CN115049645A - Solar cell panel surface defect detection method - Google Patents
Solar cell panel surface defect detection method Download PDFInfo
- Publication number
- CN115049645A CN115049645A CN202210964410.2A CN202210964410A CN115049645A CN 115049645 A CN115049645 A CN 115049645A CN 202210964410 A CN202210964410 A CN 202210964410A CN 115049645 A CN115049645 A CN 115049645A
- Authority
- CN
- China
- Prior art keywords
- gray
- area
- window
- gray level
- image
- 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
Images
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
- 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
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- 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
- G06T2207/20021—Dividing image into blocks, subimages or windows
-
- 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
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20104—Interactive definition of region of interest [ROI]
-
- 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
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/50—Photovoltaic [PV] energy
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Photovoltaic Devices (AREA)
Abstract
The invention relates to the field of image processing, in particular to a method for detecting defects on the surface of a solar cell panel. Collecting a gray level image of the surface of a battery plate to be detected; carrying out gray level grading on the gray level image of the surface of the battery panel to be detected to obtain a gray level image of the area of the battery panel; extracting all grid lines in the gray level image of the panel area, and setting window size according to the distance between adjacent grid lines to divide the grid line area to obtain a plurality of windows; constructing a gray scale area matrix of each window, and acquiring a high gray scale area emphasis characteristic value of each window according to the connected domain area formed by each gray scale pixel point in the gray scale area matrix of each window; and carrying out defect detection on the surface of the battery board to be detected according to the high-gray-scale region emphasis characteristic value of each window, and determining the defect type. The invention detects the defects in the battery panel on the basis of reserving the grid lines of the original image, can accurately distinguish spot defects and wheel mark defects, and has accurate identification effect.
Description
Technical Field
The invention relates to the field of image processing, in particular to a method for detecting defects on the surface of a solar cell panel.
Background
Solar energy is used as a main energy source on the earth, the solar energy is of great interest due to the characteristics of cleanness, safety, abundance and reproducibility, the development and utilization of solar cells make a great contribution to the realization of the protection of the earth environment, the solar cells developed by the solar energy are more and more widely applied, the defects on the surfaces of the solar cells are more common texture defects mainly comprising spots, wheel marks (fingerprints) and the like, the causes are manual operation errors or the pressure of one or some machines for producing the solar cells is too large, so that the products cannot meet the requirements of production quality, the defect types are distinguished while the defect detection is carried out on the solar cells, and then the solar cells are maintained according to specific defect types.
At present, a solar cell panel is usually subjected to defect detection by adopting a computer vision technology, but in the prior art, firstly, morphological processing needs to be carried out on the solar cell panel, grid lines and other textures in the solar cell panel are removed, and thus defect detection is carried out on an obtained shade image.
Therefore, in order to solve the problem that the defect type can not be distinguished from spot or wheel mark in the prior art, the invention provides a method for detecting the defects on the surface of the solar cell panel.
Disclosure of Invention
The invention provides a method for detecting defects on the surface of a solar cell panel, which aims to solve the problem that the prior art cannot distinguish whether the defect type is a spot or a wheel mark, and comprises the following steps: collecting a gray level image of the surface of a battery plate to be detected; carrying out gray level grading on the gray level image of the surface of the battery panel to be detected to obtain a gray level image of the area of the battery panel; extracting all grid lines in the gray level image of the panel area, and setting window size according to the distance between adjacent grid lines to divide the grid line area to obtain a plurality of windows; constructing a gray scale area matrix of each window, and acquiring a high gray scale area emphasis characteristic value of each window according to the connected domain area formed by each gray scale pixel point in the gray scale area matrix of each window; and carrying out defect detection on the surface of the battery board to be detected according to the high-gray-scale region emphasis characteristic value of each window, and determining the defect type.
According to the invention, through carrying out gray level division on the image, the characteristics of the wheel mark defect and the spot defect can be conveniently and better divided in the follow-up process, meanwhile, the grid line of the solar panel is reserved, the efficiency of image preprocessing is improved, meanwhile, the characteristics of the defect in the image can not be diluted, further, the window division is carried out according to the grid line area, the specific position of the defect can be obtained while the defect is identified, finally, the gray scale area matrix is adopted for identification according to the gray level expression characteristics of the spot defect and the wheel mark defect in the solar panel, and then the defect category is accurately distinguished through the high gray level area emphasis index, so that the defect detection of the solar panel is realized, and the detection precision and the detection efficiency are higher.
The invention adopts the following technical scheme: a method for detecting defects on the surface of a solar cell panel comprises the following steps:
and collecting a gray level image of the surface of the battery plate to be detected.
And carrying out gray level grading on the gray level image of the surface of the battery panel to be detected to obtain a gray level image of the area of the battery panel.
And extracting all grid lines in the gray image of the panel area to obtain the grid line area of the gray image of the panel area.
And setting the size of a window according to the distance between adjacent grid lines, and dividing the grid line region by using the window with the set size to obtain a plurality of windows.
And constructing a gray scale area matrix of each window, and acquiring a high gray scale area emphasis characteristic value of each window according to the connected domain area formed by each gray scale pixel point in the gray scale area matrix of each window.
And carrying out defect detection on the surface of the battery board to be detected according to the high-gray-scale region emphasis characteristic value of each window, and determining the defect type.
Further, a method for detecting defects on the surface of a solar panel, which is to perform gray scale classification on a gray scale image on the surface of the solar panel to be detected to obtain a gray scale image of a panel area, comprises the following steps:
establishing a gray level histogram of a gray level image on the surface of the battery plate to be detected, and performing curve fitting on the gray level histogram;
taking a gray value corresponding to a first trough from low to high in the curve as a background gray level;
averagely dividing all gray values larger than the background gray level into a plurality of gray levels to obtain the number of pixel points in each gray level;
and obtaining a panel area gray image according to all the pixel points with the gray values larger than the background gray level.
Further, a method for detecting defects on the surface of a solar cell panel, which is used for acquiring a grid line region of a gray level image of a cell panel region, comprises the following steps:
and extracting all grid lines in the gray image of the panel area, constructing a surrounding frame area according to the length of the grid lines and the number of the grid lines, and taking the surrounding frame area as the grid line area.
Further, a method for detecting defects on the surface of a solar cell panel, which utilizes a window with a set size to divide a grid line region, comprises the following steps:
acquiring the distance between adjacent grid lines as the width of one grid line, and taking the width of two grid lines as the side length of a window;
obtaining the size of a window according to the set side length, and continuously dividing the window in the grid line region according to the set size;
and acquiring the proportion of the set side length of the window in the length and the width of the grid line region respectively, and acquiring the number of the windows divided in the grid line region according to the proportion of the length and the width of the grid line region respectively.
Further, a method for detecting defects on the surface of a solar cell panel, which is used for acquiring the emphasized characteristic value of the high gray level area of each window, comprises the following steps:
acquiring the connected domain area formed by each gray level pixel point in the gray scale area matrix of each window;
obtaining the high gray level region emphasis characteristic value of each window according to the frequency of the connected domain area formed by each gray level pixel point, wherein the expression is as follows:
wherein the content of the first and second substances,
a high gray scale region emphasis feature value representing the mth window,
the frequency of occurrence of a connected domain with the area of j formed by the pixel points representing the ith gray level,
represents the maximum area of the connected component in the window,
indicating the number of connected components in the window.
Further, a method for detecting defects on the surface of a solar cell panel, which is used for detecting the defects on the surface of the cell panel to be detected, comprises the following steps:
acquiring a surface gray level image of the solar cell panel without defects, and calculating a high gray level region emphasis characteristic value of the surface gray level image of the solar cell panel without defects
；
Acquiring the surface gray level image of the solar cell panel with only the wheel mark defect, and calculating the intensity of a high gray level area of the surface gray level image of the solar cell panel with only the wheel mark defectAdjusting characteristic value
；
Acquiring a solar cell panel surface gray image only with spot defects, and calculating a high gray area emphasis characteristic value of the solar cell panel surface gray image only with spot defects
；
And establishing a threshold interval, and judging whether the battery board to be detected has defects or not according to the high gray area emphasis characteristic value of each window in the gray image of the surface of the battery board to be detected and the threshold interval.
Further, a method for detecting defects on the surface of a solar cell panel, the method for determining the types of the defects, comprises the following steps:
when the high gray scale area emphasis characteristic value of each window area in the battery board to be detected:
when the battery plate to be detected has no defects;
when a high gray scale region of a window region exists in the battery panel to be detected, emphasizing the characteristic value:
when the battery plate is detected to have the wheel mark defect;
when the high gray scale area of the window area exists in the battery board to be detected, the emphasized characteristic value is as follows:
and when the battery plate to be detected has spot defects.
The invention has the beneficial effects that: according to the invention, through carrying out gray level division on the image, the characteristics of the wheel mark defect and the spot defect can be conveniently and better divided in the follow-up process, meanwhile, the grid line of the solar panel is reserved, the efficiency of image preprocessing is improved, meanwhile, the characteristics of the defect in the image can not be diluted, further, the window division is carried out according to the grid line area, the specific position of the defect can be obtained while the defect is identified, finally, the gray scale area matrix is adopted for identification according to the gray level expression characteristics of the spot defect and the wheel mark defect in the solar panel, and then the defect category is accurately distinguished through the high gray level area emphasis index, so that the defect detection of the solar panel is realized, and the detection precision and the detection efficiency are higher.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a schematic structural diagram of a method for detecting defects on a surface of a solar cell panel according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of window division according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a gate line region enclosure according to an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
As shown in fig. 1, a schematic structural diagram of a method for detecting defects on a surface of a solar cell panel according to an embodiment of the present invention is shown, including:
101. and collecting a gray level image of the surface of the battery plate to be detected.
The invention aims at detecting the surface texture defects of the solar cell panel, so that the surface imaging of the solar cell panel needs to be acquired by arranging a camera to prevent the acquired image from being seriously distorted, the camera is arranged right above the solar cell panel, and the camera is used for performing weak light striking by adopting a light source at two sides to acquire the surface texture image of the solar cell panel.
The collected image of the surface area of the battery panel is grayed firstly, information of three layers of channels is converted into one layer of channel, the calculated amount of image processing is reduced, white pixels are 255, black pixels are 0, texture defects and surface grid lines are displayed as high gray values (partial white), and the background area of the battery panel is low gray values (partial gray).
102. And carrying out gray level grading on the gray level image of the surface of the battery panel to be detected to obtain a gray level image of the area of the battery panel.
The method for carrying out gray level grading on the gray level image on the surface of the battery panel to be detected to obtain the gray level image of the area of the battery panel comprises the following steps:
establishing a gray level histogram of a gray level image on the surface of the battery plate to be detected, and performing curve fitting on the gray level histogram;
taking a gray value corresponding to a first trough from low to high in the curve as a background gray level;
averagely dividing all gray values larger than the background gray level into a plurality of gray levels to obtain the number of pixel points in each gray level;
and obtaining a panel area gray image according to all the pixel points with the gray values larger than the background gray level.
The invention further carries out gray level degradation processing on the gray level image, firstly the whole gray level value (0-255) on the surface of the battery plate is distributed comparatively disorderly, if the gray level image is directly processed, the calculation amount is large, no pertinence exists, and if the relevant gray level degradation operation is directly carried out on the gray level image, part of gray level values of the wheel mark and the spot under part of conditions can be easily divided into the same level compared with the similar pixels, the accuracy of defect detection is reduced, therefore, in order to prevent the confusion, the invention firstly carries out gray level histogram statistics on the whole image, thereby determining the threshold range, leading the background part lower than the threshold to be divided into a gray level as a whole, and then carries out further gray level division on the residual pixels, leading the more similar gray level representations in the wheel mark and the spot to be separated more clearly.
The invention counts the gray level histogram of the whole image, can obtain obvious peak value distribution, and when the image is free of defects, the grid line and the electrode are high gray level pixels with a concentration trend; when the spot defect exists, the grid line, the spot and the electrode have a gray peak value with a concentration trend; when there is a rolling mark defect, the grid line and the electrode can present a more concentrated peak value, and the gray distribution of the rolling mark is more biased to the middle gray peak, no matter whether there is a texture defect or which kind of defect, the background with low gray value always has obvious peak value distribution.
Curve fitting is carried out on the gray level histogram, then Gaussian fitting is carried out, the gray level on the image battery panel is recorded as H, and the gray value corresponding to the first trough from low gray level to high gray level in the curve is recorded as H
Make the gray value less than or equal to
Of a pixel
All convert to the same gray value 1, the remaining gray pixels convert to 15 gray levels (corresponding to 2-16 gray values), and the single range of 15 gray levels is calculated as:
in the above formula, the first and second carbon atoms are,
is correspondingly higher than
Is divided by 15, which means that the remaining gray is divided into 15 gray partsWithin each respective portion, i.e.
。
The remaining gray pixels are then graded, the invention takes note of larger than
Is represented by the gray scale of
(i.e. the
) Gradation, the number of corresponding gradation levels
Comprises the following steps:
in the above formula, the first and second carbon atoms are,
representing the corresponding gray scale
Minus
After that, divide by
That is to say, all correspond to several
Due to the first one
Within the gray scale range, the gray scale value 2 after grading is correspondedWith the invention using rounding-up symbols
And +1, the gray value of the surface of the battery plate is reasonably divided into 16 gray levels with 1-16 gray values, and the contrast between the surface textures is pulled away to the greatest extent possible.
Further, the surface gray level image of the cell panel to be detected is subjected to binarization processing, the background of the collecting device is set to be black, the grid lines and the electrode parts of the solar cell panel are high-gray-level whitish, and the low-gray-level parts are except the grid lines, so that the automatic threshold value selection binarization processing is performed by an OSTU Otsu method.
The grid lines, the electrodes and the possibly existing texture defect parts are displayed to be white, the other parts of the battery plate and the background are all black, and in order to better position the battery plate part, connected domain analysis is needed.
103. And extracting all grid lines in the gray image of the panel area to obtain the grid line area of the gray image of the panel area.
The method for acquiring the grid line region of the gray image of the panel region comprises the following steps:
extracting all grid lines in the gray image of the panel area, constructing an enclosure frame area according to the length of the grid lines and the number of the grid lines, and taking the enclosure frame area as a grid line area.
The texture defects of the solar cell panel are mainly detected by a grid line region on the surface of the solar cell panel, so that the solar cell panel in an image is positioned firstly, a peripheral electrode part is removed, and then the grid line texture region is segmented and extracted.
Recording the lower left corner of the battery plate area image as the origin of a coordinate system
Horizontal right of the image as the horizontal axis
In the positive direction, the vertical direction is the longitudinal axis
Positive direction of (1), pixel coordinates are noted as
Then, the length, width and center of the maximum rectangle are counted, and the center point is recorded as
Rectangular shape
Is marked as four corner points
、
、
、
Require
(wherein
、
Respectively representing the distance between the corresponding two corner points).
By arranging the image acquisition device, the length of the pixel from the 1 st grid line to the last grid line is recorded as
The transverse length of the individual grid lines (not including electrodes) is
The invention sets up the rectangular frame in the grating area
(is a rectangular frame)
After being scaled equally) four angular points are respectively
、
、
、
The concrete parameters are as follows: the center point is still
Width and length of each
、
And the sides of the two rectangles are parallel to each other, namely:
then the resulting rectangular frame is shown in fig. 3
The grid line area in the image of the panel area is obtained.
104. And setting the size of a window according to the distance between adjacent grid lines, and dividing the grid line region by using the window with the set size to obtain a plurality of windows.
The method for dividing the grid line region by using the window with the set size comprises the following steps:
acquiring the distance between adjacent grid lines as the width of one grid line, and taking the width of two grid lines as the side length of a window;
obtaining the size of a window according to the set side length, and continuously dividing the window in the grid line region according to the set size;
and acquiring the proportion of the set side length of the window in the length and the width of the grid line region respectively, and acquiring the number of the windows divided in the grid line region according to the proportion of the length and the width of the grid line region respectively.
Since the grid line region is divided into a plurality of windows and the obvious gray level difference between the windows is studied, it is ensured that the textures in the divided windows have relatively similar periodicity (the analysis basis of the subsequent gray level distribution), the gray level difference between non-defective windows is not large (the periodicity is presented, and the included contents are basically consistent), and the difference between the defective windows and the normal texture windows is large.
If the total number of grid lines in the grid line region is L, the side length of the window obtained by the method is as follows:
in the above formula, the first and second carbon atoms are,
is (a)
) The sum of the widths of the grid lines, the distance between two adjacent grid lines is
Therefore using
Rounding down and multiplying by 2 to obtain the side length of the window
The width of the two grid lines is fixed.
The invention marks the size of the window as
In a
The number of windows on one side of the direction is:
in the above formula
Length divided by side length of window
Then rounding down to obtain the window in the rectangle
In
Alongside, or otherwise at
Total number of divisions in direction
。
In that
The number on one side of the direction is:
in the above formula, if the number of gate lines is L, (b) exists between the gate lines
) The width of each grid line is used because the division number of the window can only be an integer and the window is required to have images, and the method has the advantages of simple structure, low cost and high efficiency
Rounding down and the invention takes every two grid line widths as the length of the side length of the window, then we will (
) Divide by 2 and round down to achieve the window at
The number of the division on one side of the direction is
。
The total number of resulting windows of the solar panel is then
Meanwhile, the invention can ensure that one side of the window is parallel to the direction of the grid line (
The other side is perpendicular to the grid line (
Corresponding edges) and that the images in each window are substantially identical (that is, the images in each window are substantially similar), facilitating the subsequent calculation and analysis of the present invention, and a schematic diagram of window division of the gate line region is shown in fig. 2.
105. And constructing a gray scale area matrix of each window according to the number of gray levels in each window, and acquiring a high gray scale area emphasis characteristic value of each window according to the connected domain area formed by each gray level pixel point in the gray scale area matrix of each window.
The texture defects in the solar cell panel mainly comprise spots and fingerprints, the characterization in the grid line region is similar high-gray bright spots, but compared with grid lines with obvious regularity, the texture defects seriously damage the regular characterization of the grid lines, the influence degree of the two defects on the cell panel is different, the fingerprint defects can be regarded as surface stains, the defects are basically removed through subsequent cleaning operation, and the influence degree is small; the spot defect is surface damage caused by overlarge pressure of the device, and generally, waste products are regarded as needing to be sorted out for treatment, so that the influence degree is large.
The method mainly combines a gray scale area matrix (GLSZM), calculates the related characteristic representative information by the corresponding GLSZM obtained from the image in the window, and further obtains the numerical indexes of the spots and the wheel prints (fingerprints) to perform related description, thereby effectively describing the spot or the wheel print characteristics.
And (3) analyzing according to the characteristics of spot defects and wheel mark defects: the spot represents a local highlight part on the grid line texture on the surface of the battery panel, although the gray scale of the wheel print (fingerprint) is slightly lower than that of the spot, the wheel print is more particularly represented by locally discrete and fuzzy gray scale, and the characteristic index of the used gray scale matrix is determined to be the characteristic of a high gray scale region emphasis index according to the characteristics.
High gray area emphasis (HGLZE): the HGLZE measures the distribution of higher gray values, and a larger value indicates that the proportion of the higher gray value and a large area in an image is larger, and the method for acquiring the high gray area emphasis characteristic value of each window comprises the following steps:
acquiring the connected domain area formed by each gray level pixel point in the gray scale area matrix of each window;
obtaining the high gray level region emphasis characteristic value of each window according to the frequency of the connected domain area formed by each gray level pixel point, wherein the expression is as follows:
wherein the content of the first and second substances,
a high gray scale region emphasis feature value representing the mth window,
the frequency of occurrence of a connected domain with the area of j formed by the pixel points representing the ith gray level,
represents the maximum area of the connected component in the window,
16 is the gray scale obtained after the gray scale grading is carried out by the invention (the contrast degree of the gray scales among different defects of spots and wheel prints is improved, the reliability of the calculation result is increased),
、
、
、
the gray scale area matrix is a parameter which can be directly calculated and obtained by the prior art, for spot defects and wheel mark defects in the solar panel, spots with higher gray scale and concentrated areas have wheel marks with more discrete gray scale and lower gray scale, the emphasized characteristic value of the high gray scale area is obviously much larger than that of the wheel marks, and the emphasized characteristic value of the high gray scale area are both larger than the value of a defect-free area, so the characteristic value is used as a judgment basis for distinguishing the defects.
106. And carrying out defect detection on the surface of the battery board to be detected according to the high-gray-scale region emphasis characteristic value of each window, and determining the defect type.
Compared with the wheel mark defect, the spot defect has higher severity, and the wheel mark defect can be removed through subsequent cleaning operation without influencing the conversion efficiency and quality of the cell panel; the spot defect is usually a damage caused by an overlarge pressure of a production machine, cannot be simply processed, and seriously affects the conversion efficiency of the solar panel, so that the spot defect and the solar panel need to be accurately identified.
For a normal window, a fingerprint window and a window with spots, the high-gray-scale region emphasis characteristic values of the three are certainly increased in sequence, and the difference degree between the three is higher, so that the method calculates the high-gray-scale region emphasis characteristic values in each window
And (4) the characteristic value is checked with the difference degree between the characteristic value and a normal window so as to determine whether the window has spots and fingerprints.
The method for detecting the defects on the surface of the battery plate to be detected comprises the following steps:
acquiring the surface gray level image of the solar cell panel without defects, and calculating the solar cell panel without defectsHigh-gray-scale region emphasis characteristic value of gray-scale image on surface of battery plate
；
Acquiring a solar cell panel surface gray image only with the wheel mark defect, and calculating a high gray area emphasis characteristic value of the solar cell panel surface gray image only with the wheel mark defect
；
Acquiring a solar cell panel surface gray image only with spot defects, and calculating a high gray area emphasis characteristic value of the solar cell panel surface gray image only with spot defects
；
And establishing a threshold interval, and judging whether the battery panel to be detected has defects or not according to the high gray area emphasis characteristic value of each window in the gray image of the surface of the battery panel to be detected and the threshold interval.
The method for determining the defect type comprises the following steps:
when the high gray scale area emphasis characteristic value of each window area in the battery board to be detected:
when the battery plate to be detected has no defects;
when a high gray scale region of a window region exists in the battery panel to be detected, emphasizing the characteristic value:
if the window is characterized by small high gray values and small and discrete areas, the battery panel to be detected has a wheel mark defect;
when the high gray scale area of the window area exists in the battery board to be detected, the emphasized characteristic value is as follows:
and if the window has a part with a high gray value and a large area, the panel to be detected has a spot defect.
After specific defects in the solar cell panel are detected, each defect is marked, and related treatment opinions are given according to the marks, specifically:
marking panels with wheel print (fingerprint) defects as
Subsequently all of
Cleaning and drying the defective cell panel, and then carrying out secondary detection;
marking panel with spot defect
Control the pair of transfer devices
The label cell panel is separated and removed, and is recycled and maintained.
Therefore, the solar cell panel defect detection method and the solar cell panel defect detection device can realize the defect detection of the solar cell panel and determine the specific defect type while retaining the grid line of the solar cell panel.
According to the invention, through carrying out gray level division on the image, the characteristics of the wheel mark defect and the spot defect can be conveniently and better divided in the follow-up process, meanwhile, the grid line of the solar panel is reserved, the efficiency of image preprocessing is improved, meanwhile, the characteristics of the defect in the image can not be diluted, further, the window division is carried out according to the grid line area, the specific position of the defect can be obtained while the defect is identified, finally, the gray scale area matrix is adopted for identification according to the gray level expression characteristics of the spot defect and the wheel mark defect in the solar panel, and then the defect category is accurately distinguished through the high gray level area emphasis index, so that the defect detection of the solar panel is realized, and the detection precision and the detection efficiency are higher.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (7)
1. A method for detecting defects on the surface of a solar cell panel is characterized by comprising the following steps:
collecting a gray level image of the surface of a battery plate to be detected;
carrying out gray level grading on the gray level image of the surface of the battery panel to be detected to obtain a gray level image of the area of the battery panel;
extracting all grid lines in the gray level image of the panel area to obtain a grid line area of the gray level image of the panel area;
setting a window size according to the distance between adjacent grid lines, and dividing the grid line region by using the window with the set size to obtain a plurality of windows;
constructing a gray scale area matrix of each window, and acquiring a high gray scale area emphasis characteristic value of each window according to a connected domain area formed by each gray scale pixel point in the gray scale area matrix of each window;
and carrying out defect detection on the surface of the battery board to be detected according to the high-gray-scale region emphasis characteristic value of each window, and determining the defect type.
2. The method for detecting the surface defects of the solar cell panel according to claim 1, wherein the method for obtaining the gray image of the panel area by performing gray grading on the gray image of the surface of the cell panel to be detected comprises the following steps:
establishing a gray level histogram of a gray level image on the surface of the battery plate to be detected, and performing curve fitting on the gray level histogram;
taking a gray value corresponding to a first trough from low to high in the curve as a background gray level;
averagely dividing all gray values larger than the background gray level into a plurality of gray levels to obtain the number of pixel points in each gray level;
and obtaining a panel area gray image according to all the pixel points with the gray values larger than the background gray level.
3. The method for detecting the surface defects of the solar cell panel according to claim 1, wherein the method for acquiring the grid line region of the gray image of the panel region comprises the following steps:
and extracting all grid lines in the gray image of the panel area, constructing a surrounding frame area according to the length of the grid lines and the number of the grid lines, and taking the surrounding frame area as the grid line area.
4. The method for detecting the defects on the surface of the solar cell panel according to claim 1, wherein the method for dividing the grid line region by using the window with the set size comprises the following steps:
acquiring the distance between adjacent grid lines as the width of one grid line, and taking the width of two grid lines as the side length of a window;
obtaining the size of a window according to the set side length, and continuously dividing the window in the grid line region according to the set size;
and acquiring the proportion of the set side length of the window in the length and the width of the grid line region respectively, and acquiring the number of the windows divided in the grid line region according to the proportion of the length and the width of the grid line region respectively.
5. The method for detecting the surface defects of the solar cell panel according to claim 1, wherein the method for acquiring the high-gray-scale region emphasis feature value of each window comprises the following steps:
acquiring the connected domain area formed by each gray level pixel point in the gray scale area matrix of each window;
obtaining the high gray level region emphasis characteristic value of each window according to the frequency of the connected domain area formed by each gray level pixel point, wherein the expression is as follows:
wherein the content of the first and second substances,
a high gray scale region emphasis feature value representing the mth window,
the frequency of occurrence of a connected domain with the area of j formed by the pixel points representing the ith gray level,
represents the maximum area of the connected component in the window,
indicating the number of connected components in the window.
6. The method for detecting the defects on the surface of the solar cell panel according to claim 1, wherein the method for detecting the defects on the surface of the solar cell panel to be detected comprises the following steps:
acquiring a surface gray level image of the solar cell panel without defects, and calculating a high gray level region emphasis characteristic value of the surface gray level image of the solar cell panel without defects
；
Acquiring a solar cell panel surface gray image only with the wheel mark defect, and calculating a high gray area emphasis characteristic value of the solar cell panel surface gray image only with the wheel mark defect
；
Acquiring a solar cell panel surface gray image only with spot defects, and calculating a high gray area emphasis characteristic value of the solar cell panel surface gray image only with spot defects
；
And establishing a threshold interval, and judging whether the battery board to be detected has defects or not according to the high gray area emphasis characteristic value of each window in the gray image of the surface of the battery board to be detected and the threshold interval.
7. The method for detecting the surface defects of the solar cell panel according to claim 6, wherein the method for determining the defect type comprises the following steps:
when the high gray scale area emphasis characteristic value of each window area in the battery board to be detected:
when the battery plate to be detected has no defects;
when a high gray scale region of a window region exists in the battery panel to be detected, emphasizing the characteristic value:
when the battery plate is detected to have the wheel mark defect;
when the high gray scale area of the window area exists in the battery board to be detected, the emphasized characteristic value is as follows:
and when the battery plate to be detected has spot defects.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210964410.2A CN115049645B (en) | 2022-08-12 | 2022-08-12 | Solar cell panel surface defect detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210964410.2A CN115049645B (en) | 2022-08-12 | 2022-08-12 | Solar cell panel surface defect detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115049645A true CN115049645A (en) | 2022-09-13 |
| CN115049645B CN115049645B (en) | 2022-11-04 |
Family
ID=83166814
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210964410.2A Active CN115049645B (en) | 2022-08-12 | 2022-08-12 | Solar cell panel surface defect detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115049645B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115330768A (en) * | 2022-10-12 | 2022-11-11 | 江苏跃格智能装备有限公司 | Quality grading method for solar cell panel |
| CN116485785A (en) * | 2023-06-15 | 2023-07-25 | 无锡斯达新能源科技股份有限公司 | Surface defect detection method for solar cell |
| CN116977335A (en) * | 2023-09-22 | 2023-10-31 | 山东贞元汽车车轮有限公司 | Intelligent detection method for pitting defects on surface of mechanical part |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010117337A (en) * | 2008-11-12 | 2010-05-27 | Nippon Electro Sensari Device Kk | Surface defect inspection device |
| US20150053869A1 (en) * | 2013-08-23 | 2015-02-26 | Dainippon Screen Mfg. Co., Ltd. | Inspection apparatus and inspection method |
| CN107274393A (en) * | 2017-06-12 | 2017-10-20 | 郑州轻工业学院 | The monocrystaline silicon solar cell piece detection method of surface flaw detected based on grid line |
| 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 |
| CN112233101A (en) * | 2020-10-26 | 2021-01-15 | 钟竞 | Photovoltaic cell panel quality evaluation method and system based on artificial intelligence |
| CN113379703A (en) * | 2021-06-08 | 2021-09-10 | 西安理工大学 | Photovoltaic panel dark spot defect detection method based on Yolo-v4 network structure |
| CN113837994A (en) * | 2021-07-29 | 2021-12-24 | 尚特杰电力科技有限公司 | Photovoltaic panel defect diagnosis method based on edge detection convolutional neural network |
| CN114842009A (en) * | 2022-07-04 | 2022-08-02 | 江苏奥派电气科技有限公司 | Cable defect detection optimization method based on gray level run matrix |
| CN114882275A (en) * | 2022-04-24 | 2022-08-09 | 江苏南通二建集团讯腾云创智能科技有限公司 | Building board classification method using electronic equipment data processing |
-
2022
- 2022-08-12 CN CN202210964410.2A patent/CN115049645B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010117337A (en) * | 2008-11-12 | 2010-05-27 | Nippon Electro Sensari Device Kk | Surface defect inspection device |
| US20150053869A1 (en) * | 2013-08-23 | 2015-02-26 | Dainippon Screen Mfg. Co., Ltd. | Inspection apparatus and inspection method |
| CN107274393A (en) * | 2017-06-12 | 2017-10-20 | 郑州轻工业学院 | The monocrystaline silicon solar cell piece detection method of surface flaw detected based on grid line |
| 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 |
| CN112233101A (en) * | 2020-10-26 | 2021-01-15 | 钟竞 | Photovoltaic cell panel quality evaluation method and system based on artificial intelligence |
| CN113379703A (en) * | 2021-06-08 | 2021-09-10 | 西安理工大学 | Photovoltaic panel dark spot defect detection method based on Yolo-v4 network structure |
| CN113837994A (en) * | 2021-07-29 | 2021-12-24 | 尚特杰电力科技有限公司 | Photovoltaic panel defect diagnosis method based on edge detection convolutional neural network |
| CN114882275A (en) * | 2022-04-24 | 2022-08-09 | 江苏南通二建集团讯腾云创智能科技有限公司 | Building board classification method using electronic equipment data processing |
| CN114842009A (en) * | 2022-07-04 | 2022-08-02 | 江苏奥派电气科技有限公司 | Cable defect detection optimization method based on gray level run matrix |
Non-Patent Citations (4)
| Title |
|---|
| SIVA RAMAKRISHNA MADETI ETAL.: "A comprehensive study on different types of faults and detection techniques for solar photovoltaic system", 《SOLAR ENERGY》 * |
| 伍李春等: "基于人工神经网络的太阳能电池片表面质量检测系统", 《合肥工业大学学报(自然科学版)》 * |
| 张舞杰等: "硅太阳能电池纹理缺陷检测", 《计算机应用》 * |
| 陈智强等: "硅电池片自动串焊表面缺陷在线视觉检测研究", 《机电工程》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115330768A (en) * | 2022-10-12 | 2022-11-11 | 江苏跃格智能装备有限公司 | Quality grading method for solar cell panel |
| CN116485785A (en) * | 2023-06-15 | 2023-07-25 | 无锡斯达新能源科技股份有限公司 | Surface defect detection method for solar cell |
| CN116485785B (en) * | 2023-06-15 | 2023-08-18 | 无锡斯达新能源科技股份有限公司 | Surface defect detection method for solar cell |
| CN116977335A (en) * | 2023-09-22 | 2023-10-31 | 山东贞元汽车车轮有限公司 | Intelligent detection method for pitting defects on surface of mechanical part |
| CN116977335B (en) * | 2023-09-22 | 2023-12-12 | 山东贞元汽车车轮有限公司 | Intelligent detection method for pitting defects on surface of mechanical part |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115049645B (en) | 2022-11-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115049645B (en) | Solar cell panel surface defect detection method | |
| CN111179225B (en) | Test paper surface texture defect detection method based on gray gradient clustering | |
| CN110097034B (en) | Intelligent face health degree identification and evaluation method | |
| CN108355987B (en) | A kind of screen printing of battery quality determining method based on piecemeal template matching | |
| CN115239704B (en) | Accurate detection and repair method for wood surface defects | |
| CN116645367B (en) | Steel plate cutting quality detection method for high-end manufacturing | |
| CN107067000B (en) | A kind of recognition methods of bank note degree and device | |
| CN101739549B (en) | Face detection method and system | |
| CN113646801A (en) | Defect detection method and device for defect image and computer readable storage medium | |
| CN104239850A (en) | Answer sheet filling information automatic recognition method based on comprehensive features of images | |
| CN108318491A (en) | A kind of fabric defect detection method based on frequency spectrum curvature analysis | |
| CN113313677A (en) | Quality detection method for X-ray image of wound lithium battery | |
| CN113325001A (en) | Automatic distinguishing and detecting equipment and method for surface appearance flaws of copper foil | |
| CN113362276B (en) | Visual detection method and system for plates | |
| CN109886954A (en) | A kind of printed matter defect inspection method | |
| CN112819844B (en) | Image edge detection method and device | |
| CN115131353B (en) | Flat screen printing textile production abnormity identification and positioning method and system | |
| CN117152161A (en) | Shaving board quality detection method and system based on image recognition | |
| CN115115643A (en) | Method for detecting production defects of photovoltaic cell panel | |
| CN113962929A (en) | Photovoltaic cell assembly defect detection method and system and photovoltaic cell assembly production line | |
| CN102313740B (en) | Solar panel crack detection method | |
| CN111476792B (en) | Extraction method of strip steel image contour | |
| CN112132823A (en) | Photovoltaic cell image local uneven illumination processing method based on artificial intelligence | |
| CN104655642A (en) | Automatic measurement, characterization and classification method for steel cracking defects and system of method | |
| CN115311266B (en) | Printing and dyeing identification method in textile printing and dyeing process |
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 |