CN107845090A - A kind of silicon chip detection method and silicon chip detection means - Google Patents
A kind of silicon chip detection method and silicon chip detection means Download PDFInfo
- Publication number
- CN107845090A CN107845090A CN201711025953.3A CN201711025953A CN107845090A CN 107845090 A CN107845090 A CN 107845090A CN 201711025953 A CN201711025953 A CN 201711025953A CN 107845090 A CN107845090 A CN 107845090A
- Authority
- CN
- China
- Prior art keywords
- silicon chip
- gray
- image
- row
- pixel
- 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
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 title claims abstract description 270
- 229910052710 silicon Inorganic materials 0.000 title claims abstract description 268
- 239000010703 silicon Substances 0.000 title claims abstract description 268
- 238000001514 detection method Methods 0.000 title claims abstract description 50
- 238000002310 reflectometry Methods 0.000 claims abstract description 62
- 239000000126 substance Substances 0.000 claims abstract description 20
- 238000012545 processing Methods 0.000 claims description 23
- 230000032798 delamination Effects 0.000 claims description 12
- 238000001914 filtration Methods 0.000 claims description 9
- 238000004458 analytical method Methods 0.000 claims description 7
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 230000001678 irradiating effect Effects 0.000 claims description 2
- 230000000007 visual effect Effects 0.000 abstract description 4
- 238000000034 method Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 5
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 4
- 230000008859 change Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000003475 lamination Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000377 silicon dioxide Substances 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000013078 crystal Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 229910021420 polycrystalline silicon Inorganic materials 0.000 description 1
- 229920005591 polysilicon Polymers 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000035807 sensation Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000013517 stratification Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation 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
-
- 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
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30148—Semiconductor; IC; Wafer
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Testing Or Measuring Of Semiconductors Or The Like (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The present invention provides a kind of silicon chip detection method, comprises the following steps：Step 1：The gray value of each pixel determines the particular location of silicon chip in gray level image based on silicon chip, so as to obtain accurately silicon chip image to be analyzed；Step 2：Grey value profile statistics is carried out to the accurately silicon chip image to be analyzed, and/or described in calculating accurately in silicon chip image to be analyzed all pixels point average gray；And step 3：The reflectivity of silicon chip, and/or the crystalline substance flower situation based on the grey value profile statistical decision silicon chip are calculated based on the average gray.The silicon chip detection method of the present invention can quantify the visual condition of silicon chip, avoid the otherness of subjective judgement, speed is fast, precision is high, repeated height.
Description
Technical field
The invention belongs to area of solar cell, more particularly to a kind of silicon chip detection method and silicon chip detection means.
Background technology
Solar cell is the device for directly luminous energy being changed into by photoelectric effect or photochemical effect electric energy, with light
The silica-based solar cell of electrical effect work is main flow.In the preparation process of silica-based solar cell, the processing and detection of silicon chip
It is extremely important.For example, using different processes, the difference of reflectivity occurs in polysilicon silicon chip surface different crystal orientations, outside
Brilliant flower phenomenon occurs in sight, if crystalline substance spends the outward appearance and yield rate that seriously, will have a strong impact on solar cell, so as to influence
The photoelectric transformation efficiency and cost of solar cell.Therefore, it is necessary to silicon chip, particularly the reflectivity and outward appearance of silicon chip are entered
Row detection.
At present, the method for testing of silicon chip reflectivity is that a certain zonule of silicon chip is shone using the light source calibrated
Penetrate, the light intensity reflected received according to sensor, draw the reflectivity in the region.What this method was drawn is this
The average reflectance of zonule, it is necessary to by multiple zonules during if necessary to be measured to the reflectivity of whole silicon chip
The method for measuring and averaging is tested.Although this method can be obtained for the small less reflection of domain error
Rate value, but it is larger for the reflectance value error of whole silicon chip, and the result tested every time is different from, and testing time root
Increase according to number of test points purpose and increase, if it is desired to obtaining relatively accurate result, take longer.And for the inspection of silicon chip outward appearance
Survey, be currently the difference in reflectivity by contrasting multiple spot, the subjective sensation plus technical staff to whole silicon chip outward appearance, Ran Houjin
The yield of row silicon chip outward appearance judges.Due to the average value that the reflectance measurement of each point is zonule, the judge to outward appearance
Effect is poor, plus the subjective judgement standard disunity of different technologies personnel, therefore it is current for the detection method of silicon chip outward appearance
Ununified standard.
Therefore, urgent need is a kind of more accurately with easily silicon chip detection method, and it can be to the reflectivity of large area silicon sheet
Accurately detect and silicon chip outward appearance is quantitatively judged.
The content of the invention
Therefore, it is an object of the invention to overcome above-mentioned prior art the defects of, there is provided a kind of silicon chip detection method, including
Following steps：
Step 1：The gray value of each pixel determines the particular location of silicon chip in gray level image based on silicon chip, so as to
Obtain accurately silicon chip image to be analyzed；
Step 2：Grey value profile statistics is carried out to the silicon chip image accurately to be analyzed, and/or calculated described accurate
Silicon chip image to be analyzed in all pixels point average gray；And
Step 3：The reflectivity of silicon chip is calculated based on the average gray, and/or is counted based on the grey value profile
Judge the crystalline substance flower situation of silicon chip.
According to the silicon chip detection method of the present invention, it is preferable that in step 1, the gray level image is directly gathered by camera
Obtain, or obtained after the coloured image of camera collection is carried out into gray processing processing.
According to the silicon chip detection method of the present invention, it is preferable that in step 1, based on each pixel in the gray level image
The comparison of the gray value and gray threshold of point obtains accurately silicon chip image gray threshold to be analyzed.
According to the silicon chip detection method of the present invention, it is preferable that obtaining accurately silicon chip image to be analyzed includes following sub-step
Suddenly：
1-1：The pixel of every a line of the gray level image is scanned, gray value is less than gray scale in recording per a line
The number of the pixel of threshold value, choose maximum number therein and refer to number as row, gray value is less than to the picture of gray threshold
The number of vegetarian refreshments is more than the row of a certain ratio of row reference number purpose as silicon chip row；
1-2：The pixel of each row of the gray level image is scanned, gray value in each row is recorded and is less than gray scale
The number of the pixel of threshold value, choose maximum number therein and refer to number as row, gray value is less than to the picture of gray threshold
The row that the number of vegetarian refreshments is more than a certain ratio of row reference number purpose arrange as silicon chip；
1-3：The common region of the silicon chip row and silicon chip row is chosen, so as to obtain the silicon chip accurately to be analyzed
Image.
According to the silicon chip detection method of the present invention, it is preferable that in step 3, calculating silicon chip reflectivity includes following sub-step
Suddenly：
Sub-step A：It is respectively R based on n reflectivity1-RnKnown silicon chip fitting reflectivity and average gray function
Relational expressionWherein, n is the integer more than or equal to 2, and R represents the reflectivity of silicon chip,Represent average gray；And
Sub-step B：According to the functional relationCalculate the reflectivity of silicon chip.
According to the silicon chip detection method of the present invention, it is preferable that sub-step A includes：It is respectively R to n reflectivity1-Rn
Know that silicon chip carries out the operation of step 1 and step 2, draw all pixels point in the accurate silicon chip image of n known silicon chips respectively
Average graySo as to fit the functional relation of reflectivity and average gray
According to the silicon chip detection method of the present invention, it is preferable that in step 3, judge the crystalline substance flower situation of silicon chip including as follows
Sub-step：
Sub-step a：Pixel number corresponding to each gray value that step 2 is calculated accounts for total pixel number purpose ratio
Value is abnormal gray value less than gray value corresponding to the ratio of the fractional threshold, higher than described compared with fractional threshold
Gray value corresponding to the ratio of fractional threshold is normal gray value；
Sub-step b：Calculate the width of normal gray value；And
Sub-step c：By the width compared with preset standard, the crystalline substance flower situation of silicon chip is judged.
Second aspect, present invention also offers another silicon chip detection method, for detecting the delamination of silicon chip, including
Following steps：
Step 1：The gray value of each pixel determines the particular location of silicon chip in gray level image based on silicon chip, so as to
Obtain accurately silicon chip image to be analyzed；
Step 2：By the silicon chip image accurately to be analyzed vertically or horizontal direction is divided into m one's share of expenses for a joint undertaking images, its
In, m is the integer more than or equal to 2；
Step 3：Grey value profile statistics is carried out to each one's share of expenses for a joint undertaking image in the m one's shares of expenses for a joint undertaking image, and is calculated per a
The average gray of all pixels point in subgraph；
Step 4：The reflectivity of the silicon chip corresponding to each one's share of expenses for a joint undertaking image is calculated based on the average gray, and is based on
The crystalline substance flower situation of silicon chip corresponding to each one's share of expenses for a joint undertaking image of grey value profile statistical decision；And
Step 5：The reflectivity of silicon chip corresponding to m one's share of expenses for a joint undertaking images that step 4 obtains and brilliant flower situation are compared,
Judge the delamination of silicon chip.
Silicon chip detection method according to a second aspect of the present invention, it is preferable that in step 1, the gray level image is by camera
Directly collection obtains, or is obtained after the coloured image of camera collection is carried out into gray processing processing.
Silicon chip detection method according to a second aspect of the present invention, it is preferable that in step 1, based in the gray level image
The comparison of the gray value and gray threshold of each pixel obtains accurately silicon chip image to be analyzed, wherein, in the gray-scale map
The gray value of pixel corresponding with silicon chip should be less than the gray threshold as in.
Silicon chip detection method according to a second aspect of the present invention, it is preferable that obtaining accurately silicon chip image to be analyzed includes
Following sub-step：
1-1：The pixel of every a line of the gray level image is scanned, gray value is less than gray scale in recording per a line
The number of the pixel of threshold value, choose maximum number therein and refer to number as row, gray value is less than to the picture of gray threshold
The number of vegetarian refreshments is more than the row of a certain ratio of row reference number purpose as silicon chip row；
1-2：The pixel of each row of the gray level image is scanned, gray value in each row is recorded and is less than gray scale
The number of the pixel of threshold value, choose maximum number therein and refer to number as row, gray value is less than to the picture of gray threshold
The row that the number of vegetarian refreshments is more than a certain ratio of row reference number purpose arrange as silicon chip；
1-3：The common region of the silicon chip row and silicon chip row is chosen, so as to obtain the silicon chip accurately to be analyzed
Image.
Silicon chip detection method according to a second aspect of the present invention, it is preferable that in step 4, calculating silicon chip reflectivity includes
Following sub-step：
Sub-step A：It is respectively R based on n reflectivity1-RnKnown silicon chip fitting reflectivity and average gray function
Relational expressionWherein, n is the integer more than or equal to 2, and R represents the reflectivity of silicon chip,Represent average gray；And
Sub-step B：According to the functional relationCalculate the reflection of the silicon chip corresponding to each one's share of expenses for a joint undertaking image
Rate.
Silicon chip detection method according to a second aspect of the present invention, it is preferable that in step 4, judge the crystalline substance flower situation of silicon chip
Including following sub-step：
Sub-step a：Pixel number corresponding to each gray value that step 3 is calculated accounts for the total of each one's share of expenses for a joint undertaking image
Pixel number purpose ratio is abnormal grey less than gray value corresponding to the ratio of the fractional threshold compared with fractional threshold
Angle value, it is normal gray value higher than the gray value corresponding to the ratio of the fractional threshold；
Sub-step b：Calculate the width of normal gray value；And
Sub-step c：By the width compared with preset standard, the crystalline substance of the silicon chip corresponding to each one's share of expenses for a joint undertaking image is judged
Flower situation.
Silicon chip detection method according to a second aspect of the present invention, it is preferable that in step 5, by analyzing m one's share of expenses for a joint undertaking images
The difference of the reflectivity of corresponding silicon chip and brilliant flower situation, to judge the delamination of silicon chip.
The third aspect, the invention provides a kind of silicon chip detection means, including：
Image capture module, for obtaining the gray level image of silicon chip；
Image processing module, its gray value based on each pixel in the gray level image determine the specific position of silicon chip
Put, so as to obtain accurately silicon chip image to be analyzed；
Data processing module, grey value profile statistics is carried out to the silicon chip image accurately to be analyzed, and/or calculate institute
State the average gray of all pixels point in accurately silicon chip image to be analyzed；Or will described in accurately silicon chip image to be analyzed
Vertically or horizontal direction is divided into m one's share of expenses for a joint undertaking images, and grey value profile system is carried out to each one's share of expenses for a joint undertaking image in m one's share of expenses for a joint undertaking images
Meter, and the average gray of all pixels point in each one's share of expenses for a joint undertaking image is calculated, wherein, m is the integer more than or equal to 2, and
Determination module, for calculating the reflectivity of silicon chip, and/or judge the crystalline substance flower situation of silicon chip, or judge silicon chip
Delamination.
According to the silicon chip detection means of the present invention, it is preferable that also including filtration module, for filtering out described accurately treat point
Analyse the noise in silicon chip image.
According to the silicon chip detection means of the present invention, it is preferable that described image acquisition module includes：
Light source, it is fixed on the table, for irradiating the silicon chip；
Camera, for gathering the image of the silicon chip.
According to the silicon chip detection means of the present invention, it is preferable that also including demarcating module, for being fitted average gray-anti-
Penetrate rate functional relation.
Compared with prior art, silicon chip detection method of the invention can quantify the visual condition of silicon chip, avoid subjectivity from sentencing
Disconnected otherness, speed is fast, precision is high, repeated height.
Brief description of the drawings
Embodiments of the present invention is further illustrated referring to the drawings, wherein：
Fig. 1 is the schematic diagram according to the silicon chip detection means of the embodiment of the present invention；
Fig. 2 is that it is silicon chip image to be analyzed that white box is interior according to the position analysis figure of the silicon chip of the embodiment of the present invention；
Fig. 3 is the position analysis figure of another silicon chip in irregular shape, is silicon chip image to be analyzed in white box；
Fig. 4 is to remove the accurate silicon chip image after noise；
Fig. 5 is the grey level histogram of silicon chip；And
Fig. 6 is the silicon chip vertically reflectivity R of each layer and normal gray value width L change curve.
Embodiment
In order that the purpose of the present invention, technical scheme and advantage are more clearly understood, pass through below in conjunction with accompanying drawing specific real
Applying example, the present invention is described in more detail.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention,
It is not intended to limit the present invention.
First embodiment
As shown in figure 1, LED annular light sources and camera are fixed on the table, silicon chip is placed on sample stage, with LED rings
Shape light irradiation silicon chip sample, and with camera gather silicon chip coloured image, then the colo r image transmission is entered to computer
Row processing.In this embodiment, workbench, sample stage and camera are all placed in shell.The cromogram that computer gathers to camera
As being handled as follows：
(1) data message of each pixel in Matlab software collection coloured images is used, according to each pixel
Tri- color components of R, G, B carry out gray processing processing, conventional gray processing processing formula is I=0.3R+0.59G+0.11B,
So as to which coloured image is converted into gray level image.If it should be understood readily by those skilled in this art, directly collect the ash of silicon chip
Image (that is, black-and-white photograph) is spent, the step of converting can omit.
(2) particular location of silicon chip is determined, so as to obtain accurately silicon chip image to be analyzed.Specifically, in the embodiment
In, silicon chip is placed in white background, and the gray value of the pixel of silicon chip position is less than (this area skill of gray threshold 150
Art personnel it is understood that the gray threshold have with device parameter it is directly related, it is therefore desirable to determine gray scale as the case may be
Threshold value), according to this feature, the pixel of every a line of gray level image is scanned, gray value is less than in recording per a line
The number of 150 pixel, a line of number maximum of pixel of the gray value less than 150 is chosen as reference line, reference line
The number of pixel of the middle gray value less than 150 refers to number as row, and the number of pixel of the gray value less than 150 is more than
0.75 times of row reference number purpose (is " silicon chip proportion ", is adjusted according to the inclined degree of silicon chip, span is
0.5~1, inclination is more serious, and silicon chip proportion as silicon chip row, scans closer to row 0.5), determine silicon chip institute successively
The minimum line number and maximum number of lines to occupy, so that it is determined that longitudinal particular location of silicon chip minimum line number and maximum number of lines it
Between；
Same reason, the pixel of each row of gray level image is scanned, records gray value in each row and be less than
The number of 150 pixel, the row for choosing the number maximum of pixel of the gray value less than 150 are used as reference columns, reference columns
The number of pixel of the middle gray value less than 150 refers to number as row, and the number of pixel of the gray value less than 150 is more than
The row of 0.75 times of the row reference number purpose arrange as silicon chip, scan successively, determine minimum columns that silicon chip occupied and most
Big columns, so that it is determined that the horizontal particular location of silicon chip is between minimum columns and maximum number of column；
By the letter in gray level image between minimum line number and maximum number of lines and between minimum columns and maximum number of column
Breath extracts, so as to obtain silicon chip image accurately to be analyzed, as shown in Fig. 2 white block diagram.This judgement silicon chip essence
The method that level is put can also be directed to other shapes, the silicon chip in orientation is identified, position analysis result figure as shown in Figure 3.
Pass through the micro-adjustment to gray threshold and silicon chip proportion, it is ensured that the final pictorial information analyzed all is from silicon
Piece in itself, while reduces follow-up amount of calculation, improves analysis time, in addition, this method for judging silicon chip elaborate position is not required to
Manually to choose, it is only necessary to suitable parameter is set, judges speed and precision so as to improve.
(3) alternatively, the image in Fig. 2 white box is filtered using such as middle finger filtering mode, made an uproar so as to remove
Sound, while can greatly preserve the detailed information of silicon chip.The expression formula of middle finger filtering is R=mediun (Ii).Wave filter window
It is mouthful different, the detailed information retained difference.Due to the difference very little of each brilliant flower in silicon chip and detailed information is heavier
Will, therefore the window size that 3*3 is chosen at this is filtered processing to silicon chip image, draws removal noise as shown in Figure 4
Silicon chip image.Filtered picture greatly reduces noise point value, is advantageous to follow-up numerical analysis.
(4) average gray of all pixels point in the silicon chip image shown in Fig. 4 is calculated, in this embodiment, gray scale is put down
Average is
(5) average gray-reflectivity function relational expression is determined：Based on fixed light source parameters and camera parameter, to n
Reflectivity is respectively R1~RnKnown silicon chip carry out the operations of abovementioned steps (1)~(4), draw the essence of n known silicon chips respectively
The average gray of all pixels point in eka-silicon pictureDraw out reflectivity R and average grayGraph of a relation,
The functional relations such as linear fit, quadratic fit or cubic fit can be used to draw the functional relation between two amountsPreferably, the silicon chip difference in reflectivity value of selected demarcation reflectivity is larger, quantity is more.In the embodiment
In, the relational expression for drawing reflectivity and average gray using linear fit isIf illumination
Condition and camera parameter are not changed, and step (5) only needs once demarcate.In one embodiment, step (5)
It can first carry out in advance.
(6) relational expression for drawing the average gray substitution step (5) that step (4) is drawn, is calculated the anti-of silicon chip
Penetrate rate.Using above-mentioned relation formulaIn the case of, the reflectivity for the silicon chip being calculated is
8.26%.
The silicon chip measuring reflectance of the present embodiment is realized by following device：
Image capture module, for obtaining the image of silicon chip, including：Fixed annular LED light source on the table, place
Silicon chip sample on sample stage, fixed camera on the table, it is preferably close that light source, sample and camera are placed in realization in shell
The shooting environmental closed.Image capture module can directly gather the gray level image of silicon chip.Alternatively, if the figure that camera collects
As being coloured image, coloured image can be converted into by gray level image by gray scale conversion module；
Image processing module, its gray value based on each pixel in the gray level image determine the specific position of silicon chip
Put, so as to obtain accurate silicon chip image；
Filtration module, for filtering out the noise in accurate silicon chip image；
Data processing module, calculate the average gray of all pixels point in filtered precisely silicon chip image
Determination module, for calculating the reflectivity of silicon chip according to average gray-reflectivity function relational expression；And
Demarcating module, for fitting the average gray-reflectivity based on known reflectivity and average gray
Functional relation.
Second embodiment
The embodiment is used for the visual condition for quantitatively determining silicon chip.The step of repeating first embodiment (1)~(3), it is right
Silicon chip image shown in Fig. 4 carries out grey value profile statistics, calculates pixel corresponding with each gray value I or gray value interval
Count out and account for total pixel number purpose ratio (referred to as " pixel institute accounting "), preferably draw grey value profile curve (such as Fig. 5
It is shown).Pixel number corresponding to each gray value is accounted for into total pixel number purpose ratio with fractional threshold (in the embodiment
In, fractional threshold 0.001) it is compared, gray value corresponding to the ratio less than 0.001 is abnormal gray value, higher than 0.001
Ratio corresponding to gray value be normal gray value；Calculate the width L of normal gray value, i.e., the normal gray value of highest with most
The difference of low normal gray value；And by poor L compared with preset standard, judge the crystalline substance flower situation of silicon chip, preset standard is as follows
Table 1, its is exemplary only, and nonrestrictive.
Table 1
| L values | The brilliant flower situation of silicon chip |
| 0~20 | It is outstanding |
| 21-40 | Well |
| 41~60 | Typically |
| 61~255 | It is unqualified |
In this embodiment, width L=30, therefore it is all right to judge that silicon chip crystalline substance is spent.
The silicon chip outward appearance determining device of the present embodiment is similar with the device of first embodiment, and difference is data processing module
Grey value profile statistics is carried out to resulting silicon chip image, calculates picture corresponding with each gray value I or gray value interval
Vegetarian refreshments number accounts for total pixel number purpose ratio, preferably draws grey value profile curve；And determination module is embedded with pre- bidding
Standard, the width L based on normal gray value determine the crystalline substance flower situation of silicon chip.
3rd embodiment
Lamination is also a kind of phenomenon for often occurring in silicon chip production, i.e. the top half of silicon chip and the latter half
Outward appearance has the phenomenon of larger difference.The lamination of silicon chip can cause the outward appearance inhomogeneities of solar battery sheet to increase, from
And reduce the yield rate of solar battery sheet.The present embodiment is used for the degeee of stratification of quantitative analysis silicon chip.
The step of repeating first embodiment (1)~(3), obtain filtering out the accurate silicon chip image to be analyzed of noise.By institute
Obtained silicon chip image is vertically divided into 300 parts, is determined according to the method for first embodiment and second embodiment per a
The reflectivity R of corresponding silicon chip and normal gray value width L, as a result as shown in fig. 6, Fig. 6 is the silicon according to 3rd embodiment
The reflectivity R (dotted line) of piece and the change curve of normal gray value width L (solid line), by normal gray value width L and reflectivity R
Change compared with criterion, so that it is determined that the delamination of silicon chip.Table 2 and 3 shows to judge the normal ash of delamination
Angle value width L and reflectivity R standard, its is exemplary only, and nonrestrictive.
Table 2
| The difference of L values | Silicon chip visual condition |
| 0~10 | Without layering |
| 11~20 | Slightly it is layered |
| 21~30 | Layering |
| 41~255 | Severe delamination |
Table 3
In the present embodiment, the difference that the reflectivity R of silicon chip sample different layers is calculated is 1.98%, width L difference
Different is 16, therefore for reflectivity R angles, the sample does not have lamination, for crystalline substance spends poor L angles, exists and slightly divides
Layer, in general, the silicon chip sample is slightly layered.Shown in solid as shown in Figure 6, lamination mainly appears on silicon chip
Tip portion.It should be understood readily by those skilled in this art, only R and L result of determination is all no layering, silicon chip can be just drawn
Conclusion of the sample without layering.
According to other embodiments of the invention, it is determined that silicon chip delamination when, silicon chip image is entered in the horizontal direction
Row portioning.
The device of delamination for judging silicon chip is similar with the device of first embodiment, and difference is：
Data processing module, first by resulting silicon chip image in the horizontal direction or vertical direction portioning, then to every
A image carries out grey value profile statistics, calculates pixel number corresponding with each gray value I and accounts for total pixel number
Ratio, draw the grey value profile curve per a image, and calculate and be averaged per the gray scale of all pixels point in a image
ValueAnd the normal gray value width L per a image；
Determination module, the silicon chip corresponding to every a image is calculated based on average gray-reflectivity function relational expression
Reflectivity R, and the difference of normal gray value width L and reflectivity R between each part image are compared with criterion, so as to
Determine the delamination of silicon chip sample.
According to other embodiments of the invention, when gathering the image of silicon chip, silicon chip sample is irradiated using backlight.
According to other embodiments of the invention, silicon chip is placed on a dark background, the ash of the pixel of silicon chip position
Angle value is higher than predetermined gray threshold.
According to other embodiments of the invention, at using the progress such as Matlab, VB, C language or C++ image procossing, data
Reason and judgement.
Although the present invention be described by means of preferred embodiments, but the present invention be not limited to it is described here
Embodiment, also include made various changes and change without departing from the present invention.
Claims (18)
1. a kind of silicon chip detection method, comprises the following steps：
Step 1：The gray value of each pixel determines the particular location of silicon chip in gray level image based on silicon chip, so as to obtain
Accurately silicon chip image to be analyzed；
Step 2：Grey value profile statistics is carried out to the silicon chip image accurately to be analyzed, and/or accurately treated described in calculating
Analyze the average gray of all pixels point in silicon chip image；And
Step 3：The reflectivity of silicon chip is calculated based on the average gray, and/or based on the grey value profile statistical decision
The crystalline substance flower situation of silicon chip.
2. silicon chip detection method according to claim 1, wherein, in step 1, the gray level image is direct by camera
Collection obtains, or is obtained after the coloured image of camera collection is carried out into gray processing processing.
3. silicon chip detection method according to claim 1, in step 1, based on each pixel in the gray level image
Gray value and the comparison of gray threshold obtain accurately silicon chip image to be analyzed.
4. silicon chip detection method according to claim 3, wherein, obtaining accurately silicon chip image to be analyzed includes following son
Step：
1-1：The pixel of every a line of the gray level image is scanned, gray value is less than gray threshold in recording per a line
Pixel number, choose maximum number therein and refer to number as row, by pixel of the gray value less than gray threshold
Number be more than a certain ratio of row reference number purpose row as silicon chip row；
1-2：The pixel of each row of the gray level image is scanned, gray value in each row is recorded and is less than gray threshold
Pixel number, choose maximum number therein and refer to number as row, by pixel of the gray value less than gray threshold
Number be more than a certain ratio of row reference number purpose row as silicon chip row；
1-3：The common region of the silicon chip row and silicon chip row is chosen, so as to obtain the silicon chip figure accurately to be analyzed
Picture.
5. the silicon chip detection method according to any one of claim 1-4, in step 3, calculating silicon chip reflectivity includes
Following sub-step：
Sub-step A：It is respectively R based on n reflectivity1-RnKnown silicon chip fitting reflectivity and average gray functional relation
FormulaWherein, n is the integer more than or equal to 2, and R represents the reflectivity of silicon chip,Represent average gray；And
Sub-step B：According to the functional relationCalculate the reflectivity of silicon chip.
6. silicon chip detection method according to claim 5, sub-step A include：It is respectively R to n reflectivity1-RnIt is known
Silicon chip carries out the operation of step 1 and step 2, draws all pixels point in the accurate silicon chip image of n known silicon chips respectively
Average graySo as to fit the functional relation of reflectivity and average gray
7. the silicon chip detection method according to any one of claim 1-4, in step 3, judge the crystalline substance flower situation of silicon chip
Including following sub-step：
Sub-step a：Pixel number corresponding to each gray value that step 2 is calculated account for total pixel number purpose ratio with
Fractional threshold is compared, and is abnormal gray value less than gray value corresponding to the ratio of the fractional threshold, higher than the ratio
Gray value corresponding to the ratio of threshold value is normal gray value；
Sub-step b：Calculate the width of normal gray value；And
Sub-step c：By the width compared with preset standard, the crystalline substance flower situation of silicon chip is judged.
8. a kind of silicon chip detection method, comprises the following steps：
Step 1：The gray value of each pixel determines the particular location of silicon chip in gray level image based on silicon chip, so as to obtain
Accurately silicon chip image to be analyzed；
Step 2：By the silicon chip image accurately to be analyzed vertically or horizontal direction is divided into m one's share of expenses for a joint undertaking images, wherein, m
For the integer more than or equal to 2；
Step 3：Grey value profile statistics is carried out to each one's share of expenses for a joint undertaking image in the m one's shares of expenses for a joint undertaking image, and calculates each one's share of expenses for a joint undertaking figure
The average gray of all pixels point as in；
Step 4：The reflectivity of the silicon chip corresponding to each one's share of expenses for a joint undertaking image is calculated based on the average gray, and based on described
The crystalline substance flower situation of silicon chip corresponding to each one's share of expenses for a joint undertaking image of grey value profile statistical decision；And
Step 5：The reflectivity of silicon chip corresponding to m one's share of expenses for a joint undertaking images that step 4 obtains and brilliant flower situation are compared, judged
The delamination of silicon chip.
9. silicon chip detection method according to claim 8, wherein, in step 1, the gray level image is direct by camera
Collection obtains, or is obtained after the coloured image of camera collection is carried out into gray processing processing.
10. silicon chip detection method according to claim 9, wherein, in step 1, based on each in the gray level image
The comparison of the gray value and gray threshold of pixel obtains accurately silicon chip image to be analyzed, wherein, in the gray level image
The gray value of pixel corresponding with silicon chip should be less than the gray threshold.
11. silicon chip detection method according to claim 10, wherein, obtain accurately silicon chip image to be analyzed include it is as follows
Sub-step：
1-1：The pixel of every a line of the gray level image is scanned, gray value is less than gray threshold in recording per a line
Pixel number, choose maximum number therein and refer to number as row, by pixel of the gray value less than gray threshold
Number be more than a certain ratio of row reference number purpose row as silicon chip row；
1-2：The pixel of each row of the gray level image is scanned, gray value in each row is recorded and is less than gray threshold
Pixel number, choose maximum number therein and refer to number as row, by pixel of the gray value less than gray threshold
Number be more than a certain ratio of row reference number purpose row as silicon chip row；
1-3：The common region of the silicon chip row and silicon chip row is chosen, so as to obtain the silicon chip figure accurately to be analyzed
Picture.
12. the silicon chip detection method according to any one of claim 8-11, in step 4, calculate silicon chip reflectivity bag
Include following sub-step：
Sub-step A：It is respectively R based on n reflectivity1-RnKnown silicon chip fitting reflectivity and average gray functional relation
FormulaWherein, n is the integer more than or equal to 2, and R represents the reflectivity of silicon chip,Represent average gray；And
Sub-step B：According to the functional relationCalculate the reflectivity of the silicon chip corresponding to each one's share of expenses for a joint undertaking image.
13. the silicon chip detection method according to any one of claim 8-11, in step 4, judge the crystalline substance flower feelings of silicon chip
Condition includes following sub-step：
Sub-step a：Pixel number corresponding to each gray value that step 3 is calculated accounts for total pixel of each one's share of expenses for a joint undertaking image
The ratio counted out is abnormal gray scale less than gray value corresponding to the ratio of the fractional threshold compared with fractional threshold
Value, is normal gray value higher than the gray value corresponding to the ratio of the fractional threshold；
Sub-step b：Calculate the width of normal gray value；And
Sub-step c：By the width compared with preset standard, the crystalline substance flower feelings of the silicon chip corresponding to each one's share of expenses for a joint undertaking image are judged
Condition.
14. the silicon chip detection method according to any one of claim 8-11, in step 5, by analyzing m one's share of expenses for a joint undertaking figures
As the reflectivity of corresponding silicon chip and the difference of brilliant flower situation, to judge the delamination of silicon chip.
15. a kind of silicon chip detection means, including：
Image capture module, for obtaining the gray level image of silicon chip；
Image processing module, its gray value based on each pixel in the gray level image determine the particular location of silicon chip, from
And obtain accurately silicon chip image to be analyzed；
Data processing module, grey value profile statistics is carried out to the silicon chip image accurately to be analyzed, and/or calculate the essence
The average gray of all pixels point in accurate silicon chip image to be analyzed；Or accurately silicon chip image edge to be analyzed is perpendicular by described in
Nogata to or horizontal direction be divided into m one's share of expenses for a joint undertaking images, in m one's share of expenses for a joint undertaking images each one's share of expenses for a joint undertaking image carry out grey value profile statistics,
And the average gray of all pixels point in each one's share of expenses for a joint undertaking image is calculated, wherein, m is the integer more than or equal to 2, and
Determination module, for calculating the reflectivity of silicon chip, and/or judge the crystalline substance flower situation of silicon chip, or judge the layering of silicon chip
Situation.
16. silicon chip detection means according to claim 15, in addition to filtration module, for filtering out described accurately treat point
Analyse the noise in silicon chip image.
17. silicon chip detection means according to claim 15, wherein, described image acquisition module includes：
Light source, it is fixed on the table, for irradiating the silicon chip；
Camera, for gathering the image of the silicon chip.
18. silicon chip detection means according to claim 15, in addition to demarcating module, for being fitted average gray-anti-
Penetrate rate functional relation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711025953.3A CN107845090B (en) | 2017-10-27 | 2017-10-27 | Silicon wafer detection method and silicon wafer detection device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711025953.3A CN107845090B (en) | 2017-10-27 | 2017-10-27 | Silicon wafer detection method and silicon wafer detection device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107845090A true CN107845090A (en) | 2018-03-27 |
| CN107845090B CN107845090B (en) | 2021-07-20 |
Family
ID=61680792
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711025953.3A Active CN107845090B (en) | 2017-10-27 | 2017-10-27 | Silicon wafer detection method and silicon wafer detection device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107845090B (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108961218A (en) * | 2018-06-11 | 2018-12-07 | 无锡维胜威信息科技有限公司 | Solar power silicon platelet spends extracting method |
| CN109611906A (en) * | 2018-12-04 | 2019-04-12 | 宁波吉吉娃电子科技有限公司 | Schema adaptation mechanism |
| 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 |
| CN110514627A (en) * | 2019-08-26 | 2019-11-29 | 松山湖材料实验室 | Silicon wafer method for measuring reflectance and its measuring device |
| CN110970318A (en) * | 2019-11-12 | 2020-04-07 | 浙江爱旭太阳能科技有限公司 | Measurement characterization method for solar cell color difference and plating winding |
| CN111325731A (en) * | 2020-02-20 | 2020-06-23 | 珠海格力智能装备有限公司 | Installation detection method and device of remote control device |
| CN111553871A (en) * | 2019-03-27 | 2020-08-18 | 上海交通大学 | Method for judging concentration of metal element in binary alloy by utilizing gray value of synchrotron radiation in-situ imaging image |
| CN112304904A (en) * | 2019-07-15 | 2021-02-02 | 松山湖材料实验室 | Silicon wafer reflectivity detection method based on filter array |
| CN112864034A (en) * | 2019-11-27 | 2021-05-28 | 上海先进半导体制造有限公司 | Aluminum corrosion treatment method and system |
| CN113063793A (en) * | 2021-03-25 | 2021-07-02 | 中山大学 | Graphical processing analysis method for service condition of titanium alloy surface coating and application thereof |
| CN113554637A (en) * | 2021-07-30 | 2021-10-26 | 上海集成电路制造创新中心有限公司 | Method and device for detecting dirt on camera cover plate |
| CN113689410A (en) * | 2021-08-25 | 2021-11-23 | 浙江浙能技术研究院有限公司 | Detection method of anti-reflection film layer on surface of photovoltaic module |
| CN114111722A (en) * | 2020-08-31 | 2022-03-01 | 上海微电子装备（集团）股份有限公司 | Detection device, detection method of silicon wafer placement state and wafer library management system |
| CN114659451A (en) * | 2020-12-22 | 2022-06-24 | 松山湖材料实验室 | Film thickness measuring method and measuring device |
-
2017
- 2017-10-27 CN CN201711025953.3A patent/CN107845090B/en active Active
Non-Patent Citations (7)
| Title |
|---|
| LIXIA YANG等: "Interface Engineering of High Efficiency Organic-Silicon Heterojunction Solar Cells", 《ACS APPLIED MATERIALS & INTERFACES》 * |
| 张玮华: "太阳能电池硅片缺陷自动检测分类方法研究", 《中国优秀硕士学位论文全文数据库-工程科技II辑》 * |
| 李桥等: "硅片缺陷检测中的图像分割方法", 《计算机应用》 * |
| 盛小春: "基于图像处理的太阳能单晶硅片表面质量检测", 《中国优秀硕士学位论文全文数据库-信息科技辑》 * |
| 赵慧阳: "基于机器视觉的太阳能电池片表面缺陷检测的研究", 《中国优秀硕士学位论文全文数据库-信息科技辑》 * |
| 邵奇等: "硅片缺陷自动检测分类方法探究", 《电子测试》 * |
| 陈洪博等: "图像法测定煤岩组分反射率工作曲线的建立与应用", 《煤炭学报》 * |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108961218B (en) * | 2018-06-11 | 2021-07-02 | 无锡维胜威信息科技有限公司 | Solar silicon wafer crystal flower extraction method |
| CN108961218A (en) * | 2018-06-11 | 2018-12-07 | 无锡维胜威信息科技有限公司 | Solar power silicon platelet spends extracting method |
| CN109611906A (en) * | 2018-12-04 | 2019-04-12 | 宁波吉吉娃电子科技有限公司 | Schema adaptation mechanism |
| CN111553871B (en) * | 2019-03-27 | 2023-06-16 | 上海交通大学 | Method for judging concentration of metal element in binary alloy by using gray value of synchronous radiation in-situ imaging image |
| CN111553871A (en) * | 2019-03-27 | 2020-08-18 | 上海交通大学 | Method for judging concentration of metal element in binary alloy by utilizing gray value of synchrotron radiation in-situ imaging image |
| 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 |
| CN112304904A (en) * | 2019-07-15 | 2021-02-02 | 松山湖材料实验室 | Silicon wafer reflectivity detection method based on filter array |
| CN112304904B (en) * | 2019-07-15 | 2023-11-03 | 松山湖材料实验室 | Silicon wafer reflectivity detection method based on filter array |
| CN110514627A (en) * | 2019-08-26 | 2019-11-29 | 松山湖材料实验室 | Silicon wafer method for measuring reflectance and its measuring device |
| CN110514627B (en) * | 2019-08-26 | 2024-06-07 | 松山湖材料实验室 | Silicon wafer reflectivity measuring method and measuring device thereof |
| CN110970318A (en) * | 2019-11-12 | 2020-04-07 | 浙江爱旭太阳能科技有限公司 | Measurement characterization method for solar cell color difference and plating winding |
| CN112864034A (en) * | 2019-11-27 | 2021-05-28 | 上海先进半导体制造有限公司 | Aluminum corrosion treatment method and system |
| CN112864034B (en) * | 2019-11-27 | 2023-09-01 | 上海先进半导体制造有限公司 | Aluminum corrosion treatment method and system |
| CN111325731A (en) * | 2020-02-20 | 2020-06-23 | 珠海格力智能装备有限公司 | Installation detection method and device of remote control device |
| CN114111722A (en) * | 2020-08-31 | 2022-03-01 | 上海微电子装备（集团）股份有限公司 | Detection device, detection method of silicon wafer placement state and wafer library management system |
| CN114659451A (en) * | 2020-12-22 | 2022-06-24 | 松山湖材料实验室 | Film thickness measuring method and measuring device |
| CN113063793A (en) * | 2021-03-25 | 2021-07-02 | 中山大学 | Graphical processing analysis method for service condition of titanium alloy surface coating and application thereof |
| CN113554637A (en) * | 2021-07-30 | 2021-10-26 | 上海集成电路制造创新中心有限公司 | Method and device for detecting dirt on camera cover plate |
| CN113689410A (en) * | 2021-08-25 | 2021-11-23 | 浙江浙能技术研究院有限公司 | Detection method of anti-reflection film layer on surface of photovoltaic module |
| CN113689410B (en) * | 2021-08-25 | 2024-04-26 | 浙江浙能技术研究院有限公司 | Detection method for anti-reflection film layer on surface of photovoltaic module |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107845090B (en) | 2021-07-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107845090A (en) | A kind of silicon chip detection method and silicon chip detection means | |
| CN111855664B (en) | Adjustable three-dimensional tunnel defect detection system | |
| CN102288613B (en) | Surface defect detecting method for fusing grey and depth information | |
| CN102713582B (en) | Inclusion detection in polished gemstones | |
| CN106023158B (en) | The fresh water pipless pearl pearly layer defect identification method of SD-OCT images | |
| CN104537651B (en) | Proportion detecting method and system for cracks in road surface image | |
| CN103617611B (en) | A kind of automatic threshold segmentation spot center and size detecting method | |
| CN106442420A (en) | Qualitative and quantitative combination water quality monitoring method | |
| CN104008550A (en) | Wafer surface defect feature analysis method and system and wafer surface detect feature classification method and system | |
| CN102680494B (en) | Based on arcuation face, the polishing metal flaw real-time detection method of machine vision | |
| CN102974551A (en) | Machine vision-based method for detecting and sorting polycrystalline silicon solar energy | |
| Sharma et al. | Grain quality detection by using image processing for public distribution | |
| CN109671058A (en) | A kind of defect inspection method and system of big image in different resolution | |
| CN109741314A (en) | A kind of visible detection method and system of part | |
| CN101949819A (en) | Cell counting method based on image identification | |
| CN101063659A (en) | System for detecting paper flaw of paper sheet | |
| CN110189375A (en) | A kind of images steganalysis method based on monocular vision measurement | |
| CN105241811B (en) | Multi-level focus adopts drawing method and system automatically | |
| CN106530281A (en) | Edge feature-based unmanned aerial vehicle image blur judgment method and system | |
| CN108152295A (en) | Wind generator set blade surface defect intelligent checking system | |
| CN101793840A (en) | Diamond cutting parameter measurement method and measuring device | |
| CN106156758B (en) | A kind of tidal saltmarsh method in SAR seashore image | |
| CN108186051A (en) | A kind of image processing method and processing system of the automatic measurement fetus Double Tops electrical path length from ultrasonoscopy | |
| CN104256882A (en) | Method for measuring proportion of reconstituted tobacco in cut tobacco on basis of computer vision | |
| CN108960070A (en) | A kind of water level elevation detection system and its method |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |