CN116363097A - Defect detection method and system for photovoltaic panel - Google Patents
Defect detection method and system for photovoltaic panel Download PDFInfo
- Publication number
- CN116363097A CN116363097A CN202310330348.6A CN202310330348A CN116363097A CN 116363097 A CN116363097 A CN 116363097A CN 202310330348 A CN202310330348 A CN 202310330348A CN 116363097 A CN116363097 A CN 116363097A
- Authority
- CN
- China
- Prior art keywords
- image
- photovoltaic panel
- defect detection
- type
- grid line
- 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 186
- 238000001514 detection method Methods 0.000 title claims abstract description 130
- 238000000034 method Methods 0.000 claims abstract description 65
- 238000012545 processing Methods 0.000 claims abstract description 47
- 238000007781 pre-processing Methods 0.000 claims abstract description 16
- 238000004422 calculation algorithm Methods 0.000 claims description 23
- 238000004590 computer program Methods 0.000 claims description 18
- 230000011218 segmentation Effects 0.000 claims description 16
- 238000003860 storage Methods 0.000 claims description 15
- 230000002708 enhancing effect Effects 0.000 claims description 14
- 230000004927 fusion Effects 0.000 claims description 13
- 230000008569 process Effects 0.000 claims description 13
- 230000009466 transformation Effects 0.000 claims description 12
- 230000000149 penetrating effect Effects 0.000 claims description 9
- 238000012937 correction Methods 0.000 claims description 4
- 230000001131 transforming effect Effects 0.000 claims description 2
- 230000000694 effects Effects 0.000 description 25
- 238000010586 diagram Methods 0.000 description 23
- 238000005286 illumination Methods 0.000 description 9
- 230000002950 deficient Effects 0.000 description 7
- 238000002474 experimental method Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 5
- 238000011156 evaluation Methods 0.000 description 4
- 238000001914 filtration Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 238000005424 photoluminescence Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000005401 electroluminescence Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000005284 excitation Effects 0.000 description 2
- 238000003702 image correction Methods 0.000 description 2
- 238000007500 overflow downdraw method Methods 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 201000004569 Blindness Diseases 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 235000019800 disodium phosphate Nutrition 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000000241 photoluminescence detection Methods 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 238000011895 specific detection Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
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
- G06T7/0006—Industrial image inspection using a design-rule based approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- 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
-
- 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)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a defect detection method and a defect detection system for a photovoltaic panel, wherein the defect detection method comprises the steps of obtaining a photovoltaic panel image to be detected and preprocessing the photovoltaic panel image; performing image enhancement processing on the photovoltaic panel image by adopting an MSRCR enhancement method fused with a high-pass filter to obtain a grid line image of the photovoltaic panel; performing first-type defect detection on the grid line image by adopting a defect detection method based on region growth to obtain a first-type defect detection result containing cracks and unfilled corners; the method provided by the invention is used for processing the photovoltaic panel image, so that the grid information in the image can be more accurately obtained, the noise influence is reduced as much as possible, the grid line image which is more suitable for the detection of the photovoltaic panel is obtained, and the defect detection precision is improved.
Description
Technical Field
The invention belongs to the technical field of defect detection of photovoltaic panels, and particularly relates to a defect detection method and system of a photovoltaic panel.
Background
Along with the increase of energy demand and the improvement of environmental awareness, solar photovoltaic cells (hereinafter referred to as photovoltaic panels) are widely used and popularized. The quality and performance of the photovoltaic panel, which is an important component of solar power generation, directly affect the power generation efficiency and service life of the solar photovoltaic system. Therefore, accurate and efficient detection of the quality and performance of photovoltaic panels is critical. At present, the commonly used photovoltaic panel detection method mainly comprises two types of manual detection and automatic detection. The manual detection method is usually visual inspection, microscopic detection and other methods, but the manual detection efficiency is low and the detection result is unstable. In contrast, with the development and perfection of machine vision and image processing technologies, the application of the machine vision technology to perform automated defect detection has the advantages of high efficiency and accuracy, and is widely applied in the market.
The detection of Electroluminescence (EL) is preferred because it is prone to damage to the photovoltaic panel. Although some research efforts have been made to detect defects in photovoltaic panels, many challenges and difficulties remain. Among them, the complexity of the defects such as uneven illumination of the photovoltaic panel image, grid lines and cracks is a difficulty to be solved in the current research. Some researchers have also proposed partial solutions to these difficulties. Ma Xiaolong and the like, aiming at the uneven illumination phenomenon in the acquired image, an image enhancement algorithm based on partial overlapping of areas is provided, the operation amount is small, the effect is obvious, the algorithm obtains the adjustment scale of the pixel gray value of the area according to the brightness difference value, then an image interpolation method is adopted to expand the adjustment range of the pixel gray value from the area to the pixel point, and finally each pixel point is adjusted so as to eliminate the uneven illumination phenomenon. Hu Mojie and the like, performing image segmentation by combining a histogram equalization method and an adaptive threshold method after removing grid lines on an image, detecting connected domains in the binary segmented image by adopting a Two-Pass algorithm, marking the connected domains with different colors, and marking the connected domains with the areas larger than the threshold as defects. Although the above researches have achieved a certain effect, there are problems of few detection defect types, excessive loss of image details and the like.
