CN116740073A - Solar cell defect detection method and system based on visual image of graphite boat - Google Patents
Solar cell defect detection method and system based on visual image of graphite boat Download PDFInfo
- Publication number
- CN116740073A CN116740073A CN202311031443.2A CN202311031443A CN116740073A CN 116740073 A CN116740073 A CN 116740073A CN 202311031443 A CN202311031443 A CN 202311031443A CN 116740073 A CN116740073 A CN 116740073A
- Authority
- CN
- China
- Prior art keywords
- image
- representing
- graphite boat
- value
- gray
- 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
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 title claims abstract description 96
- 229910002804 graphite Inorganic materials 0.000 title claims abstract description 95
- 239000010439 graphite Substances 0.000 title claims abstract description 94
- 230000007547 defect Effects 0.000 title claims abstract description 64
- 238000001514 detection method Methods 0.000 title claims abstract description 41
- 230000000007 visual effect Effects 0.000 title claims abstract description 28
- 230000011218 segmentation Effects 0.000 claims abstract description 64
- 238000000034 method Methods 0.000 claims abstract description 27
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 239000011159 matrix material Substances 0.000 claims description 23
- 238000006073 displacement reaction Methods 0.000 claims description 16
- 238000009499 grossing Methods 0.000 claims description 10
- 238000001914 filtration Methods 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 7
- 230000008859 change Effects 0.000 claims description 6
- 230000036961 partial effect Effects 0.000 claims description 6
- 230000009467 reduction Effects 0.000 claims description 6
- 239000013598 vector Substances 0.000 claims description 6
- 230000002401 inhibitory effect Effects 0.000 claims description 2
- 230000005764 inhibitory process Effects 0.000 claims description 2
- 230000001629 suppression Effects 0.000 claims description 2
- 238000012545 processing Methods 0.000 abstract description 11
- 230000006870 function Effects 0.000 description 9
- 238000004519 manufacturing process Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 6
- 238000003708 edge detection Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000010191 image analysis Methods 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000003908 quality control method Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 239000003575 carbonaceous material Substances 0.000 description 1
- 239000012876 carrier material Substances 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000008021 deposition Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000003628 erosive effect Effects 0.000 description 1
- 239000007770 graphite material Substances 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000002829 reductive effect Effects 0.000 description 1
- 230000000452 restraining effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- 239000010409 thin film Substances 0.000 description 1
- 238000011179 visual inspection Methods 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
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
- G06T2207/20192—Edge enhancement; Edge preservation
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P70/00—Climate change mitigation technologies in the production process for final industrial or consumer products
- Y02P70/50—Manufacturing or production processes characterised by the final manufactured product
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a solar cell defect detection method and a system based on a visual image of a graphite boat, belonging to the technical field of image data processing, wherein the method comprises the following steps: acquiring a graphite boat image; preprocessing the graphite boat image; calculating an optimal segmentation threshold of the graphite boat image; dividing a plurality of battery slice images from the graphite boat image according to the optimal dividing threshold; according to the positions of the plurality of separated battery piece images, when no battery piece exists in a boat slot at a certain position, determining that the boat slot has an empty piece condition; extracting the outline of the battery piece in the battery piece image; performing linear fitting on the outline of the battery piece; when the linear fitting result of the cell outline is a straight line, determining that the cell is defect-free; when the linear fitting result of the contour of the battery piece is a curve, calculating the curvature of the curve; when the curvature of the curve is larger than the preset curvature, determining that the battery piece has edge warping defects. The invention can improve the detection efficiency of appearance defects of large-scale battery plates.
Description
Technical Field
The invention belongs to the technical field of image data processing, and particularly relates to a method and a system for detecting defects of a solar cell slice based on a visual image of a graphite boat.
Background
Along with the rapid development of clean energy, the photovoltaic power generation is promoted and applied more with the flexibility and environmental protection. The quality of the solar cell as the smallest unit of the photovoltaic system has a huge influence on the whole photovoltaic system. The internal defects and the appearance defects of the solar cell not only affect the photoelectric conversion efficiency, but also damage the photovoltaic module under severe conditions, so that the quality of the solar cell is necessarily monitored by a photovoltaic enterprise.
With the continuous innovation of the technology, the appearance defect detection of the solar cell is gradually changed from a manual visual inspection mode to an automatic detection mode, and the appearance defect algorithm based on deep learning is widely applied due to high accuracy.
However, the appearance defect algorithm based on deep learning needs to photograph the battery pieces one by one when detecting the defects of the battery pieces, and then sends the battery pieces into a trained appearance defect detection model for detection, but photographing one by one still needs a lot of time cost, is difficult to be qualified for detecting the appearance defects of the battery pieces on a large scale, and has low detection efficiency for the appearance defects of the battery pieces on a large scale.
Disclosure of Invention
In order to solve the technical problems that in the prior art, a deep learning-based appearance defect algorithm needs to photograph the battery pieces one by one, a large amount of time and cost are still needed, appearance defect detection of large-scale battery pieces is difficult to be qualified, and the detection efficiency of the appearance defects of the large-scale battery pieces is low, the invention provides a solar battery piece defect detection method and system based on a graphite boat visual image.
First aspect
The invention provides a solar cell defect detection method based on a visual image of a graphite boat, which comprises the following steps:
s101: acquiring an image of a graphite boat, wherein a plurality of boat grooves are arranged in the graphite boat in an array manner, and the boat grooves are used for placing solar cells;
s102: preprocessing the graphite boat image, wherein the preprocessing comprises the following steps: filtering noise reduction, brightness equalization and contrast equalization;
s103: calculating an optimal segmentation threshold of the graphite boat image;
s104: dividing a plurality of battery piece images with gray values larger than the optimal dividing threshold value from the graphite boat image according to the optimal dividing threshold value;
s105: according to the positions of the plurality of separated battery piece images, when no battery piece exists in a boat slot at a certain position, determining that the boat slot has an empty piece condition;
s106: extracting a battery piece outline in the battery piece image;
s107: performing linear fitting on the contour of the battery piece;
s108: when the linear fitting result of the battery piece outline is a straight line, determining that the battery piece is defect-free; when the linear fitting result of the battery piece outline is a curve, calculating the curvature of the curve;
s109: and when the curvature of the curve is larger than the preset curvature, determining that the battery piece has the edge warping defect.
Second aspect
The invention provides a solar cell defect detection system based on a visual image of a graphite boat, which is used for executing the solar cell defect detection method based on the visual image of the graphite boat in the first aspect.
Compared with the prior art, the invention has at least the following beneficial technical effects:
(1) According to the invention, a large number of solar cells can be arranged in the boat groove of the graphite boat, the outline of the cells in the graphite boat image is extracted by acquiring the graphite boat image, whether the cells have edge warping defects or not is determined according to the outline of the cells, the large number of solar cells can be detected simultaneously, photographing one by one is not needed, the time cost is saved, the appearance defect detection of the large-scale cells can be realized, and the detection efficiency of the appearance defect of the large-scale cells is improved.
(2) According to the method and the device for detecting the defects of the solar cells, the optimal segmentation threshold value of the graphite boat image is calculated, and a plurality of cell images with gray values larger than the optimal segmentation threshold value are segmented from the graphite boat image according to the optimal segmentation threshold value, so that the segmentation accuracy of the cell images can be improved, and the defect detection accuracy of the solar cells can be further improved.
Drawings
The above features, technical features, advantages and implementation of the present invention will be further described in the following description of preferred embodiments with reference to the accompanying drawings in a clear and easily understood manner.
Fig. 1 is a schematic flow chart of a method for detecting defects of a solar cell sheet based on a visual image of a graphite boat.
Fig. 2 is a schematic diagram of a visual image of a graphite boat provided by the invention.
Detailed Description
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following description will explain the specific embodiments of the present invention with reference to the accompanying drawings. It is evident that the drawings in the following description are only examples of the invention, from which other drawings and other embodiments can be obtained by a person skilled in the art without inventive effort.
For simplicity of the drawing, only the parts relevant to the invention are schematically shown in each drawing, and they do not represent the actual structure thereof as a product. Additionally, in order to simplify the drawing for ease of understanding, components having the same structure or function in some of the drawings are shown schematically with only one of them, or only one of them is labeled. Herein, "a" means not only "only this one" but also "more than one" case.
It should be further understood that the term "and/or" as used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
In this context, it should be noted that the terms "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected, unless otherwise explicitly stated and defined. Either mechanically or electrically. Can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.
In addition, in the description of the present invention, the terms "first," "second," and the like are used merely to distinguish between descriptions and are not to be construed as indicating or implying relative importance.
Example 1
In one embodiment, referring to fig. 1 of the specification, a schematic flow chart of a method for detecting defects of a solar cell sheet based on a visual image of a graphite boat provided by the invention is shown.
The invention provides a solar cell defect detection method based on a visual image of a graphite boat, which comprises the following steps:
s101: and obtaining an image of the graphite boat.
The graphite boat is also called a graphite boat or a carbon boat, and is a special carrier material for preparing solar cells. Graphite boats are typically made of graphite or carbon materials, have good thermal conductivity and chemical stability, and are suitable for use in high temperature processes. In the production process of solar cells, graphite boats play an important role. In the manufacture of solar cells, it is necessary to place the solar cells or other substrate materials in the boat slot of a graphite boat and then perform a series of process steps such as heat treatment, deposition of thin films, etc. The graphite boat plays roles of supporting the base material and conducting heat in the process, so that the stability and the efficiency of the solar cell are ensured.
The graphite boat is provided with a plurality of boat grooves in an array, and the boat grooves are used for placing solar cells.
Specifically, graphite boat images may be captured by an industrial camera.
Referring to fig. 2 of the specification, a schematic diagram of a visual image of a graphite boat provided by the invention is shown.
In fig. 2, the reference number 1 refers to a graphite boat, wherein the reference number 2 of the array distribution refers to solar cells, and the image of the solar cells is generally rectangular.
S102: and preprocessing the graphite boat image.
Wherein the preprocessing comprises the following steps: filtering noise reduction, brightness equalization, and contrast equalization.
In one possible implementation, S102 specifically includes substeps S1021 to S1028:
s1021: and filtering and denoising the graphite boat image according to the following formula:
wherein, f m (x,y) Representing pixel points [ ]x,y) The filtered and denoised pixel values at that location,f(x,y) Representing pixel points [ ]x,y) The original pixel value at which is to be found,Arepresenting the length of the image of the graphite boat,Brepresenting the width of the image of the graphite boat,PIan image of the graphite boat is shown.
It should be noted that the neighborhood average filtering may reduce noise in the image, especially some random noise or small-sized noise. By calculating the average of the pixel values in the neighborhood around the pixel point, irregular fluctuations and burst noise in the image can be smoothed, thereby reducing the influence of noise. Further, detail can be smoothed, edge information can be protected, and meanwhile, the method has the characteristics of simplicity and easiness in implementation.
S1022: the radius of the luminance balancing circular structure is determined according to the following formula:
wherein, Rthe radius of the luminance balancing circular structure is shown,r min representing the radius minimum of the luminance balancing circular structure,r max represents the maximum value of the radius of the luminance balancing circular structure,S min representing the minimum area of the cell area,S max indicating the maximum surface of the cell areaAnd (3) accumulation.
It will be appreciated that the radius of the luminance balancing circular structure is within [r min ,r max ]Is dependent on the area extent of the cell region.
S1023: with preset step lengthsDetermining a plurality of brightness balance circular structures, wherein the number of the brightness balance circular structures is as follows:
wherein, Lthe number of luminance equalizing circular structures is represented,the representation is rounded up and down to the top,srepresenting a preset step size.
It will be appreciated that the step size is presetsDetermining a plurality of brightness equalization circular structures can ensure that a sufficient number of brightness equalization circular structures are obtained to cover the entire graphite boat image.
Wherein, the person skilled in the art can set the preset step length according to the actual situationsThe size of (3) is not limited in the present invention.
S1024: through a plurality of brightness equalization circular structures, carrying out brightness equalization on the graphite boat image after filtering and noise reduction:
wherein, f t (x,y) Representing pixel points [ ]x,y) The pixel value after the brightness is balanced is located,f m (x,y) Representing pixel points [ ]x,y) The filtered and denoised pixel values at that location,the gray scale closing operation is represented by a gray scale closing operation,c(r i ) Represent the firstiA circular structure with balanced brightness.
It will be appreciated that for each luminance balancing circular structurec(r i ) Calculate its radiusr i And applies the method to each pixel point in the graphite boat imagex,y) Obtaining pixel values after brightness equalization through brightness equalization operationft(x,y)。
The gray level closing operation is an image processing operation, and is commonly used for removing small noise points in an image, filling small holes in the image or connecting broken image areas. The gray scale closing operation combines two operations of gray scale expansion and gray scale erosion, and can maintain the overall shape of the image and smooth the boundary of the image.
The processing steps of the brightness equalization circular structure can equalize the brightness distribution of the image, enhance details, reduce noise influence and protect edge information. These advantages can improve the quality and visual effect of the image, and facilitate the accuracy and reliability of subsequent hidden crack detection and analysis tasks.
S1025: constructing an energy change function:
wherein, E(W) The energy variation function is represented by a function of the energy variation,Xrepresenting pixel points [ ]x,y) At the position oftThe gray value of the time in question,Wthe displacement vector is represented by a vector of the displacement,urepresentation ofxThe amount of displacement in the direction is determined,vrepresentation ofyThe amount of displacement in the direction is determined,lrepresentation oftThe amount of displacement in the direction is determined,representing a three-dimensional gradient sign>Consistency parameter representing gray value, +.>Representing displacement vectorsWGradient of->Representing euclidean normsSymbol (S)>A smoothness parameter representing the gray value.
Wherein the energy variation function is typically calculated by comparing pixel values at different points in time at different locations in the image. The energy variation function may be used to describe image features such as moving objects, texture variations, edge extraction, illumination variations, etc.
S1026: constructing an Euler Lagrangian equation according to the energy change function:
wherein, 。
s1027: solving Euler Lagrangian equation, calculating horizontal component of energy change function in space domainu(x,y) And a vertical componentv(x,y)：
Wherein, representation ofuIs used for the average value of (a),urepresentation ofxDisplacement of direction,/->Representation ofvIs used for the average value of (a),vrepresentation ofyThe amount of displacement in the direction is determined,F x representation ofFFor a pair ofxIs used for the partial derivative of (a),F y representation ofFFor a pair ofyIs used for the partial derivative of (a),F t representation ofFFor a pair oftIs a partial derivative of (c).
S1028: horizontal component in the spatial domain according to an energy variation functionu(x,y) And a vertical componentv(x,y) Contrast equalization is performed:
wherein, f d (x,y) Representing pixel points [ ]x,y) Pixel values after contrast equalization.
It should be noted that, by solving the energy change function and performing the contrast balancing operation, the visual effect of the image can be improved, the edge information and details can be enhanced, and the quality and the observability of the image can be improved. The contrast balance has important significance for image processing tasks such as hidden crack detection and the like.
In the invention, the battery pieces of different production processes and the graphite boats of different specifications can be uniformly standardized by adopting filtering noise reduction, brightness balance and contrast balance, so that the applicability is high, and the popularization of the surface defect detection technology is facilitated.
S103: and calculating an optimal segmentation threshold of the graphite boat image.
In particular, the gray scale distribution in the image may be different for different specifications of graphite boats and different production processes of the battery plate. By calculating the optimal segmentation threshold, a proper threshold can be adaptively determined according to the gray characteristic of the image, so that the segmentation accuracy is improved.
In one possible embodiment, S103 specifically includes substeps S1031 to S1037:
s1031: and converting the pretreated graphite boat image into a gray level histogram.
Specifically, the values of RGB channels of the graphite boat image are added and converted into a gray-scale histogram by a certain weight, for example, using an average method or a weighting method.
S1032: an initial segmentation threshold is set, an area with a gray value larger than the segmentation threshold is used as a foreground area, and an area with a gray value smaller than or equal to the segmentation threshold is used as a background area.
It should be noted that the gray level of the solar cell in the image is often larger than the gray level of the background, the foreground region may be understood as a cell region, and the background region may be understood as a non-cell region.
S1033: calculating the gray probability of a foreground region and a background region under the current segmentation threshold value:
wherein, arepresenting the current segmentation threshold value of the current segment,，nrepresenting the total number of gray levels>Indicating when the segmentation threshold isaGray probability of background area, < >>Indicating when the segmentation threshold isaThe gray probability of the foreground region at the time,ithe gray level index value is represented and,N i representing gray scale levels asiThe number of pixels in the time period,Nthe total number of pixels of the graphite image is represented.
S1034: calculating the gray average value of the foreground region and the background region under the current segmentation threshold value:
wherein, indicating when the segmentation threshold isaGray mean value of background area +.>Indicating when the segmentation threshold isaGray level mean value of foreground region +.>Indicating when the segmentation threshold isaAnd the gray average value of the whole graphite boat image.
S1035: calculating an inter-class variance between the background region and the foreground region at a current segmentation threshold:
wherein, representing the inter-class variance.
Wherein the inter-class variance is used to measure the degree of difference between the classes when a set of data (or image pixels) is divided into different classes.
Further, in one possible implementation, S1035 is specifically:
when the gray scale weight and neighborhood information are introduced, the formula for calculating the inter-class variance between the background region and the foreground region at the current segmentation threshold can be updated as:
wherein, representing the inter-class variance when introducing gray scale weights and neighborhood information, < >>Representing when the gray level is a segmentation thresholdaThe average value of the number of pixels in the neighborhood,N a m- representing gray level as segmentation thresholda-mThe number of pixels in the time period,N a-1 representing gray level as segmentation thresholdaThe number of pixels at-1,N a representing gray level as segmentation thresholdaThe number of pixels in the time period,N a+1 representing gray level as segmentation thresholdaThe number of pixels at +1,N a m+ representing gray level as segmentation thresholda+mThe number of pixels in the time period,Nthe total number of pixels representing the graphite image,mrepresenting neighborhood parameters>Representing the gray scale weights.
In conventional inter-class variance calculations, the weights of each pixel are considered equal, i.e., all pixels contribute the same amount to the inter-class variance. However, in some cases, the contributions of the pixels of different gray levels to the segmentation may not be the same. Some pixels may be more important or more representative in the image, while other pixels may be more noisy or less important.
In the invention, the gray scale weights are introduced to the pixel points according to the importance or the representativeness of the pixel points, so that the important pixel points are more influenced when the inter-class variance is calculated, and the accuracy of image segmentation is enhanced. Further, the neighborhood information is introduced to further consider the relationship between the pixel point and the surrounding pixel points. In an image, there is typically a high similarity between adjacent pixels, because the gray scale variation of an object is typically smooth. Therefore, by introducing neighborhood information, the gray value difference between the pixel point and the surrounding pixel points can be considered, so that the boundary between the target object and the background in the image can be better described.
S1036: other segmentation thresholds are sequentially selected, and S1033 to S1036 are repeated, and an inter-class variance between the background region and the foreground region is calculated under each segmentation threshold.
S1037: and taking the corresponding segmentation threshold value when the inter-class variance is maximum as the optimal segmentation threshold value.
When the inter-class variance is maximum, this means that the gray scale difference between the foreground and the background reaches the maximum, that is, the segmentation effect is optimal. Therefore, the segmentation threshold corresponding to the maximum inter-class variance is reasonably selected as the optimal segmentation threshold, because the threshold can separate the foreground from the background to the greatest extent, and the optimal image segmentation result is obtained.
S104: and according to the optimal segmentation threshold, segmenting a plurality of battery piece images with gray values larger than the optimal segmentation threshold from the graphite boat image.
It should be noted that, the gray level of the solar cell in the graphite boat image is often larger than the gray level of the background, and therefore, an image with a gray level value larger than the optimal segmentation threshold value may be understood as a cell image.
S105: and determining that the boat slot has empty pieces according to the positions of the plurality of divided battery piece images when no battery piece exists in the boat slot at a certain position.
In the production process of solar cells, empty sheets are a common defect. By detecting whether empty chips exist in the boat groove, defects in production can be found in time, and quality control is facilitated. And the empty sheet problem is found and treated early, so that the quantity of unqualified products is reduced, and the quality and the overall yield of the solar cell sheet are improved.
It can be understood that under normal conditions, one boat slot corresponds to one solar cell, and whether each boat slot should be provided with a cell is judged according to the boat slot layout rule of the graphite boat. If a battery piece image exists in a certain boat slot, the battery piece is correctly placed in the boat slot; and if no battery piece image exists in a certain boat slot, indicating that the boat slot has empty piece condition.
S106: and extracting the outline of the battery piece in the battery piece image.
Specifically, the battery slice contours in the battery slice images may be extracted by a Canny operator, a Sobel operator, or the like.
In one possible embodiment, S106 specifically includes substeps S1061 to S1066:
s1061: introducing a Sobel operator to acquire a horizontal feature matrix and a vertical feature matrix of the battery piece image:
wherein, S x representing the horizontal feature matrix of the device,S y representing a vertical feature matrix.
Specifically, a horizontal feature matrixS x Representing each pixel point in an imageHorizontal direction gradient values, horizontal feature matrixS x We can be helped to find edges in the horizontal direction in the image. Vertical feature matrixS y Representing the vertical gradient value of each pixel point in the image and the vertical feature matrixS y We can be helped to find edges in the vertical direction in the image.
S1062: calculating a horizontal gradient and a vertical gradient according to the horizontal feature matrix and the vertical feature matrix:
wherein, G x a gradient in the horizontal direction is indicated,G y a gradient in the vertical direction is indicated,Iand a gray value matrix representing the cell image.
S1063: calculating the gradient intensity and gradient direction of the pixel points:
wherein, Gthe intensity of the gradient is indicated and,θindicating the gradient direction.
It should be noted that, by introducing a Sobel operator and calculating the horizontal and vertical feature matrices of the battery slice image, gradient information of the image in the horizontal and vertical directions can be obtained. Then, the gradient intensity and the gradient direction of the pixel point are calculated according to the gradient characteristics, so that the edge detection and the characteristic extraction of the image are realized. Such gradient information is important for subsequent image analysis and processing steps.
S1064: when the battery piece image has a plurality of gradient information, reserving a maximum value pixel point and restraining a non-maximum value pixel point.
In one possible embodiment, S1064 specifically includes:
four preset edge angles are selected to be respectively。
When the gradient strength of the pixel points is larger than the gradient of the preset edge angle, determining that the current pixel point is the maximum value pixel point to be reserved, otherwise, determining that the current pixel point is the non-maximum value pixel point to be restrained.
In the invention, the processing mode of reserving the maximum pixel point and inhibiting the non-maximum pixel point is beneficial to improving the image edge detection effect, enhancing the image characteristics and reducing the influence of noise, thereby obtaining more accurate and clear image edge information and providing a better foundation for the subsequent image processing and analysis tasks.
S1065: setting a high threshold and a low threshold, and determining the pixel point as a strong-edge pixel point when the gradient strength of the pixel point is greater than the high threshold; when the gradient strength of the pixel points is between the low threshold value and the high threshold value, determining the pixel points as weak edge pixel points; and when the edge pixel gradient value of the pixel point is smaller than the low threshold value, suppressing the pixel point.
Further, in one possible implementation, the high threshold and the low threshold are determined in the following manner:
wherein, THindicating a high threshold value of the value,TLindicating a low threshold value of the value,H* Gray value matrix representing battery plate image after non-maximum value inhibition, max #H* ) The maximum value in the gray value matrix of the cell image after non-maximum value suppression is represented.
It should be noted that, setting the high threshold to be a certain multiple of the maximum value in the gray value matrix and the low threshold to be another smaller multiple of the high threshold can clearly distinguish between the strong edge pixel and the weak edge pixel in the edge detection, so as to improve the accuracy and the reliability of the edge detection.
S1066: and connecting the strong edge pixel points to obtain the outline of the battery piece.
According to the invention, the outline of the battery piece in the image can be extracted, the battery piece is distinguished from the background, and a foundation is provided for subsequent tasks such as image analysis and defect detection. Edge detection is a key step in image processing, can help us capture important characteristic and structural information in images, and has important significance for analysis and quality control of battery slice images. The extracted outline of the battery piece can be used for carrying out tasks such as shape analysis, size measurement, defect detection and the like of the battery piece, so that the quality of the battery piece is evaluated and controlled.
S107: and performing linear fitting on the contour of the battery piece.
In the traditional fitting mode, the least square method is the most commonly used one, however, the least square method may be sensitive to noise points in the fitting process and is easy to be interfered by abnormal points, so that the fitting result is not smooth enough. Thus, the present invention innovatively provides a novel linear fit approach.
In one possible embodiment, S107 specifically includes substeps S1071 to S1076:
s1071: four sides in the cell outline are determined.
S1072: for the start point and the end point of each edge, adjacent edges are selectedpAnd (3) carrying out coordinate smoothing on the contour points:
wherein, represents the smoothed abscissa, +.>Representing the ordinate after the smoothing process, (-)x 1 ,y 1 )、(x 2 ,y 2 ) And%x p ,y p ) Representing the coordinates of the contour points involved in the coordinate smoothing process,prepresenting the number of adjacent contour points.
It should be noted that some noise may be in the start point or end point of the cell profileBy selecting adjacent points or incomplete edge pointspThe contour points are subjected to coordinate smoothing processing, so that noise points can be removed or weakened, and the obtained contour is smoother and more accurate.
S1073: according to the coordinates of the starting point and the ending point after the coordinate smoothing treatment, a straight line is fitted, and the straight line can be expressed as:
wherein, ythe vertical coordinate is indicated by the vertical coordinate,xthe horizontal axis of the drawing is indicated,kthe slope is indicated as such,brepresenting the intercept.
S1074: uniformly selecting contour points except for a starting point and an ending point, and calculating the distance from each contour point to a fitted straight line:
wherein, d i represent the firstiDistance from each contour point to the fitted straight line, (-)x i ,y i ) Represent the firstiCoordinates of the contour points.
S1075: and counting the number of contour points with the distance to the fitted straight line smaller than the preset distance.
The size of the preset distance can be set by a person skilled in the art according to practical situations, and the invention is not limited.
S1076: when the number of contour points smaller than the preset distance is larger than the preset number, determining that the edge to be fitted is a straight line, and finishing fitting; and when the number of the contour points smaller than the preset distance is smaller than or equal to the preset number, determining the edge to be fitted as a curve, equally dividing the edge to be fitted into two sections, re-executing S1072 to S1076, and continuing fitting until the fitting of the whole edge is completed.
Wherein, the person skilled in the art can set the preset number of sizes according to the actual situation, and the invention is not limited.
In the present invention, by selectingAdjacent to each otherpThe contour points are subjected to coordinate smoothing processing, so that noise points can be removed or weakened, and the obtained contour is smoother and more accurate. And moreover, the edge to be fitted can be judged to be a straight line or a curve according to the fitting result, the fitting is continued according to the situation, and the adaptability of the algorithm is stronger.
S108: when the linear fitting result of the cell outline is a straight line, the cell is determined to be defect-free. When the linear fitting result of the cell outline is a curve, the curvature of the curve is calculated.
In one possible implementation, S108 specifically includes:
the curvature of the curve is calculated by the following formula:
wherein, Kthe curvature is indicated by the fact that,f(x) The curve to be fitted is shown as such,represents the first derivative of the curve,/>Representing the second derivative of the curve.
S109: when the curvature of the curve is larger than the preset curvature, determining that the battery piece has edge warping defects.
The edge-lifting defect refers to that the edge or the surrounding area of the battery piece is bent upwards or downwards, namely the edge or the surrounding area of the battery piece is uneven or convex. Such edge lifting may be a defect or a bad phenomenon in the production process, which may affect the quality and performance of the battery sheet.
It should be noted that, the occurrence of edge warpage means that the curvature of the profile curve becomes larger, so that when the curvature of the curve is larger than the preset curvature, it can be determined that the edge warpage defect exists in the battery piece.
In one possible embodiment, the method for detecting defects of solar cells based on the visual image of the graphite boat further comprises:
s110: and outputting a blank condition detection result and a edge warping condition detection result of the graphite boat image.
Compared with the prior art, the invention has at least the following beneficial technical effects:
(1) According to the invention, a large number of solar cells can be arranged in the boat groove of the graphite boat, the outline of the cells in the graphite boat image is extracted by acquiring the graphite boat image, whether the cells have edge warping defects or not is determined according to the outline of the cells, the large number of solar cells can be detected simultaneously, photographing one by one is not needed, the time cost is saved, the appearance defect detection of the large-scale cells can be realized, and the detection efficiency of the appearance defect of the large-scale cells is improved.
(2) According to the method and the device for detecting the defects of the solar cells, the optimal segmentation threshold value of the graphite boat image is calculated, and the plurality of cell images with gray values larger than the optimal segmentation threshold value are segmented from the graphite boat image according to the optimal segmentation threshold value, so that the segmentation accuracy of the cell images can be improved, and further the defect detection accuracy of the solar cells can be improved.
Example 2
In one embodiment, the invention provides a solar cell defect detection system based on a visual image of a graphite boat, which is used for executing the solar cell defect detection method based on the visual image of the graphite boat in the embodiment 1.
The solar cell defect detection system based on the visual image of the graphite boat provided by the invention can realize the steps and effects of the solar cell defect detection method based on the visual image of the graphite boat in the embodiment 1, and the invention is not repeated for avoiding repetition.
Compared with the prior art, the invention has at least the following beneficial technical effects:
(1) According to the invention, a large number of solar cells can be arranged in the boat groove of the graphite boat, the outline of the cells in the graphite boat image is extracted by acquiring the graphite boat image, whether the cells have edge warping defects or not is determined according to the outline of the cells, the large number of solar cells can be detected simultaneously, photographing one by one is not needed, the time cost is saved, the appearance defect detection of the large-scale cells can be realized, and the detection efficiency of the appearance defect of the large-scale cells is improved.
(2) According to the method and the device for detecting the defects of the solar cells, the optimal segmentation threshold value of the graphite boat image is calculated, and a plurality of cell images with gray values larger than the optimal segmentation threshold value are segmented from the graphite boat image according to the optimal segmentation threshold value, so that the segmentation accuracy of the cell images can be improved, and the defect detection accuracy of the solar cells can be further improved.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples illustrate only a few embodiments of the invention, which are described in detail and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.
Claims (7)
1. The solar cell defect detection method based on the visual image of the graphite boat is characterized by comprising the following steps of:
s101: acquiring an image of a graphite boat, wherein a plurality of boat grooves are arranged in the graphite boat in an array manner, and the boat grooves are used for placing solar cells;
s102: preprocessing the graphite boat image, wherein the preprocessing comprises the following steps: filtering noise reduction, brightness equalization and contrast equalization;
s103: calculating an optimal segmentation threshold of the graphite boat image;
s104: dividing a plurality of battery piece images with gray values larger than the optimal dividing threshold value from the graphite boat image according to the optimal dividing threshold value;
s105: according to the positions of the plurality of separated battery piece images, when no battery piece exists in a boat slot at a certain position, determining that the boat slot has an empty piece condition;
s106: extracting a battery piece outline in the battery piece image;
s107: performing linear fitting on the contour of the battery piece;
s108: when the linear fitting result of the battery piece outline is a straight line, determining that the battery piece is defect-free; when the linear fitting result of the battery piece outline is a curve, calculating the curvature of the curve;
s109: when the curvature of the curve is larger than a preset curvature, determining that the battery piece has edge warping defects;
wherein, the step S103 specifically includes:
s1031: converting the pretreated graphite boat image into a gray level histogram;
s1032: setting an initial segmentation threshold, taking a region with a gray value larger than the segmentation threshold as a foreground region, and taking a region with a gray value smaller than or equal to the segmentation threshold as a background region;
s1033: calculating the gray probability of the foreground region and the background region under the current segmentation threshold value:
wherein, arepresenting the current segmentation threshold value of the current segment,，nrepresenting the total number of gray levels>Indicating when the segmentation threshold isaGray probability of background area, < >>Indicating when the segmentation threshold isaThe gray probability of the foreground region at the time,ithe gray level index value is represented and,N i representing gray scale levels asiThe number of pixels in the time period,Nrepresenting the total number of pixels of the graphite image;
s1034: calculating the gray average value of the foreground region and the background region under the current segmentation threshold value:
wherein, indicating when the segmentation threshold isaGray mean value of background area +.>Indicating when the segmentation threshold isaGray level mean value of foreground region +.>Indicating when the segmentation threshold isaThe gray average value of the whole graphite boat image;
s1035: calculating an inter-class variance between the background region and the foreground region at a current segmentation threshold:
wherein, representing the inter-class variance;
s1036: sequentially selecting other segmentation thresholds, repeating the steps S1033 to S1036, and calculating the inter-class variance between the background area and the foreground area under each segmentation threshold;
s1037: taking a corresponding segmentation threshold value when the inter-class variance is maximum as the optimal segmentation threshold value;
wherein, the step S107 specifically includes:
s1071: determining four edges in the outline of the battery piece;
s1072: for the start point and the end point of each edge, adjacent edges are selectedpAnd (3) carrying out coordinate smoothing on the contour points:
wherein, represents the smoothed abscissa, +.>Representing the ordinate after the smoothing process, (-)x 1 ,y 1 )、(x 2 ,y 2 ) And%x p ,y p ) Representing the coordinates of the contour points involved in the coordinate smoothing process,prepresenting the number of adjacent contour points;
s1073: according to the coordinates of the starting point and the ending point after the coordinate smoothing treatment, a straight line is fitted, and the straight line can be expressed as:
wherein, ythe vertical coordinate is indicated by the vertical coordinate,xthe horizontal axis of the drawing is indicated,kthe slope is indicated as such,brepresenting the intercept;
s1074: uniformly selecting contour points except for a starting point and an ending point, and calculating the distance from each contour point to a fitted straight line:
wherein, d i represent the firstiDistance from each contour point to the fitted straight line, (-)x i ,y i ) Represent the firstiCoordinates of the contour points;
s1075: counting the number of contour points with the distance to the fitted straight line smaller than a preset distance;
s1076: when the number of contour points smaller than the preset distance is larger than the preset number, determining that the edge to be fitted is a straight line, and finishing fitting; and when the number of the contour points smaller than the preset distance is smaller than or equal to the preset number, determining the edge to be fitted as a curve, equally dividing the edge to be fitted into two sections, re-executing S1072 to S1076, and continuing fitting until the fitting of the whole edge is completed.
2. The method for detecting defects of a solar cell based on a visual image of a graphite boat according to claim 1, wherein the step S102 specifically comprises:
s1021: and filtering and denoising the graphite boat image according to the following formula:
wherein, f m (x,y) Representing pixel points [ ]x,y) The filtered and denoised pixel values at that location,f(x,y) Representing pixel points [ ]x,y) The original pixel value at which is to be found,Arepresenting the length of the image of the graphite boat,Brepresenting the width of the image of the graphite boat,PIrepresenting a graphite boat image;
s1022: the radius of the luminance balancing circular structure is determined according to the following formula:
wherein, Rthe radius of the luminance balancing circular structure is shown,r min representing the radius minimum of the luminance balancing circular structure,r max represents the maximum value of the radius of the luminance balancing circular structure,S min representing the minimum area of the cell area,S max representing the maximum area of the cell area;
s1023: with preset step lengthsDetermining a plurality of brightness balance circular structures, wherein the number of the brightness balance circular structures is as follows:
wherein, Lthe number of luminance equalizing circular structures is represented,the representation is rounded up and down to the top,srepresenting a preset step length;
s1024: through a plurality of brightness equalization circular structures, carrying out brightness equalization on the graphite boat image after filtering and noise reduction:
wherein, f t (x, y) Representing pixel points [ ]x,y) The pixel value after the brightness is balanced is located,f m (x,y) Representing pixel points [ ]x,y) The filtered and denoised pixel values at that location,the gray scale closing operation is represented by a gray scale closing operation,c(r i ) Represent the firstiA circular structure with balanced brightness;
s1025: constructing an energy change function:
wherein, E(W) The energy variation function is represented by a function of the energy variation,Xrepresenting pixel points [ ]x,y) At the position oftThe gray value of the time in question,Wthe displacement vector is represented by a vector of the displacement,urepresentation ofxThe amount of displacement in the direction is determined,vrepresentation ofyThe amount of displacement in the direction is determined,lrepresentation oftThe amount of displacement in the direction is determined,representing a three-dimensional gradient sign>Consistency parameter representing gray value, +.>Representing displacement vectorsWGradient of->Representing euclidean norm sign, +.>A smoothness parameter representing a gray value;
s1026: constructing an Euler Lagrangian equation according to the energy variation function:
wherein, ；
s1027: solving the Euler Lagrangian equation, and calculating the horizontal component of the energy change function in the space domainu(x, y) And a vertical componentv(x, y)：
Wherein, representation ofuIs used for the average value of (a),urepresentation ofxDisplacement of direction,/->Representation ofvIs used for the average value of (a),vrepresentation ofyThe amount of displacement in the direction is determined,F x representation ofFFor a pair ofxIs used for the partial derivative of (a),F y representation ofFFor a pair ofyIs used for the partial derivative of (a),F t representation ofFFor a pair oftIs a partial derivative of (2);
s1028: horizontal component in the spatial domain according to an energy variation functionu(x, y) And a vertical componentv(x, y) Contrast equalization is performed:
wherein, f d (x,y) Representing pixel points [ ]x,y) Pixel values after contrast equalization.
3. The method for detecting defects of a solar cell sheet based on a visual image of a graphite boat according to claim 1, wherein the step S106 specifically comprises:
s1061: introducing a Sobel operator, and acquiring a horizontal feature matrix and a vertical feature matrix of the battery piece image:
wherein, S x representing the horizontal feature matrix of the device,S y representing a vertical feature matrix;
s1062: calculating a horizontal direction gradient and a vertical direction gradient according to the horizontal feature matrix and the vertical feature matrix:
wherein, G x a gradient in the horizontal direction is indicated,G y a gradient in the vertical direction is indicated,Ia gray value matrix representing the cell image;
s1063: calculating the gradient intensity and gradient direction of the pixel points:
wherein, Gthe intensity of the gradient is indicated and,θrepresenting the gradient direction;
s1064: when the battery piece image has a plurality of gradient information, reserving maximum value pixel points and inhibiting non-maximum value pixel points;
s1065: setting a high threshold and a low threshold, and determining the pixel point as a strong edge pixel point when the gradient strength of the pixel point is greater than the high threshold; when the gradient strength of the pixel points is between the low threshold value and the high threshold value, determining that the pixel points are weak edge pixel points; when the edge pixel gradient value of the pixel point is smaller than the low threshold value, suppressing the pixel point;
s1066: and connecting the strong edge pixel points to obtain the contour of the battery piece.
4. The method for detecting defects of a solar cell based on a visual image of a graphite boat according to claim 3, wherein the determination modes of the high threshold and the low threshold are as follows:
wherein, THindicating a high threshold value of the value,TLindicating a low threshold value of the value,H* Gray value matrix representing battery plate image after non-maximum value inhibition, max #H* ) The maximum value in the gray value matrix of the cell image after non-maximum value suppression is represented.
5. The method for detecting defects of a solar cell sheet based on a visual image of a graphite boat according to claim 1, wherein S108 specifically comprises:
the curvature of the curve is calculated by the following formula:
wherein, Kthe curvature is indicated by the fact that,f(x) The curve to be fitted is shown as such,represents the first derivative of the curve,/>Representing the second derivative of the curve.
6. The method for detecting defects of a solar cell sheet based on a visual image of a graphite boat according to claim 1, further comprising:
s110: and outputting a blank condition detection result and a edge warping condition detection result of the graphite boat image.
7. A solar cell defect detection system based on a visual image of a graphite boat, which is used for executing the solar cell defect detection method based on a visual image of a graphite boat according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311031443.2A CN116740073B (en) | 2023-08-16 | 2023-08-16 | Solar cell defect detection method and system based on visual image of graphite boat |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311031443.2A CN116740073B (en) | 2023-08-16 | 2023-08-16 | Solar cell defect detection method and system based on visual image of graphite boat |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116740073A true CN116740073A (en) | 2023-09-12 |
| CN116740073B CN116740073B (en) | 2023-10-20 |
Family
ID=87919088
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311031443.2A Active CN116740073B (en) | 2023-08-16 | 2023-08-16 | Solar cell defect detection method and system based on visual image of graphite boat |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116740073B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117420152A (en) * | 2023-12-19 | 2024-01-19 | 聚时科技(深圳)有限公司 | Visual inspection method, visual inspection device, computer equipment and computer readable storage medium |
| CN117708689A (en) * | 2024-02-05 | 2024-03-15 | 锦浪科技股份有限公司 | Photovoltaic module fault detection method based on distance characteristic method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113643264A (en) * | 2021-08-18 | 2021-11-12 | 哲为(上海)仪器科技有限公司 | Image processing method for graphite boat whole boat image warping detection |
| WO2022000931A1 (en) * | 2020-07-03 | 2022-01-06 | 宁夏隆基乐叶科技有限公司 | Graphite boat |
| CN115294016A (en) * | 2022-06-23 | 2022-11-04 | 镇江苏仪德科技有限公司 | Method for detecting existence of warping sheet of photovoltaic cell in PECVD graphite boat |
| CN115565902A (en) * | 2022-09-23 | 2023-01-03 | 无锡尚德太阳能电力有限公司 | Solar cell semi-finished product state evaluation method |
-
2023
- 2023-08-16 CN CN202311031443.2A patent/CN116740073B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022000931A1 (en) * | 2020-07-03 | 2022-01-06 | 宁夏隆基乐叶科技有限公司 | Graphite boat |
| CN113643264A (en) * | 2021-08-18 | 2021-11-12 | 哲为(上海)仪器科技有限公司 | Image processing method for graphite boat whole boat image warping detection |
| CN115294016A (en) * | 2022-06-23 | 2022-11-04 | 镇江苏仪德科技有限公司 | Method for detecting existence of warping sheet of photovoltaic cell in PECVD graphite boat |
| CN115565902A (en) * | 2022-09-23 | 2023-01-03 | 无锡尚德太阳能电力有限公司 | Solar cell semi-finished product state evaluation method |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117420152A (en) * | 2023-12-19 | 2024-01-19 | 聚时科技(深圳)有限公司 | Visual inspection method, visual inspection device, computer equipment and computer readable storage medium |
| CN117420152B (en) * | 2023-12-19 | 2024-03-19 | 聚时科技(深圳)有限公司 | Visual inspection method, visual inspection device, computer equipment and computer readable storage medium |
| CN117708689A (en) * | 2024-02-05 | 2024-03-15 | 锦浪科技股份有限公司 | Photovoltaic module fault detection method based on distance characteristic method |
| CN117708689B (en) * | 2024-02-05 | 2024-04-30 | 锦浪科技股份有限公司 | Photovoltaic module fault detection method based on distance characteristic method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116740073B (en) | 2023-10-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116740073B (en) | Solar cell defect detection method and system based on visual image of graphite boat | |
| Tsai et al. | Mean shift-based defect detection in multicrystalline solar wafer surfaces | |
| Li et al. | Automatic saw-mark detection in multicrystalline solar wafer images | |
| US20060029257A1 (en) | Apparatus for determining a surface condition of an object | |
| CN116977329B (en) | Photovoltaic grid line detection method based on machine vision | |
| CN115063423B (en) | Self-adaptive identification method for cold and hot cracks of mechanical castings based on computer vision | |
| CN116664565A (en) | Hidden crack detection method and system for photovoltaic solar cell | |
| CN109632808A (en) | Seamed edge defect inspection method, device, electronic equipment and storage medium | |
| CN112381759B (en) | Monocrystalline silicon solar wafer defect detection method based on optical flow method and confidence coefficient method | |
| CN115953398B (en) | Defect identification method for strip steel surface | |
| CN115082477B (en) | Semiconductor wafer processing quality detection method based on light reflection removing effect | |
| CN115029704B (en) | Intelligent control method and device for stainless steel pickling process | |
| CN106651831B (en) | Bamboo block defect detection method and system | |
| CN115797813B (en) | Water environment pollution detection method based on aerial image | |
| CN110619146A (en) | Polycrystalline silicon cell crack defect detection method based on structural similarity measurement | |
| CN115274486B (en) | Semiconductor surface defect identification method | |
| Liu et al. | Iterating tensor voting: A perceptual grouping approach for crack detection on EL images | |
| CN112561875A (en) | Photovoltaic cell panel coarse grid detection method based on artificial intelligence | |
| CN104915951B (en) | A kind of stippled formula DPM two-dimension code area localization methods | |
| CN116823725A (en) | Aeroengine blade surface defect detection method based on support vector machine | |
| CN117541582B (en) | IGBT insulation quality detection method for high-frequency converter | |
| CN112381794B (en) | Printing defect detection method based on deep convolution generation network | |
| CN110766675B (en) | Solar cell panel defect detection method | |
| CN116758080A (en) | Method and system for detecting screen printing defects of solar cell | |
| CN114581407B (en) | Self-adaptive defect detection method for photovoltaic module |
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 |