CN115954291A - Crack monitoring system of TOPCon structure photovoltaic cell panel - Google Patents
Crack monitoring system of TOPCon structure photovoltaic cell panel Download PDFInfo
- Publication number
- CN115954291A CN115954291A CN202310199676.7A CN202310199676A CN115954291A CN 115954291 A CN115954291 A CN 115954291A CN 202310199676 A CN202310199676 A CN 202310199676A CN 115954291 A CN115954291 A CN 115954291A
- Authority
- CN
- China
- Prior art keywords
- angle
- straight
- line segment
- initial
- amplitude
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 71
- 230000009466 transformation Effects 0.000 claims description 15
- 238000012216 screening Methods 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 10
- 230000000873 masking effect Effects 0.000 claims description 4
- 230000001105 regulatory effect Effects 0.000 claims description 2
- 230000007547 defect Effects 0.000 description 37
- 238000000034 method Methods 0.000 description 15
- 238000001228 spectrum Methods 0.000 description 10
- 238000001514 detection method Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 9
- 230000000694 effects Effects 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 238000013519 translation Methods 0.000 description 6
- 230000014616 translation Effects 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 230000000737 periodic effect Effects 0.000 description 4
- 230000011218 segmentation Effects 0.000 description 4
- 230000003247 decreasing effect Effects 0.000 description 3
- 238000005286 illumination Methods 0.000 description 3
- 230000000452 restraining effect Effects 0.000 description 3
- 239000000126 substance Substances 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000002829 reductive effect Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
Images
Classifications
-
- 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
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to the technical field of photovoltaic system monitoring, in particular to a crack monitoring system of a TOPCon structure photovoltaic cell panel. The crack monitoring system of the photovoltaic cell panel can accurately identify the surface cracks of the photovoltaic cell panel in the photovoltaic system, and realizes reliable monitoring of crack faults of the photovoltaic cell panel.
Description
Technical Field
The invention relates to the technical field of photovoltaic system monitoring, in particular to a crack monitoring system of a TOPCon structure photovoltaic cell panel.
Background
The photovoltaic cell panel is an important device in a photovoltaic system, and the photovoltaic cell panel with a Tunnel Oxide Passivated Contact (TOPCon) structure is different from a traditional common photovoltaic cell panel.
In the production process of the photovoltaic cell panel with the TOPCon structure, the surface of the produced photovoltaic cell panel has crack defects due to improper operation of related technical personnel or loss of production and processing equipment. Since such crack defects are small and sometimes not easily noticeable, the problem of missing detection and error detection is likely to occur during manual visual inspection. However, these cracks that are not easily perceived on the photovoltaic cell panel can cause the inherent structure of the photovoltaic cell panel surface design to be abnormal, thereby affecting the flow of internal electrons under the illumination condition, and finally affecting the electric energy conversion efficiency of the photovoltaic cell panel, so monitoring the cracks on the surface of the photovoltaic cell panel with the TOPCon structure is an important link.
In the existing monitoring process of cracks on the surface of a photovoltaic cell panel, binaryzation processing is carried out by carrying out a traditional threshold segmentation method on a surface image of a single photovoltaic cell panel, so that crack defects are identified, and the identification of the cracks on the surface of the photovoltaic cell panel is realized. However, the final crack identification effect is directly affected by the selection of the threshold in the conventional threshold segmentation method, and the setting of the threshold excessively depends on the experience of the related technical staff, so that the crack monitoring result is not accurate enough.
Disclosure of Invention
In order to solve the problem that the surface crack monitoring result of the existing photovoltaic cell panel is inaccurate, the invention aims to provide a crack monitoring system of a photovoltaic cell panel with a TOPCon structure, and the adopted technical scheme is as follows:
the invention provides a crack monitoring system of a TOPCon structure photovoltaic cell panel, which comprises:
a monitoring acquisition module for: collecting a surface image of a battery panel to be detected, and sending the surface image to a monitoring and identifying module;
a monitor identification module to: receiving the surface image sent by the monitoring acquisition module, further acquiring an initial gray-scale image corresponding to the surface image, and performing time-frequency transformation on the initial gray-scale image to obtain an initial frequency spectrogram; determining a central point and each initial angle in the initial spectrogram, determining a straight-line segment passing through the central point in the initial spectrogram and corresponding to each initial angle, and determining an angle amplitude corresponding to each initial angle according to pixel points on the straight-line segment corresponding to each initial angle; screening all the initial angles according to the angle amplitude corresponding to each initial angle and the size of the initial spectrogram to obtain each first angle; determining each constraint straight-line segment corresponding to each straight-line segment of the first angle, and determining the angle amplitude of each constraint straight-line segment according to pixel points on each constraint straight-line segment; screening each first angle according to the angle amplitude of each constraint straight-line segment corresponding to each straight-line segment of each first angle to obtain each second angle; masking the pixel points on the straight line segments corresponding to the second angles in the initial spectrogram to obtain a mask spectrogram; and performing time-frequency inverse transformation on the mask frequency spectrogram to obtain a mask gray-scale image, and determining a crack area of the panel according to the mask gray-scale image and the initial gray-scale image so as to realize the monitoring of the panel.
Further, the monitoring identification module is configured to:
determining a gray threshold according to the gray value of each pixel point on the straight-line segment corresponding to each initial angle, and determining the pixel point on the straight-line segment corresponding to each initial angle, wherein the gray value of the pixel point is larger than the gray threshold, as a target pixel point;
and calculating the accumulated sum of all distances corresponding to each initial angle according to the distance from each target pixel point on the straight line segment corresponding to each initial angle to the central point, and determining the accumulated sum as the angle amplitude corresponding to the initial angle.
Further, the monitoring identification module is configured to:
and determining the median of all gray values as a gray threshold according to the gray value of each pixel point on the straight line segment corresponding to each initial angle.
Further, the monitoring identification module is configured to:
calculating an angle amplitude mean value according to the angle amplitude corresponding to each initial angle, and calculating a difference value between the angle amplitude corresponding to each initial angle and the angle amplitude mean value;
calculating a threshold intermediate value according to the difference value of the angle amplitude corresponding to each initial angle and the angle amplitude mean value and the size of the initial spectrogram, and calculating an angle amplitude threshold according to the angle amplitude mean value and the threshold intermediate value;
the angle amplitude corresponding to each initial angle is compared with an angle amplitude threshold value, and the initial angle corresponding to the angle amplitude larger than the angle amplitude threshold value is determined as a first angle.
Further, the calculation formulas corresponding to the intermediate value of the calculation threshold and the angle amplitude threshold are respectively as follows:
wherein the content of the first and second substances,is an angle magnitude threshold value>Is in the middle of the thresholdValue,. Or>And &>Length and width of the initial spectrogram respectively>To correspond toiThe angular amplitude of the initial angle is such that,mis the total number of initial angles>Is the average value of the amplitude values of the angles,Kis a regulatory factor.
Further, the monitoring identification module is configured to:
taking the straight line segment corresponding to each first angle as a target straight line segment, symmetrically arranging at least one pair of straight line segments on two sides of the target straight line segment, wherein each pair of arranged straight line segments is parallel to the corresponding target straight line segment, and determining all the arranged straight line segments as each restraint straight line segment corresponding to the target straight line segment.
Further, the monitoring identification module is configured to:
for the target straight-line segment corresponding to each first angle, determining the distance from each constraint straight-line segment of the target straight-line segment corresponding to each first angle to the target straight-line segment;
calculating the absolute value of the difference of the angle amplitude of each pair of constrained straight-line segments of the target straight-line segment corresponding to each first angle, and determining the accumulated sum of all the absolute values of the differences as the constrained difference value of the target straight-line segment corresponding to each first angle;
judging whether each constraint straight-line segment of the target straight-line segment corresponding to each first angle meets constraint conditions, and if so, determining the corresponding first angle as a second angle; the constraint conditions are as follows: for each constraint straight-line segment on each side of the target straight-line segment corresponding to each first angle, the larger the distance from the constraint straight-line segment to the target straight-line segment corresponding to the first angle is, the smaller the angle amplitude of the constraint straight-line segment is, and the angle amplitudes of the constraint straight-line segments are all smaller than the angle amplitude of the target straight-line segment; and the constraint difference value of the target straight-line segment corresponding to each first angle is smaller than a set constraint difference threshold value.
Further, the monitoring identification module is configured to:
the constraint straight line segments corresponding to each side of the target straight line segment of each first angle are arranged at equal intervals.
Further, the monitoring identification module is configured to:
and setting the gray value of the pixel point on the straight line segment corresponding to each second angle in the initial frequency spectrogram as 0, so as to obtain the mask frequency spectrogram.
Further, the monitoring identification module is configured to:
and performing difference operation on the initial gray level image and the mask gray level image to obtain a difference gray level image, and performing crack area identification on the difference gray level image to obtain a crack area of the battery panel.
The invention has the beneficial effects that: according to the invention, the monitoring acquisition module is used for acquiring the surface image of the battery panel to be detected and sending the surface image to the monitoring identification module, and the monitoring identification module is used for accurately identifying the crack area on the surface of the photovoltaic battery panel by carrying out crack identification on the surface image, so that the reliable monitoring of the crack fault of the photovoltaic battery panel is realized. The specific process of identifying the crack area by the monitoring and identifying module is as follows: and acquiring an initial gray scale image and an initial frequency spectrogram according to the surface image, wherein the initial gray scale image and the initial frequency spectrogram both contain crack region information. And then, determining each initial angle in the initial spectrogram, wherein the initial angles do not comprise the angle corresponding to the grid line of the photovoltaic cell panel, so that the influence of the grid line on the subsequent crack identification of the photovoltaic cell panel is effectively avoided. The pixel points on the straight-line segments, passing through the central point, of the initial spectrogram and corresponding to each initial angle are analyzed, the angle amplitude corresponding to each initial angle is determined, and the crack defects on the photovoltaic cell panel correspond to the bright lines in the initial spectrogram, the angle amplitude corresponding to the bright lines is usually large, so that each first angle possibly corresponding to the crack can be accurately screened according to the angle amplitude corresponding to each initial angle, and therefore the accuracy and reliability of follow-up crack identification on the TOPCon structure photovoltaic cell panel are effectively improved. And then determining each constraint line segment straight-line segment corresponding to the line segment straight-line segment of each first angle according to the characteristics of the crack defect, namely that the energy of the crack defect in the central part of the corresponding spectrogram is highest, the gray value of the corresponding pixel point is larger, and the gray value is gradually decreased towards the periphery, and further accurately screening each second angle corresponding to the crack in each first angle according to the angle amplitude of the constraint line segment straight-line segments. The mask gray-scale image is obtained through time-frequency inverse transformation by performing mask processing on straight-line segments corresponding to the second angles in the initial frequency spectrogram, and original crack information is accurately removed from the mask gray-scale image, so that the surface crack area of the photovoltaic cell panel can be accurately determined by comparing the mask gray-scale image with the initial gray-scale image, and the reliable monitoring of the photovoltaic cell panel is realized. The crack monitoring system automatically and accurately screens the angle corresponding to the crack by converting the surface image of the photovoltaic cell panel into the frequency domain and based on the characteristics of the crack defect in the frequency spectrum, thereby finally and accurately identifying the crack area on the surface of the cell panel, effectively avoiding the unreliability of manually setting a threshold value for area segmentation in the prior art, and effectively improving the accuracy and reliability of crack monitoring on the TOPCon structure photovoltaic cell panel while ensuring the overall intelligent effect of the scheme.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a schematic structural diagram of a crack monitoring system of a TOPCon structured photovoltaic panel according to an embodiment of the present invention;
FIG. 2 is a flowchart of a process for identifying cracks by the monitoring and identifying module according to the embodiment of the present invention;
FIG. 3 shows each initial angle corresponding to an over-center point in the initial spectrogramA schematic diagram of a straight line segment of (a);
Description of the preferred embodiment
To further explain the technical means and effects of the present invention adopted to achieve the predetermined objects, the following detailed description of the embodiments, structures, features and effects of the technical solutions according to the present invention will be given with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" refers to not necessarily the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
In order to monitor crack faults on the surface of a produced TOPCon structure photovoltaic cell panel, the embodiment provides a crack monitoring system for a TOPCon structure photovoltaic cell panel. As shown in fig. 1, the system comprises a monitoring and collecting module and a monitoring and recognizing module, wherein the monitoring and collecting module is used for collecting the surface image of the battery panel to be detected and sending the surface image to the monitoring and recognizing module, and the monitoring and recognizing module is used for receiving the surface image sent by the monitoring and collecting module and carrying out crack recognition on the surface image, so that the crack area on the surface of the battery panel is accurately recognized, and the reliable monitoring of the battery panel is realized. The specific implementation process of the monitoring and identifying module for crack identification is shown in fig. 2, and specifically comprises the following steps:
step S1: and acquiring an initial gray-scale image corresponding to the surface image, and performing time-frequency transformation on the initial gray-scale image to obtain an initial spectrogram.
Because the surface image quality of the TOPCon structure photovoltaic cell panel collected by shooting has great influence on the detection accuracy of the subsequent photovoltaic cell panel surface crack defects, in order to avoid great influence on the final imaging quality caused by the internal structure unit of the shooting and collecting equipment in the shooting and collecting process, the CCD camera with high imaging quality and clear image detail retention is selected as the equipment for shooting and collecting, and therefore the surface image of the TOPCon structure photovoltaic cell panel is obtained by shooting and collecting. Meanwhile, in order to avoid the influence of the finally shot and collected photovoltaic cell panel image on the crack detection precision due to uneven illumination, the LED annular structure lamp source is used for carrying out illumination processing on the surface of the photovoltaic cell panel in the embodiment, so that the surface of the photovoltaic cell panel is uniformly illuminated.
For the surface image of the photovoltaic cell panel in the RGB color space obtained by shooting and collecting, in order to reduce the calculation cost and improve the identification precision of the crack defects on the surface of the photovoltaic cell panel in the subsequent crack identification process, the surface image of the photovoltaic cell panel in the RGB color space is converted into a gray image by using a weighted average algorithm, so that the initial gray image of the photovoltaic cell panel is obtained.
Meanwhile, random noise may occur in the obtained initial gray level image of the photovoltaic cell panel under the influence of a shooting and collecting working environment, and the noise can cause great influence on the detection of surface cracks of the subsequent photovoltaic cell panel.
Further observation of the surface structure of the photovoltaic cell panel with the TOPCon structure shows that inherent grid lines can appear on the surface of the photovoltaic cell panel, and the grid lines are lead wires collected from the surface electrodes of the cell to the main grid lines and used for collecting and transmitting current obtained after conversion through the photovoltaic cell panel. The inherent grid line and the surface crack have higher similar overlapping degree, so that in a traditional crack monitoring system, a grid line area on the surface of the photovoltaic cell panel needs to be separated firstly, and then the processed surface image of the photovoltaic cell panel is further calculated and analyzed, so that the crack defect on the surface of the photovoltaic cell panel is detected.
Further observation of grid lines on the surface of the photovoltaic cell panel with the TOPCon structure shows that the grid line structure inherent to the surface of the photovoltaic cell panel appears periodically, the gray image on the surface of the photovoltaic cell panel is converted into a spectrum image of the photovoltaic cell panel through Fourier transform, the grid lines which show a certain periodic rule in a space domain can also show a corresponding periodic transformation rule on the spectrum image of the corresponding photovoltaic cell panel, but cracks on the surface of the photovoltaic cell panel have randomness, the periodic transformation rule cannot appear on the spectrum image of the corresponding photovoltaic cell panel, and the crack straight lines cannot be always parallel to the grid lines, so that crack detection can be carried out on the photovoltaic cell panel with the TOPCon structure according to the condition.
Based on the analysis, the time-frequency transformation is carried out on the initial gray-scale image of the photovoltaic cell panel obtained after the collection processing through two-dimensional Fourier transformation, and therefore an initial frequency spectrogram is obtained. Since the specific transformation process of the two-dimensional fourier transform is a well-known technique, it will not be described in detail here.
Step S2: determining a central point and each initial angle in the initial spectrogram, determining a straight-line segment passing through the central point in the initial spectrogram and corresponding to each initial angle, and determining an angle amplitude corresponding to each initial angle according to pixel points on the straight-line segment corresponding to each initial angle.
For the initial spectrogram, the size of the initial spectrogram is assumed to be,/>And &>Respectively the length and width of the initial spectrogram, the lower left corner of the initial spectrogram is taken as the origin, and the long side of the initial spectrogram is taken as the straight lineThe line is an x axis, and a straight line where the short side of the initial spectrogram is located is a y axis, so that a two-dimensional coordinate system is constructed. At this time, there is a center point in the initial spectrogram, />. Hypothesis->Is the coordinate of a pixel point at a certain position on the initial frequency spectrum chart, then>For the pixel point at this position and a central point +>The angle between the straight line of the connecting line and the positive direction of the x axis,, />has a value range of->。
Considering that grid lines exist on the surface of the photovoltaic cell panel, the grid lines are usually horizontal or vertical grid lines. According to the correlation property of two-dimensional Fourier transform, the linear direction of the grid line of the photovoltaic cell panel under the spatial domain and the linear direction of the grid line of the photovoltaic cell panel under the frequency domain are in reverse mutual orthogonality, and the crack under the spatial domain is transformed to pass through the center of the centralized spectrum image. Based on the characteristic, the direction of the grid line in the frequency spectrum image is determined according to the direction of the grid line in the initial gray scale image, and the angle corresponding to the direction of the grid line in the frequency spectrum image does not participate in subsequent calculation. In this embodiment, since the gate lines in the initial gray scale map include horizontal gate lines and vertical gate lines, that is, the direction of the gate lines is the horizontal direction or the vertical direction, the horizontal gate lines are represented as horizontal lines after two-dimensional discrete fourier transformThe vertical grid line appears to be ≥ after a two-dimensional discrete fourier transform>Thus, therefore, it isAnd &>The two angles do not participate in the subsequent calculation, and the value range of the initial angle is ^ or ^> . In order to reduce the amount of calculation while ensuring the detection accuracy, in the present embodiment, sampling is performed at equal intervals within the range of the initial angle, for example, every 1 ° from-89 °, so that each initial angle can be determined, and as another embodiment, every 0.5 ° from-89.5 °, so that each initial angle can be determined.
The crack defect of the photovoltaic cell panel in the airspace image has the characteristic of periodic transformation different from horizontal and vertical grid lines on the photovoltaic cell panel, and the crack defect appears as a bright line passing through the center point of a spectrogram on a corresponding Fourier transformed spectrogram image. In order to determine bright line lines corresponding to the crack defects of the photovoltaic panel, after a central point and each initial angle in an initial spectrogram are determined, a straight line segment passing through the central point and corresponding to each initial angle in the initial spectrogram is determined, the central point of the straight line segment is the central point of the initial spectrogram, and a straight line equation corresponding to the straight line segmentIs composed of. FIG. 3 shows the initial spectrogramOver-center point->Corresponding to each initial angle>Is straight line segment->。/>
Because the bright lines on the spectrogram represent straight lines on the gray level image of the photovoltaic cell panel under the corresponding airspace, the straight lines comprise crack defects of the photovoltaic cell panel, and the higher the frequency on the spectrogram is, the more likely the corresponding crack information is represented on the corresponding original airspace image, therefore, the angle amplitude corresponding to each initial angle is determined by analyzing pixel points on the straight line segment corresponding to each initial angle, so that the angle amplitude corresponding to the bright lines can be screened out subsequently, and the implementation process comprises the following steps:
step S21: determining a gray threshold according to the gray value of each pixel point on the straight-line segment corresponding to each initial angle, and determining the pixel point of which the gray value on the straight-line segment corresponding to each initial angle is larger than the gray threshold as a target pixel point;
step S22: and calculating the accumulated sum of all distances corresponding to each initial angle according to the distance from each target pixel point on the straight line segment corresponding to each initial angle to the central point, and determining the accumulated sum as the angle amplitude corresponding to the initial angle.
In particular, for each initial angleBased on the correspondence with the initial angle>The gray values of all pixel points on the straight line segment are sorted from small to large, and the median is taken as the corresponding gray threshold value ^ greater than or equal to>. Then get the correspondingThe initial angle->Greater than a gray threshold value on a straight line segment>And taking the pixel points as target pixel points. Calculating the accumulated sum of the distance values from all the target pixel points to the central point in the initial frequency spectrogram, and taking the accumulated sum as the angle amplitude value corresponding to the initial angle:
wherein the content of the first and second substances,in correspondence with an initial angle>Is greater than or equal to>) Is corresponding to the initial angle->On the straight line segment of (4) pixel point->I.e. corresponds to the initial angle->On the straight line segment of (4) pixel point->To a center point in an initial spectrogram>The distance of (c).
Because the cracks under the airspace form bright lines in the corresponding frequency domain part after Fourier transformation, the corresponding gray threshold value is setCan be combined with a corresponding initial angle of>Screening the pixel points on the straight line segment to avoid corresponding initial angle->The invalid points on the straight line segments interfere with subsequent further calculation and analysis, and meanwhile, the calculation amount is reduced so as to further improve the real-time effect of the scheme.
And step S3: and screening all the initial angles according to the angle amplitude corresponding to each initial angle and the size of the initial spectrogram to obtain each first angle.
At some initial angleLower corresponding angle amplitude->If the value is greater, then the initial angle is assigned>The lower straight line segment is represented as a bright line in the spectrogram, and is represented as a line on the corresponding airspace initial gray level image, and the line is possibly a crack defect. Based on this, the angle amplitude corresponding to each initial angle is analyzed, so as to screen out the initial angle corresponding to the bright line, and the implementation steps include:
step S31: calculating an angle amplitude mean value according to the angle amplitude corresponding to each initial angle, and calculating the difference value between the angle amplitude corresponding to each initial angle and the angle amplitude mean value;
step S32: calculating a threshold intermediate value according to the difference value of the angle amplitude corresponding to each initial angle and the angle amplitude mean value and the size of the initial spectrogram, and calculating an angle amplitude threshold according to the angle amplitude mean value and the threshold intermediate value;
step S33: and comparing the angle amplitude corresponding to each initial angle with an angle amplitude threshold value, and determining the initial angle corresponding to the angle amplitude greater than the angle amplitude threshold value as the first angle.
In particular, in order to better distinguish the corresponding angle amplitude values under different initial anglesThe value of (2) is required to set a corresponding angle amplitude threshold value T so as to screen out all bright line angles from the initial angle. In order to avoid that the fixed empirical threshold value excessively depends on the operation experience of the related technical personnel, the present embodiment adaptively determines the angle amplitude threshold value T according to the initial spectrogram of the corresponding photovoltaic cell panel, where the determination formula is as follows:
wherein the content of the first and second substances,is an angle magnitude threshold value>Is a threshold value median value>And &>Respectively the length and width of the initial spectrogram,to correspond toiThe angular amplitude of the initial angle is such that,mis the total number of initial angles>Is the average value of the amplitude values of the angles,Kis used for adjusting factors and controlling the final screening and dividing effects,the adjustment factorKThe value of (a) can be determined according to relevant experimental tests, and K =2 is set in this embodiment.
In the embodiment, by evaluating the angle amplitude corresponding to each initial angle in the initial spectrogram, when the distribution of all the angle amplitudes is relatively consistent, it is indicated that the frequency spectrum image has less high-frequency characteristic information and the corresponding defect in the airspace is not serious, the median value of the threshold is relatively small, and the angle amplitude threshold T is close to the average value of the angle amplitudes; when the difference of all the angle amplitudes is larger, it is indicated that the high-frequency characteristic information on the spectrogram image is more at the moment, the defect in the corresponding spatial domain is more serious, the median value of the threshold value is larger at the moment, and the angle amplitude threshold value T is larger than the angle amplitude average value. According to the method, the angle amplitude threshold can be determined more accurately, subjectivity of manually determining the angle amplitude threshold is avoided, screening accuracy of subsequent bright line angles is effectively improved, and accuracy and reliability of crack detection of the TOPCon structure photovoltaic cell panel are improved.
After adaptively determining the angle amplitude threshold value T through the above calculation formula, the angle amplitude is determinedGreater than the angle magnitude threshold value>Corresponding initial angle->Extracting to obtain each first angle, and constructing to obtain a corresponding bright and dark angle set of the photovoltaic cell panel>。
And step S4: and determining each constraint straight-line segment corresponding to each straight-line segment of the first angle, and determining the angle amplitude of each constraint straight-line segment according to the pixel point on each constraint straight-line segment.
Typically, the photovoltaic panel surface cracks are of pixel-level width, corresponding toStraight line equation of corresponding bright lines of central pixel points of cracksThe straight line within a certain offset distance around should be further analyzed to determine whether it is also part of the crack, thereby further screening the angle corresponding to the crack.
In the embodiment, the further screening of the corresponding angles of the cracks is realized by determining each constraint straight-line segment corresponding to each straight-line segment of the first angle, determining the angle amplitude of each constraint straight-line segment and further according to the change condition of the angle amplitude. Wherein the step of determining each constraint straight-line segment corresponding to each straight-line segment of the first angle comprises:
and for the straight line segments corresponding to each first angle, symmetrically arranging at least one pair of straight line segments on two sides of each straight line segment, wherein each pair of the arranged straight line segments is parallel to the corresponding straight line segment, and determining all the arranged straight line segments as each restraining straight line segment of the corresponding straight line segment.
Specifically, for each straight line segment corresponding to each first angle, the straight line segment is sequentially translated to one side by a distance、 、/>……/>,nFor the total number of line segment translations to one side, corresponding to the constrained line segment set for that sideSimultaneously, the straight line segment is sequentially translated to the other side by the distance->、/>、/>……/>So as to obtain the constrained straight line segment set of the other side correspondingly. In this embodiment, as shown in FIG. 4, the difference between any two adjacent translation distances is distance ≧ for either side of the straight line segment corresponding to each first angle>That is, the restraining straight line segments corresponding to each side of the straight line segment of each first angle are arranged at equal intervals, and the distance between any adjacent restraining straight line segments is ≥ h>. In addition, according to the distribution condition of the cracks on the photovoltaic cell panel, the size range of the translation distance needs to be set, when the cracks are narrow, the range of the translation distance should be set to be small, when the cracks are wide, the range of the translation distance should be set to be large, and the range of the translation distance is set to be (0, 15) in the embodiment]。
Step S5: and screening each first angle according to the angle amplitude of each constraint straight-line segment of the straight-line segment corresponding to each first angle to obtain each second angle.
Further observation and analysis of the initial spectrogram of the photovoltaic cell panel can find that the energy of the crack defect in the central part of the corresponding spectrogram is highest, and the gray value of the corresponding pixel point is larger and gradually reduced towards the periphery. Meanwhile, the crack defects should be symmetrically distributed on the positive and negative portions of the corresponding spectrogram, that is, the crack defects should be symmetrically distributed on both sides of the central portion of the corresponding spectrogram. Based on the distribution characteristics of the crack defects in the corresponding spectrogram, each constraint straight-line segment corresponding to each straight-line segment of the first angle needs to be judged so as to determine which straight-line segments corresponding to the first angles are the straight-line segment parts of the crack defects.
Based on the above analysis, after obtaining each constraint straight-line segment corresponding to the straight-line segment of each first angle through the above step S4, the angle amplitude of each constraint straight-line segment corresponding to the straight-line segment of each first angle is determined in a manner of determining the angle amplitude of the straight-line segment corresponding to each initial angle in the step S2. Screening each first angle based on the angle amplitude of each constraint straight-line segment of the straight-line segment corresponding to each first angle to obtain each second angle, wherein the implementation steps comprise:
step S51: for the target straight-line segment corresponding to each first angle, determining the distance from each constraint straight-line segment of the target straight-line segment corresponding to each first angle to the target straight-line segment;
step S52: calculating the absolute value of the difference of the angle amplitude of each pair of constrained straight-line segments of the target straight-line segment corresponding to each first angle, and determining the accumulated sum of all the absolute values of the differences as the constrained difference value of the target straight-line segment corresponding to each first angle;
step S53: judging whether each constraint straight-line segment of the target straight-line segment corresponding to each first angle meets constraint conditions, and if so, determining the corresponding first angle as a second angle; the constraint conditions are as follows: for each constraint straight-line segment on each side of the target straight-line segment corresponding to each first angle, the larger the distance from the constraint straight-line segment to the target straight-line segment corresponding to the first angle is, the smaller the angle amplitude of the constraint straight-line segment is, and the angle amplitudes of the constraint straight-line segments are all smaller than the angle amplitude of the target straight-line segment; and the constraint difference value of the target straight-line segment corresponding to each first angle is smaller than a set constraint difference threshold value.
In particular, for each first angleIs straight line segment->Based onThe angle amplitude of each restriction straight line segment on each side is collected and/or judged by the angle amplitude value of each restriction straight line segment on each side>The corresponding angle amplitude is recorded as>Gathering the constrained straight line segments on the other side of the linear arrayThe corresponding angle amplitude is recorded asAnd judging whether the following constraint conditions are met:
constraint 1: to correspond to each first angleIn a linear section of>At the center, the angular amplitude of each corresponding straight constraint segment on each side is decreased, i.e. in a manner that +>And->。
Constraint 2: corresponding to each first angleIn a linear section of>Constraint difference value ofLess than a set constraint variance threshold.
For each first angleIs straight line segment->If the angle amplitude of each corresponding constraint straight-line segment meets the constraint condition 1, the corresponding first angle +is determined>In a linear section of>Each corresponding constraint straight line segment meets the crack defect characteristics in the analysis, namely the crack defect has the highest energy in the central part of the corresponding spectrogram, and the gray value of the corresponding pixel point is larger and gradually decreased towards the periphery; if the constraint 2 is satisfied, it indicates that the angle corresponds to the first angle->Straight line segment ofEach corresponding binding line corresponds to the first angle->In a linear section of>The two sides of the photovoltaic panel are symmetrically distributed, so that the photovoltaic panel crack defect characteristics are met. The set constraint difference threshold may be set according to actual conditions, and in the present embodiment, the set constraint difference threshold is set to 0.2.
Using the constraint conditions 1 and 2 to collect the bright line angles of the photovoltaic cell panelThe angles in (1) are screened, so that second angles can be screened, and the second angles form a crack defect angle set ^ or ^ of the photovoltaic cell panel>. Follow-up crack defect angle set according to photovoltaic cell panelShut and/or>Can determine the crack defect line bundle in the spectrogram of the corresponding photovoltaic cell panel>。
Step S6: and carrying out masking processing on the pixel points on the straight line segments corresponding to the second angles in the initial spectrogram so as to obtain a masking spectrogram.
For the straight line segment corresponding to each second angle in the initial spectrogram, namely the crack defect linear beam in the spectrogram of the solar panelAnd generating crack defects at corresponding positions in the time domain images corresponding to the straight-line segments, and setting corresponding crack masks according to the straight-line segments corresponding to the second angles in the initial spectrogram, namely assigning pixel points on the straight-line segments to be 0, and keeping the values of the pixel points at the other positions unchanged, so as to obtain the mask spectrogram of the photovoltaic cell panel.
Step S7: and performing time-frequency inverse transformation on the mask frequency spectrogram to obtain a mask gray-scale image, and determining the crack area of the panel according to the mask gray-scale image and the initial gray-scale image.
Because the corresponding crack information is removed by setting the crack mask in the mask spectrogram obtained in the step S6, the mask spectrogram is subjected to inverse fourier transform, and a mask gray-scale image corresponding to the photovoltaic cell panel can be obtained. However, the mask gray-scale image of the photovoltaic cell panel obtained after the inverse Fourier transform lacks original crack information, so that corresponding crack area information can be obtained by subtracting the mask gray-scale image from the initial gray-scale image of the photovoltaic cell panel, so as to complete crack detection of the TOPCon structure photovoltaic cell panel.
The crack monitoring system of the TOPCon structure photovoltaic cell panel provided by the invention can realize reliable identification and monitoring of cracks of the photovoltaic cell panel, and the specific process is as follows: according to the system, each initial angle is determined in the initial spectrogram by acquiring the initial grey-scale image and the initial spectrogram corresponding to the surface image of the cell panel to be detected, and the initial angles do not include the angle corresponding to the grid line of the cell panel, so that the influence of the grid line on the subsequent crack identification of the photovoltaic cell panel is effectively avoided. The method comprises the steps that pixel points on straight-line segments, passing through a central point, of an initial spectrogram and corresponding to each initial angle are analyzed, so that the angle amplitude corresponding to each initial angle is determined, the crack defects on the photovoltaic cell panel correspond to bright grains in the initial spectrogram, the angle amplitude corresponding to the bright grains is usually large, an angle amplitude threshold value T is determined in a self-adaptive mode according to the angle amplitude corresponding to each initial angle, each first angle possibly corresponding to the crack can be accurately screened in the initial angle based on the angle amplitude threshold value T, and therefore the accuracy and the reliability of follow-up TOPCon structure photovoltaic cell panel crack defect identification are effectively improved. And then determining each constraint line segment straight-line segment of the line segment straight-line segments corresponding to each first angle according to the characteristics of the crack defects, namely the central part of the crack defects in the corresponding spectrogram has the highest energy, the gray value of the corresponding pixel point is larger, and gradually decreases towards the periphery, and the two sides of the central part of the crack defects in the corresponding spectrogram are symmetrically distributed, and further accurately screening each second angle corresponding to the crack in each first angle according to the angle amplitude of the constraint line segment straight-line segments. The mask gray-scale image is obtained through the time-frequency inverse transformation, original crack information is accurately removed from the mask gray-scale image, and therefore the surface crack area of the photovoltaic cell panel can be accurately determined through comparing the mask gray-scale image with the initial gray-scale image. According to the method, the surface image of the photovoltaic cell panel is converted into a frequency domain, and the angle corresponding to the crack is automatically and accurately screened out based on the characteristics of the crack defect in the frequency spectrum, so that the crack defect on the surface of the cell panel is accurately identified, the unreliability of region segmentation performed by artificially setting a threshold value in the prior art is effectively avoided, and the accuracy and reliability of crack monitoring on the photovoltaic cell panel with the TOPCon structure are effectively improved while the overall intelligent effect of the scheme is ensured.
It should be noted that: the precedence order of the above embodiments of the present invention is only for description, and does not represent the merits of the embodiments. And specific embodiments thereof have been described above. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments.
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 (10)
1. The utility model provides a crack monitoring system of TOPCon structure photovoltaic cell board which characterized in that includes:
a monitoring acquisition module for: collecting a surface image of a battery panel to be detected, and sending the surface image to a monitoring and identifying module;
a monitor identification module to: receiving the surface image sent by the monitoring acquisition module, further acquiring an initial gray-scale image corresponding to the surface image, and performing time-frequency transformation on the initial gray-scale image to obtain an initial frequency spectrogram; determining a central point and each initial angle in the initial spectrogram, determining a straight-line segment passing through the central point in the initial spectrogram and corresponding to each initial angle, and determining an angle amplitude corresponding to each initial angle according to pixel points on the straight-line segment corresponding to each initial angle; screening all the initial angles according to the angle amplitude corresponding to each initial angle and the size of the initial spectrogram to obtain each first angle; determining each constraint straight-line segment corresponding to each straight-line segment of the first angle, and determining the angle amplitude of each constraint straight-line segment according to pixel points on each constraint straight-line segment; screening each first angle according to the angle amplitude of each constraint straight-line segment corresponding to each straight-line segment of each first angle to obtain each second angle; masking the pixel points on the straight line segments corresponding to the second angles in the initial spectrogram to obtain a mask spectrogram; and performing time-frequency inverse transformation on the mask frequency spectrogram to obtain a mask gray-scale image, and determining a crack area of the panel according to the mask gray-scale image and the initial gray-scale image so as to realize the monitoring of the panel.
2. Crack monitoring system for TOPCon structured photovoltaic panels according to claim 1, characterised in that said monitoring identification module is adapted to:
determining a gray threshold according to the gray value of each pixel point on the straight-line segment corresponding to each initial angle, and determining the pixel point of which the gray value on the straight-line segment corresponding to each initial angle is larger than the gray threshold as a target pixel point;
and calculating the accumulated sum of all distances corresponding to each initial angle according to the distance from each target pixel point on the straight line segment corresponding to each initial angle to the central point, and determining the accumulated sum as the angle amplitude corresponding to the initial angle.
3. Crack monitoring system for TOPCon structured photovoltaic panels according to claim 1, characterised in that said monitoring identification module is adapted to:
and determining the median of all gray values as a gray threshold according to the gray value of each pixel point on the straight line segment corresponding to each initial angle.
4. Crack monitoring system for TOPCon structured photovoltaic panels according to claim 1, characterised in that said monitoring identification module is adapted to:
calculating an angle amplitude mean value according to the angle amplitude corresponding to each initial angle, and calculating a difference value between the angle amplitude corresponding to each initial angle and the angle amplitude mean value;
calculating a threshold intermediate value according to the difference value of the angle amplitude corresponding to each initial angle and the angle amplitude mean value and the size of the initial spectrogram, and calculating an angle amplitude threshold according to the angle amplitude mean value and the threshold intermediate value;
the angle amplitude corresponding to each initial angle is compared with an angle amplitude threshold value, and the initial angle corresponding to the angle amplitude larger than the angle amplitude threshold value is determined as a first angle.
5. The crack monitoring system for photovoltaic cell panels with a TOPCon structure as claimed in claim 4, wherein the calculation formulas for calculating the intermediate threshold value and the angular amplitude threshold value are respectively:
wherein is present>Is an angle magnitude threshold value>Is a median threshold value>And &>Length and width of the initial spectrogram respectively>To correspond toiThe angular amplitude of the initial angle is such that,mis the total number of initial angles>Is the average value of the amplitude values of the angles,Kis a regulatory factor.
6. Crack monitoring system for TOPCon structured photovoltaic panels according to claim 1, characterised in that said monitoring identification module is adapted to:
taking the straight line segment corresponding to each first angle as a target straight line segment, symmetrically arranging at least one pair of straight line segments on two sides of the target straight line segment, enabling each pair of the arranged straight line segments to be parallel to the corresponding target straight line segment, and determining all the arranged straight line segments as each restraint straight line segment corresponding to the target straight line segment.
7. Crack monitoring system for TOPCon structured photovoltaic panels according to claim 6, characterised in that said monitoring identification module is adapted to:
for the target straight-line segment corresponding to each first angle, determining the distance from each constraint straight-line segment of the target straight-line segment corresponding to each first angle to the target straight-line segment;
calculating the absolute value of the difference value of the angle amplitude of each pair of constrained straight-line segments corresponding to the target straight-line segments of each first angle, and determining the accumulated sum of all the absolute values of the difference values as the constrained difference value of the target straight-line segments corresponding to each first angle;
judging whether each constraint straight-line segment of the target straight-line segment corresponding to each first angle meets constraint conditions, and if so, determining the corresponding first angle as a second angle; the constraint conditions are as follows: for each constraint straight-line segment on each side of the target straight-line segment corresponding to each first angle, the larger the distance from the constraint straight-line segment to the target straight-line segment corresponding to the first angle is, the smaller the angle amplitude of the constraint straight-line segment is, and the angle amplitudes of the constraint straight-line segments are all smaller than the angle amplitude of the target straight-line segment; and the constraint difference value of the target straight-line segment corresponding to each first angle is smaller than a set constraint difference threshold value.
8. Crack monitoring system for TOPCon structured photovoltaic panels according to claim 6, characterised in that said monitoring identification module is adapted to:
the constraint straight line segments corresponding to each side of the target straight line segment of each first angle are arranged at equal intervals.
9. Crack monitoring system for TOPCon structured photovoltaic panels according to claim 1, characterised in that said monitoring identification module is adapted to:
and setting the gray value of the pixel point on the straight line segment corresponding to each second angle in the initial frequency spectrogram as 0, thereby obtaining the mask frequency spectrogram.
10. Crack monitoring system for photovoltaic panels of the TOPCon structure according to claim 1, characterised in that said monitoring and identification module is adapted to:
and performing difference operation on the initial gray level image and the mask gray level image to obtain a difference gray level image, and performing crack area identification on the difference gray level image to obtain a crack area of the battery panel.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310199676.7A CN115954291B (en) | 2023-03-04 | 2023-03-04 | Crack monitoring system of TOPCON structure photovoltaic cell panel |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310199676.7A CN115954291B (en) | 2023-03-04 | 2023-03-04 | Crack monitoring system of TOPCON structure photovoltaic cell panel |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115954291A true CN115954291A (en) | 2023-04-11 |
CN115954291B CN115954291B (en) | 2023-11-14 |
Family
ID=87287996
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310199676.7A Active CN115954291B (en) | 2023-03-04 | 2023-03-04 | Crack monitoring system of TOPCON structure photovoltaic cell panel |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115954291B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117585399A (en) * | 2024-01-19 | 2024-02-23 | 汶上义桥煤矿有限责任公司 | Coal mine conveyor belt tearing detection method based on image processing |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120268733A1 (en) * | 2009-11-04 | 2012-10-25 | Massimo Balducci | Apparatus, system and method for detecting defects of metallic lids |
US20130343632A1 (en) * | 2006-07-14 | 2013-12-26 | Hitachi High-Technologies Corporation | Defect inspection method and apparatus |
US20140210997A1 (en) * | 2013-01-25 | 2014-07-31 | The Boeing Company | System and Method for Automated Crack Inspection and Repair |
JP2014163805A (en) * | 2013-02-26 | 2014-09-08 | Hitachi Ltd | Ultrasonic inspection method and device |
CN110455827A (en) * | 2019-08-13 | 2019-11-15 | 高瑶 | A kind of smart new energy solar panel qualification detection system and method |
CN111369511A (en) * | 2020-02-28 | 2020-07-03 | 浙江大学 | Optical element surface weak scratch detection method based on spectral characteristics |
CN114743115A (en) * | 2022-03-31 | 2022-07-12 | 启东晶尧光电科技有限公司 | Shelter recognition method and fault early warning system for photovoltaic electronic equipment |
-
2023
- 2023-03-04 CN CN202310199676.7A patent/CN115954291B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130343632A1 (en) * | 2006-07-14 | 2013-12-26 | Hitachi High-Technologies Corporation | Defect inspection method and apparatus |
US20120268733A1 (en) * | 2009-11-04 | 2012-10-25 | Massimo Balducci | Apparatus, system and method for detecting defects of metallic lids |
US20140210997A1 (en) * | 2013-01-25 | 2014-07-31 | The Boeing Company | System and Method for Automated Crack Inspection and Repair |
JP2014163805A (en) * | 2013-02-26 | 2014-09-08 | Hitachi Ltd | Ultrasonic inspection method and device |
CN110455827A (en) * | 2019-08-13 | 2019-11-15 | 高瑶 | A kind of smart new energy solar panel qualification detection system and method |
CN111369511A (en) * | 2020-02-28 | 2020-07-03 | 浙江大学 | Optical element surface weak scratch detection method based on spectral characteristics |
CN114743115A (en) * | 2022-03-31 | 2022-07-12 | 启东晶尧光电科技有限公司 | Shelter recognition method and fault early warning system for photovoltaic electronic equipment |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117585399A (en) * | 2024-01-19 | 2024-02-23 | 汶上义桥煤矿有限责任公司 | Coal mine conveyor belt tearing detection method based on image processing |
CN117585399B (en) * | 2024-01-19 | 2024-04-09 | 汶上义桥煤矿有限责任公司 | Coal mine conveyor belt tearing detection method based on image processing |
Also Published As
Publication number | Publication date |
---|---|
CN115954291B (en) | 2023-11-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110334750A (en) | Iron tower of power transmission line bolt corrosion degree image classification recognition methods | |
CN116721107B (en) | Intelligent monitoring system for cable production quality | |
CN115954291A (en) | Crack monitoring system of TOPCon structure photovoltaic cell panel | |
CN115082744B (en) | Artificial intelligence-based solar heat collection efficiency analysis method and system | |
CN106709529B (en) | Visual detection method for photovoltaic cell color difference classification | |
CN109239073A (en) | A kind of detection method of surface flaw for body of a motor car | |
CN111739012A (en) | Camera module white spot detecting system based on turntable | |
CN112880837A (en) | Equipment fault analysis method | |
CN112508911A (en) | Rail joint touch net suspension support component crack detection system based on inspection robot and detection method thereof | |
CN114511519A (en) | Train bottom bolt loss detection method based on image processing | |
CN113487563B (en) | EL image-based self-adaptive detection method for hidden cracks of photovoltaic module | |
CN115940809A (en) | Solar panel fault detection method based on power data and visual analysis | |
CN114663393A (en) | Convolutional neural network-based detection method for defect of poor spacing between photovoltaic modules | |
CN113591973B (en) | Intelligent comparison method for appearance state change of track plate | |
CN107977959B (en) | Respirator state identification method suitable for electric power robot | |
CN105787955A (en) | Sparse segmentation method and device of strip steel defect | |
CN113962929A (en) | Photovoltaic cell assembly defect detection method and system and photovoltaic cell assembly production line | |
CN117351017A (en) | Visual defect detection method for photovoltaic steel structure component | |
US20230042106A1 (en) | System and method for the statistical analysis of images of photovoltaic panels | |
CN114638822B (en) | Method and system for detecting surface quality of automobile cover plate by using optical means | |
CN116167999A (en) | Distribution line zero-value insulator infrared thermal image detection method based on image matching | |
CN110189301B (en) | Foreign matter detection method for generator stator core steel sheet stacking platform | |
CN113947563A (en) | Cable process quality dynamic defect detection method based on deep learning | |
CN117351021B (en) | Intelligent detection method for production quality of photovoltaic connecting wire | |
CN110688919A (en) | Intelligent line patrol fault identification method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |