect heart straight line l3, third heart straight line l4, the corresponding l in the same method calculating outer edge center of circle1’、l2’、l3’、
l4'；Finally calculate separately l1With l1’、l2With l2’、l3With l3’、l4With l4' between angle, seek the average value of this four angles
As the printing deviation value calculated by crescent shape.
In fact, both the above calculates the method for printing deviation in no particular order, computer can be to its parallel processing, to improve
Efficiency.Here description merely for convenience.When printing drift angle is smaller, to " inclined with the printing that grid line width data is calculated
Difference " is equipped with biggish weight, is equipped with lesser weight to " the printing deviation value calculated by crescent shape "；When printing drift angle
When larger, it is equipped with lesser weight to " printing deviation being calculated with grid line width data ", " is counted by crescent shape
The printing deviation value of calculation " is equipped with biggish weight.Finally using the weighted average of the two as final printing deviation and output.
The judgement of drift angle size therein and the proportion of weight, need to rule of thumb set.
The detection device of the crystal silicon photovoltaic electrode of solar battery secondary printing precision visible detection method of the present embodiment, such as
Shown in Fig. 2 and Fig. 3, the device include control system 7, rack 6, the side of being mounted on the rack rotation positioning platform 5, array camera
1 and be mounted on rack side and be used for fixed array camera camera fixed frame 2 and optical calibrating plate 3；The rotation
Four bearing ink pads of outer edge equal distribution 4,8,9,10 on locating platform, by rotation positioning platform can make each bearing ink pad according to
Secondary rotation is to the lower section of array camera, and the array camera uses the industrial camera of four distributed arrangements, and camera position can
Transfer so that array camera can arranged orthogonal in the top of bearing ink pad；The optical calibrating plate 3 or silicon wafer pass through vacuum suction
Mode is fixed on bearing ink pad.
The array camera 1 is for taking pictures to the optical calibrating plate or silicon wafer that are placed on bearing ink pad, the optics
Scaling board 3 is used to carry out location position to array camera；The control system be used for control rotation positioning platform 5 rotation and
It controls array camera 1 and acquires image and image procossing.
As shown in figure 3, the control system 7 includes central processing unit and the camera shooting connecting respectively with central processing unit
Controller, image pick-up card, register control, image processor, memory, man-machine interface；
The register control connection rotation positioning platform reaches bearing ink pad specified to control rotation positioning platform rotation
Position；
The camera controller connects array camera with image pick-up card, controls triggering array camera by camera controller
It takes pictures, obtains camera and take pictures and be converted into digital signal input central processing unit through image pick-up card after image, pass through central processing
Device and image processor handle the digital picture of silicon wafer, complete the secondary printing accuracy detection of target silicon wafer.
The embodiments described above only express several embodiments of the present invention, and the description thereof is more specific and detailed, but simultaneously
Limitations on the scope of the patent of the present invention therefore cannot be interpreted as.It should be pointed out that for those of ordinary skill in the art
For, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to guarantor of the invention
Protect range.Therefore, the scope of protection of the patent of the present invention should subject to the claims.
Claims (10)
1. a kind of crystal silicon photovoltaic electrode of solar battery secondary printing precision visible detection method, which is characterized in that including following
Step:
(1) mounted array camera and parameter is set, is completed at the same time calibration of the optical calibrating plate to array camera；
(2) after the every printing of printing machine is primary, the silicon wafer after printing is rotated by 90 ° through rotation positioning platform to be sent under array camera
Side takes pictures on station, while triggering array camera and taking pictures, and the silicon chip image taken pictures then is transmitted to central processing unit；
(3) central processing unit extracts silicon chip edge by machine vision Processing Algorithm, detects the width of grid line on silicon wafer,
And identify and position the position of mark point in silicon wafer when printing twice；
(4) position of the grid line width after the one-step print detected according to step (3) and after secondary printing and mark point is believed
Breath calculates the offset between secondary printing grid line and one-step print grid line, secondary printing accuracy detection can be completed.
2. crystal silicon photovoltaic electrode of solar battery secondary printing precision visible detection method according to claim 1, special
Sign is, in the step (1), the calibration for completing optical calibrating plate to array camera, detailed process is as follows:
The optical calibrating plate of customization is fixed on the bearing ink pad for station of taking pictures, triggering array camera carry out to optical calibrating plate into
Row is taken pictures, and the image that then will take pictures is transmitted to central processing unit by image pick-up card；The central processing unit utilizes internal base
Quasi- demarcating module handles optical calibrating plate image, to calibrate between the pixel equivalent and each camera of each camera
Relative positional relationship enables camera fields of view when its normal photographing to unify into a coordinate system.
3. crystal silicon photovoltaic electrode of solar battery secondary printing precision visible detection method according to claim 1, special
Sign is, in the step (3), the central processing unit extracts silicon chip edge by machine machine vision processing algorithm, detects
The width of grid line on silicon wafer, specifically include the following steps:
(311) silicon chip image is filtered first, the picture of silicon chip edge feature is then obtained using arithmetic operators
Position of the element under camera image coordinate system；
(312) centered on the pixel of silicon chip edge feature, and along the tangential direction and gradient direction of the edge feature of the pixel,
The interpolation processing for carrying out selected range, obtains interpolation image；
(313) tangential direction to interpolation image along silicon chip edge feature seeks mean value, to obtain one-dimensional pattern matrix；
(314) edge feature positioning is carried out using the edge Locating operator of sub-pix to one-dimensional pattern matrix, obtains silicon wafer side
Coordinate of the edge feature under interpolation image coordinate system, and the coordinate is converted by homogeneous transformation to camera image coordinate system；
(315) according to the silicon chip edge feature obtained after step (311)~step (314) processing under camera image coordinate system
Position calculates coordinate of the silicon wafer grid line feature under camera image coordinate system, to calculate silicon wafer grid line width.
4. crystal silicon photovoltaic electrode of solar battery secondary printing precision visible detection method according to claim 1, special
Sign is, in the step (3), identifies and position the position of mark point in silicon wafer when printing twice, the specific steps are as follows:
(321) it after each camera gets silicon chip image, is scanned in image, if detecting " crescent moon " shape in the picture
Pattern, then enter in next step；Otherwise it is judged as correct printing, exits the detection method；
(322) outer edge and inward flange of crescent moon pattern are identified by matching algorithm, wherein outer peripheral radian is greater than inner edge
The radian of edge；
(323) two edges of crescent moon are fitted respectively, fit two circles, and determine the position in the two centers of circle；
(324) parameter for combining known printing screen, determines that two center locations are belonging respectively to one-step print or secondary print
The mark point position of brush.
5. crystal silicon photovoltaic electrode of solar battery secondary printing precision visible detection method according to claim 4, special
Sign is, described to be fitted respectively to two edges of crescent moon in step (323), specially respectively in the outer edge of crescent moon
With multi-point sampling is carried out on inward flange, and carry out circular fitting with least square method.
6. crystal silicon photovoltaic electrode of solar battery secondary printing precision visible detection method according to claim 4, special
Sign is, in step (4), the offset calculated between secondary printing grid line and one-step print grid line, specifically: according to
Step (3) grid line width data obtained and locating mark points data, the grid line width ginseng of comprehensive first and second printing screen
Number is equipped with weight to the grid line width data and locating mark points data respectively, calculate one-step print and secondary printing it
Between offset.
7. crystal silicon photovoltaic electrode of solar battery secondary printing precision visible detection method according to claim 4, special
Sign is that the offset calculated between secondary printing grid line and one-step print grid line should use the algorithm of following steps:
(411) grid line for passing through secondary printing in camera fields of view is sampled first and is taken a little, utilize the width threshold value of setting
The width of each sampled point is judged, sampled point can be divided into three by the serial number and width value to more each sampled point
Class: 1. sequence number value is smaller and width value is greater than the threshold value set；2. width value is less than the threshold value of setting；3. sequence number value is larger
And width value is greater than the threshold value of setting；
(412) point by first kind sampled point in top edge fits first straight line L with least square method1；It will be except after the first kind
The point of the top edge of sampled point fits second straight line L with least square method2；The point of the lower edge of third class sampled point is used
Least square method fits third straight line L3；The point of lower edge other than third class sampled point is fitted using least square method
4th straight line L out4；If the number of sampled point only has two or three in same class in above classification, then directly give up this
The straight line fitting of class calculates, and directly enables L1=L3Or L2And L4；Then L is calculated1And L3Angle average value, and and L2And L4's
It is poor that the average value of angle is made, and the deviation of angle between the printing of this grid line first and second can be obtained；Again with same step and
Method calculates other grid lines, the deviation in other camera fields of view, is averaged, finally obtains and calculated with grid line width data
The printing deviation arrived；
(413) after obtaining the printing deviation being calculated with grid line width data, printing is calculated by calculating the shape of crescent moon
Deviation, specifically: calculating the first of this two o'clock by the magazine inward flange central coordinate of circle of any two of array camera
Heart straight line l1, can similarly obtain the second company heart straight line l2, third connect heart straight line l3, third heart straight line l4, same method
Calculate the corresponding l in the outer edge center of circle1’、l2’、l3’、l4'；Finally calculate separately l1With l1’、l2With l2’、l3With l3’、l4With l4' it
Between angle, seek the average value of this four angles as the printing deviation value calculated by crescent shape；
It is equipped with weight according to the printing deviation that printing drift angle obtains above-mentioned steps (412) and step (413) two methods, finally
Using the weighted average of the two as final printing deviation and output.
8. crystal silicon photovoltaic electrode of solar battery secondary printing precision vision described in any one is examined according to claim 1~7
The detection device of survey method, which is characterized in that including control system, rack, the side of being mounted on the rack rotation positioning platform, with
And it is mounted on rack side and is used for the camera fixed frame of fixed array camera；Outer edge on the rotation positioning platform is impartial
Four bearing ink pads are distributed, each bearing ink pad can be made successively to rotate to the lower section of array camera, the battle array by rotation positioning platform
Column camera uses the industrial camera of four distributed arrangements, and camera position can be transferred so that array camera can arranged orthogonal in
The top of bearing ink pad；
The array camera is for taking pictures to the optical calibrating plate or silicon wafer that are placed on bearing ink pad, the optical calibrating plate
For carrying out location position to array camera；The control system is used to control rotation and the control array of rotation positioning platform
Camera acquires image and image procossing.
9. crystal silicon photovoltaic electrode of solar battery secondary printing precision vision inspection apparatus according to claim 8, special
Sign is that the control system includes that central processing unit and the camera controller connecting respectively with central processing unit, image are adopted
Truck, register control, image processor, memory, man-machine interface；
The register control connection rotation positioning platform makes bearing ink pad reach specific bit to control rotation positioning platform rotation
It sets；
The camera controller connects array camera with image pick-up card, controls triggering array camera by camera controller and claps
According to obtaining camera and take pictures and be converted into digital signal input central processing unit through image pick-up card after image, pass through central processing unit
It is handled with digital picture of the image processor to silicon wafer, completes the secondary printing accuracy detection of target silicon wafer.
10. crystal silicon photovoltaic electrode of solar battery secondary printing precision vision detection system according to claim 8, special
Sign is that the optical calibrating plate or silicon wafer are fixed on bearing ink pad by way of vacuum suction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811181356.4A CN109360794B (en) | 2018-10-11 | 2018-10-11 | Visual detection method and device for secondary printing precision of crystalline silicon photovoltaic solar cell electrode |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811181356.4A CN109360794B (en) | 2018-10-11 | 2018-10-11 | Visual detection method and device for secondary printing precision of crystalline silicon photovoltaic solar cell electrode |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109360794A true CN109360794A (en) | 2019-02-19 |
| CN109360794B CN109360794B (en) | 2023-10-20 |
Family
ID=65349074
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811181356.4A Active CN109360794B (en) | 2018-10-11 | 2018-10-11 | Visual detection method and device for secondary printing precision of crystalline silicon photovoltaic solar cell electrode |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109360794B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110008946A (en) * | 2019-04-09 | 2019-07-12 | 苏州阿特斯阳光电力科技有限公司 | A kind of recognition methods of laser labelling and device, printing equipment |
| CN110211890A (en) * | 2019-06-20 | 2019-09-06 | 通威太阳能（安徽）有限公司 | A kind of detection method for preventing SE cell production process from printing mixed piece |
| CN112172338A (en) * | 2020-09-24 | 2021-01-05 | 山西潞安太阳能科技有限责任公司 | Automatic positioning system for silk-screen printing Mark points |
| CN112713102A (en) * | 2019-10-25 | 2021-04-27 | 苏州阿特斯阳光电力科技有限公司 | Pattern alignment detection method |
| CN112819764A (en) * | 2021-01-25 | 2021-05-18 | 湖南省计量检测研究院 | Method for detecting electrode position of membrane electrode MEA (membrane electrode assembly) of stacked fuel cell |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008179029A (en) * | 2007-01-24 | 2008-08-07 | Matsushita Electric Ind Co Ltd | Screen process printing device/method |
| JP2013188921A (en) * | 2012-03-13 | 2013-09-26 | Mitsubishi Electric Corp | Apparatus and method of screen printing, and substrate position correction method in screen printing method |
| CN103640349A (en) * | 2013-11-26 | 2014-03-19 | 常州天合光能有限公司 | Secondary printing alignment method of silicon solar cell |
| CN104009113A (en) * | 2013-05-22 | 2014-08-27 | 江苏爱多光伏科技有限公司 | Method for manufacturing precision-overprinted selective emitter cell |
| CN204674194U (en) * | 2015-02-12 | 2015-09-30 | 深圳市纳研科技有限公司 | Solar silicon wafers secondary printing navigation system |
| CN106891629A (en) * | 2017-02-06 | 2017-06-27 | 苏州润阳光伏科技有限公司 | A kind of grid line printing process of solar cell |
| CN107554050A (en) * | 2017-09-08 | 2018-01-09 | 华南理工大学 | The vision calibration device and its localization method of a kind of photovoltaic cell printing equipment |
-
2018
- 2018-10-11 CN CN201811181356.4A patent/CN109360794B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008179029A (en) * | 2007-01-24 | 2008-08-07 | Matsushita Electric Ind Co Ltd | Screen process printing device/method |
| JP2013188921A (en) * | 2012-03-13 | 2013-09-26 | Mitsubishi Electric Corp | Apparatus and method of screen printing, and substrate position correction method in screen printing method |
| CN104009113A (en) * | 2013-05-22 | 2014-08-27 | 江苏爱多光伏科技有限公司 | Method for manufacturing precision-overprinted selective emitter cell |
| CN103640349A (en) * | 2013-11-26 | 2014-03-19 | 常州天合光能有限公司 | Secondary printing alignment method of silicon solar cell |
| CN204674194U (en) * | 2015-02-12 | 2015-09-30 | 深圳市纳研科技有限公司 | Solar silicon wafers secondary printing navigation system |
| CN106891629A (en) * | 2017-02-06 | 2017-06-27 | 苏州润阳光伏科技有限公司 | A kind of grid line printing process of solar cell |
| CN107554050A (en) * | 2017-09-08 | 2018-01-09 | 华南理工大学 | The vision calibration device and its localization method of a kind of photovoltaic cell printing equipment |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110008946A (en) * | 2019-04-09 | 2019-07-12 | 苏州阿特斯阳光电力科技有限公司 | A kind of recognition methods of laser labelling and device, printing equipment |
| CN110008946B (en) * | 2019-04-09 | 2022-03-15 | 苏州阿特斯阳光电力科技有限公司 | Laser mark identification method and device and printing equipment |
| CN110211890A (en) * | 2019-06-20 | 2019-09-06 | 通威太阳能（安徽）有限公司 | A kind of detection method for preventing SE cell production process from printing mixed piece |
| CN112713102A (en) * | 2019-10-25 | 2021-04-27 | 苏州阿特斯阳光电力科技有限公司 | Pattern alignment detection method |
| CN112713102B (en) * | 2019-10-25 | 2022-07-15 | 苏州阿特斯阳光电力科技有限公司 | Pattern alignment detection method |
| CN112172338A (en) * | 2020-09-24 | 2021-01-05 | 山西潞安太阳能科技有限责任公司 | Automatic positioning system for silk-screen printing Mark points |
| CN112172338B (en) * | 2020-09-24 | 2022-03-15 | 山西潞安太阳能科技有限责任公司 | Automatic positioning system for silk-screen printing Mark points |
| CN112819764A (en) * | 2021-01-25 | 2021-05-18 | 湖南省计量检测研究院 | Method for detecting electrode position of membrane electrode MEA (membrane electrode assembly) of stacked fuel cell |
| CN112819764B (en) * | 2021-01-25 | 2022-07-19 | 湖南省计量检测研究院 | Method for detecting electrode position of membrane electrode MEA (membrane electrode assembly) of stacked fuel cell |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109360794B (en) | 2023-10-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109360794A (en) | A kind of crystal silicon photovoltaic electrode of solar battery secondary printing precision visible detection method and device | |
| CN107167169A (en) | Readings of pointer type meters identification measuring method based on NI Vision Builder for Automated Inspection | |
| CN106989669B (en) | Big visual field high-precision vision system calibrating method based on virtual three-dimensional target | |
| CN103267491B (en) | The method and system of automatic acquisition complete three-dimensional data of object surface | |
| WO2015154487A1 (en) | Grouped holes verticality detection system and method based on visual measurement | |
| CN107256568B (en) | High-precision mechanical arm hand-eye camera calibration method and calibration system | |
| CN101117043B (en) | Silk screen printing device | |
| CN102673106B (en) | Silk screen print positioning equipment and method for photovoltaic solar silicon chip | |
| CN102063718A (en) | Field calibration and precision measurement method for spot laser measuring system | |
| CN103134425B (en) | Information processor and information processing method | |
| WO2020073940A1 (en) | Method and device for calibrating print positioning platform of crystalline silicon photovoltaic solar cell on the basis of machine vision | |
| CN102778207A (en) | Method, device and system for measuring stress strain of structural member | |
| CN110065075A (en) | A kind of spatial cell robot external status cognitive method of view-based access control model | |
| CN106767810A (en) | The indoor orientation method and system of a kind of WIFI and visual information based on mobile terminal | |
| WO2024011981A1 (en) | Method for verifying light spot of cylindrical target | |
| CN105043720B (en) | The measuring method of infrared fileter refractive index based on single camera | |
| CN107507174A (en) | Power plant's instrument equipment drawing based on hand-held intelligent inspection is as recognition methods and system | |
| CN105278260B (en) | A kind of PCB exposure figures correctness verification method | |
| CN106382886A (en) | Online detection device and online detection method for indexable cutter machining | |
| CN107084671A (en) | A kind of recessed bulb diameter measuring system and measuring method based on three wire configuration light | |
| CN107554050A (en) | The vision calibration device and its localization method of a kind of photovoltaic cell printing equipment | |
| CN105091755B (en) | A kind of detection method of screen process press secondary printing precision | |
| CN111604916A (en) | Machine room IT equipment fault cabinet U-position positioning system and method | |
| CN105423975A (en) | Calibration system and method of large-size workpiece | |
| CN107560541A (en) | The measuring method and device of picture centre deviation |
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 | ||
| CB02 | Change of applicant information |
Address after: 510640 No. five, 381 mountain road, Guangzhou, Guangdong, Tianhe District Applicant after: SOUTH CHINA University OF TECHNOLOGY Applicant after: Guangdong kelongwei Intelligent Equipment Co.,Ltd. Address before: 510640 No. five, 381 mountain road, Guangzhou, Guangdong, Tianhe District Applicant before: SOUTH CHINA University OF TECHNOLOGY Applicant before: FOLUNGWIN AUTOMATIC EQUIPMENT Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20221012 Address after: Building 2, No. 11, Jinfu West Road, Tanglip, Liaobu Town, Dongguan City, Guangdong Province, 523405 Applicant after: Guangdong kelongwei Intelligent Equipment Co.,Ltd. Address before: 510640 No. five, 381 mountain road, Guangzhou, Guangdong, Tianhe District Applicant before: SOUTH CHINA University OF TECHNOLOGY Applicant before: Guangdong kelongwei Intelligent Equipment Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |