CN117116799B - Visual detection method and system for silicon wafer - Google Patents

Visual detection method and system for silicon wafer Download PDF

Info

Publication number
CN117116799B
CN117116799B CN202311338995.8A CN202311338995A CN117116799B CN 117116799 B CN117116799 B CN 117116799B CN 202311338995 A CN202311338995 A CN 202311338995A CN 117116799 B CN117116799 B CN 117116799B
Authority
CN
China
Prior art keywords
laser
monocrystalline silicon
defect
image
vortex
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202311338995.8A
Other languages
Chinese (zh)
Other versions
CN117116799A (en
Inventor
陈川南
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuxi Jingyuntong Technology Co Ltd
Original Assignee
Wuxi Jingyuntong Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Wuxi Jingyuntong Technology Co Ltd filed Critical Wuxi Jingyuntong Technology Co Ltd
Priority to CN202311338995.8A priority Critical patent/CN117116799B/en
Publication of CN117116799A publication Critical patent/CN117116799A/en
Application granted granted Critical
Publication of CN117116799B publication Critical patent/CN117116799B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L22/00Testing 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
    • H01L22/10Measuring as part of the manufacturing process
    • H01L22/12Measuring 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
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/0002Inspection of images, e.g. flaw detection
    • G06T7/0004Industrial image inspection
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/60Analysis of geometric attributes
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L22/00Testing 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
    • H01L22/20Sequence of activities consisting of a plurality of measurements, corrections, marking or sorting steps
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10004Still image; Photographic image
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20024Filtering details
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20048Transform domain processing
    • G06T2207/20056Discrete and fast Fourier transform, [DFT, FFT]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30108Industrial image inspection
    • G06T2207/30148Semiconductor; IC; Wafer

Abstract

The invention provides a visual detection method and a visual detection system for a silicon wafer, wherein the method comprises the following steps: emitting laser to the detected monocrystalline silicon piece in real time; identifying a plurality of concentric elliptical light spots in the monocrystalline silicon wafer image, and analyzing the laser spot pulse focusing condition of each laser projection point; image filtering and denoising are carried out; calculating the gray value of the maximum concentric elliptical light spot in the image, judging whether the maximum concentric elliptical light spot is an impure defect integral area formed by a plurality of vortex defects, if so, carrying out the next step, otherwise, repeating the steps; and carrying out secondary identification on the vortex defect area in the monocrystalline silicon wafer image after image filtering denoising, and judging the impurity defect degree of the detected monocrystalline silicon wafer caused by a plurality of vortex defects. The invention provides a technical scheme capable of clearly and visually detecting the black center defect degree of a monocrystalline silicon piece of a solar cell by machine vision and further judging whether vortex defects in the black center region are the whole regions of the impurity defects with obvious contrast.

Description

Visual detection method and system for silicon wafer
Technical Field
The invention belongs to the technical field of visual detection of monocrystalline silicon wafers, and particularly relates to a visual detection method and a visual detection system of a silicon wafer.
Background
In the prior art, in the solar cell production process, the detection of solar cells and cell components mainly depends on the traditional manual visual detection. The speed and the precision of manual detection are far from the requirements of high-quality solar cell detection because of the limitation of various factors such as self conditions, subjective emotion and the like. And with increasing market demands, more efficient process flows and production accuracy are required for the production of solar cells. In recent years, the rapid development of machine vision technology is very suitable for the requirement, and the application of the machine vision technology in the photovoltaic industry is becoming wider and wider.
The monocrystalline silicon wafer is a monocrystalline silicon polished wafer obtained by growing a silicon ingot, cooling, polishing, cutting, chamfering, front and back grinding and thinning, and is round, as disclosed in Chinese patent application publication No. 201010218718.X, and a doped cast monocrystalline silicon and preparation method thereof, impurities are introduced in the process of growing the monocrystalline silicon ingot in the production process of the monocrystalline silicon ingot in the prior art, so that the solar cell monocrystalline silicon wafer produced by the steps of polishing, cutting, grinding and thinning the monocrystalline silicon ingot can generate vortex defects as shown in figure 1.
In the prior art, a micro defect detection is generally performed on a solar cell monocrystalline silicon wafer obtained by a series of production steps of a monocrystalline silicon ingot by adopting a chemical corrosion technology, for example, a detection method of a quasi-monocrystalline silicon wafer micro defect is disclosed in a Chinese patent application document with the application number of 201510272000.1, wherein the method comprises the steps of performing manual mechanical polishing and chemical corrosion polishing, performing preferential corrosion of the micro defect, and performing minority carrier lifetime and iron-boron opposite scanning on the corroded silicon wafer to further judge the cutting line mark on the surface of the silicon wafer and the mechanical scratch degree. In addition, the application number 202211601912.5 of China discloses a method and a system for detecting defects in a monocrystalline silicon rod, wherein after a silicon wafer to be detected is obtained by cutting through a cutting device, after the silicon wafer to be detected is corroded through an acid corrosion device and a Saike corrosion device, whether defect aggregation of high-density symmetrical patterns and whether distribution, morphology and density of defects meet the condition of NDP defects are further detected on the surface of the silicon wafer to be detected through a visual detection device and a microscopic detection device, and then whether the silicon wafer to be detected belongs to the monocrystalline silicon wafer with the defects is detected.
However, in the prior art, the detection of the monocrystalline silicon piece of the solar cell needs to be advanced by chemical reagent corrosion, then image acquisition and detection judgment are performed in the aspect of machine vision, the specific technical means of image detection are not clear at the same time when the process is complex, and various imaging parameter conditions of machine vision images for judging that the monocrystalline silicon piece has defects are lacking.
Disclosure of Invention
The invention provides a visual detection method and a visual detection system for a silicon wafer aiming at the defects. The visual detection method and the visual detection system for the silicon wafer are used for machine visual detection of the monocrystalline silicon wafer insert of the solar cell, after the condition of focusing a plurality of concentric elliptical light spots and laser spot pulses of each laser projection point in an image of the monocrystalline silicon wafer is obtained by identification, the gray value of the largest elliptical light spot is calculated after image filtering denoising according to the condition, whether the largest elliptical light spot is an impurity integral area with serious black centers or not is judged, secondary identification is carried out according to the filtered and denoised image, the integral impurity integral area (black center area) formed by vortex defects with unobvious contrast inside the monocrystalline silicon wafer insert and the impurity integral area formed by vortex defects with obvious contrast are identified, and the technical scheme that whether the vortex defects in the black center area are the impurity integral area with obvious contrast can be further judged when the degree of the black center defect of the monocrystalline silicon wafer of the machine visual detection of the solar cell is further determined, and the classification of various black center monocrystalline silicon wafers caused by different standard screening can be definitely carried out on the monocrystalline silicon wafer for many times: the method comprises the steps of a region with a less serious black center, an impure defect whole region with a more serious black center, an impure defect whole region with obvious vortex defect contrast, an impure defect whole region with an unobvious vortex defect contrast, an unbalanced minority carrier particle flow strong composite region with obvious vortex defect contrast and an unbalanced minority carrier particle flow weak composite region with obvious vortex defect contrast.
The invention provides the following technical scheme: the visual detection method of the silicon wafer is used for detecting vortex defects in the monocrystalline silicon wafer in real time and comprises the following steps of:
s1: emitting laser to the detected monocrystalline silicon piece in real time;
s2: recognizing and obtaining a plurality of concentric elliptical light spots in the monocrystalline silicon wafer image, and analyzing the laser spot pulse focusing condition of each laser projection point in the monocrystalline silicon wafer image acquired by a CCD camera;
s3: performing image filtering denoising on the monocrystalline silicon wafer image acquired by the CCD camera according to the analysis result of the step S2;
s4: calculating the gray value of the maximum concentric elliptical light spot in the monocrystalline silicon wafer image, judging whether the gray value is an impurity defect integral area formed by a plurality of vortex defects, if so, carrying out the next step, otherwise, repeating the steps S1-S3;
s5: and (3) carrying out secondary identification on the vortex defect area in the monocrystalline silicon piece image after filtering and denoising the S3 step image, calculating to obtain elliptical ring light spot laser intensity values of a plurality of vortex defects and elliptical ring light spot laser intensity contrast of two adjacent vortex defects, and judging the impurity defect degree of the detected monocrystalline silicon piece caused by the plurality of vortex defects.
Further, the step S2 includes the following steps:
s21: construction of laser spot energy distribution functionAnd calculating the deflection angle of the vertical axis from the center point of the monocrystalline silicon piece as +.>Is +.>,/>
S22: according to the calculation result of the step S21, calculating and identifying the major axis radius of the j concentric elliptical light spots formed by the j elliptical boundary lines collected by the object plane of the CCD camera and converged on the object plane of the CCD camera by the Q laser image pointsAnd minor axis radius->I is an i-th laser projection point in a j-th elliptical boundary line acquired by a CCD camera object plane, i=1, 2, … and Q; j=1, 2, …, J;
s23: laser spot emitted by semiconductor laser emitterAn i-th laser projection point in a j-th elliptical boundary line in a single-crystal silicon image acquired at the object plane of a CCD camera>Performing Fourier transform to obtain the ith laser projection Fourier transform point +.>Wherein->For the ith laser projection point in the jth elliptical boundary line +.>Abscissa after fourier transformation, +.>For the ith laser projection point in the jth elliptical boundary line +.>An ordinate after the Fourier transform;
S24: calculating an ith laser projection Fourier transform point in the jth elliptical boundary line in the single crystal silicon imageImpulse response function value +.>And constructing an ith laser projection point +.in the jth elliptical boundary line in the single crystal silicon image>Fourier transform object plane domain model->
S25: computing a fourier transform object plane domain model in constructionThe ith laser projection point +_in the jth elliptical boundary line of (a)>Laser spot point pulse focusing function value of laser spot in monocrystalline silicon image obtained by CCD camera transmission imaging>
Further, the laser spot energy distribution function constructed in the step S21 is as follows:
the deflection angle of the vertical axis where the center point of the monocrystalline silicon piece is located isIs +.>The following are provided:
wherein, the liquid crystal display device comprises a liquid crystal display device,Wfor the laser emission power,is the radius of the laser beam at the waist, +.>For the radius of the laser light wave at different positions along the optical axis z direction +.>Z is the coordinate of the laser light wave at the optical axis,δis the laser emission frequency; />Is the radius of the laser equiphase surfaces from the emitting point of the semiconductor laser emitter to different positions of the optical axis z,lambda is the laser wavelength.
Further, in the step S22, a j-th concentric elliptical light formed by converging Q laser image points collected by the object plane of the CCD camera on a j-th elliptical boundary line of the object plane of the CCD camera is calculated and identified Major axis radius of plaqueAnd minor axis radius->The formula of (2) is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,is the radius of the laser beam at the waist, +.>For the radius of the laser light wave at different positions along the optical axis z direction +.>Z is the coordinate of the laser light wave at the optical axis,δis the laser emission frequency; />Is the radius of the laser equiphase surfaces from the emitting point of the semiconductor laser emitter to different positions of the optical axis z,
further, in the step S24, the ith laser projection fourier transform point within the jth elliptical boundary line is calculatedImpulse response function value +.>The formula of (2) is as follows:
an ith laser image point within a jth elliptical boundary line in the single crystal silicon image constructed in the step S24Fourier transform object plane domain model->The following are provided:
wherein, the liquid crystal display device comprises a liquid crystal display device,for the ith laser projection point in the jth elliptical boundary line +.>Fourier transform frequency domain values in the object plane of the CCD camera.
Further, the step S25 is to construct a Fourier transform object plane domain modelThe ith laser projection point +_in the jth elliptical boundary line of (a)>Laser spot point pulse focusing function value of laser spot in monocrystalline silicon image obtained by CCD camera transmission imaging>The formula of (2) is as follows:
Wherein, the liquid crystal display device comprises a liquid crystal display device,εin order to adjust the factor(s),ε=0.1 to 0.35, c is the light velocity, NA is the numerical aperture of the CCD camera,Is a Gaussian factor depending on the laser irradiation level, +.>Has a value of 1 to 2.5, (-)>As a function of the attenuation of the lasing frequency,
further, the step S3 includes the steps of:
s31: according to the laser spot pulse focusing function calculated in the step S2Calculating the i laser projection point of the CCD camera in the j-th elliptic boundary line +.>Transmitting laser spot pixel values of laser spots obtained by imaging:
s32: further calculating an ith laser image point of the CCD camera pair within the jth elliptical boundary lineTransmitting the pixel gray value of the imaged laser spot +.>
Wherein, the liquid crystal display device comprises a liquid crystal display device,is the laser spot pixel value +.>The value of red in the RGB color channel, +.>Is the laser spot pixel value +.>The value of green in the RGB color channel, < >>Is the laser spot pixel value +.>Blue values in the RGB color channel;
s33: the CCD camera pair obtained according to the step S32 is used for imaging the ith laser in the jth elliptical boundary linePixel gray value of laser spot obtained by transmission imaging +.>Constructing a noise pixel intensity calculation model>
Wherein q=1, 2, …;
S34: judging whether the result of the noise pixel intensity calculation model calculated in the step S33 is larger than 0.85, if so, judging that the CCD camera is used for carrying out image point setting on the ith laser in the jth elliptical boundary lineTransmitting the laser light spots obtained by imaging as noise points, and filtering and removing the noise points; otherwise, protectLeaving the CCD camera at the i laser projection point +.>Transmitting the imaged laser spot and repeating the steps S31-S33.
Further, the calculation formula for calculating the gray value of the largest concentric elliptical light spot in the monocrystalline silicon wafer image in the step S4 is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,the gray value of the largest concentric elliptical light spot in the monocrystalline silicon wafer image is obtained; when->And judging that the maximum concentric ellipse in the monocrystalline silicon wafer image is an impurity defect integral area formed by a plurality of vortex defects, or judging that the maximum concentric ellipse in the monocrystalline silicon wafer image is not an impurity defect integral area formed by a plurality of vortex defects.
Further, the step S5 includes the steps of:
s51: calculating an elliptical ring spot laser intensity value of an elliptical ring area of an mth vortex defect in the multiple vortex defects in the monocrystalline silicon wafer image after filtering and denoising the S3 step image The mth vortex defect is surrounded by the jth elliptic boundary line and the jth+p elliptic boundary line, m=1, 2, …, M; j is more than 1 and less than p is more than J;
wherein, the laser intensity value of the elliptical circular ring light spot with the mth vortex defectIn the calculation formula, i=A, A+1, A+2, …, B,1 < A < B < Q,>calculating standard deviation adjustment coefficients for the laser intensity of the elliptical circular ring light spot with the mth vortex defect, +.>,/>Is->And->Is a correlation coefficient of (2); />For the ith laser projection point in the mth vortex defect +.>And the peak point of the laser spot in the laser beam>The difference between the x-axis coordinates of (c); />For the ith laser projection point in the mth vortex defect +.>And the peak point of the laser spot in the laser beamThe difference between the y-axis coordinates of (c); />For the m-th vortex defect a plurality of laser projection points and the peak point of the laser spot therein +.>Standard deviation between x-axis coordinates of (c); />A plurality of laser projection points +.>Laser spot peak point +.>Standard deviation between y-axis coordinates of (c);
s52: calculating the laser intensity contrast of elliptical circular ring light spots with a plurality of adjacent vortex defects
Wherein, k is more than 1 and less than J is more than p and less than o is more than J,for the (m-1) th vortex defect surrounded by the (j-k) th elliptic boundary line and the (j) th elliptic boundary line, >For the (m+1) th vortex defect surrounded by the (j+p) th elliptic boundary line and the (j+q) th elliptic boundary line,/v>The laser intensity contrast ratio of the mth vortex defect to the adjacent mth-1 vortex defect which is close to the center of the ellipse on the inner side of the mth vortex defect and the adjacent mth+1 vortex defect which is far from the center of the ellipse on the outer side of the mth vortex defect;
s53: judging the laser intensity contrast of the elliptical ring light spots of the adjacent vortex defects calculated in the step S52Whether or not it is greater than 0.9, if +.>Judging that the detected monocrystalline silicon piece is a monocrystalline silicon piece with I type defects, wherein the monocrystalline silicon piece with I type defects is a monocrystalline silicon piece with obvious contrast of a plurality of vortex defects;
if it isJudging that the detected monocrystalline silicon piece is a class II defect monocrystalline silicon piece and the class II defect monocrystalline silicon piece is a monocrystalline silicon piece with a plurality of vortex defects and unobvious contrast;
s54: when (when)And (3) further calculating the image gray value of the elliptical ring light spots with a plurality of adjacent vortex defects, wherein the gray value is greater than 0.9:
wherein, the liquid crystal display device comprises a liquid crystal display device,the m-1 eddy defect is an image gray value of an elliptical circular light spot with the m-1 eddy defect, and the m-1 eddy defect is an eddy defect formed by the G-th laser projection point to the H-th laser projection point; />The image gray value of the elliptical circular light spot with the mth vortex defect is the vortex defect formed by the A-th laser image point to the B-th laser image point; / >The m+1st vortex defect is the vortex defect formed by the T-th laser projection point to the U-th laser projection point; n is the n-th elliptic boundary line; g is more than 1 and less than H is more than 1 and less than B is more than 1 and less than U is more than 0;
s55: comparison whenWhen the gray value is more than 0.9, the gray value of the image of the elliptical circular ring light spot with a plurality of adjacent vortex defects is larger than the gray value of the image of the elliptical circular ring light spot;
if it isAnd->Judging the mth vortex defect as a strong composite defect area of unbalanced minority carrier flow;
if it isAnd->And judging the mth vortex defect as a weak composite defect area of the unbalanced minority carrier flow.
The invention also provides a silicon wafer visual detection system adopting the method, which comprises a laser emission module, a CCD image acquisition module, a central control module, an image laser spot pulse focusing analysis module, an image filtering denoising module, an impurity defect integral region screening module and an impurity defect degree calculation module;
the laser emission module is used for emitting laser to the detected monocrystalline silicon piece in real time; the laser emitting module comprises a semiconductor laser emitter;
the CCD image acquisition module is used for acquiring monocrystalline silicon wafer images converged in the object plane area of the CCD camera after being filtered by the optical filter; the CCD image acquisition module comprises an optical filter, a CCD camera and a light angle sensor, wherein the optical filter The center point of the CCD camera and the center point of the detected monocrystalline silicon piece are always kept coincident; the light angle sensor is used for monitoring deflection angle of the vertical axis where the center point of the monocrystalline silicon piece is located in real time
The central control module is used for controlling the laser emission module to emit laser to the detected monocrystalline silicon piece and controlling the CCD image acquisition module to acquire the monocrystalline silicon piece image;
the image laser spot pulse focusing analysis module is used for identifying and obtaining a plurality of concentric elliptical light spots in the monocrystalline silicon wafer image and analyzing the laser spot pulse focusing condition of each laser projection point in the monocrystalline silicon wafer image acquired by the CCD camera;
the image filtering denoising module is used for performing image filtering denoising on the monocrystalline silicon wafer image acquired by the CCD camera;
the impurity defect integral region screening module is used for calculating the gray value of the largest concentric ellipse in the monocrystalline silicon wafer image and judging whether the gray value is an impurity defect integral region formed by a plurality of vortex defects;
the impurity defect degree calculation module is used for carrying out secondary identification on vortex defect areas in the monocrystalline silicon wafer image after image filtering denoising, calculating to obtain elliptical ring light spot laser intensity values of a plurality of vortex defects and elliptical ring light spot laser intensity contrast of two adjacent vortex defects, and judging impurity defect degree of the detected monocrystalline silicon wafer caused by the vortex defects.
The beneficial effects of the invention are as follows:
1. the visual detection method of the silicon wafer provided by the invention further carries out image filtering denoising according to the focusing condition of a plurality of concentric elliptical light spots and laser spot pulses of each laser projection point in the monocrystalline silicon wafer image during laser irradiation acquired by the CCD camera, and then calculates the gray value of the maximum elliptical light spotAnd is further dependent on the gray value->Judging whether the maximum concentric ellipse in the monocrystalline silicon piece image is an impurity defect integral area formed by a plurality of vortex defects or not, namely judging whether the maximum concentric ellipse is an impurity defect integral area with serious black centers or not;
the visual detection method of the silicon wafer provided by the invention also carries out secondary identification on the result obtained by screening the gray value according to the filtered and denoised image, and identifies the integral impure defect integral area (black heart area) formed by vortex defects with unobvious contrast inside the monocrystalline silicon wafer insert and the integral impure defect area formed by vortex defects with obvious contrast;
furthermore, the invention can effectively and clearly classify the types of the black heart monocrystalline silicon wafers caused by various vortex defects: the method comprises the steps of a region with a less serious black center, an impure defect whole region with a more serious black center, an impure defect whole region with obvious vortex defect contrast, an impure defect whole region with an unobvious vortex defect contrast, an unbalanced minority carrier particle flow strong composite region with obvious vortex defect contrast and an unbalanced minority carrier particle flow weak composite region with obvious vortex defect contrast. The method provides an effective technical scheme for black core detection and internal vortex defect degree of the monocrystalline silicon piece of the solar cell in the prior art.
2. The invention constructs the laser facula energy distribution function when identifying the monocrystalline silicon wafer image acquired by the CCD cameraAnd the deflection angle is +.about.about.the vertical axis from the center point of the monocrystalline silicon piece>Calculating laser spot energy according to the difference of (2)>The total of Q laser image points are converged on the CCDA plurality of concentric elliptical light spots in the object plane of the camera are subjected to ellipse fitting, so that the major axis radius of the plurality of concentric elliptical light spots can be identified>And minor axis radius->Then, the impulse response function of the laser projection Fourier transform point of the laser projection point is calculated according to the impulse response function value +.>Fourier transform is carried out on laser projection points, and then the Fourier transform is converted into spot image conversion of laser beams in the object plane of the CCD camera image sensor>Furthermore, the laser spot point pulse focusing function value of the laser spot corresponding to each laser projection point of the CCD camera image sensor in the object plane can be obtained>The method provides a basis for the calculation of the pixel value of the laser spot required by the subsequent image filtering denoising and gray value calculation, and converts the laser projection point in the three-dimensional coordinate system of the semiconductor laser transmitter into the object plane Fourier domain in the CCD camera through Fourier transformation, thereby being beneficial to accurate filtering denoising and gray value calculation.
3. The visual detection method of the silicon wafer provided by the invention filters and denoises the monocrystalline silicon wafer image acquired by the CCD camera, and calculates the ith laser projection point in the jth elliptical boundary linePixel gray value of laser spot obtained by transmission imaging +.>And further construct noisePixel intensity calculation model->The noise pixel intensity calculation model allows the pixel gray value +.>According with poisson distribution, further defining 0.85 as a noise threshold according to the probability mass density function distribution condition of poisson distribution, and if the probability mass density function distribution condition is larger than the threshold, indicating that the CCD camera is about the ith laser projection point in the jth elliptical boundary line->The laser light spots obtained by transmission imaging are noise spots, and the gray value of pixels is effectively improved through the steps S31-S34>The filtering denoising effect of the device improves the refinement degree and accuracy of filtering denoising, and avoids the condition that the accuracy of the subsequent secondary identification is reduced due to the fact that noise points are judged by transition errors in the filtering denoising process.
4. In the visual detection method of the silicon wafer, among a plurality of vortex defects in a single crystal silicon wafer image after filtering and denoising, an elliptical ring spot laser intensity value of an elliptical ring area where an mth vortex defect is located is surrounded by a jth elliptical boundary line and a jth+p elliptical boundary line Calculating, and further constructing the intensity contrast of elliptical circular ring light spots with a plurality of adjacent vortex defects>Elliptical ring spot laser intensity contrast +.A plurality of adjacent vortex defects calculated by breaking the S52 step>Whether the detected monocrystalline silicon piece is larger than 0.9 or not is determined, whether the detected monocrystalline silicon piece is a monocrystalline silicon piece with obvious or unobvious contrast ratio of a plurality of vortex defects, and according to the fact that the image gray level value of an elliptical circular light spot of one vortex defect in the plurality of vortex defects is larger than the image gray level value of an elliptical circular light spot of two adjacent vortex defects (>And->) Also the image gray value of elliptical ring spot smaller than adjacent two vortex defects (+.>And->) Further judging whether the target vortex defect is a strong composite defect area of unbalanced minority carrier particle flow or a weak composite defect area of unbalanced minority carrier particle flow, and if the target vortex defect is the strong composite defect area of unbalanced minority carrier particle flow, indicating that the target vortex defect (mth vortex defect) is a vortex defect area with more impurities; if the target vortex defect (mth vortex defect) is a vortex defect region with less impurities, the target vortex defect (mth vortex defect) is indicated as a non-equilibrium minority in the particle flow weak recombination defect region.
Further, the contrast of the laser intensity of the elliptical ring spot passing through the adjacent plurality of vortex defectsThe comparison of the image gray values of the elliptical ring light spots with the adjacent vortex defects is used for assisting in the calculation and judgment, so that when the detected monocrystalline silicon wafer is a defective product with obvious contrast of the vortex defects, the vortex defect area with more impurities is further judged, and the method can be used for debugging parameters in the production heating process of the monocrystalline silicon ingot, providing directions for the changes of the parameters and debugging and checking the parameters after the subsequent changes of the parameters.
Drawings
The invention will be described in more detail hereinafter on the basis of embodiments and with reference to the accompanying drawings. Wherein:
FIG. 1 is a schematic view of the vortex defect of a monocrystalline silicon piece as set forth in the background of the invention;
FIG. 2 is a schematic flow chart of a visual detection method for a silicon wafer;
FIG. 3 is a schematic flow chart of identifying the focusing condition of a plurality of concentric elliptical light spots and laser spot pulses in the visual detection method of a silicon wafer;
FIG. 4 shows that the deflection angle of the vertical axis of the semiconductor laser emitter and the CCD camera, which are located away from the center point of the monocrystalline silicon piece in the monocrystalline silicon piece image acquired by the visual inspection method of the silicon piece, is Is a laser spot schematic diagram of (1);
FIG. 5 is a schematic view of a symmetrical confocal cavity field formed by laser emitted by a semiconductor laser emitter in a visual inspection method of a silicon wafer;
FIG. 6 is a schematic diagram of a process for removing noise by image filtering on a monocrystalline silicon wafer image collected by a CCD camera in step S3 in the visual detection method of a silicon wafer;
FIG. 7 is a schematic flow chart of step S5 of a visual inspection method for silicon wafer according to the present invention;
FIG. 8 is a schematic diagram of a visual inspection system for silicon wafers according to the present invention;
fig. 9 is a schematic diagram of the structure of the visual inspection system for silicon wafers, which is provided by the invention, when detecting monocrystalline silicon wafers.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 2, a flow chart of a visual inspection method for silicon wafers according to the present invention is shown, and the method is used for detecting vortex defects inside monocrystalline silicon wafers shown in fig. 1 in real time, and comprises the following steps:
S1: emitting laser to the detected monocrystalline silicon piece in real time; after the laser is filtered by the monocrystalline silicon piece, the ground state electrons in the monocrystalline silicon piece are excited into an excited state, near infrared light with the peak of about 1100nm to 1200nm is released, and then a high-sensitivity high-resolution CCD camera is used for carrying out sensitization and imaging;
preferably, the step S1 is implemented by the central control module controlling the semiconductor laser emitters in the laser emitting module;
s2: recognizing and obtaining a plurality of concentric elliptical light spots in the monocrystalline silicon wafer image, and analyzing the laser spot pulse focusing condition of each laser projection point in the monocrystalline silicon wafer image acquired by the CCD camera;
s3: performing image filtering denoising on the monocrystalline silicon wafer image acquired by the CCD camera according to the analysis result of the step S2;
s4: calculating the gray value of the maximum concentric elliptical spot in the monocrystalline silicon wafer image, judging whether the maximum concentric elliptical spot is an impure defect integral area formed by a plurality of vortex defects, if so, carrying out the next step, otherwise, repeating S1-S3;
s5: and S3, carrying out secondary identification on vortex defect areas in the monocrystalline silicon wafer image subjected to filtering denoising in the step S3, calculating to obtain elliptical ring light spot laser intensity values of a plurality of vortex defects and elliptical ring light spot laser intensity contrast of two adjacent vortex defects, and judging the impurity defect degree of the detected monocrystalline silicon wafer caused by the plurality of vortex defects.
After laser is emitted by a semiconductor laser emitter, the laser irradiates the surface of a detected monocrystalline silicon piece, and is refracted by the monocrystalline silicon piece, and because the monocrystalline silicon piece is made of semiconductor materials, laser emitted by the semiconductor laser emitter is converged on a lens of a CCD camera after being filtered by a filter, and then is converged on an object plane (a physical imaging plane) of the CCD camera through a pinhole imaging principle to form a plurality of concentric elliptical light spots, in order to analyze the sizes of the concentric elliptical light spots obtained by photosensitive imaging on the object plane of the CCD camera after the laser is refracted at different positions of the monocrystalline silicon piece, gray values and laser intensity values of vortex defects of an elliptical ring area surrounded by boundary lines of two adjacent elliptical light spots formed by later image edge filtering noise reduction, further processing is needed to analyze the monocrystalline silicon piece image, so as to analyze laser spot energy and laser spot pulse focusing conditions of each laser projection point as another preferred embodiment of the invention, as shown in fig. 3, and the step S2 comprises the following steps:
s21: construction of laser spot energy distribution functionThe energy distribution of the laser spot is that the laser emitted by the laser emission point is projected on the object plane of the CCD camera, the energy distribution of the laser spot formed by a plurality of laser projection points on the optical axis z-axis, and the deflection angle of the vertical axis which is far from the center point of the monocrystalline silicon piece is calculated to be +. >Is +.>,/>The method comprises the steps of carrying out a first treatment on the surface of the The z axis of the laser emission point coincides with the axis of the vertical direction of the center point of the monocrystalline silicon piece;
s22: according to the calculation result of the step S21, calculating and identifying the major axis radius of the j concentric elliptical light spots formed by the j elliptical boundary lines collected by the object plane of the CCD camera and converged on the object plane of the CCD camera by Q laser image pointsAnd minor axis radius->I is an i-th laser projection point in a j-th elliptical boundary line acquired by a CCD camera object plane, i=1, 2, … and Q; j=1, 2, …, J;
s23: laser spot emitted by semiconductor laser emitterAn i-th laser projection point in a j-th elliptical boundary line in a single-crystal silicon image acquired at the object plane of a CCD camera>Performing Fourier transform to obtain the ith laser projection Fourier transform point +.>Wherein->For the ith laser projection point in the jth elliptical boundary line +.>Abscissa after fourier transformation, +.>For the ith laser projection point in the jth elliptical boundary line +.>An ordinate after the Fourier transform;
s24: calculating an ith laser projection Fourier transform point in a jth elliptical boundary line in a single crystal silicon image Impulse response function value +.>And constructing an ith laser projection point +.>Fourier transform object plane domain model->
S25: computing a fourier transform object plane domain model in constructionThe ith laser projection point +_in the jth elliptical boundary line of (a)>Laser spot point pulse focusing function value of laser spot in monocrystalline silicon image obtained by CCD camera transmission imaging>
It should be further noted that the laser spot emitted by the semiconductor laser emitterThe x and y of the laser point are respectively the abscissa and the ordinate of the laser point in the horizontal plane of the laser point parallel to the object plane of the CCD camera, and z is the vertical coordinate of the laser point on the z axis of the optical axis emitted by the semiconductor laser transmitter; an ith laser projection point +.in a jth elliptical boundary line in a single crystal silicon image acquired at the object plane of a CCD camera>Is->The object plane abscissa of the imaging sensor for a CCD camera,/->An object plane ordinate of an imaging sensor of the CCD camera; an ith laser projection Fourier transform point in a jth elliptical boundary line in a single crystal silicon image +.>Is->For the ith laser image point within the jth elliptical boundary lineFourier transform frequency domain in CCD camera object plane area after Fourier transform >Abscissa in (fourier domain),>for passing the ith laser projection point +.>Fourier transform frequency domain in CCD camera object plane area after Fourier transform>And the ordinate in the inner.
The semiconductor laser transmitter generates laser by the transition of electron hole pairs in a pn junction, so more photons are generated closer to the center line of the pn junction, the pn junction is in a straight shape in the horizontal axis of an abscissa and the vertical axis of an ordinate, and the deflection angle of the semiconductor laser transmitter from the vertical axis of the center point of a monocrystalline silicon wafer is as follows due to the difference of waveguide limits because the conventional splicing method is mostly adopted for reducing the cost of the semiconductor laser transmitterAlso, the divergence angle in the horizontal axis direction is usually larger than that in the vertical axis direction, as shown in FIG. 4,/->、 />And->Three laser rays emitted by three semiconductor lasers with different distances from the center point of the monocrystalline silicon piece to the deflection angle of the vertical axis are respectively on the j-th elliptic boundary line and the +.>Closer to the horizontal axis direction,>closer to the longitudinal axis direction,)>Between the horizontal and vertical axes, it can be seen from fig. 4 that +. >Therefore, the laser spots formed by the laser projection points projected on the object plane of the CCD camera after refraction of the monocrystalline silicon wafer are elliptical, and a plurality of concentric elliptical spots are formed due to the existence of vortex defects.
Further preferably, in step S2 provided by the present invention, as shown in fig. 5, the laser emitted by the semiconductor laser emitter forms a symmetrical confocal cavity field, and the laser spot energy distribution function constructed in step S21 is as follows:
the deflection angle from the vertical axis of the center point of the monocrystalline silicon piece isIs +.>The following are provided:
wherein, the liquid crystal display device comprises a liquid crystal display device,Wfor the laser emission power,is the radius of the laser beam at the waist as shown in fig. 5,/->For the radius of the laser light wave at different positions along the optical axis z direction as shown in FIG. 5, +.>Z is the coordinate of the laser light wave at the optical axis,δis the laser emission frequency; />Is the radius of the equiphase surface of the laser from the emitting point of the semiconductor laser emitter to the different positions of the optical axis z, +.>Lambda is the laser wavelength.
Still preferably, in the step S2 provided by the present invention, in the step S22, the major axis radius of the jth concentric elliptical spot formed by the jth elliptical boundary line, in which the Q laser projection points collected by the object plane of the identification CCD camera are converged on the object plane of the CCD camera, is calculated And minor axis radius->The formula of (2) is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,is the radius of the laser beam at the waist, +.>For the radius of the laser light wave at different positions along the optical axis z direction +.>Z is the coordinate of the laser light wave at the optical axis,δis the laser emission frequency; />Is the radius of the equiphase surface of the laser from the emitting point of the semiconductor laser emitter to the different positions of the optical axis z, +.>
Deflection angle of vertical axis of center point of monocrystalline silicon piece of semiconductor laser transmitterIs a laser spot of (2)Focusing on an object plane of a laser projection point image acquired by a CCD camera, wherein the i-th laser projection point which is acquired by the object plane of the CCD camera and is Q in total is +.>The light spot formed by converging on the object plane of the CCD camera is a long axis with radius ofThe minor axis radius is->The j-th spot of the ellipse, i.e. the j-th spot has an oval borderline, which is the j-th oval borderline, Q laser projection points +.>The j concentric elliptical spot solid projections in the j elliptical boundary line are formed together, and the impulse response function of the laser projection points is as follows: i=1, 2, …, Q;/>
Thus, the ith laser image point in the jth elliptical boundary lineFourier transform of the ith laser projection Fourier transform point in the jth elliptical boundary line >Is in line with the long axis +.>The minor axis is->Is a elliptical radius shape object plane domain;
still preferably, in step S2 provided by the present invention, the ith laser projection fourier transform point in the jth elliptical boundary line is calculated in step S24Impulse response function value +.>The formula of (2) is as follows:
an ith laser projection point within the jth elliptical boundary line in the single crystal silicon image constructed in step S24Fourier transform object plane domain model->The following are provided:
wherein, the liquid crystal display device comprises a liquid crystal display device,for the ith laser projection point in the jth elliptical boundary line +.>Fourier transform frequency domain values in the object plane of the CCD camera.
Further preferably, in the step S2 provided by the present invention, the fourier transform object plane domain model is constructed in the step S25The ith laser projection point +_in the jth elliptical boundary line of (a)>Laser spot point pulse focusing function value of laser spot in monocrystalline silicon image obtained by CCD camera transmission imaging>The formula of (2) is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,εin order to adjust the factor(s),ε=0.1 to 0.35, c is the speed of light, NA is the numerical aperture of the CCD camera,is a Gaussian factor depending on the laser irradiation level, +.>Has a value of 1 to 2.5, (-)>As a function of the attenuation of the lasing frequency,
As another preferred embodiment provided by the present invention, as shown in fig. 6, the step S3 includes the steps of:
s31: according to the laser spot pulse focusing function calculated in the step S2Calculating the i laser projection point of the CCD camera in the j-th elliptic boundary line +.>Transmitting laser spot pixel values of laser spots obtained by imaging:
/>
s32: further calculating an ith laser image point of the CCD camera pair within the jth elliptical boundary linePixel gray value of laser spot obtained by transmission imaging +.>
Wherein, the liquid crystal display device comprises a liquid crystal display device,is the laser spot pixel value +.>The value of red in the RGB color channel, +.>Is the laser spot pixel value +.>Green in RGB color channelColor value->Is the laser spot pixel value +.>Blue values in the RGB color channel;
s33: the ith laser image point in the jth elliptical boundary line of the CCD camera pair obtained according to step S32Pixel gray value of laser spot obtained by transmission imaging +.>Constructing a noise pixel intensity calculation model
Wherein q=1, 2, …; i.e., q is a discrete natural number;namely, the discrete natural number is factorized;
s34: judging whether the result of the noise pixel intensity calculation model calculated in the step S33 is larger than 0.85, if so, judging that the CCD camera is used for carrying out image point setting on the ith laser in the jth elliptical boundary line Transmitting the laser light spots obtained by imaging as noise points, and filtering and removing the noise points; otherwise, the i laser projection point of the CCD camera pair in the j-th elliptical boundary line is reserved +.>The imaged laser spot is transmitted and steps S31-S33 are repeated.
In order to determine whether the oval area formed by the plurality of concentric oval light spots obtained by the step S2 is an impure defect area with serious black center, namely whether the detected monocrystalline silicon piece is a black center piece with serious black center area, as another preferred embodiment of the invention, in the method for visualizing the silicon piece, a calculation formula for calculating the gray value of the largest concentric oval light spot in the monocrystalline silicon piece image in the step S4 is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,the gray value of the largest concentric elliptical light spot in the monocrystalline silicon wafer image; when->And judging that the maximum concentric ellipse in the monocrystalline silicon wafer image is an integral area of the impurity defect formed by a plurality of vortex defects, otherwise, judging that the maximum concentric ellipse in the monocrystalline silicon wafer image is not an integral area of the impurity defect formed by a plurality of vortex defects.
The whole area of the impurity defect represents that the detected monocrystalline silicon piece is a defective product with serious black heart condition whenIn this case, the single crystal silicon wafer to be inspected is a defective product whose black core condition is not particularly serious.
In order to further identify whether the intensity contrast of the elliptical ring light spots of the adjacent multiple vortex defects is obvious or not, and further determine that a certain vortex defect is a region doped with impurities, as another preferred embodiment of the present invention, based on the above embodiment, it is further preferred that, as shown in fig. 7, step S5 includes the following steps:
s51: calculating an elliptical ring spot laser intensity value of an elliptical ring area of an mth vortex defect in the multiple vortex defects in the monocrystalline silicon wafer image after filtering and denoising of the S3 step imageThe mth vortex defect is surrounded by the jth elliptic boundary line and the jth+p elliptic boundary line, m=1, 2, …, M; j is more than 1 and less than p is more than J;
wherein the number of laser projection points in the mth vortex defect is B-A+1, namely the mth laser projection point to the mth laser projection point collected by the object plane of the CCD camerase:Sub>A form the mth vortex defect, and the elliptical circular ring light spot laser intensity value of the mth vortex defectIn the calculation formula, i=A, A+1, A+2, …, B,1 < A < B < Q,>calculating standard deviation adjustment coefficients for the laser intensity of the elliptical circular ring light spot with the mth vortex defect, +.>, />Is->And->Correlation coefficient of>,/>Is->Is- >Covariance between; />Is the ith laser projection point +.>And the peak point of the laser spot in the laser beam>Difference between x-axis coordinates of +.>;/>For the ith laser projection point in the mth vortex defect +.>And the peak point of the laser spot in the laser beam>Difference between y-axis coordinates of +.>;/>Is the mth vortex defect and the peak point of the laser spot in the mth vortex defect, wherein the mth vortex defect is formed by the mth elliptic boundary line and the jth+p elliptic boundary line>Standard deviation between x-axis coordinates of +.>;/>A plurality of laser projection points +.>Peak point of laser spot in boundary line with jth ellipseStandard deviation between y-axis coordinates of +.>
Peak point of laser spotIs the peak point with the maximum laser emission fluorescence intensity in the m-th vortex defect area
S52: calculating the laser intensity contrast of elliptical circular ring light spots with a plurality of adjacent vortex defects
Wherein, k is more than 1 and less than J is more than p and less than o is more than J,for the (m-1) th vortex defect surrounded by the (j-k) th elliptic boundary line and the (j) th elliptic boundary line, >For the (m+1) th vortex defect surrounded by the (j+p) th elliptic boundary line and the (j+q) th elliptic boundary line,/v>The laser intensity contrast ratio of the mth vortex defect to the adjacent mth-1 vortex defect which is close to the center of the ellipse on the inner side of the mth vortex defect and the adjacent mth+1 vortex defect which is far from the center of the ellipse on the outer side of the mth vortex defect;
s53: judgment S52 step is calculatedElliptical ring spot laser intensity contrast to adjacent multiple vortex defectsWhether or not it is greater than 0.9, if +.>More than 0.9, the laser spot intensity value of the mth vortex defect is compared with the laser spot intensity values of two adjacent vortex defects to have larger difference, the detected monocrystalline silicon piece is judged to be the I-type defect monocrystalline silicon piece, the I-type defect monocrystalline silicon piece is the monocrystalline silicon piece with obvious contrast ratio of a plurality of vortex defects in the whole area of the largest concentric ellipse impure defect in the monocrystalline silicon piece image, namely, the elliptical ring spot intensity contrast ratio of the adjacent vortex defects is obvious, the gray value contrast ratio is obvious, namely, the like black, dark gray, white, black and light gray interval black heart monocrystalline silicon piece is formed by the vortex defects;
if it isLess than or equal to 0.9, the laser spot intensity value of the mth vortex defect is shown to have little difference compared with the laser spot intensity values of two adjacent vortex defects, the detected monocrystalline silicon piece is judged to be a monocrystalline silicon piece with II type defects, and the monocrystalline silicon piece with II type defects is a monocrystalline silicon piece with unobvious contrast ratio and consistent overall chromaticity of a plurality of vortex defects in the overall area of the largest concentric ellipse impure defect in the monocrystalline silicon piece image;
S54: when (when)And (3) further calculating the image gray value of the elliptical ring light spots with a plurality of adjacent vortex defects, wherein the gray value is greater than 0.9:
wherein, the liquid crystal display device comprises a liquid crystal display device,the m-1 eddy defect is an image gray value of an elliptical circular light spot with the m-1 eddy defect, and the m-1 eddy defect is an eddy defect formed by the G-th laser projection point to the H-th laser projection point; />The image gray value of the elliptical circular light spot with the mth vortex defect is the vortex defect formed by the A-th laser image point to the B-th laser image point; />The m+1st vortex defect is the vortex defect formed by the T-th laser projection point to the U-th laser projection point; n is the n-th elliptic boundary line; g is more than 1 and less than H is more than 1 and less than B is more than 1 and less than U is more than 0;
s55: comparison whenWhen the gray value is more than 0.9, the gray value of the image of the elliptical circular ring light spot with a plurality of adjacent vortex defects is larger than the gray value of the image of the elliptical circular ring light spot;
when (when)At > 0.9, if->And->Judging that the mth vortex defect is a strong composite defect area of unbalanced minority carrier flow, and has low concentration and weak luminous intensity, namely the image gray value of the mth vortex area is maximum relative to the image gray values of the mth-1 vortex area and the (m+1) th vortex area, and forming the black heart defect The impurity doped in the m-1 vortex area and the m+1 vortex area in the production process is relatively more;
when (when)At > 0.9, if->And->And judging that the mth vortex defect is a weak composite defect area of unbalanced minority carrier flow, wherein the concentration is high, the luminous intensity is high, namely, the image gray value of the mth vortex area is minimum relative to the image gray values of the mth-1 vortex area and the (m+1) th vortex area, namely, the mth vortex area is used as one area of the monocrystalline silicon wafer forming the black center defect, and impurities doped in the mth vortex area in the production process are relatively less than impurities doped in the mth-1 vortex area and the (m+1) th vortex area.
As shown in fig. 8, the present invention further provides a silicon wafer visualization detection system using the silicon wafer visualization detection method provided in any one of the embodiments, which includes a laser emission module, a CCD image acquisition module, a central control module, an image laser spot pulse focusing analysis module, an image filtering denoising module, an impurity defect overall region screening module, and an impurity defect degree calculation module;
the laser emission module is used for emitting laser to the detected monocrystalline silicon piece in real time; the laser emission module comprises a semiconductor laser emitter;
The CCD image acquisition module is used for acquiring monocrystalline silicon wafer images converged in the object plane area of the CCD camera after being filtered by the optical filter; the CCD image acquisition module comprises an optical filter, a CCD camera and a light angle sensor, and the center point of the optical filter and the center point of the CCD camera are always kept coincident with the center point of the detected monocrystalline silicon piece; the light angle sensor is used for monitoring deflection angle of vertical axis where the center point of the monocrystalline silicon piece is located in real time
As shown in fig. 9, when the silicon wafer visualization detection system provided by the invention is used for detecting monocrystalline silicon wafers, the monocrystalline silicon wafers are placed in a camera bellows, a semiconductor laser emitter of a laser emitting module is arranged below the monocrystalline silicon wafers, and a light filter and a CCD camera of a CCD image acquisition module are sequentially arranged above the monocrystalline silicon wafers;
the central control module is used for controlling the laser emission module to emit laser to the detected monocrystalline silicon piece and controlling the CCD image acquisition module to acquire the monocrystalline silicon piece image;
the image laser spot pulse focusing analysis module is used for identifying and obtaining a plurality of concentric elliptical light spots in the monocrystalline silicon wafer image and analyzing the laser spot pulse focusing condition of each laser projection point in the monocrystalline silicon wafer image acquired by the CCD camera;
The image filtering denoising module is used for performing image filtering denoising on the monocrystalline silicon wafer image acquired by the CCD camera;
the impurity defect integral region screening module is used for calculating the gray value of the largest concentric ellipse in the monocrystalline silicon wafer image and judging whether the gray value is an impurity defect integral region formed by a plurality of vortex defects;
the impurity defect degree calculation module is used for carrying out secondary identification on vortex defect areas in the monocrystalline silicon wafer image after image filtering denoising, calculating to obtain elliptical ring light spot laser intensity values of a plurality of vortex defects and elliptical ring light spot laser intensity contrast of two adjacent vortex defects, and judging the impurity defect degree of the detected monocrystalline silicon wafer caused by the vortex defects.
Further preferably, the laser emission module further comprises a light homogenizing sheet arranged below the detected monocrystalline silicon piece, and the center point of the light homogenizing sheet is always coincident with the center point of the detected monocrystalline silicon piece.
Further preferably, the laser emission module is carried by a three-dimensional moving platform, and the three-dimensional moving platform is used for driving the semiconductor laser emitter to emit light rays to be aligned with the vertical direction where the receiving light converging point of the CCD camera is located.
It should be noted that, the foregoing reference numerals of the embodiments of the present invention are merely for describing the embodiments, and do not represent the advantages and disadvantages of the embodiments. And the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, apparatus, article or method that comprises the element.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) as above, comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method of the embodiments of the present invention.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.

Claims (6)

1. The visual detection method of the silicon wafer is used for detecting vortex defects in the monocrystalline silicon wafer in real time and is characterized by comprising the following steps of:
s1: emitting laser to the detected monocrystalline silicon piece in real time;
s2: recognizing and obtaining a plurality of concentric elliptical light spots in the monocrystalline silicon wafer image, and analyzing the laser spot pulse focusing condition of each laser projection point in the monocrystalline silicon wafer image acquired by a CCD camera;
s3: performing image filtering denoising on the monocrystalline silicon wafer image acquired by the CCD camera according to the analysis result of the step S2;
s4: calculating the gray value of the maximum concentric elliptical light spot in the monocrystalline silicon wafer image, judging whether the gray value is an impurity defect integral area formed by a plurality of vortex defects, if so, carrying out the next step, otherwise, repeating the steps S1-S3;
s5: performing secondary identification on the eddy defect area in the monocrystalline silicon piece image after filtering and denoising the S3 step image, calculating to obtain elliptical ring light spot laser intensity values of a plurality of eddy defects and elliptical ring light spot laser intensity contrast of two adjacent eddy defects, and judging the impurity defect degree of the detected monocrystalline silicon piece caused by the eddy defects;
The step S2 comprises the following steps:
s21: construction of laser spot energy distribution functionAnd calculating the deflection angle of the vertical axis from the center point of the monocrystalline silicon piece as +.>Is +.>,/>
S22: according to the calculation result of the step S21, calculating and identifying the major axis radius of the j concentric elliptical light spots formed by the j elliptical boundary lines collected by the object plane of the CCD camera and converged on the object plane of the CCD camera by the Q laser image pointsAnd minor axis radius->I is an i-th laser projection point in a j-th elliptical boundary line acquired by a CCD camera object plane, i=1, 2, … and Q; j=1, 2, …, J;
s23: laser spot emitted by semiconductor laser emitterAn i-th laser projection point in a j-th elliptical boundary line in a single-crystal silicon image acquired at the object plane of a CCD camera>Performing Fourier transform to obtain the ith laser projection Fourier transform point +.>Wherein->For the ith laser projection point in the jth elliptical boundary line +.>Abscissa after fourier transformation, +.>For the ith laser projection point in the jth elliptical boundary line +.>An ordinate after the Fourier transform;
S24: calculating an ith laser projection Fourier transform point in the jth elliptical boundary line in the single crystal silicon imageImpulse response function value +.>And constructing an ith laser projection point +.in the jth elliptical boundary line in the single crystal silicon image>Fourier transform object plane domain model->
S25: computing a fourier transform object plane domain model in constructionThe ith laser projection point +_in the jth elliptical boundary line of (a)>Laser spot point pulse focusing function value of laser spot in monocrystalline silicon image obtained by CCD camera transmission imaging>
The Fourier transform object plane domain model constructed in the step S25The ith laser projection point +_in the jth elliptical boundary line of (a)>Laser spot point pulse focusing function value of laser spot in monocrystalline silicon image obtained by CCD camera transmission imaging>The formula of (2) is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,εin order to adjust the factor(s),ε=0.1 to 0.35, c is the speed of light, NA is the numerical aperture of the CCD camera,is a Gaussian factor depending on the laser irradiation level, +.>Has a value of 1 to 2.5, (-)>As a function of the attenuation of the lasing frequency,
the step S3 comprises the following steps:
s31: according to the laser spot pulse focusing function calculated in the step S2 Calculating the i laser projection point of the CCD camera in the j-th elliptic boundary line +.>Transmitting laser spot pixel values of laser spots obtained by imaging:
s32: further calculating an ith laser image point of the CCD camera pair within the jth elliptical boundary linePixel gray value of laser spot obtained by transmission imaging +.>
Wherein, the liquid crystal display device comprises a liquid crystal display device,is the laser spot pixel value +.>The value of red in the RGB color channel, +.>Is the laser spot pixel value +.>The value of green in the RGB color channel, < >>Is the laser spot pixel value +.>Blue values in the RGB color channel;
s33: the CCD camera pair obtained according to the step S32 is used for imaging the ith laser in the jth elliptical boundary linePixel gray value of laser spot obtained by transmission imaging +.>Constructing a noise pixel intensity calculation model
Wherein q=1, 2, …;
s34: judging whether the result of the noise pixel intensity calculation model calculated in the step S33 is larger than 0.85, if so, judging that the CCD camera is used for carrying out image point setting on the ith laser in the jth elliptical boundary lineTransmitting the laser light spots obtained by imaging as noise points, and filtering and removing the noise points; otherwise, the i laser projection point of the CCD camera pair in the j-th elliptical boundary line is reserved +. >Transmitting the imaged laser light spots, and repeating the steps S31-S33;
and in the step S4, the calculation formula for calculating the gray value of the maximum concentric elliptical light spot in the monocrystalline silicon wafer image is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,the gray value of the largest concentric elliptical light spot in the monocrystalline silicon wafer image is obtained; when->And judging that the maximum concentric ellipse in the monocrystalline silicon wafer image is an impurity defect integral area formed by a plurality of vortex defects, or judging that the maximum concentric ellipse in the monocrystalline silicon wafer image is not an impurity defect integral area formed by a plurality of vortex defects.
2. The visual inspection method of silicon wafer according to claim 1, wherein the laser spot energy distribution function constructed in the step S21 is as follows:
the deflection angle of the vertical axis where the center point of the monocrystalline silicon piece is located isIs +.>The following are provided:
wherein, the liquid crystal display device comprises a liquid crystal display device,Wfor the laser emission power,is the radius of the laser beam at the waist, +.>For the radius of the laser light wave at different positions along the optical axis z direction +.>Z is the coordinate of the laser light wave at the optical axis,δis the laser emission frequency; />Is the radius of the laser equiphase surfaces from the emitting point of the semiconductor laser emitter to different positions of the optical axis z, Lambda is the laser wavelength.
3. The visual inspection method of a silicon wafer according to claim 1, wherein in the step S22, the major axis radius of the j concentric elliptical spots defined by the j elliptical boundary lines collected by the object plane of the identified CCD camera and collected by the object plane of the identified CCD camera is calculated and counted to be Q laser projection points converged on the object plane of the CCD cameraAnd minor axis radius->The formula of (2) is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,is the radius of the laser beam at the waist, +.>To radius the laser light wave at different positions along the optical axis z,z is the coordinate of the laser light wave at the optical axis,δis the laser emission frequency; />Is the radius of the equiphase surface of the laser from the emitting point of the semiconductor laser emitter to the different positions of the optical axis z, +.>
4. The method according to claim 1, wherein the step S24 is performed by calculating the ith laser projection fourier transform point within the jth elliptical boundary lineImpulse response function value of (2)The formula of (2) is as follows:
an ith laser image point within a jth elliptical boundary line in the single crystal silicon image constructed in the step S24Fourier transform object plane domain model->The following are provided:
wherein, the liquid crystal display device comprises a liquid crystal display device,for the ith laser projection point in the jth elliptical boundary line +. >Fourier transform frequency domain values in the object plane of the CCD camera.
5. The visual inspection method of silicon wafer according to claim 1, wherein the step S5 comprises the steps of:
s51: calculating an elliptical ring spot laser intensity value of an elliptical ring area of an mth vortex defect in the multiple vortex defects in the monocrystalline silicon wafer image after filtering and denoising the S3 step imageThe mth vortex defect is surrounded by the jth elliptic boundary line and the jth+p elliptic boundary line, m=1, 2, …, M; j is more than 1 and less than p is more than J;
wherein, the laser intensity value of the elliptical circular ring light spot with the mth vortex defectIn the calculation formula, i=A, A+1, A+2, …, B,1 < A < B < Q,>calculating standard deviation adjustment coefficients for the laser intensity of the elliptical circular ring light spot with the mth vortex defect,,/>is->And->Is a correlation coefficient of (2); />For the ith laser projection point in the mth vortex defect +.>And the peak point of the laser spot in the laser beam>The difference between the x-axis coordinates of (c); />For the ith laser projection point in the mth vortex defect +.>And the peak point of the laser spot in the laser beamThe difference between the y-axis coordinates of (c); />For the m-th vortex defect a plurality of laser projection points and the peak point of the laser spot therein +. >Standard deviation between x-axis coordinates of (c); />A plurality of laser projection points +.>Laser spot peak point +.>Standard deviation between y-axis coordinates of (c);
s52: calculating the laser intensity contrast of elliptical circular ring light spots with a plurality of adjacent vortex defects
Wherein, k is more than 1 and less than J is more than p and less than o is more than J,for the (m-1) th vortex defect surrounded by the (j-k) th elliptic boundary line and the (j) th elliptic boundary line,>is formed by j+p elliptic boundary lines and j+qM+1st vortex defect surrounded by elliptic boundary line,>the laser intensity contrast ratio of the mth vortex defect to the adjacent mth-1 vortex defect which is close to the center of the ellipse on the inner side of the mth vortex defect and the adjacent mth+1 vortex defect which is far from the center of the ellipse on the outer side of the mth vortex defect;
s53: judging the laser intensity contrast of the elliptical ring light spots of the adjacent vortex defects calculated in the step S52Whether or not it is greater than 0.9, if +.>Judging that the detected monocrystalline silicon piece is a monocrystalline silicon piece with I type defects, wherein the monocrystalline silicon piece with I type defects is a monocrystalline silicon piece with obvious contrast of a plurality of vortex defects;
if it isJudging that the detected monocrystalline silicon piece is a class II defect monocrystalline silicon piece and the class II defect monocrystalline silicon piece is a monocrystalline silicon piece with a plurality of vortex defects and unobvious contrast;
S54: when (when)And (3) further calculating the image gray value of the elliptical ring light spots with a plurality of adjacent vortex defects, wherein the gray value is greater than 0.9:
wherein, the liquid crystal display device comprises a liquid crystal display device,the m-1 eddy defect is an image gray value of an elliptical circular light spot with the m-1 eddy defect, and the m-1 eddy defect is an eddy defect formed by the G-th laser projection point to the H-th laser projection point; />The image gray value of the elliptical circular light spot with the mth vortex defect is the vortex defect formed by the A-th laser image point to the B-th laser image point; />The m+1st vortex defect is the vortex defect formed by the T-th laser projection point to the U-th laser projection point; n is the n-th elliptic boundary line; g is more than 1 and less than H is more than 1 and less than B is more than 1 and less than U is more than 0;
s55: comparison whenWhen the gray value is more than 0.9, the gray value of the image of the elliptical circular ring light spot with a plurality of adjacent vortex defects is larger than the gray value of the image of the elliptical circular ring light spot;
if it isAnd->Judging the mth vortex defect as a strong composite defect area of unbalanced minority carrier flow;
if it isAnd->Then determine the mth vortexThe defect is a weak composite defect area of unbalanced minority carrier flow.
6. The visual detection system for the silicon wafer adopting the method as claimed in any one of claims 1 to 5 is characterized by comprising a laser emission module, a CCD image acquisition module, a central control module, an image laser spot pulse focusing analysis module, an image filtering denoising module, an impurity defect integral area screening module and an impurity defect degree calculation module;
The laser emission module is used for emitting laser to the detected monocrystalline silicon piece in real time; the laser emitting module comprises a semiconductor laser emitter;
the CCD image acquisition module is used for acquiring monocrystalline silicon wafer images converged in the object plane area of the CCD camera after being filtered by the optical filter; the CCD image acquisition module comprises an optical filter, a CCD camera and a light angle sensor, wherein the center point of the optical filter and the center point of the CCD camera are always coincided with the center point of the detected monocrystalline silicon piece; the light angle sensor is used for monitoring deflection angle of the vertical axis where the center point of the monocrystalline silicon piece is located in real time
The central control module is used for controlling the laser emission module to emit laser to the detected monocrystalline silicon piece and controlling the CCD image acquisition module to acquire the monocrystalline silicon piece image;
the image laser spot pulse focusing analysis module is used for identifying and obtaining a plurality of concentric elliptical light spots in the monocrystalline silicon wafer image and analyzing the laser spot pulse focusing condition of each laser projection point in the monocrystalline silicon wafer image acquired by the CCD camera;
the image filtering denoising module is used for performing image filtering denoising on the monocrystalline silicon wafer image acquired by the CCD camera;
The impurity defect integral region screening module is used for calculating the gray value of the largest concentric ellipse in the monocrystalline silicon wafer image and judging whether the gray value is an impurity defect integral region formed by a plurality of vortex defects;
the impurity defect degree calculation module is used for carrying out secondary identification on vortex defect areas in the monocrystalline silicon wafer image after image filtering denoising, calculating to obtain elliptical ring light spot laser intensity values of a plurality of vortex defects and elliptical ring light spot laser intensity contrast of two adjacent vortex defects, and judging impurity defect degree of the detected monocrystalline silicon wafer caused by the vortex defects.
CN202311338995.8A 2023-10-17 2023-10-17 Visual detection method and system for silicon wafer Active CN117116799B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311338995.8A CN117116799B (en) 2023-10-17 2023-10-17 Visual detection method and system for silicon wafer

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311338995.8A CN117116799B (en) 2023-10-17 2023-10-17 Visual detection method and system for silicon wafer

Publications (2)

Publication Number Publication Date
CN117116799A CN117116799A (en) 2023-11-24
CN117116799B true CN117116799B (en) 2024-02-13

Family

ID=88800371

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311338995.8A Active CN117116799B (en) 2023-10-17 2023-10-17 Visual detection method and system for silicon wafer

Country Status (1)

Country Link
CN (1) CN117116799B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103147122A (en) * 2006-05-19 2013-06-12 Memc电子材料有限公司 Controlling agglomerated point defect and oxygen cluster formation induced by the lateral surface of silicon single crystal during CZ growth
CN112113977A (en) * 2011-07-12 2020-12-22 科磊股份有限公司 Wafer inspection
CN115266758A (en) * 2022-09-27 2022-11-01 苏州高视半导体技术有限公司 Wafer detection system, wafer detection method, electronic device and storage medium
CN116235205A (en) * 2020-09-04 2023-06-06 科磊股份有限公司 Grading enhancement defect detection method for three-dimensional wafer structure
CN116564844A (en) * 2023-05-26 2023-08-08 江苏第三代半导体研究院有限公司 Semiconductor test method, electronic device, storage medium, and program product

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103147122A (en) * 2006-05-19 2013-06-12 Memc电子材料有限公司 Controlling agglomerated point defect and oxygen cluster formation induced by the lateral surface of silicon single crystal during CZ growth
CN112113977A (en) * 2011-07-12 2020-12-22 科磊股份有限公司 Wafer inspection
CN116235205A (en) * 2020-09-04 2023-06-06 科磊股份有限公司 Grading enhancement defect detection method for three-dimensional wafer structure
CN115266758A (en) * 2022-09-27 2022-11-01 苏州高视半导体技术有限公司 Wafer detection system, wafer detection method, electronic device and storage medium
CN116564844A (en) * 2023-05-26 2023-08-08 江苏第三代半导体研究院有限公司 Semiconductor test method, electronic device, storage medium, and program product

Also Published As

Publication number Publication date
CN117116799A (en) 2023-11-24

Similar Documents

Publication Publication Date Title
WO2010125911A1 (en) Defect inspection device and defect inspection method
US20060146316A1 (en) Method for monitoring micro-lens curvature in-line
CN106770128A (en) Quick three-dimensional detects the detection means and detection method of subsurface defect of optical element
US20150029324A1 (en) Substrate inspection method, substrate manufacturing method and substrate inspection device
CN206348270U (en) Quick three-dimensional detects the detection means of subsurface defect of optical element
EP3699576A1 (en) Image reconstruction method and device, and microscope imaging device
CN109974648B (en) Method for evaluating geometric accuracy of micro-hole based on coordinate value
CN111899215A (en) Method for extracting optical element body defects
CN109584259B (en) Quartz crucible bubble layered counting device and method
CN115428125A (en) Inspection device and inspection method
CN117116799B (en) Visual detection method and system for silicon wafer
US20150030230A1 (en) Substrate inspection method, substrate manufacturing method and substrate inspection device
CN113781458A (en) Artificial intelligence based identification method
CN112129734B (en) Depth-distinguishable fluorescence imaging system
CN106814074B (en) A kind of cladding nuclear fuels surface defects detection system and method
CN116091505B (en) Automatic defect detection and classification method and system for sapphire substrate
TW201812705A (en) Optical measurement of step size and plated metal thickness
JP2010025836A (en) Visual inspection method, visual inspection device, semiconductor inspection device, and semiconductor wafer cross-section inspection device
CN116452519A (en) Tyre bulge flaw detection method based on multi-mode information fusion
CN102890421B (en) The optimization method of detection method, detection system and photoetching process that photoetching defocuses
CN112581424B (en) Classification extraction method for surface and subsurface defects of optical element
CN112858341B (en) Detection method, shooting system and detection system
CN107037335A (en) A kind of ultraviolet power equipment measures the method and device of discharge fault true photon number
CN103884711B (en) The method of testing in a kind of oxygen/iodine supersonic speed heat of mixing flow field
CN113091898A (en) Laser beam quality measuring method based on scattered light imaging 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