CN113834816A - Machine vision-based photovoltaic cell defect online detection method and system - Google Patents
Machine vision-based photovoltaic cell defect online detection method and system Download PDFInfo
- Publication number
- CN113834816A CN113834816A CN202111165075.1A CN202111165075A CN113834816A CN 113834816 A CN113834816 A CN 113834816A CN 202111165075 A CN202111165075 A CN 202111165075A CN 113834816 A CN113834816 A CN 113834816A
- Authority
- CN
- China
- Prior art keywords
- image
- photovoltaic cell
- defect
- machine vision
- defects
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000007547 defect Effects 0.000 title claims abstract description 112
- 238000001514 detection method Methods 0.000 title claims abstract description 77
- 238000000034 method Methods 0.000 claims abstract description 60
- 238000005424 photoluminescence Methods 0.000 claims abstract description 24
- 238000003384 imaging method Methods 0.000 claims abstract description 22
- 238000003709 image segmentation Methods 0.000 claims abstract description 17
- 230000008569 process Effects 0.000 claims abstract description 16
- 238000001914 filtration Methods 0.000 claims abstract description 14
- 238000007781 pre-processing Methods 0.000 claims abstract description 12
- 230000011218 segmentation Effects 0.000 claims abstract description 7
- 238000005516 engineering process Methods 0.000 claims abstract description 6
- 238000003702 image correction Methods 0.000 claims abstract description 6
- 230000009467 reduction Effects 0.000 claims abstract description 6
- 230000003287 optical effect Effects 0.000 claims description 17
- 230000009466 transformation Effects 0.000 claims description 9
- 230000000694 effects Effects 0.000 claims description 8
- 230000003044 adaptive effect Effects 0.000 claims description 6
- 230000002902 bimodal effect Effects 0.000 claims description 5
- 239000003086 colorant Substances 0.000 claims description 4
- 238000005520 cutting process Methods 0.000 claims description 4
- 230000002708 enhancing effect Effects 0.000 claims description 4
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 3
- 238000009434 installation Methods 0.000 claims description 3
- 230000001678 irradiating effect Effects 0.000 claims description 3
- 238000012790 confirmation Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000005401 electroluminescence Methods 0.000 description 2
- 238000002073 fluorescence micrograph Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000012880 independent component analysis Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 208000037656 Respiratory Sounds Diseases 0.000 description 1
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 229910052710 silicon Inorganic materials 0.000 description 1
- 239000010703 silicon Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/02—Affine transformations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
-
- 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
- G06T7/11—Region-based segmentation
-
- 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
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
-
- 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
- G06T2207/20112—Image segmentation details
- G06T2207/20132—Image cropping
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Geometry (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a machine vision-based photovoltaic cell defect online detection method and a machine vision-based photovoltaic cell defect online detection system, which comprise the following steps: s2, acquiring an image: collecting an original image of the photovoltaic cell passing through a region to be detected by utilizing a photoluminescence imaging technology, wherein the original image is a photoluminescence image; s3, image preprocessing: carrying out image correction, image filtering and noise reduction, grid line removal and image enhancement on the acquired original image to finish image preprocessing; s4, image segmentation: combining a histogram dual-peak method and a self-adaptive threshold method to carry out image segmentation on the preprocessed image to obtain a binary image; s5, detecting a connected domain: and performing connected domain detection on the binary image obtained after segmentation by adopting a connected domain detection algorithm, identifying the defects of the photovoltaic cell, and outputting a defect detection result. Through the process, the defects of different types of the photovoltaic cells can be quickly and accurately identified, and the method has the advantages of being accurate in detection and wide in detection range.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a photovoltaic cell defect online detection method and system based on machine vision.
Background
The photovoltaic cell is one of important products in the photovoltaic industry, and has a wide view field prospect. However, as the photovoltaic cell has many production and processing links, various defects such as unfilled corners, cracks, black spots and black cores are inevitably generated in the production process, and the performance and the stability of the cell are seriously influenced by the defects. Therefore, it is very important to detect defects of photovoltaic cells during the production process, and it is an indispensable part of industrial production.
Liu Lei et al have proposed a positioning algorithm that integrates integral projection and gray level gravity center to process binary segmentation image, and have designed the Support Vector Machine (SVM) classifier with Radial Basis (RBF) as the kernel function, have realized the detection (Liu Lei, Wang Chong, Zhao Zhang, Li Hai Bian) to several common defects such as unfilled corner, broken grid, crackle [ J ] research of the defect detection technique of solar cell based on machine vision and instrument report, 2018,32(10):47-52), but the accuracy rate that this method detects the defect of photovoltaic cell still needs to be improved. ANWAR S et al propose an improved anisotropic diffusion filter algorithm and an Image segmentation technique, which accurately detect micro-crack defects in a polycrystalline silicon solar cell (Said Amirul Anwar, Mohd Zaid Abdullah. micro-crack detection of polycrystalline silicon cells and improving and diagnosing filter and Image segmentation technique [ J ]. EURASIP Journal on Image and Video Processing,2014 (1)), but the method can only be used for detecting micro-crack defects and has no capability for defects such as grid breakage, black heart and the like. Peng Xu et al use MACHINE VISION METHODS such as image segmentation, Gaussian filtering, Hough transformation, etc. to detect MICRO-cracks OF a PHOTOVOLTAIC module (Peng Xu, Wenju Zhou, Minrui Fei.DETECTION METHODS FOR MICRO-CRACKED DEFECTS OF PHOTOVOLTAIC MODULES BASED ON MACHINE VISION [ A ]. IEEE Beijing selection OF proceedings OF 2014 3rd International Conference ON Cloud Computing and integration Systems [ C ]. IEEE Beijing selection OF 2014 5). however, MICRO-cracks detected by the method are suitable FOR cracks with larger size and have poor detection effect ON fine cracks. Song et al successfully identifies Broken Corner And Black Edge defects in photovoltaic cells (Song, M.P., et al, "Research On Brown Corner And Black Edge Detection Of Solar cell," 2018International Conference On Machine Learning And Cybernetics (ICLCs) 2018.), but the types Of defects Of photovoltaic cells that can be detected by the method are limited. Tsai et al propose a Defect Detection method for electroluminescence image Solar Modules based on independent component analysis (D.Tsai, S.Wu and W.Chiu, "Defect Detection in Solar Modules Using ICA Basis Images," in IEEE Transactions on Industrial information, vol.9, No.1, pp.122-131, Feb.2013, doi:10.1109/TII.2012.2209663.), which first finds a set of independent Basis Images among a set of Defect-free Solar photovoltaic cell Images Using ICA in a learning phase, reconstructs each detected Solar photovoltaic cell image into a linear combination of Basis Images in a Detection phase, classifies the coefficients of the linear combination as feature vectors, and evaluates the reconstruction error between a test image and the reconstructed image of the ICA Basis image, which can effectively judge whether the image meets the features of the Defect, but gives the specific positions of the defects, meanwhile, the method can only be used for detecting and training images with the same size and similar types of samples, but not solar photovoltaic cell images with any size and type, and has certain limitation.
Disclosure of Invention
In view of the above, the invention provides a machine vision-based photovoltaic cell defect online detection method and system, so as to solve the technical problems that secondary damage may be caused to a photovoltaic cell by power-on detection in imaging modes such as uneven illumination, electroluminescence and the like, and factors such as noise, grid lines and the like influence a defect detection result.
In one aspect, the invention provides a machine vision-based photovoltaic cell defect online detection method, which comprises the following steps:
s2, acquiring an image: collecting an original image of the photovoltaic cell passing through a region to be detected by utilizing a photoluminescence imaging technology, wherein the original image is a photoluminescence image;
s3, image preprocessing: carrying out image correction, image filtering and noise reduction, grid line removal and image enhancement on the acquired original image to finish image preprocessing;
s4, image segmentation: combining a histogram dual-peak method and a self-adaptive threshold method to carry out image segmentation on the preprocessed image to obtain a binary image;
s5, detecting a connected domain: and performing connected domain detection on the binary image obtained after segmentation by adopting a connected domain detection algorithm, identifying the defects of the photovoltaic cell, and outputting a defect detection result.
Further, the step 2 is embodied as: and respectively placing a light source at two sides of the area to be detected, and simultaneously starting the two light sources when the photovoltaic cell passes by so as to collect the original image of the photovoltaic cell.
Further, the image preprocessing in the step S3 includes the following processes:
s31, detecting straight lines in an original image of the photovoltaic cell by using Hough transformation to obtain a deflection angle, correcting the image by using affine transformation, obtaining four corner points of the photovoltaic cell by using a corner point detection method, cutting out irrelevant black background areas, and obtaining an image only retaining the photovoltaic cell;
s32, denoising the input image by adopting a Gaussian low-pass filter;
s33, removing the densely distributed grid lines in the photovoltaic cell by adopting a local low-pass filter, and simultaneously keeping the details and the content of the image;
and S34, enhancing the image by adopting logarithmic transformation to improve the brightness and contrast of the image.
Further, the gaussian low-pass filter is:
in the formula, (u, v) represents the spatial coordinate of a certain pixel point in the image, u represents the abscissa of the pixel point, v represents the ordinate of the pixel point, H (u, v) is the value of the Gaussian low-pass filter at the pixel point (u, v), D0Is a preset constant, and D (u, v) is the distance between the center point (u, v) of the spectrogram of the photovoltaic cell and the center point of the spectrogram.
Further, the local low-pass filter is:
in the formula, G (u, v) is a value of the local low-pass filter at the pixel (u, v), d is a preset constant, ω is a bandwidth of the local low-pass filter, ω is greater than 0 and less than or equal to N, and M, N represents the width and length of the image respectively.
Further, the image segmentation in the step 4 comprises:
s41, respectively adopting a histogram bimodal method and an adaptive threshold method to carry out image segmentation on the preprocessed photovoltaic cell image, and segmenting a binary image of a large-area defect and a small-area defect;
and S42, multiplying the binary images of the large-area defect and the small-area defect which are segmented by a histogram dual-peak method and an adaptive threshold method, and fusing to obtain a complete binary image.
Further, the connected component detection in step 5 is divided into two steps:
s51, representing the defect-based defect in the integrated binary image obtained after fusion as a black area, marking the defect area in the integrated binary image with different colors after detecting the defect area by adopting a connected domain detection algorithm, and obtaining the area and coordinate information of the defect area;
and S52, selecting a defect area according to the defect area and the coordinate information, and outputting a defect detection result in the original image.
Further, the step S2 is preceded by:
s1, photo confirmation: the method comprises the steps of confirming photographing of an industrial camera provided with an optical filter under the condition of no photovoltaic cell so as to ensure that a light source cannot interfere imaging, wherein the industrial camera is used for collecting an original image of the photovoltaic cell.
Further, the step S1 is embodied as: the method comprises the steps of confirming photographing of an industrial camera provided with an optical filter under the condition of no photovoltaic cell, and judging whether a light source can generate interference on imaging or not by analyzing photographed pictures; if the light source is judged to interfere the imaging, the installation of the optical filter is checked, or the optical filter is replaced; otherwise, the process proceeds to step S2.
On the other hand, the invention also provides a photovoltaic cell defect online detection system based on machine vision, which comprises an industrial personal computer, an industrial camera, a light source and a light filter, wherein the industrial personal computer is respectively connected with the industrial camera, the light filter is arranged in front of a lens of the industrial camera, and the light source is used for irradiating the photovoltaic cell to enable the internal electrons of the photovoltaic cell to jump so as to generate a fluorescence effect; the industrial camera is used for acquiring a photoluminescence image of the photovoltaic cell when receiving a photographing instruction of the industrial personal computer and returning the photoluminescence image to the industrial personal computer; the optical filter is used for filtering light emitted by the light source, so that an image obtained by the industrial personal computer is completely formed by the fluorescence effect of the photovoltaic cell; the industrial personal computer is used for sending a photographing instruction to the industrial camera and receiving a photoluminescence image fed back by the industrial camera, and the online photovoltaic cell defect detection system based on machine vision adopts the online photovoltaic cell defect detection method based on machine vision to detect defects.
Therefore, compared with the prior art, the invention has the advantages that:
(1) the grid lines can be effectively removed, and the details and the content of the image can be better reserved. A large number of densely distributed grid lines in the photovoltaic cell can be detected together with the defects in the defect detection process, so that the grid lines need to be removed in image preprocessing, and the defect detection result is prevented from being affected. The grid lines are represented as high-frequency signals in the spectrogram, if a traditional low-pass filter is adopted to carry out global low-pass filtering on the image, although the high-frequency signals representing the grid lines can be removed, other high-frequency signals can be filtered at the same time, so that the processed image is serious in fuzzification degree, and more details are lost. According to the invention, a local low-pass filter is constructed to accurately filter the high-frequency signals corresponding to the grid lines, so that the fuzzification degree of the image is reduced while the grid lines are effectively removed, and the details and the content of the image are better reserved.
(2) The recognition precision is higher, and the recognition range is wider. The invention adopts an image segmentation method combining a histogram bimodal method and a self-adaptive threshold method, effectively segments various defects in the photovoltaic cell and has higher identification precision.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate an embodiment of the invention and, together with the description, serve to explain the invention and not to limit the invention. In the drawings:
fig. 1 is a flowchart of a method for online detecting defects of a photovoltaic cell based on machine vision according to an embodiment of the present invention;
FIG. 2 is a flowchart of a method for online defect detection of a photovoltaic cell based on machine vision according to another embodiment of the present invention;
FIG. 3 is a schematic diagram of a partial low pass filter;
FIG. 4 is a process diagram for processing a photovoltaic cell image using the machine vision based on-line detection method for defects in a photovoltaic cell of the present invention;
fig. 5 is a schematic diagram of a photoluminescence imaging technique.
Detailed Description
It should be noted that, in order to make those skilled in the art better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure. It is to be understood that the described embodiments are merely exemplary of some, and not all, of the present disclosure. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
Fig. 1 is a flowchart of a method for online detecting defects of a photovoltaic cell based on machine vision according to an embodiment of the present invention. In one embodiment, as shown in FIG. 1, the method comprises the steps of:
s2, acquiring an image: collecting an original image of the photovoltaic cell passing through a region to be detected by utilizing a photoluminescence imaging technology, wherein the original image is a photoluminescence image;
s3, image preprocessing: carrying out image correction, image filtering and noise reduction, grid line removal and image enhancement on the acquired original image to finish image preprocessing;
s4, image segmentation: combining a histogram dual-peak method and a self-adaptive threshold method to carry out image segmentation on the preprocessed image to obtain a binary image;
s5, detecting a connected domain: and performing connected domain detection on the binary image obtained after segmentation by adopting a connected domain detection algorithm, identifying the defects of the photovoltaic cell, and outputting a defect detection result.
According to the invention, photoluminescence image data of the photovoltaic cell is firstly collected, the imaging speed and sample data quality of the photovoltaic cell are improved by a photoluminescence imaging technology, and a process sheet of the photovoltaic cell can be detected; secondly, in order to solve the problem that factors such as noise, grid lines and the like influence the defect detection result, image correction and image filtering noise reduction are provided, a local low-pass filter is adopted to remove the grid lines, and the image is enhanced; then combining a histogram bimodal method and a self-adaptive threshold method to carry out image segmentation; and finally, carrying out connected domain detection on the binary image obtained after segmentation through a connected domain detection algorithm. The method can remove the grid lines under the condition of better retaining image details and information, and can accurately identify the defects of different types of photovoltaic cells.
Meanwhile, referring to fig. 2, in this embodiment, in order to ensure that the acquired image is generated by the fluorescence effect of the photovoltaic cell and prevent the light source from interfering with the imaging, before detecting the defect of the photovoltaic cell, the online detection method for the defect of the photovoltaic cell based on machine vision further includes, in step S1, taking a picture to confirm: the method comprises the steps of confirming photographing of an industrial camera provided with an optical filter under the condition of no photovoltaic cell so as to ensure that a light source cannot interfere imaging, wherein the industrial camera is used for collecting original images of the photovoltaic cell. Specifically, in the step, the industrial camera provided with the optical filter is photographed and confirmed under the condition of no photovoltaic battery, and whether the light source can interfere with imaging or not is judged by analyzing the photographed picture; if the light source is judged to interfere the imaging, the installation of the optical filter is checked, or the optical filter is replaced; otherwise, the process proceeds to step S2.
In one embodiment, step 2 is embodied as: and respectively placing a light source at two sides of the area to be detected, and simultaneously starting the two light sources when the photovoltaic cell passes by so as to collect the original image of the photovoltaic cell. Namely, the degree of uneven illumination of the image is reduced by the double-light-source visual imaging.
In one embodiment, the image preprocessing in step S3 includes the following processes:
s31, detecting straight lines in an original image of the photovoltaic cell by using Hough transformation to obtain a deflection angle, correcting the image by using affine transformation, obtaining four corner points of the photovoltaic cell by using a corner point detection method, cutting out irrelevant black background areas, and obtaining an image only retaining the photovoltaic cell;
the reason why the image needs to be corrected is that the photovoltaic cells are inclined at a certain angle in the image due to improper placement of the photovoltaic cells.
S32, denoising the input image by adopting a Gaussian low-pass filter; through the process, noise pollution to the photo-luminescent image of the photovoltaic cell in the collection, formation and transmission processes is eliminated;
the Gaussian low-pass filter is as follows:
in the formula, (u, v) represents the spatial coordinate of a certain pixel point in the image, u represents the abscissa of the pixel point, v represents the ordinate of the pixel point, H (u, v) is the value of the Gaussian low-pass filter at the pixel point (u, v), u represents v, D0Is a preset constant, and D (u, v) is the distance between the center point (u, v) of the spectrogram of the photovoltaic cell and the center point of the spectrogram. It should be noted that, because gaussian low-pass filtering is a frequency domain filtering method, a frequency domain filtering method is involved, that is, a spectrogram is defaulted, and therefore, a spectrogram of a photovoltaic cell exists.
Because there are a large amount of densely distributed grid lines in the photovoltaic cell, it can be detected out together with the defect in the defect detection of follow-up, cause the serious influence to the defect detection result, therefore need to remove the grid line when preconditioning, have proposed a kind of local low-pass filter, while removing the grid line effectively, has kept the detail and content of the picture better, specifically:
s33, removing the densely distributed grid lines in the photovoltaic cell by adopting a local low-pass filter, and simultaneously keeping the details and the content of the image; the local low-pass filter is:
in the formula, G (u, v) is a value of the local low-pass filter at the pixel (u, v), d is a preset constant, generally representing a certain threshold of the filtering range, ω is a bandwidth of the local low-pass filter, ω is greater than 0 and less than or equal to N, and M, N represents a width and a length of the image respectively. Fig. 3 is a schematic diagram of a local low-pass filter.
And S34, enhancing the image by adopting logarithmic transformation to improve the brightness and contrast of the image.
In one embodiment, the image segmentation in step 4 comprises:
s41, respectively adopting a histogram bimodal method and an adaptive threshold method to carry out image segmentation on the preprocessed photovoltaic cell image, and segmenting a binary image of a large-area defect and a small-area defect;
the large-area defect specifically means an area defect having an area ratio of 0.03% or more in the entire image, and the small-area defect means an area defect having an area ratio of less than 0.03% in the entire image.
And S42, multiplying the binary images of the large-area defect and the small-area defect which are segmented by a histogram dual-peak method and an adaptive threshold method, and fusing to obtain a complete binary image.
Meanwhile, as a preferred embodiment of the present invention, the connected component detection in step 5 is divided into two steps:
s51, representing the defect-based defect in the integrated binary image obtained after fusion as a black area, marking the defect area in the integrated binary image with different colors after detecting the defect area by adopting a connected domain detection algorithm, and obtaining the area and coordinate information of the defect area;
and S52, selecting a defect area according to the defect area and the coordinate information, and outputting a defect detection result in the original image.
FIG. 5 is a process diagram of processing a photovoltaic cell image by the online detection method for defects of a photovoltaic cell based on machine vision. The specific process is as follows: collecting original images of photovoltaic cells, correcting images, cutting, reducing noise, removing grid lines, enhancing images, dividing images, marking defects with different colors, and obtaining a defect detection result. According to the process, the method has the advantages of being accurate in detection and capable of rapidly and accurately identifying different types of defects of the photovoltaic cell.
In one embodiment, the invention also provides a photovoltaic cell defect online detection system based on machine vision, which comprises an industrial personal computer, an industrial camera, a light source and a light filter, wherein the industrial personal computer is respectively connected with the industrial camera, the light filter is arranged in front of a lens of the industrial camera, and the light source is used for irradiating the photovoltaic cell to enable the internal electrons of the photovoltaic cell to jump so as to generate a fluorescence effect; the industrial camera is used for acquiring a photoluminescence image of the photovoltaic cell when receiving a photographing instruction of the industrial personal computer and returning the photoluminescence image to the industrial personal computer; the optical filter is used for filtering light emitted by the light source, so that an image obtained by the industrial personal computer is completely formed by the fluorescence effect of the photovoltaic cell; the industrial personal computer is used for sending a photographing instruction to the industrial camera and receiving a photoluminescence image fed back by the industrial camera, and the machine vision-based photovoltaic cell defect online detection system adopts the machine vision-based photovoltaic cell defect online detection method to detect defects. Preferably, the industrial personal computer comprises an I7 processor and an RTX3060 display card; the light source is a blue laser light source.
Specifically, in the detection system, an optical filter is arranged in front of a lens of an industrial camera so as to effectively filter blue light emitted by the light source to obtain infrared band fluorescence generated by a photovoltaic cell, then two light sources are respectively arranged at two sides of a region to be detected, when the photovoltaic cell passes through the region to be detected, blue laser light emitted by the light sources at the two sides irradiates the photovoltaic cell, the industrial camera collects fluorescence images generated by the photovoltaic cell and transmits the fluorescence images to an industrial personal computer to form a photoluminescence image of the photovoltaic cell, the imaging speed and the sample data quality of the photovoltaic cell are improved through a photoluminescence imaging technology, a photovoltaic cell process sheet can be detected, in order to solve the problem that the factors such as noise, grid lines and the like influence the defect detection result, image correction and image filtering noise reduction are adopted, a local low-pass filter is provided to remove grid lines and image enhancement is adopted, and then image segmentation is carried out by combining a histogram dual-peak method and a self-adaptive threshold method, and finally, carrying out connected domain detection on the binary image obtained after segmentation through a connected domain detection algorithm. The invention can remove the grid line under the condition of better retaining image details and information, can accurately identify the defects of different types of photovoltaic cells, and has higher identification precision and wider identification range.
Fig. 5 is a schematic diagram of the photoluminescence imaging technique. Specifically, the photovoltaic cell 4 comprises a conduction band 41, a silicon wafer 42 and a valence band 43 which are sequentially stacked from top to bottom, the optical filter 2 is arranged in front of a lens of the industrial camera 1, when the photovoltaic cell 4 passes through a region to be detected, the light source 3 emits blue laser light to irradiate the photovoltaic cell 4, electrons in the photovoltaic cell 4 are transited to generate a fluorescence effect, and then a photoluminescence image of the photovoltaic cell is generated.
The online detection method and system for defects of photovoltaic cells based on machine vision provided by the invention are described in detail above. The principles and embodiments of the present invention are explained herein using specific examples, which are presented only to assist in understanding the core concepts of the present invention. It should be noted that, for those skilled in the art, it is possible to make various improvements and modifications to the present invention without departing from the principle of the present invention, and those improvements and modifications also fall within the scope of the claims of the present invention.
Claims (10)
1. The online detection method for the defects of the photovoltaic cell based on machine vision is characterized by comprising the following steps:
s2, acquiring an image: collecting an original image of the photovoltaic cell passing through a region to be detected by utilizing a photoluminescence imaging technology, wherein the original image is a photoluminescence image;
s3, image preprocessing: carrying out image correction, image filtering and noise reduction, grid line removal and image enhancement on the acquired original image to finish image preprocessing;
s4, image segmentation: combining a histogram dual-peak method and a self-adaptive threshold method to carry out image segmentation on the preprocessed image to obtain a binary image;
s5, detecting a connected domain: and performing connected domain detection on the binary image obtained after segmentation by adopting a connected domain detection algorithm, identifying the defects of the photovoltaic cell, and outputting a defect detection result.
2. The machine vision-based online detection method for defects of photovoltaic cells according to claim 1, wherein the step 2 is embodied as: and respectively placing a light source at two sides of the area to be detected, and simultaneously starting the two light sources when the photovoltaic cell passes by so as to collect the original image of the photovoltaic cell.
3. The online detection method for defects of photovoltaic cells based on machine vision according to claim 2, wherein the image preprocessing in step S3 includes the following processes:
s31, detecting straight lines in an original image of the photovoltaic cell by using Hough transformation to obtain a deflection angle, correcting the image by using affine transformation, obtaining four corner points of the photovoltaic cell by using a corner point detection method, cutting out irrelevant black background areas, and obtaining an image only retaining the photovoltaic cell;
s32, denoising the input image by adopting a Gaussian low-pass filter;
s33, removing the densely distributed grid lines in the photovoltaic cell by adopting a local low-pass filter, and simultaneously keeping the details and the content of the image;
and S34, enhancing the image by adopting logarithmic transformation to improve the brightness and contrast of the image.
4. The machine vision-based online detection method for defects of photovoltaic cells according to claim 3, wherein the Gaussian low-pass filter is:
in the formula, (u, v) represents the spatial coordinate of a certain pixel point in the image, u represents the abscissa of the pixel point, v represents the ordinate of the pixel point, H (u, v) is the value of the Gaussian low-pass filter at the pixel point (u, v), D0Is a preset constant, and D (u, v) is the distance between the center point (u, v) of the spectrogram of the photovoltaic cell and the center point of the spectrogram.
5. The machine-vision-based online defect detection method for photovoltaic cells according to claim 3, wherein the local low-pass filter is:
in the formula, G (u, v) is a value of the local low-pass filter at the pixel (u, v), d is a preset constant, ω is a bandwidth of the local low-pass filter, ω is greater than 0 and less than or equal to N, and M, N represents the width and length of the image respectively.
6. The machine vision-based online detection method for defects of photovoltaic cells of claim 3, wherein the image segmentation in the step 4 comprises:
s41, respectively adopting a histogram bimodal method and an adaptive threshold method to carry out image segmentation on the preprocessed photovoltaic cell image, and segmenting a binary image of a large-area defect and a small-area defect;
and S42, multiplying the binary images of the large-area defect and the small-area defect which are segmented by a histogram dual-peak method and an adaptive threshold method, and fusing to obtain a complete binary image.
7. The online detection method for defects of photovoltaic cells based on machine vision according to claim 6, characterized in that the connected component detection in step 5 is divided into two steps:
s51, representing the defect-based defect in the integrated binary image obtained after fusion as a black area, marking the defect area in the integrated binary image with different colors after detecting the defect area by adopting a connected domain detection algorithm, and obtaining the area and coordinate information of the defect area;
and S52, selecting a defect area according to the defect area and the coordinate information, and outputting a defect detection result in the original image.
8. The machine vision-based online defect detection method for photovoltaic cells according to any one of claims 1 to 7, wherein the step S2 is preceded by:
s1, photo confirmation: the method comprises the steps of confirming photographing of an industrial camera provided with an optical filter under the condition of no photovoltaic cell so as to ensure that a light source cannot interfere imaging, wherein the industrial camera is used for collecting an original image of the photovoltaic cell.
9. The on-line machine vision-based photovoltaic cell defect detection method of claim 8, wherein the step S1 is embodied as: the method comprises the steps of confirming photographing of an industrial camera provided with an optical filter under the condition of no photovoltaic cell, and judging whether a light source can generate interference on imaging or not by analyzing photographed pictures; if the light source is judged to interfere the imaging, the installation of the optical filter is checked, or the optical filter is replaced; otherwise, the process proceeds to step S2.
10. The online detection system for the defects of the photovoltaic cell based on machine vision is characterized by comprising an industrial personal computer, an industrial camera, a light source and a light filter, wherein the industrial personal computer is respectively connected with the industrial camera, the light filter is arranged in front of a lens of the industrial camera, and the light source is used for irradiating the photovoltaic cell to enable electrons in the photovoltaic cell to jump so as to generate a fluorescence effect; the industrial camera is used for acquiring a photoluminescence image of the photovoltaic cell when receiving a photographing instruction of the industrial personal computer and returning the photoluminescence image to the industrial personal computer; the optical filter is used for filtering light emitted by the light source, so that an image obtained by the industrial personal computer is completely formed by the fluorescence effect of the photovoltaic cell; the industrial personal computer is used for sending a photographing instruction to the industrial camera and receiving a photoluminescence image fed back by the industrial camera, and the online photovoltaic cell defect detection system based on machine vision carries out defect detection by adopting the online photovoltaic cell defect detection method based on machine vision as claimed in any one of claims 1 to 9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111165075.1A CN113834816A (en) | 2021-09-30 | 2021-09-30 | Machine vision-based photovoltaic cell defect online detection method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111165075.1A CN113834816A (en) | 2021-09-30 | 2021-09-30 | Machine vision-based photovoltaic cell defect online detection method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113834816A true CN113834816A (en) | 2021-12-24 |
Family
ID=78967988
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111165075.1A Pending CN113834816A (en) | 2021-09-30 | 2021-09-30 | Machine vision-based photovoltaic cell defect online detection method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113834816A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114463327A (en) * | 2022-04-08 | 2022-05-10 | 深圳市睿阳精视科技有限公司 | Multi-shooting imaging detection equipment and method for watermark defect of electronic product lining package |
CN114701083A (en) * | 2022-06-06 | 2022-07-05 | 广东高景太阳能科技有限公司 | Silicon wafer cutting raw material processing method, system, medium and equipment |
CN116245794A (en) * | 2022-12-02 | 2023-06-09 | 广州市儒兴科技股份有限公司 | Solar cell back surface field appearance test method and device and readable storage medium |
CN117664984A (en) * | 2023-12-01 | 2024-03-08 | 上海宝柏新材料股份有限公司 | Defect detection method, device, system and storage medium |
WO2024055253A1 (en) * | 2022-09-15 | 2024-03-21 | 华为技术有限公司 | Imaging apparatus and imaging method |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100002932A1 (en) * | 2008-07-01 | 2010-01-07 | Nisshinbo Holdings Inc. | Photovoltaic devices inspection apparatus and method of determining defects in photovoltaic device |
CN102812347A (en) * | 2010-01-04 | 2012-12-05 | Bt成像股份有限公司 | Improved illumination systems and methods for photoluminescence imaging of photovoltaic cells and wafers |
CN108355981A (en) * | 2018-01-08 | 2018-08-03 | 西安交通大学 | A kind of battery connector quality determining method based on machine vision |
CN109084957A (en) * | 2018-08-31 | 2018-12-25 | 华南理工大学 | The defects detection and color sorting process and its system of photovoltaic solar crystal-silicon battery slice |
WO2019104767A1 (en) * | 2017-11-28 | 2019-06-06 | 河海大学常州校区 | Fabric defect detection method based on deep convolutional neural network and visual saliency |
CN110866916A (en) * | 2019-11-29 | 2020-03-06 | 广州大学 | Machine vision-based photovoltaic cell black-core black-corner detection method, device and equipment |
CN111229648A (en) * | 2020-01-19 | 2020-06-05 | 青岛滨海学院 | Solar cell panel flaw detection system and detection method based on machine vision |
CN111242892A (en) * | 2019-12-27 | 2020-06-05 | 西安理工大学 | Method for detecting defects of solar photovoltaic cell |
CN112821868A (en) * | 2020-12-31 | 2021-05-18 | 浙江爱旭太阳能科技有限公司 | Control method of PL detection system |
-
2021
- 2021-09-30 CN CN202111165075.1A patent/CN113834816A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100002932A1 (en) * | 2008-07-01 | 2010-01-07 | Nisshinbo Holdings Inc. | Photovoltaic devices inspection apparatus and method of determining defects in photovoltaic device |
CN102812347A (en) * | 2010-01-04 | 2012-12-05 | Bt成像股份有限公司 | Improved illumination systems and methods for photoluminescence imaging of photovoltaic cells and wafers |
WO2019104767A1 (en) * | 2017-11-28 | 2019-06-06 | 河海大学常州校区 | Fabric defect detection method based on deep convolutional neural network and visual saliency |
CN108355981A (en) * | 2018-01-08 | 2018-08-03 | 西安交通大学 | A kind of battery connector quality determining method based on machine vision |
CN109084957A (en) * | 2018-08-31 | 2018-12-25 | 华南理工大学 | The defects detection and color sorting process and its system of photovoltaic solar crystal-silicon battery slice |
CN110866916A (en) * | 2019-11-29 | 2020-03-06 | 广州大学 | Machine vision-based photovoltaic cell black-core black-corner detection method, device and equipment |
CN111242892A (en) * | 2019-12-27 | 2020-06-05 | 西安理工大学 | Method for detecting defects of solar photovoltaic cell |
CN111229648A (en) * | 2020-01-19 | 2020-06-05 | 青岛滨海学院 | Solar cell panel flaw detection system and detection method based on machine vision |
CN112821868A (en) * | 2020-12-31 | 2021-05-18 | 浙江爱旭太阳能科技有限公司 | Control method of PL detection system |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114463327A (en) * | 2022-04-08 | 2022-05-10 | 深圳市睿阳精视科技有限公司 | Multi-shooting imaging detection equipment and method for watermark defect of electronic product lining package |
CN114701083A (en) * | 2022-06-06 | 2022-07-05 | 广东高景太阳能科技有限公司 | Silicon wafer cutting raw material processing method, system, medium and equipment |
CN114701083B (en) * | 2022-06-06 | 2022-08-05 | 广东高景太阳能科技有限公司 | Silicon wafer cutting raw material processing method, system, medium and equipment |
WO2024055253A1 (en) * | 2022-09-15 | 2024-03-21 | 华为技术有限公司 | Imaging apparatus and imaging method |
CN116245794A (en) * | 2022-12-02 | 2023-06-09 | 广州市儒兴科技股份有限公司 | Solar cell back surface field appearance test method and device and readable storage medium |
CN117664984A (en) * | 2023-12-01 | 2024-03-08 | 上海宝柏新材料股份有限公司 | Defect detection method, device, system and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113834816A (en) | Machine vision-based photovoltaic cell defect online detection method and system | |
CN111307818B (en) | Visual online detection method for laser welding spot of lithium battery tab | |
CN106650770B (en) | Mura defect detection method based on sample learning and human eye visual characteristics | |
CN111369545B (en) | Edge defect detection method, device, model, equipment and readable storage medium | |
CN101464418B (en) | Flaw detection method and apparatus | |
US8750596B2 (en) | System and method for identifying defects in a material | |
CN109632808B (en) | Edge defect detection method and device, electronic equipment and storage medium | |
JP2010522316A (en) | Method for recognizing array region in die formed on wafer, and setting method for such method | |
CN108918093B (en) | Optical filter mirror surface defect detection method and device and terminal equipment | |
CN115908415B (en) | Edge-based defect detection method, device, equipment and storage medium | |
CN110648330B (en) | Defect detection method for camera glass | |
CN115205223B (en) | Visual inspection method and device for transparent object, computer equipment and medium | |
CN116363140B (en) | Method, system and device for detecting defects of medium borosilicate glass and storage medium | |
CN116773548B (en) | Wafer surface defect detection method and system | |
CN109378279A (en) | Wafer detection method and wafer detection system | |
CN112381759B (en) | Monocrystalline silicon solar wafer defect detection method based on optical flow method and confidence coefficient method | |
CN117635565A (en) | Semiconductor surface defect detection system based on image recognition | |
CN117115130A (en) | Wafer edge defect detection method and device | |
CN112308843A (en) | Camera overexposure abnormity adjusting method and system for photovoltaic cell panel image acquisition | |
CN117392112A (en) | Method, device and system for detecting wafer surface defects | |
CN113129260B (en) | Automatic detection method and device for internal defects of lithium battery cell | |
US11493453B2 (en) | Belt inspection system, belt inspection method, and recording medium for belt inspection program | |
CN109448012A (en) | A kind of method for detecting image edge and device | |
CN116503324A (en) | Black glue module appearance detection method and system based on machine vision | |
JP3788586B2 (en) | Pattern inspection apparatus and 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 |