By observing the region to be tested of the photovoltaic panel, defects can be divided into two types, wherein the first type comprises cracks and unfilled corners, the defects are characterized by crossing and connecting a plurality of grid lines, the crack growth direction can be a fold line in any direction, and the unfilled corners are necessarily positioned at four corners of the photovoltaic panel and are expressed as straight lines or curves. The second type is stain, dead spot, etc., and the defects are represented as black spots with different sizes randomly distributed on each position of the photovoltaic panel. In the use process, the first type of defects have higher destructiveness, which is usually caused by internal factors such as lattice defects, material stress and the like, so that in the quality evaluation of the photovoltaic panel, the detection and the evaluation of the first type of defects are more important.
Disclosure of Invention
Aiming at the technical problem that the first type of defects are difficult to accurately identify in the prior art, the invention provides a defect detection method and system for a photovoltaic panel. The defect detection method provided by the invention provides the MSRCR enhancement method fused with the high-pass filter, so that the photovoltaic panel image is processed, the grid information in the image can be more accurately obtained, the influence of noise is reduced as much as possible, the grid line image more suitable for the detection of the photovoltaic panel is obtained, and the defect detection precision is improved.
A method for defect detection of a photovoltaic panel, comprising the steps of:
s1: acquiring a photovoltaic panel image to be detected and preprocessing;
s2: performing image enhancement processing on the photovoltaic panel image by adopting an MSRCR enhancement method fused with a high-pass filter to obtain a grid line image of the photovoltaic panel;
the method comprises the steps of firstly enhancing the photovoltaic panel image by using an MSRCR algorithm, then converting the enhanced image from a space domain to a frequency domain to obtain a spectrogram, inputting the spectrogram into the high-pass filter, and then converting the filtered spectrogram to the space domain to obtain a grid image;
s3: performing first-type defect detection on the grid line image by adopting a defect detection method based on region growth to obtain a first-type defect detection result;
wherein the first type of defect comprises at least a crack and a unfilled corner.
Photoluminescence is the excitation of semiconductor materials in photovoltaic panels by a light source to produce fluorescence of a certain wavelength, which is weaker at defect sites and appears as darker areas. When a photovoltaic panel with shooting defects exists, illumination of the whole image is uneven, noise in the image is increased, and misjudgment is easily generated on the broken position of the grid line during detection. In order to make the grid lines in the image clearer and easy to process subsequent steps, the technical scheme of the invention provides image enhancement processing combining high-pass filtering and MSRCR algorithm, so as to obtain the grid line image of the photovoltaic panel.
Further alternatively, the high pass filter is represented as follows:
in the above formula, H (u, v) represents a value of the high-pass filter at a point (u, v) on the spectrogram, 1 represents pass, and 0 represents no pass; d (u, v) represents the distance of a point (u, v) in the spectrogram from the center of the image, M and N represent the length and width of the spectrogram, wherein,
is the coordinates of the image center of the spectrogram; d represents the radius of a circle drawn by the center of the image of the spectrogram, and is a constant; w represents the bandwidth of the high-pass filter and +.>
Further optionally, the defect detection method further includes: the method further comprises the following steps of transforming gray values of the grid line image:
B(x,y)＝f(x,y)I(x,y)
wherein B (x, y) represents the gray value of the pixel point (x, y) after transformation, f (x, y) represents the transformation coefficient, and I (x, y) represents the pixel value of the pixel point (x, y) coordinate in the grid line image; taking 5 pixels at any point in the grid line image and in the up-down direction, calculating the difference between the average value of gray values of the 4 pixels except the center point and the average value of gray values of the 5 selected pixels, and recording the difference as S; q, Q are specific values of the transformation coefficients under the corresponding conditions, and Q is more than 1 and less than 1.
Further alternatively, the process of performing the first type defect detection on the grid line image by using a defect detection method based on region growth to obtain a first type defect detection result is as follows:
firstly, generating a penetrating connecting line perpendicular to a grid line in the middle of an image of the grid line image;
starting a seed point at the intersection point of the penetrating connecting line and the leftmost or right grid line, and growing in the upper, lower and right/left directions;
and if the top or bottom positions of the grid lines and the top/bottom positions of the rest grid lines are different in the height direction by more than a set threshold value in the growth process, the corresponding top/bottom positions are included in the first type of defect positions.
Further optionally, the defect detection method further includes:
carrying out grid line removal treatment on the photovoltaic panel image after the preprocessor;
then, carrying out enhancement treatment on the image from which the grid lines are removed;
threshold segmentation processing is carried out on the image after the enhancement processing, so that the stain position is determined;
splicing and fusing the stain position and the first type defect detection result to obtain a detection result containing the first type defect and the second type defect;
the second type of defects are black dots with different sizes which are randomly distributed on each position of the photovoltaic panel.
Further optionally, preprocessing the photovoltaic panel image in step 1 includes: performing inclination correction on the photovoltaic panel image by adopting a Hough linear transformation algorithm; and then, performing ROI region clipping on the corrected photovoltaic panel image.
In a second aspect, the present invention provides a detection system based on the defect detection method, which includes:
the image acquisition and preprocessing module is used for acquiring the image of the photovoltaic panel to be detected and preprocessing the image;
the grid line image generation module is used for carrying out image enhancement processing on the photovoltaic panel image by adopting an MSRCR enhancement method fused with a high-pass filter to obtain a grid line image of the photovoltaic panel;
the method comprises the steps of firstly enhancing the photovoltaic panel image by using an MSRCR algorithm, then converting the enhanced image from a space domain to a frequency domain to obtain a spectrogram, inputting the spectrogram into the high-pass filter, and then converting the filtered spectrogram to the space domain to obtain a grid image;
the first type defect detection module is used for detecting the first type defects of the grid line image by adopting a defect detection method based on region growth to obtain a first type defect detection result; wherein the first type of defect comprises at least a crack and a unfilled corner.
Further optionally, the system further includes a second type defect detection module and a fusion module, the second type defect detection module is configured to detect a second type defect in the photovoltaic panel image, where the second type defect detection module includes: the grid line removing module, the enhancing module and the threshold dividing module;
the grid line removing module is used for removing grid lines from the photovoltaic panel image after the preprocessor;
the enhancement module is used for enhancing the image with the grid lines removed;
the threshold segmentation module is used for carrying out threshold segmentation processing on the image after the enhancement processing so as to determine the stain position;
the fusion module is used for splicing and fusing the stain position and the first type defect detection result to obtain a detection result containing the first type defect and the second type defect;
the second type of defects are black dots with different sizes which are randomly distributed on each position of the photovoltaic panel.
In a third aspect, the present invention provides an electronic terminal, comprising:
one or more processors;
a memory storing one or more computer programs;
the processor invokes the computer program to perform:
a method for detecting defects of a photovoltaic panel.
In a fourth aspect, the present invention provides a readable storage medium storing a computer program, the computer program being invoked by a processor to perform:
a method for detecting defects of a photovoltaic panel.
Advantageous effects
1. According to the defect detection method and system for the photovoltaic panel, provided by the technical scheme of the invention, aiming at the first type of defects, the MSRCR enhancement method fused with the high-pass filter is set, and then the photovoltaic panel image is processed to obtain the grid line image, so that the grid information in the image can be more accurately obtained, the illumination information is improved, the grid line detail information is locally enhanced, and then the grid line image which is more suitable for the detection of the photovoltaic panel is obtained, and the defect detection precision is improved.
2. The technical scheme of the invention is further optimized, and the second type of defect detection is realized on the basis of the first type of defect detection, namely, the defects including cracks and unfilled corners and the other type of defects including stains, dead spots and the like are detected more comprehensively.
3. The technical scheme of the invention is further optimized, and a penetrating connecting line perpendicular to the grid line is generated in the middle of the image of the grid line image; and then starting a seed point at the intersection point of the penetrating connecting line and the leftmost or right grid line, and growing in the upper, lower and right/left directions. Through the technical means, the area growth is improved, the blindness of determining the seed points is solved, the calculated amount is reduced, and the detection efficiency is improved.
Drawings
FIG. 1 is an overall flow chart of the defect detection method provided by the present invention;
FIG. 2 is a flowchart of preprocessing a photovoltaic panel image, wherein (a) is an original image, (b) is a schematic diagram of inclination correction, and (c) is a schematic diagram of ROI clipping;
FIG. 3 is an image enhancement effect diagram, wherein (a), (b), (c), (d), (e), and (f) correspond to an original image, an effect diagram after logarithmic transformation, an effect diagram after histogram equalization, an effect diagram after CLAHE processing, an effect diagram after SSR processing, and an effect diagram after MSRCR processing, respectively;
FIG. 4 is a schematic diagram of a high-pass filter, wherein (a) and (b) correspond to a conventional high-pass filter and a high-pass filter of the present invention, respectively;
fig. 5 is a diagram of a grid line extraction effect diagram based on high-pass filtering, wherein (a), (b) and (c) respectively correspond to a photovoltaic cell spectrogram, an effect diagram after spectrum filtering, and a grid line extraction schematic diagram;
FIG. 6 is a view of the result of image enhancement fused with high-pass filtering, wherein (a), (b), (c), and (d) correspond to the original image and the schematic view of linear superposition processing, the schematic view of differential processing, and the schematic view of the processing according to the method of the present invention, respectively;
FIG. 7 is a gate line connection diagram, wherein (a) and (b) correspond to the schematic diagrams of the through connection line and the first type defect respectively, and the schematic diagrams of the through connection line are drawn;
FIG. 8 is a graph of the first type of defect detection results based on region growth, wherein (a), (b), (c), and (d) correspond to the region growth schematic, the image to be segmented, the region growth result, and the first type of defect detection result schematic, respectively;
FIG. 9 is a diagram of the final defect detection result, wherein (a), (b), (c), (d), (e), and (f) correspond to the original image, the schematic drawing of the grid, the schematic drawing after SSR enhancement, the schematic drawing after threshold segmentation, the schematic drawing of stain location, and the schematic drawing of fusion location result, respectively;
fig. 10 is a graph showing the contrast of the image enhancement effect, wherein (a), (b) and (c) correspond to the original image, the MSRCR and the method of the present invention, respectively.
Detailed Description
The invention provides a defect detection method of a photovoltaic panel, which is used for detecting the defects of the photovoltaic panel and has the core that an MSRCR enhancement method of a fusion high-pass filter is set, and then a grid line image is obtained by processing a photovoltaic panel image, so as to obtain grid line image improved illumination information and local enhancement grid line detail information. The invention will be further illustrated with reference to examples.
Example 1:
the defect detection method of the photovoltaic panel provided by the embodiment of the invention comprises the following steps of:
s1: and acquiring a photovoltaic panel image to be detected and preprocessing.
As shown in fig. 2, when the image of the photovoltaic panel is actually photographed, a part of the photovoltaic panel is improperly placed and may be inclined to some extent, and correction is required for facilitating subsequent processing. Because the photoluminescence detection images are shot in a darkroom, other parts except the photovoltaic panel are black, four-side edge information can be effectively extracted from the photovoltaic panel image through a Hough linear transformation algorithm, and a good image correction effect can be obtained by calculating the average value of four-side edge slopes. And finally determining ROI (region of interest) areas according to coordinates of four corner points of the photovoltaic panel in the corrected image for cutting.
Since the hough straight line transformation algorithm is the prior art for image correction and ROI region clipping, it will not be specifically described.
S2: performing image enhancement processing on the photovoltaic panel image by adopting an MSRCR enhancement method fused with a high-pass filter to obtain a grid line image of the photovoltaic panel; the method comprises the steps of enhancing the photovoltaic panel image by using an MSRCR algorithm, converting the enhanced image from a spatial domain to a frequency domain to obtain a spectrogram, inputting the spectrogram into the high-pass filter, and converting the filtered spectrogram to the spatial domain to obtain a grid line image, as shown in fig. 5.
The high pass filter is represented as follows:
in the above formula, H (u, v) represents a value of the high-pass filter at a point (u, v) on the spectrogram, 1 represents pass, and 0 represents no pass; d (u, v) represents the distance of a point (u, v) in the spectrogram from the center of the image, M and N represent the size, i.e., length and width, of the spectrogram, wherein,
is the coordinates of the image center of the spectrogram; d represents the radius of a circle drawn by the center of the image of the spectrogram, and is a constant; w represents the bandwidth of the high-pass filter and +.>
Through experiments, the grid line with the parameter of w= 5,d =55, which is used for extracting the high-frequency component of the photovoltaic cell, has good extraction effect, namely the grid line is used as the optimal value.
The invention combines the high-pass filter and the MSRCR to obtain the image more suitable for the detection of the photovoltaic panel. In order to prevent the grid line from affecting the original defects during fusion, a weighted fusion method is designed according to the texture structure characteristics of the grid line and the first type of defects. In the defective area, the gray value of the grid line is obviously weaker than that of the normal part, because the grid line is positioned in the vertical direction, 5 pixels are taken from any point in the image and in the up-down direction, the difference between the average value of the gray values of 4 pixels and the average value of the gray values of 5 pixels except the center point is calculated, and is recorded as S, and the calculation formula is as follows:
in the above formula, I (x, y) represents a pixel value of an (x, y) coordinate in the grid line image, and since the gray value of the defective portion is significantly lower than that of the grid line portion, when S is greater than a certain threshold value λ (empirical value), the defective portion is considered to be a possible position of the defect, and the value of λ is determined experimentally. Because the illumination is uneven, the grid line extracted from the original image is a line with uneven brightness, so that when S is approximately equal to 0, the background area except the grid line and the defect is represented, when S is other values, the grid line is represented, and according to the characteristic, the gray value of the image after MSRCR enhancement is transformed as follows:
B(x,y)＝f(x,y)I(x,y)
in the above expression, B (x, y) represents the gradation value of the pixel point (x, y) after conversion, and f (x, y) represents the conversion coefficient. For defective positions, the gray value is increased, so that the defective positions look brighter; for the normal part of the grid lines, the gray value is reduced, so that the grid lines look darker; for the background portion, the gray value is not changed, so that the grid line of the whole image is more prominent. It should be noted that, the specific value of the above-mentioned transformation coefficient is the best example of the present invention, but the present invention is not limited to this, and in other possible examples, the gray value is increased based on the "defective position; for the grid line of the normal part, reducing the gray value; the background part is adjusted without changing the gray value. In addition, the standard that S is equal to about 0 is set according to the actually required precision, that is, about equal range centered on 0 is set according to the precision requirement, and the specific value of the present invention is not limited thereto.
The MSRCR enhancement method adopting the fusion high-pass filter in the technical scheme of the invention considers that:
photoluminescence is the excitation of semiconductor materials in photovoltaic panels by a light source to produce fluorescence of a certain wavelength, which is weaker at defect sites and appears as darker areas. When a photovoltaic panel with shooting defects exists, illumination of the whole image is uneven, noise in the image is increased, and misjudgment is easily generated on the broken position of the grid line during detection. To make the gate lines in the image clearer for processing in subsequent steps. However, in the image enhancement method, the histogram equalization, logarithmic change, the CLAHE, the SSR and the MSRCR cannot achieve the better effect, wherein the logarithmic change only can treat the darker parts around in the image, but the enhancement of the grid lines does not have much effect, but the similarity between the grid lines of part of the highlight areas and the background becomes high, the histogram equalization, the CLAHE and the SSR have excellent enhancement effect on the grid lines, but the dark parts around the photovoltaic panel are not treated in place, the halation is generated to influence the normal area of part of the photovoltaic panel, the erroneous judgment of edge area defects is extremely easy to be caused, the MSRCR suppresses the appearance of halation, and most of details of the image are reserved, but the enhancement of the grid lines in the image is still insufficient, and the distinction between the enhancement and the background is not high. In addition, the traditional high-pass filter can pass high frequencies and filter or attenuate low frequencies, so that the image sharpens the salient boundary, but because the illumination in the photovoltaic panel image is not uniform, if the traditional high-pass filter is used, noise information in the image is amplified together during enhancement, and the final detection result is affected. The specific effects of each algorithm are shown in fig. 3. Therefore, the MSRCR enhancement method fused with the high-pass filter can more accurately acquire grid information in the image and reduce the influence of noise as much as possible.
It should be noted that, compared with some other image fusion methods, the linear superposition does not consider the characteristics of the photovoltaic panel image, and the defect area is excessively enhanced; while the manner of differential fusion over emphasizes the gate line characteristics. The method comprehensively considers the two points, and performs specific fusion of the images by utilizing the process aiming at the characteristics of the photovoltaic panel, so that a better image enhancement effect can be obtained.
S3: and performing first-type defect detection on the grid line image by adopting a defect detection method based on region growth to obtain a first-type defect detection result. Firstly, generating a penetrating connecting line perpendicular to a grid line in the middle of an image of the grid line image; starting a seed point at the intersection point of the penetrating connecting line and the leftmost or right grid line, and growing in the upper, lower and right/left directions; wherein, if the top or bottom positions of the grid lines and the top/bottom positions of the rest grid lines are different by more than a threshold value in the height direction in the growth process, the corresponding top/bottom positions are included in the first type defect positions.
Among other things, the first type of defect causes the photovoltaic panel to break, i.e. the grid is broken at a certain position. A mode of combining edge detection with straight line detection is adopted to generate a plurality of broken line segments, so that the line segments are difficult to distinguish, the judgment of defect positions is greatly interfered, and the robustness is not strong. In order to accurately extract the first type of defect position, the invention adopts a region growing mode to find a breakpoint along the grid line, and when the grid can extend to the top and the bottom, the position is free of defects, and if the position is not reached, the position of the breakpoint is the coordinate of the first type of defect. Because of the different images taken, the grid line spacing will be slightly different, and finding the seed points for each grid line is a very difficult task.
For this purpose, as shown in fig. 7, the present invention generates a horizontal line perpendicular to the grid lines at the center of the image, the connecting line penetrating all the grid lines of the photovoltaic panel, and the gray value is equivalent to the grid lines. The first type of defects may be distributed above, below or on the through connection line, but in either way, the region growth will be blocked by the crack defects, so the region where the first type of defects are located can be known only by determining the positions of the upper and lower vertices on the gate line. Because the distribution condition of the through connection lines and the grid lines is known, the time and space cost of the region growth can be greatly reduced according to the prior information.
As shown in fig. 8, in this embodiment, the intersection point of the connecting line and the leftmost gate line is used as the initial seed point, and only the growth needs to be performed in the up, down and right directions. The method is used for detecting the first type of defects, only the top and bottom coordinate positions of each grid line are recorded, and when the coordinate of a certain vertex is far away from the coordinate of other vertexes, the point can be judged to be contained in the first type of defects.
Thus, if the gate line top or bottom position differs from the remaining plurality of gate line top/bottom positions by more than a threshold value (empirical value) in the height direction during the growth, the corresponding top/bottom position is included in the first type of defect position. Wherein, the "remaining majority gate line" refers to a majority gate line with top/bottom positions close to each other among the remaining gate lines, and whether the top/bottom of the gate line is a defective portion during growth is determined by the top/bottom positions of the partial gate line as a standard.
S4: carrying out grid line removal treatment on the photovoltaic panel image after the preprocessor; and then carrying out enhancement processing on the image with the grid lines removed. Among them, SSR image enhancement processing is preferable.
S5: threshold segmentation processing is carried out on the image after the enhancement processing, so that the stain position is determined; and splicing and fusing the stain position and the first type defect detection result to obtain a detection result containing the first type defect and the second type defect.
The second type of defects such as dead spots exist in the photovoltaic panel, and as the spots are not necessarily located on the grid lines, good detection results cannot be obtained by adopting the method. In order to better detect the stain, the method uses a threshold segmentation mode after grid line removal to find the stain on the photovoltaic panel. Considering that stains in an image also belong to high frequency information, the step of removing the grid in the image can cause the information to be weakened, so that the image is necessary to be enhanced after grid removal. Therefore, after grid removal, the positions of the spots can be highlighted by adopting SSR enhancement, and the spot positions with low gray values and the background parts with high gray values can be separated by adopting a threshold segmentation mode to find the spot positions. And similarly, detecting the first type of defects in the step, fusing the first type of defects with the result of the region growth, removing the overlapping part of the first type of defects, and obtaining all defect positions of the photovoltaic panel to finish the defect detection of the photovoltaic panel. The effect of the specific detection process is shown in fig. 9.
It should be noted that, the removal of the grid from the image, the threshold segmentation process, and the determination of the stain position from the image after the threshold segmentation process, i.e., how to determine the stain position, are all prior art means, and therefore, they are not specifically described.
And (3) verification:
photoluminescence image data of a photovoltaic cell is collected as a sample, and main defects in the sample image are unfilled corners, cracks, stains and other defects, and the invention has divided the defects into two main types. Two sets of experiments were performed to verify the defect detection method proposed by the present invention. The first group of experiments are carried out by selecting different photovoltaic panel images for enhancement, and the image enhancement method is evaluated by subjective comparison results and objective evaluation indexes. The second group of experiments provides defect detection evaluation indexes based on the existing data set, and the feasibility of the method provided by the invention is proved by comparing the defect detection effects of other algorithms.
In the image enhancement experiment, partial images are randomly selected to compare the improved algorithm with the classical algorithm. MSRCR enhances image contrast through multiscale Retinex algorithm, can effectively strengthen image detail information, but in the photovoltaic panel detection, the panel is mainly uneven white, and the algorithm is difficult to adapt to scene change of each photovoltaic panel in parameters, and the enhancement effect on grid lines is poor. Compared with MSRCR, the algorithm of the invention strengthens the column direction texture of the photovoltaic panel, and the condition of detail loss does not occur. In order to better evaluate the enhancement effect, the invention adopts a non-reference objective evaluation index NIQE with a stronger theoretical basis to evaluate the image, the index establishes a series of characteristics capable of representing the image quality, and the characteristics are described by utilizing a multi-element Gaussian model, and the mathematical expression is as follows:
v in the above 1 ，v 2 Sum sigma 1 ，∑ 2 The smaller the mean vector and covariance matrix of the multiple Gaussian model of the natural image and the multiple Gaussian model of the distorted image are, the better the image quality is. Five groups of images are selected for comparison test, and the NIQE values of each group of images are below MSRCR, so that the superiority of the method in improving the image quality is shown.
In a defect detection experiment, in order to evaluate the defect detection effect, the method adopts a counting mode to judge the effect, records the number of the first type of defects and the second type of defects respectively, and compares the number with the number of the defects detected by adopting a grid line removing method. Similar to image enhancement, five sets of images were still used for comparison experiments, see in particular the results in table 2: compared with the grid removing mode, the accuracy of the first type defect detection is improved by 11.76%, and the accuracy of the two types of defects is 84.1%. The reliability of the defect detection method of the present invention is further demonstrated thereby.
TABLE 1
TABLE 2
It should be noted that, in some embodiments, the defect detection method is composed of steps S1 to S3, based on/dedicated to the detection of the first type of defect; in other embodiments, after the first type of defect is detected in the manner of steps S1-S3, the second type of defect may be detected in other existing manners or in the manner of S4-S5, which is not specifically limited in the present invention. The execution of S4 and S5 may be performed in synchronization with S2 to S3, or may be performed prior to S2 to S3.
Example 2:
the embodiment provides a detection system based on the defect detection method, which comprises the following steps: the device comprises an image acquisition and preprocessing module, a grid line image generation module and a first type of defect detection method.
The image acquisition and preprocessing module is used for acquiring the photovoltaic panel image to be detected and preprocessing the photovoltaic panel image. And the grid line image generation module is used for carrying out image enhancement processing on the photovoltaic panel image by adopting an MSRCR enhancement method fused with a high-pass filter to obtain the grid line image of the photovoltaic panel. The method comprises the steps of firstly enhancing the photovoltaic panel image by using an MSRCR algorithm, then converting the enhanced image from a space domain to a frequency domain to obtain a spectrogram, inputting the spectrogram into the high-pass filter, and then converting the filtered spectrogram to the space domain to obtain a grid image. The first type defect detection module is used for detecting the first type defects of the grid line image by adopting a defect detection method based on region growth to obtain a first type defect detection result; wherein the first type of defect comprises at least a crack and a unfilled corner.
In other possible embodiments, the detection system further comprises: a second type of defect detection module and a fusion module. The second type defect detection module is used for detecting a second type defect in the photovoltaic panel image, wherein the second type defect detection module comprises: the device comprises a grid line removing module, an enhancing module and a threshold dividing module. The grid line removing module is used for removing grid lines from the photovoltaic panel image after the preprocessor; the enhancement module is used for enhancing the image after the grid lines are removed; the threshold segmentation module is used for carrying out threshold segmentation processing on the image after the enhancement processing and then determining the stain position. And the fusion module is used for splicing and fusing the stain position and the first type defect detection result to obtain a detection result containing the first type defect and the second type defect.
The implementation process of each module refers to the content of the above method, and will not be described herein. It should be understood that the above-described division of functional modules is merely a division of logic functions, and other divisions may be implemented in actual manners, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Meanwhile, the integrated units can be realized in a hardware form or a software functional unit form. For example, the hardware that acquires the photovoltaic panel image is a camera/video camera.
Example 3:
the present embodiment provides an electronic terminal including one or more processors and a memory storing one or more computer programs; wherein the processor invokes the computer program to perform: a method for detecting defects of a photovoltaic panel. The specific implementation is as follows:
s1: and acquiring a photovoltaic panel image to be detected and preprocessing.
S2: performing image enhancement processing on the photovoltaic panel image by adopting an MSRCR enhancement method fused with a high-pass filter to obtain a grid line image of the photovoltaic panel; the method comprises the steps of firstly enhancing the photovoltaic panel image by using an MSRCR algorithm, then converting the enhanced image from a space domain to a frequency domain to obtain a spectrogram, inputting the spectrogram into the high-pass filter, and then converting the filtered spectrogram to the space domain to obtain a grid image.
S3: and performing first-type defect detection on the grid line image by adopting a defect detection method based on region growth to obtain a first-type defect detection result.
In some implementations, further performing:
carrying out grid line removal treatment on the photovoltaic panel image after the preprocessor; and then carrying out enhancement processing on the image with the grid lines removed. Threshold segmentation processing is carried out on the image after the enhancement processing, so that the stain position is determined; and splicing and fusing the stain position and the first type defect detection result to obtain a detection result containing the first type defect and the second type defect.
For specific implementation, refer to the relevant statements of example 1.
The memory may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk memory.
If the memory and the processor are implemented independently, the memory, the processor, and the communication interface may be interconnected by a bus and communicate with each other. The bus may be an industry standard architecture bus, an external device interconnect bus, or an extended industry standard architecture bus, among others. The buses may be classified as address buses, data buses, control buses, etc.
Alternatively, in a specific implementation, if the memory and the processor are integrated on a chip, the memory and the processor may communicate with each other through an internal interface.
It should be appreciated that in embodiments of the present invention, the processor may be a central processing unit (Central Processing Unit, CPU), which may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSPs), application specific integrated circuits (Application Specific Integrated Circuit, ASICs), off-the-shelf programmable gate arrays (Field-Programmable Gate Array, FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The memory may include read only memory and random access memory and provide instructions and data to the processor. A portion of the memory may also include non-volatile random access memory. For example, the memory may also store information of the device type.
Example 4:
the present embodiment provides a readable storage medium storing a computer program that is called by a processor to implement: a method for detecting defects of a photovoltaic panel. The specific implementation is as follows:
s1: and acquiring a photovoltaic panel image to be detected and preprocessing.
S2: performing image enhancement processing on the photovoltaic panel image by adopting an MSRCR enhancement method fused with a high-pass filter to obtain a grid line image of the photovoltaic panel; the method comprises the steps of firstly enhancing the photovoltaic panel image by using an MSRCR algorithm, then converting the enhanced image from a space domain to a frequency domain to obtain a spectrogram, inputting the spectrogram into the high-pass filter, and then converting the filtered spectrogram to the space domain to obtain a grid image.
S3: and performing first-type defect detection on the grid line image by adopting a defect detection method based on region growth to obtain a first-type defect detection result.
In some implementations, further performing:
carrying out grid line removal treatment on the photovoltaic panel image after the preprocessor; and then carrying out enhancement processing on the image with the grid lines removed. Threshold segmentation processing is carried out on the image after the enhancement processing, so that the stain position is determined; and splicing and fusing the stain position and the first type defect detection result to obtain a detection result containing the first type defect and the second type defect.
For specific implementation, refer to the relevant statements of example 1.
The readable storage medium is a computer readable storage medium, which may be an internal storage unit of the controller according to any one of the foregoing embodiments, for example, a hard disk or a memory of the controller. For example, the terrain feature model constructed in the invention exists in a hard disk, and then the computer program for executing the fusion step is stored in a memory, so that the fusion process is realized by depending on the memory. The readable storage medium may also be an external storage device of the controller, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are provided on the controller. Further, the readable storage medium may also include both an internal storage unit and an external storage device of the controller. The readable storage medium is used to store the computer program and other programs and data required by the controller. The readable storage medium may also be used to temporarily store data that has been output or is to be output.
Based on such understanding, the technical solution of the present invention is essentially or a part contributing to the prior art, or all or part of the technical solution may be embodied in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned readable storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It should be emphasized that the examples described herein are illustrative rather than limiting, and that this invention is not limited to the examples described in the specific embodiments, but is capable of other embodiments in accordance with the teachings of the present invention, as long as they do not depart from the spirit and scope of the invention, whether modified or substituted, and still fall within the scope of the invention.
Claims (10)
1. A defect detection method of a photovoltaic panel is characterized in that: the method comprises the following steps:
s1: acquiring a photovoltaic panel image to be detected and preprocessing;
s2: performing image enhancement processing on the photovoltaic panel image by adopting an MSRCR enhancement method fused with a high-pass filter to obtain a grid line image of the photovoltaic panel;
the method comprises the steps of firstly enhancing the photovoltaic panel image by using an MSRCR algorithm, then converting the enhanced image from a space domain to a frequency domain to obtain a spectrogram, inputting the spectrogram into the high-pass filter, and then converting the filtered spectrogram to the space domain to obtain a grid image;
s3: performing first-type defect detection on the grid line image by adopting a defect detection method based on region growth to obtain a first-type defect detection result;
wherein the first type of defect comprises at least a crack and a unfilled corner.
2. The defect detection method according to claim 1, wherein: the high pass filter is represented as follows:
in the above formula, H (u, v) represents a value of the high-pass filter at a point (u, v) on the spectrogram, 1 represents pass, and 0 represents no pass; d (u, v) represents the distance of a point (u, v) in the spectrogram from the center of the image, M and N represent the length and width of the spectrogram, wherein,
is the coordinates of the image center of the spectrogram; d represents the radius of a circle drawn by the center of the image of the spectrogram, and is a constant; w represents the bandwidth of the high-pass filter and +.>
3. The defect detection method according to claim 2, wherein: the method further comprises the following steps of transforming gray values of the grid line image:
B(x,y)＝f(x,y)I(x,y)
wherein B (x, y) represents the gray value of the pixel point (x, y) after transformation, f (x, y) represents the transformation coefficient, and I (x, y) represents the pixel value of the pixel point (x, y) coordinate in the grid line image; taking 5 pixels at any point in the grid line image and in the up-down direction, calculating the difference between the average value of gray values of the 4 pixels except the center point and the average value of gray values of the 5 selected pixels, and recording the difference as S; q, Q are specific values of the transformation coefficients under the corresponding conditions, and Q is more than 1 and less than 1.
4. The defect detection method according to claim 1, wherein: the first type defect detection is carried out on the grid line image by adopting a defect detection method based on region growth, so that a first type defect detection result is obtained, wherein the process is as follows:
firstly, generating a penetrating connecting line perpendicular to a grid line in the middle of an image of the grid line image;
starting a seed point at the intersection point of the penetrating connecting line and the leftmost or right grid line, and growing in the upper, lower and right/left directions;
and if the top or bottom positions of the grid lines and the top/bottom positions of the rest grid lines are different in the height direction by more than a set threshold value in the growth process, the corresponding top/bottom positions are included in the first type of defect positions.
5. The defect detection method according to claim 1, wherein: further comprises:
carrying out grid line removal treatment on the photovoltaic panel image after the preprocessor;
then, carrying out enhancement treatment on the image from which the grid lines are removed;
threshold segmentation processing is carried out on the image after the enhancement processing, so that the stain position is determined;
splicing and fusing the stain position and the first type defect detection result to obtain a detection result containing the first type defect and the second type defect;
the second type of defects are black dots with different sizes which are randomly distributed on each position of the photovoltaic panel.
6. The defect detection method according to claim 1, wherein: the preprocessing of the photovoltaic panel image in step 1 comprises: performing inclination correction on the photovoltaic panel image by adopting a Hough linear transformation algorithm; and then, performing ROI region clipping on the corrected photovoltaic panel image.
7. A detection system based on the defect detection method according to any one of claims 1 to 6, characterized in that: comprising the following steps:
the image acquisition and preprocessing module is used for acquiring the image of the photovoltaic panel to be detected and preprocessing the image;
the grid line image generation module is used for carrying out image enhancement processing on the photovoltaic panel image by adopting an MSRCR enhancement method fused with a high-pass filter to obtain a grid line image of the photovoltaic panel;
the method comprises the steps of firstly enhancing the photovoltaic panel image by using an MSRCR algorithm, then converting the enhanced image from a space domain to a frequency domain to obtain a spectrogram, inputting the spectrogram into the high-pass filter, and then converting the filtered spectrogram to the space domain to obtain a grid image;
the first type defect detection module is used for detecting the first type defects of the grid line image by adopting a defect detection method based on region growth to obtain a first type defect detection result; wherein the first type of defect comprises at least a crack and a unfilled corner.
8. The detection system of claim 7, wherein: the system further comprises a second type defect detection module and a fusion module, wherein the second type defect detection module is used for detecting a second type defect in the photovoltaic panel image, and the second type defect detection module comprises: the grid line removing module, the enhancing module and the threshold dividing module;
the grid line removing module is used for removing grid lines from the photovoltaic panel image after the preprocessor;
the enhancement module is used for enhancing the image with the grid lines removed;
the threshold segmentation module is used for carrying out threshold segmentation processing on the image after the enhancement processing so as to determine the stain position;
the fusion module is used for splicing and fusing the stain position and the first type defect detection result to obtain a detection result containing the first type defect and the second type defect;
the second type of defects are black dots with different sizes which are randomly distributed on each position of the photovoltaic panel.
9. An electronic terminal, characterized in that: comprising the following steps:
one or more processors;
a memory storing one or more computer programs;
the processor invokes the computer program to perform:
the method for detecting a defect according to any one of claims 1 to 6.
10. A readable storage medium, characterized by: a computer program is stored, the computer program being invoked by a processor to perform:
the method for detecting a defect according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310330348.6A CN116363097A (en) | 2023-03-30 | 2023-03-30 | Defect detection method and system for photovoltaic panel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310330348.6A CN116363097A (en) | 2023-03-30 | 2023-03-30 | Defect detection method and system for photovoltaic panel |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116363097A true CN116363097A (en) | 2023-06-30 |
Family
ID=86907060
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310330348.6A Pending CN116363097A (en) | 2023-03-30 | 2023-03-30 | Defect detection method and system for photovoltaic panel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116363097A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116777917A (en) * | 2023-08-24 | 2023-09-19 | 山东东方智光网络通信有限公司 | Defect detection method and system for optical cable production |
-
2023
- 2023-03-30 CN CN202310330348.6A patent/CN116363097A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116777917A (en) * | 2023-08-24 | 2023-09-19 | 山东东方智光网络通信有限公司 | Defect detection method and system for optical cable production |
| CN116777917B (en) * | 2023-08-24 | 2023-11-21 | 山东东方智光网络通信有限公司 | Defect detection method and system for optical cable production |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107230202B (en) | Automatic identification method and system for road surface disease image | |
| CN114418957B (en) | Global and local binary pattern image crack segmentation method based on robot vision | |
| CN106780486B (en) | Steel plate surface defect image extraction method | |
| CN111179243A (en) | Small-size chip crack detection method and system based on computer vision | |
| CN111260616A (en) | Insulator crack detection method based on Canny operator two-dimensional threshold segmentation optimization | |
| WO2021109697A1 (en) | Character segmentation method and apparatus, and computer-readable storage medium | |
| CN109241973B (en) | Full-automatic soft segmentation method for characters under texture background | |
| US11151402B2 (en) | Method of character recognition in written document | |
| CN115294099B (en) | Method and system for detecting hairline defect in steel plate rolling process | |
| CN108830857B (en) | Self-adaptive Chinese character copy label image binarization segmentation method | |
| CN115908415B (en) | Edge-based defect detection method, device, equipment and storage medium | |
| CN113706490B (en) | Wafer defect detection method | |
| CN114529459A (en) | Method, system and medium for enhancing image edge | |
| CN114332026A (en) | Visual detection method and device for scratch defects on surface of nameplate | |
| CN114926407A (en) | Steel surface defect detection system based on deep learning | |
| CN116363097A (en) | Defect detection method and system for photovoltaic panel | |
| CN109716355B (en) | Particle boundary identification | |
| CN110246139B (en) | Method for rapidly extracting plankton in-situ image ROI (region of interest) based on double thresholds | |
| CN117094975A (en) | Method and device for detecting surface defects of steel and electronic equipment | |
| CN115439523A (en) | Method and equipment for detecting pin size of semiconductor device and storage medium | |
| CN111667509B (en) | Automatic tracking method and system for moving target under condition that target and background colors are similar | |
| CN108205678A (en) | A kind of nameplate Text region processing method containing speck interference | |
| CN112288726A (en) | Method for detecting foreign matters on belt surface of underground belt conveyor | |
| CN116433978A (en) | Automatic generation and automatic labeling method and device for high-quality flaw image | |
| US20060126136A1 (en) | Image inspecting apparatus, image inspecting method, control program and computer-readable storage medium |
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 |