CN117351034B - Perovskite battery laser scribing method and system - Google Patents
Perovskite battery laser scribing method and system Download PDFInfo
- Publication number
- CN117351034B CN117351034B CN202311640281.2A CN202311640281A CN117351034B CN 117351034 B CN117351034 B CN 117351034B CN 202311640281 A CN202311640281 A CN 202311640281A CN 117351034 B CN117351034 B CN 117351034B
- Authority
- CN
- China
- Prior art keywords
- sequence
- area
- pixels
- adhesion
- obtaining
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 230000007547 defect Effects 0.000 claims abstract description 151
- 230000011218 segmentation Effects 0.000 claims description 40
- 230000002093 peripheral effect Effects 0.000 claims description 37
- 230000002950 deficient Effects 0.000 claims description 36
- 238000010330 laser marking Methods 0.000 claims description 8
- 230000000877 morphologic effect Effects 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 5
- 238000007621 cluster analysis Methods 0.000 claims description 4
- 230000001174 ascending effect Effects 0.000 claims description 3
- 238000003064 k means clustering Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 6
- 238000010586 diagram Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
- G06V10/763—Non-hierarchical techniques, e.g. based on statistics of modelling distributions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
-
- 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
- Y02E10/549—Organic PV cells
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Multimedia (AREA)
- Geometry (AREA)
- Probability & Statistics with Applications (AREA)
- Image Analysis (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to the technical field of image processing, in particular to a perovskite battery laser scribing method and system, comprising the following steps: collecting laser scribing area images of the perovskite battery; obtaining segmented pixels of the laser scribing region image, and merging the segmented pixels to obtain a defect region; calculating the adhesion degree of each defect area, obtaining a plurality of adhesion areas and a plurality of independent defect areas according to the adhesion degree of each defect area, and obtaining a plurality of independent defect areas of each adhesion area; and performing laser scribing early warning control according to the areas and the number of the independent defect areas. Therefore, the accurate laser scribing early warning control is realized by accurately dividing the independent defect areas on the laser scribing.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a perovskite battery laser scribing method and system.
Background
Perovskite batteries are a novel solar cell technology and have the advantages of high efficiency conversion, low cost, easiness in preparation and the like. Laser scribing can cut perovskite cells to the desired dimensions or connect electrodes between cells, ensuring that they can function effectively in a battery. The integrity of the scribe line is critical to the performance of the perovskite cell. Defects in the scribing process may cause poor connection of internal electrodes of the battery, affecting the use efficiency. Therefore, how to accurately divide each independent defect area of the scribing area of the perovskite battery to realize accurate early warning control.
Disclosure of Invention
The invention provides a perovskite battery laser scribing method and a perovskite battery laser scribing system, which are used for solving the existing problems: how to accurately divide the scribing defects of the perovskite batteries, so that early warning control is performed according to the scribing defects.
The invention relates to a perovskite battery laser scribing method and a perovskite battery laser scribing system, which adopt the following technical scheme:
one embodiment of the invention provides a perovskite battery laser scribing method, which comprises the following steps:
collecting laser scribing area images of the perovskite battery;
obtaining a plurality of same-run program columns according to gray value continuous characteristics in the laser marking region image, obtaining a plurality of segmented pixels according to the number of pixels in the same-run sequence, obtaining a plurality of line segmented sequences of each line according to the continuity of the segmented pixels, and obtaining a sequence to be judged of each line segmented sequence; obtaining the merging degree of each line segment sequence and the sequence to be judged according to the gray value similarity of the segment pixels in each line segment sequence and the segment pixels in the sequence to be judged, and merging each line segment sequence and the sequence to be judged according to the merging degree of each line segment sequence and the sequence to be judged to obtain a plurality of defect areas;
obtaining the adhesion degree of each defect area according to the regularity and roundness of the outer edge of each defect area, obtaining a plurality of adhesion areas and a plurality of independent defect areas according to the adhesion degree of each defect area, obtaining a plurality of pit center areas of each adhesion area according to the gray value distribution characteristics in each adhesion area, and dividing each adhesion area according to the pit center areas to obtain a plurality of independent defect areas of each adhesion area;
and performing laser scribing early warning control according to the areas and the number of the independent defect areas.
Preferably, the method for obtaining a plurality of program columns at the same stream according to the gray value continuous characteristic in the laser scribing area image includes the following specific steps:
for a laser scribing area image, taking a pixel at the upper left corner of the image as a first starting point pixel, starting from the first starting point pixel, taking a horizontal direction as a run exploration direction, acquiring continuous and uninterrupted pixels which have the same gray value as the first starting point pixel as co-run pixels of the first starting point pixel, and arranging the co-run pixels of the first starting point pixel and the first starting point pixel according to a run exploration sequence to obtain a co-run program column of the first starting point pixel; taking the pixel of the last pixel of the co-running program column of the first starting point pixel at the next position in the laser marking area image as a second starting point pixel, starting from the second starting point pixel, taking the horizontal direction as a running exploration direction, obtaining continuous co-running pixels which have the same gray value as the second starting point pixel and serve as the co-running pixels of the second starting point pixel, and arranging the co-running pixels of the second starting point pixel and the second starting point pixel according to the running exploration sequence to obtain the co-running program column of the second starting point pixel; and so on, until all of the co-current program columns for the laser scribed area image are acquired.
Preferably, the method for obtaining a plurality of segmented pixels according to the number of pixels in the same run sequence and obtaining a plurality of line segment sequences of each line according to continuity of the segmented pixels includes the following specific steps:
acquiring a same-run program sequence with only one pixel from all the same-run program sequences, marking the same-run program sequence as a single same-run sequence, and taking the pixels in the single same-run sequence as segmented pixels;
the sequence of successive segmented pixels per line in the laser scribed region image is referred to as a line segmentation sequence per line.
Preferably, the specific method for obtaining the sequence to be determined of each line segment sequence includes:
the method comprises the steps of taking a line segmentation sequence of the next line of each line segmentation sequence as a selectable line segmentation sequence of each line segmentation sequence, obtaining the column number corresponding to each pixel in each line segmentation sequence, enabling a section formed by the maximum value and the minimum value of the column numbers of all pixels in each line segmentation sequence to be called as the column number section of each line segmentation sequence, obtaining the column number section of the selectable line segmentation sequence, and taking the selectable line segmentation sequence as a sequence to be judged of each line segmentation sequence when the intersection or the adjacent exists between the column number section of each line segmentation sequence and the column number section of each selectable line segmentation sequence.
Preferably, the merging degree of each line segment sequence and the sequence to be determined is obtained according to the similarity of the gray values of the segment pixels in each line segment sequence and the segment pixels in the sequence to be determined, and the merging treatment is performed on each line segment sequence and the sequence to be determined according to the merging degree of each line segment sequence and the sequence to be determined to obtain a plurality of defect areas, including the specific methods as follows:
wherein,representing the gray value mean value of all segmented pixels in each row segment sequence,/for each row segment sequence>Representing the mean gray value of all segmented pixels in the sequence to be determined of each row segment sequence, +.>Euclidean distance of the central pixel representing each row segment sequence from the central pixel of the sequence to be determined,/>Representing the variance of the gray values of all segmented pixels in each row segment sequence, +.>Representing the number of segmented pixels in each row segment sequence,/-, for example>Representing the variance of the gray values of all segmented pixels in the sequence to be determined for each line segment sequence,/-, for each line segment sequence>Representing the number of segmented pixels in the sequence to be determined for each row segment sequence,/for each row segment sequence>Representation utilization->The function is normalized, and the function is added>Representing the merging degree of each line segment sequence and the sequence to be determined, wherein I represents an absolute value symbol;
and combining the row segmentation sequence with the combination degree larger than a preset combination threshold value with the sequence to be judged to obtain a plurality of defect areas.
Preferably, the adhesion degree of each defective area is obtained according to the regularity and roundness of the outer edge of each defective area, and a plurality of adhesion areas and a plurality of independent defective areas are obtained according to the adhesion degree of each defective area, including the following specific methods:
acquiring edge pixels of each defect area as peripheral pixels of each defect area, acquiring two adjacent peripheral pixels of each peripheral pixel, acquiring an included angle between a tangent line of each peripheral pixel of each defect area and a tangent line of each adjacent peripheral pixel, recording as a tangent line included angle between each peripheral pixel of each defect area and each adjacent peripheral pixel, recording as an average value of tangent line included angles between each peripheral pixel of each defect area and two adjacent peripheral pixels, and recording as an adjacent tangent line included angle of each peripheral pixel of each defect area;
the method for calculating the adhesion degree of each defective area comprises the following steps:
wherein,adjacent tangential angle of the kth peripheral pixel representing each defective area, +.>Representing the number of peripheral pixels per defective area, < > for each defective area>Representing the area of each defective area, +.>Represents the area of the circumscribed circle of each defective area, +.>Represents the degree of adhesion of each defective area, ||represents absolute value sign, |j ++>Representation->Cosine values of (2);
comparing the adhesion degree of each defect area with a preset adhesion degree threshold, judging the defect area with the adhesion degree larger than the preset adhesion degree threshold as an adhesion area, and judging the defect area with the adhesion degree smaller than or equal to the preset adhesion degree threshold as an independent defect area.
Preferably, the method for obtaining the pit center areas of each adhesion area according to the gray value distribution characteristics in each adhesion area includes the following specific steps:
the method comprises the steps of obtaining a gray value of each pixel of each adhesion area, carrying out ascending arrangement on the gray values of all pixels of each adhesion area to obtain a gray value sequence of each adhesion area, taking a difference value between each gray value and the next gray value in the gray value sequence of each adhesion area as a first difference value of each gray value in the gray value sequence of each adhesion area, obtaining a gray value corresponding to the maximum value of the first difference value in the gray value sequence of each adhesion area, marking the gray value as a segmented gray value of the gray value sequence of each adhesion area, and taking a gray value before the segmented gray value in the gray value sequence of each adhesion area as a pit center gray value;
obtaining pixels with gray values being pit center gray values in each adhesion area, wherein the pixels are called suspected pit center pixels of each adhesion area;
and carrying out morphological analysis on the suspected pit center pixels, and marking the residual suspected pit center pixels after morphological treatment as pit center pixels to obtain a connected domain formed by the pit center pixels as pit center regions.
Preferably, the dividing each adhesion area according to the pit center area to obtain a plurality of independent defect areas of each adhesion area includes the following specific steps:
the method comprises the steps of obtaining the number L of pit center areas of each adhesion area, setting the clustering number as L, and carrying out cluster analysis on pixels of each adhesion area by using a K-means clustering method based on the gray value of each pixel, wherein a connected area formed by the pixels in each category is each independent defect area of each adhesion area.
Preferably, the laser scribing early warning control is performed according to the area and the number of the independent defect areas, and the specific method includes:
acquiring the area maximum value Q of the independent defect areas in the areas of all the independent defect areas of the laser scribing area image, and acquiring the number F of the independent defect areas;
the laser scribing early warning control method comprises the following steps:
making an early warning sound A;
forcibly stopping scribing;
wherein,representing a preset area threshold,/->Representing a preset number threshold,/->And representing a preset early warning sound.
A perovskite battery laser scribing system, the system comprising the following modules:
the image acquisition module is used for acquiring the laser scribing area image of the perovskite battery;
the defect region acquisition module is used for obtaining a plurality of same-run program columns according to the gray value continuous characteristics in the laser marking region image, obtaining a plurality of segmented pixels according to the number of pixels in the same-run sequence, obtaining a plurality of line segmented sequences of each line according to the continuity of the segmented pixels, and obtaining a sequence to be judged of each line segmented sequence; obtaining the merging degree of each line segment sequence and the sequence to be judged according to the gray value similarity of the segment pixels in each line segment sequence and the segment pixels in the sequence to be judged, and merging each line segment sequence and the sequence to be judged according to the merging degree of each line segment sequence and the sequence to be judged to obtain a plurality of defect areas;
the independent defect area acquisition module is used for obtaining the adhesion degree of each defect area according to the regularity and the roundness of the outer edge of each defect area, obtaining a plurality of adhesion areas and a plurality of independent defect areas according to the adhesion degree of each defect area, obtaining a plurality of pit center areas of each adhesion area according to the gray value distribution characteristics of each adhesion area, and dividing each adhesion area according to the pit center areas to obtain a plurality of independent defect areas of each adhesion area;
and the scribing early warning control module is used for carrying out laser scribing early warning control according to the area and the number of the independent defect areas.
The technical scheme of the invention has the beneficial effects that: because the gray scale difference of the laser scribing defect area is larger, the pixels of the laser scribing defect area should be pixels at the run-length segmentation points, the acquired laser scribing area image is subjected to run-length analysis to obtain a plurality of segmentation pixels, and the segmentation pixels with similar gray scale values are combined to obtain the defect area.
Because the defects are adhered to each other to obtain a defect area which contains a plurality of independent defect areas, in order to obtain the independent defect areas, adhesion characteristics of the defect areas are required to be analyzed to obtain adhesion areas and the independent defect areas, each small defect of the adhesion areas has pit characteristics, and each adhesion area is analyzed according to the pit characteristics to obtain the independent defect area of each adhesion area. By accurately obtaining the independent defect areas, accurate scribing early warning control can be realized.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of steps of a perovskite battery laser scribing method of the present invention;
FIG. 2 is a block diagram of a perovskite battery laser scribing system of the present invention;
FIG. 3 is an image of a perovskite battery of the invention;
FIG. 4 is a schematic view of a defective area according to the present invention;
FIG. 5 is a block region of the present invention including a number of independent defect regions.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description refers to the specific implementation, structure, characteristics and effects of a perovskite battery laser scribing method and system according to the invention with reference to the attached drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
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.
The following specifically describes a specific scheme of a perovskite battery laser scribing method and a system provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of steps of a perovskite battery laser scribing method according to an embodiment of the invention is shown, the method includes the steps of:
step S001: the current laser scribing area image is collected.
It should be noted that laser scribing can cut perovskite cells to the desired dimensions, or connect electrodes between cells, ensuring that they can function effectively in a battery. The integrity of the scribe line is critical to the performance of the perovskite cell. Therefore, scribing early warning control is needed according to the scribing defects of the perovskite battery in the scribing process. The perovskite battery image is acquired first, and a laser scribing area image in the perovskite battery image is segmented.
Specifically, in order to implement the perovskite battery laser scribing method provided in this embodiment, this embodiment needs to collect the perovskite battery image first, and divide the laser scribing area image in the perovskite battery image.
The specific operation of collecting the perovskite battery image and dividing the laser scribing area image in the perovskite battery image comprises the following steps:
a camera was placed directly above the perovskite cell, and the camera acquired perovskite cell images every 1 minute. Fig. 3 shows a perovskite battery image.
Labeling and labeling the laser scribing area in the perovskite battery image collected by manual work, forming a data set from the perovskite battery image collected by history with labels, constructing a laser scribing area segmentation network, wherein the laser scribing area segmentation network is a YoloV3 network, and completing training of the laser scribing area segmentation network by utilizing the data set.
Inputting the currently acquired perovskite battery image into a laser scribing area segmentation network for segmentation processing to obtain a laser scribing area image of the current perovskite battery image.
Step S002: and obtaining a plurality of segmented pixels in the current laser scribing area image, and merging the segmented pixels to obtain a defect area.
It should be noted that when the laser scribing has defects, that is, a series of pits with different depths exist on the laser scribing, the gray scale value of the defect area of the laser scribing has large difference, and the gray scale value of the non-defect area of the laser scribing is similar, so that the analysis can be performed according to the gray scale run.
Specifically, for the current laser scribing area image, taking a pixel at the upper left corner of the image as a first starting point pixel, taking a horizontal direction as a run exploration direction, acquiring continuous pixels which have the same gray value as the first starting point pixel and are used as the same-run pixels of the first starting point pixel, and arranging the same-run pixels of the first starting point pixel and the first starting point pixel according to a run exploration sequence to obtain the same-run sequence of the first starting point pixel. Taking the pixel of the last pixel of the same run sequence of the first starting point pixel at the next position in the current laser marking area image as a second starting point pixel, starting from the second starting point pixel, taking the horizontal direction as a run exploration direction, acquiring continuous same-run pixels which are the same as the second starting point pixel in gray scale value, and arranging the same-run pixels of the second starting point pixel and the second starting point pixel according to the run exploration sequence to obtain a same-run program sequence of the second starting point pixel; and so on until all the co-current program columns of the current laser scribe area image are acquired.
The pixel at the next position of the pixel means the pixel at the next position of each pixel in the horizontal direction. When a pixel is in the last column of the row, the pixel in the next position of the pixel is the first pixel in the next row.
Further, the co-run program columns with only one pixel are obtained in all the co-run program columns and are recorded as a single co-run sequence, and the pixels in the single co-run sequence are used as segmented pixels. The segmented pixels are typically pixels of the defective area because of differences in gray values between the segmented pixels and adjacent other pixels.
It should be noted that, the above process obtains segmented pixels, but it cannot be determined which segmented pixels belong to the same defect area, so that the segmented pixels need to be combined through similarity of gray values of the segmented pixels, and the segmented pixels belonging to the same defect area are combined.
Further, a sequence formed by each row of continuous segmented pixels is called a row segmented sequence of each row, a row segmented sequence of the next row of each row segmented sequence is used as a selectable row segmented sequence of each row segmented sequence, a column number corresponding to each pixel in each row segmented sequence is obtained, a section formed by the maximum value and the minimum value of the column numbers of all pixels in each row segmented sequence is called a column number section of each row segmented sequence, a column number section of the selectable row segmented sequence is obtained, and when the column number section of each row segmented sequence and the column number section of each selectable row segmented sequence have intersection or are adjacent, the selectable row segmented sequence is used as a sequence to be judged of each row segmented sequence.
The calculation method of the merging degree of each line segment sequence and the sequence to be judged comprises the following steps:
wherein,representing the gray value mean value of all segmented pixels in each row segment sequence,/for each row segment sequence>Representing the mean gray value of all segmented pixels in the sequence to be determined of each row segment sequence, +.>Reflecting the difference of gray values of the segmented pixels in each row segment sequence and the segmented pixels in the sequence to be determined,/for each row segment sequence>Representing the Euclidean distance between the central pixel of each line segment sequence and the central pixel of the sequence to be determined, wherein the larger the Euclidean distance is, the more the segment pixel in the line segment sequence is far from the segment pixel in the sequence to be determined, so that the segment pixel in the line segment sequence and the segment pixel in the sequence to be determined belong to the same segment pixelThe probability of a trap area is less, +.>Representing the variance of the gray values of all segmented pixels in each row segment sequence, +.>Representing the number of segmented pixels in each row segment sequence,/-, for example>Representing the variance of the gray values of all segmented pixels in the sequence to be determined for each line segment sequence,/-, for each line segment sequence>Representing the number of segmented pixels in the sequence to be determined for each row segment sequence. />The gray value distribution difference of the segmented pixels in each row segment sequence and the segmented pixels in the sequence to be determined is reflected, and the larger the gray value distribution difference is indicated, so that the probability that the segmented pixels in the row segment sequence and the segmented pixels in the sequence to be determined belong to the same defect area is smaller. />Representation utilization->And (5) carrying out normalization processing on the function. />Representing the degree of merging of each line segment sequence with the sequence to be determined, || represents the absolute value sign.
Further, the line segment sequence with the merging degree larger than the preset merging threshold Y1 is merged with the sequence to be determined to obtain a plurality of defect areas, and fig. 4 shows a schematic diagram of the defect areas.
In this embodiment, Y1 is taken as an example of 0.85, and other values may be taken in other embodiments, and the embodiment is not particularly limited.
Step S003: obtaining a bonding area and independent defect areas according to the defect areas, obtaining a plurality of concave central areas of each bonding area, and obtaining a plurality of independent defect areas of the bonding areas according to the plurality of concave central areas.
It should be noted that some of the obtained defect regions are formed by adhering a plurality of small independent defects, and in order to improve the dividing accuracy of the defect regions, adhesion determination needs to be performed on the defect regions. Wherein a defective area formed by sticking a plurality of small defects has a region edge that is not smooth and has poor regularity, and thus it is possible to perform a sticking judgment for each defective area based on this.
Further, the edge pixels of each defect area are obtained and marked as peripheral pixels of each defect area, two adjacent peripheral pixels of each peripheral pixel are obtained, the included angle between the tangent line of each peripheral pixel of each defect area and the tangent line of each adjacent peripheral pixel is obtained, the included angle between each peripheral pixel of each defect area and the tangent line of each adjacent peripheral pixel is marked, the average value of the included angles between each peripheral pixel of each defect area and the tangent line of two adjacent peripheral pixels is marked as the adjacent tangent line included angle of each peripheral pixel of each defect area.
The method for calculating the adhesion degree of each defective area comprises the following steps:
wherein,adjacent tangential angle of the kth peripheral pixel representing each defective area, +.>Represents the number of peripheral pixels per defective area, and also represents the perimeter of each defective area,/->The adjacent tangent line can be included in->Mapping to [0,2 ]]Between (I)>Representation->Maximum value that can be taken, thus byThe value of the included angle of the adjacent tangent line can be reflected, and the larger the value is, the larger the included angle of the adjacent tangent line is, namely the larger the tangential angle variation of peripheral pixels of the defect area is, because the outer surrounding edge of the defect area is irregular, the larger the adhesion degree of the defect area is. />Represents the area of each defective area and also the number of pixels contained in each defective area,/->Represents the area of the circumscribed circle of each defective area, +.>The ratio of the area of the defective area to the area of the circumscribing circle of the defective area is represented, and the larger the value is, the more similar the area of the circumscribing circle of the defective area and the defective area is, and thus the more round the defective area is, and thus the smaller the adhesion degree of the defective area is. />Represents the degree of adhesion of each defective area, ||represents absolute value sign, |j ++>Representation->Cosine values of (a) are provided.
Further, the adhesion degree of each defect area is compared with a preset adhesion degree threshold Y2, the defect area with the adhesion degree larger than the preset adhesion degree threshold Y2 is judged to be an adhesion area, and the defect area with the adhesion degree smaller than or equal to the preset adhesion degree threshold Y2 is judged to be an independent defect area.
In this embodiment, Y2 is taken as an example of 0.85, and other values may be taken in other embodiments, and the embodiment is not particularly limited.
It should be noted that, the defect of the laser scribing is generally a pit with different depths, only the independent defect area includes only one pit, the adhesion area includes a plurality of pits, the center of the pit is generally low in gray value, and the gray value difference between the center of the pit and other areas is large, so that the plurality of pit centers of the adhesion area are obtained based on the feature.
Further, the gray value of each pixel of each adhesion area is obtained, the gray values of all pixels of each adhesion area are arranged in an ascending order to obtain a gray value sequence of each adhesion area, the difference value between each gray value and the next gray value in the gray value sequence of each adhesion area is used as a first difference value of each gray value in the gray value sequence of each adhesion area, the gray value corresponding to the maximum value of the first difference value is obtained in the gray value sequence of each adhesion area, the gray value is recorded as the segmented gray value of the gray value sequence of each adhesion area, and the gray value before the segmented gray value in the gray value sequence of each adhesion area is called the pit center gray value.
And obtaining pixels with gray values being pit center gray values in each adhesion area, wherein the pixels are called suspected pit center pixels of each adhesion area.
It should be noted that, since the pixels at the center of the pseudo pit are scattered, and the pixels at the center of the pit are generally scattered, the interference of the pixels at the center of the pseudo pit, which are scattered, needs to be eliminated.
And carrying out morphological analysis on the suspected pit center pixels, and calling the residual suspected pit center pixels after morphological processing as pit center pixels. And acquiring a communication domain formed by pit center pixels, recording the communication domain as pit center regions, and acquiring the number L of pit center regions of each adhesion region.
The number of pit center areas reflects the number of independent defect areas in the stuck area. Thus, in order to obtain each of the independent defect areas in the stuck area, cluster analysis is performed based on the number of pit center areas.
Further, setting the clustering number as L, performing cluster analysis on the pixels of each adhesion region by using a K-means clustering method based on the gray value of each pixel, wherein the connected region formed by the pixels in each category is each independent defect region of each adhesion region, and fig. 5 shows a schematic diagram of the adhesion region including a plurality of independent defect regions.
So far, each independent defect area on the laser scribing is segmented.
Step S004: and performing laser scribing early warning control according to the independent defect area.
The area of each independent defect region and the number of independent defect regions on the laser scribe line can reflect the quality of the laser scribe line, and the larger the area of the independent defect region and the larger the number of the independent defect regions, the worse the quality, and thus the laser scribe processing early warning is performed based on the above.
And acquiring the area maximum value Q of the independent defect areas in the areas of all the independent defect areas of the current laser scribing area image, and acquiring the number F of the independent defect areas.
The laser scribing early warning control method comprises the following steps:
making an early warning sound A;
forcibly stopping scribing;
wherein,representing a preset area threshold,/->Representing a preset number threshold,/->And representing a preset early warning sound. In this embodiment, Y3 is taken as 10, Y4 is taken as 5, and other values may be taken in other embodiments, and the embodiment is not particularly limited. In this embodiment, a is described as "having quality abnormality, please confirm" as an example, and other embodiments may take other values, and the embodiment is not particularly limited.
Referring to fig. 2, a perovskite battery laser scribing system provided by an embodiment of the invention is shown, the system includes the following modules:
the image acquisition module is used for acquiring the laser scribing area image of the perovskite battery;
the defect region acquisition module is used for obtaining a plurality of same-run program columns according to the gray value continuous characteristics in the laser marking region image, obtaining a plurality of segmented pixels according to the number of pixels in the same-run sequence, obtaining a plurality of line segmented sequences of each line according to the continuity of the segmented pixels, and obtaining a sequence to be judged of each line segmented sequence; obtaining the merging degree of each line segment sequence and the sequence to be judged according to the gray value similarity of the segment pixels in each line segment sequence and the segment pixels in the sequence to be judged, and merging each line segment sequence and the sequence to be judged according to the merging degree of each line segment sequence and the sequence to be judged to obtain a plurality of defect areas;
the independent defect area acquisition module is used for obtaining the adhesion degree of each defect area according to the regularity and the roundness of the outer edge of each defect area, obtaining a plurality of adhesion areas and a plurality of independent defect areas according to the adhesion degree of each defect area, obtaining a plurality of pit center areas of each adhesion area according to the gray value distribution characteristics of each adhesion area, and dividing each adhesion area according to the pit center areas to obtain a plurality of independent defect areas of each adhesion area;
and the scribing early warning control module is used for carrying out laser scribing early warning control according to the area and the number of the independent defect areas.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, but any modifications, equivalent substitutions, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.
Claims (9)
1. A perovskite battery laser scribing method, characterized in that the method comprises the following steps:
collecting laser scribing area images of the perovskite battery;
obtaining same-run sequences according to gray value continuous features in the laser marking region images, obtaining segmented pixels according to the number of pixels in the same-run sequences, obtaining line segmented sequences of each line according to the continuity of the segmented pixels, and obtaining sequences to be judged of each line segmented sequence; obtaining the merging degree of each line segment sequence and the sequence to be judged according to the gray value similarity of the segment pixels in each line segment sequence and the segment pixels in the sequence to be judged, and merging each line segment sequence and the sequence to be judged according to the merging degree of each line segment sequence and the sequence to be judged to obtain a defect region;
obtaining the adhesion degree of each defect area according to the regularity and roundness of the outer edge of each defect area, obtaining adhesion areas and independent defect areas according to the adhesion degree of each defect area, obtaining a pit center area of each adhesion area according to the gray value distribution characteristics in each adhesion area, and dividing each adhesion area according to the pit center area to obtain the independent defect area of each adhesion area;
performing laser scribing early warning control according to the area and the number of the independent defect areas;
the method comprises the following specific steps of:
wherein,representing the gray value mean value of all segmented pixels in each row segment sequence,/for each row segment sequence>Representing the mean gray value of all segmented pixels in the sequence to be determined of each row segment sequence, +.>Euclidean distance of the central pixel representing each row segment sequence from the central pixel of the sequence to be determined,/>Representing the variance of the gray values of all segmented pixels in each row segment sequence,representing the number of segmented pixels in each row segment sequence,/-, for example>Representing the variance of the gray values of all segmented pixels in the sequence to be determined for each line segment sequence,/-, for each line segment sequence>Representing the number of segmented pixels in the sequence to be determined for each row segment sequence,representation utilization->The function is normalized, and the function is added>Representing the merging degree of each line segment sequence and the sequence to be determined, wherein I represents an absolute value symbol;
and combining the row segmentation sequence with the combination degree larger than a preset combination threshold value with the sequence to be judged to obtain a defect region.
2. The method for laser scribing a perovskite battery according to claim 1, wherein the method for obtaining the same-stream program line according to the gray value continuous characteristic in the laser scribing area image comprises the following specific steps:
for a laser scribing area image, taking a pixel at the upper left corner of the image as a first starting point pixel, starting from the first starting point pixel, taking a horizontal direction as a run exploration direction, acquiring continuous and uninterrupted pixels which have the same gray value as the first starting point pixel as co-run pixels of the first starting point pixel, and arranging the co-run pixels of the first starting point pixel and the first starting point pixel according to a run exploration sequence to obtain a co-run program column of the first starting point pixel; taking the pixel of the last pixel of the co-running program column of the first starting point pixel at the next position in the laser marking area image as a second starting point pixel, starting from the second starting point pixel, taking the horizontal direction as a running exploration direction, obtaining continuous co-running pixels which have the same gray value as the second starting point pixel and serve as the co-running pixels of the second starting point pixel, and arranging the co-running pixels of the second starting point pixel and the second starting point pixel according to the running exploration sequence to obtain the co-running program column of the second starting point pixel; and so on, until all of the co-current program columns for the laser scribed area image are acquired.
3. The method for laser scribing a perovskite battery according to claim 1, wherein the step of obtaining segmented pixels according to the number of pixels in the same run sequence and obtaining a row segmentation sequence of each row according to continuity of the segmented pixels comprises the following specific steps:
acquiring a same-run program sequence with only one pixel from all the same-run program sequences, marking the same-run program sequence as a single same-run sequence, and taking the pixels in the single same-run sequence as segmented pixels;
the sequence of successive segmented pixels per line in the laser scribed region image is referred to as a line segmentation sequence per line.
4. The method for laser scribing a perovskite battery according to claim 1, wherein the step of obtaining the sequence to be determined of each row segment sequence comprises the following specific steps:
the method comprises the steps of taking a line segmentation sequence of the next line of each line segmentation sequence as a selectable line segmentation sequence of each line segmentation sequence, obtaining the column number corresponding to each pixel in each line segmentation sequence, enabling a section formed by the maximum value and the minimum value of the column numbers of all pixels in each line segmentation sequence to be called as the column number section of each line segmentation sequence, obtaining the column number section of the selectable line segmentation sequence, and taking the selectable line segmentation sequence as a sequence to be judged of each line segmentation sequence when the intersection or the adjacent exists between the column number section of each line segmentation sequence and the column number section of each selectable line segmentation sequence.
5. The perovskite battery laser scribing method according to claim 1, wherein the method for obtaining the adhesion degree of each defect area according to the regularity and roundness of the outer edge of each defect area and obtaining the adhesion area and the independent defect area according to the adhesion degree of each defect area comprises the following specific steps:
acquiring edge pixels of each defect area as peripheral pixels of each defect area, acquiring two adjacent peripheral pixels of each peripheral pixel, acquiring an included angle between a tangent line of each peripheral pixel of each defect area and a tangent line of each adjacent peripheral pixel, recording as a tangent line included angle between each peripheral pixel of each defect area and each adjacent peripheral pixel, recording as an average value of tangent line included angles between each peripheral pixel of each defect area and two adjacent peripheral pixels, and recording as an adjacent tangent line included angle of each peripheral pixel of each defect area;
the method for calculating the adhesion degree of each defective area comprises the following steps:
wherein,adjacent tangential angle of the kth peripheral pixel representing each defective area, +.>Representing the number of peripheral pixels per defective area, < > for each defective area>Representing the area of each defective area, +.>Represents the area of the circumscribed circle of each defective area, +.>Represents the degree of adhesion of each defective area, ||represents absolute value sign, |j ++>Representation->Cosine values of (2);
comparing the adhesion degree of each defect area with a preset adhesion degree threshold, judging the defect area with the adhesion degree larger than the preset adhesion degree threshold as an adhesion area, and judging the defect area with the adhesion degree smaller than or equal to the preset adhesion degree threshold as an independent defect area.
6. The method for laser scribing a perovskite battery according to claim 1, wherein the method for obtaining the pit center region of each adhesion region according to the gray value distribution characteristic in each adhesion region comprises the following specific steps:
the method comprises the steps of obtaining a gray value of each pixel of each adhesion area, carrying out ascending arrangement on the gray values of all pixels of each adhesion area to obtain a gray value sequence of each adhesion area, taking a difference value between each gray value and the next gray value in the gray value sequence of each adhesion area as a first difference value of each gray value in the gray value sequence of each adhesion area, obtaining a gray value corresponding to the maximum value of the first difference value in the gray value sequence of each adhesion area, marking the gray value as a segmented gray value of the gray value sequence of each adhesion area, and taking a gray value before the segmented gray value in the gray value sequence of each adhesion area as a pit center gray value;
obtaining pixels with gray values being pit center gray values in each adhesion area, wherein the pixels are called suspected pit center pixels of each adhesion area;
and carrying out morphological analysis on the suspected pit center pixels, and marking the residual suspected pit center pixels after morphological treatment as pit center pixels to obtain a connected domain formed by the pit center pixels as pit center regions.
7. The perovskite battery laser scribing method according to claim 1, wherein the dividing each adhesion area according to the pit center area to obtain the independent defect area of each adhesion area comprises the following specific steps:
the method comprises the steps of obtaining the number L of pit center areas of each adhesion area, setting the clustering number as L, and carrying out cluster analysis on pixels of each adhesion area by using a K-means clustering method based on the gray value of each pixel, wherein a connected area formed by the pixels in each category is each independent defect area of each adhesion area.
8. The perovskite battery laser scribing method according to claim 1, wherein the laser scribing pre-warning control is performed according to the area and the number of the independent defect areas, and the specific method comprises the following steps:
acquiring the area maximum value Q of the independent defect areas in the areas of all the independent defect areas of the laser scribing area image, and acquiring the number F of the independent defect areas;
the laser scribing early warning control method comprises the following steps:
making an early warning sound A;
forcibly stopping scribing;
wherein,representing a preset area threshold,/->Representing a preset number threshold,/->And representing a preset early warning sound.
9. A perovskite battery laser scribing system, comprising the following modules:
the image acquisition module is used for acquiring the laser scribing area image of the perovskite battery;
the defect region acquisition module is used for obtaining a same-run sequence according to the gray value continuous characteristic in the laser marking region image, obtaining segmented pixels according to the number of pixels in the same-run sequence, obtaining a line segmentation sequence of each line according to the continuity of the segmented pixels, and acquiring a sequence to be judged of each line segmentation sequence; obtaining the merging degree of each line segment sequence and the sequence to be judged according to the gray value similarity of the segment pixels in each line segment sequence and the segment pixels in the sequence to be judged, and merging each line segment sequence and the sequence to be judged according to the merging degree of each line segment sequence and the sequence to be judged to obtain a defect region;
the method comprises the following specific steps of:
wherein,representing the gray value mean value of all segmented pixels in each row segment sequence,/for each row segment sequence>Representing the mean gray value of all segmented pixels in the sequence to be determined of each row segment sequence, +.>Euclidean distance of the central pixel representing each row segment sequence from the central pixel of the sequence to be determined,/>Representing the variance of the gray values of all segmented pixels in each row segment sequence,representing the number of segmented pixels in each row segment sequence,/-, for example>Representing the variance of the gray values of all segmented pixels in the sequence to be determined for each line segment sequence,/-, for each line segment sequence>Representing the number of segmented pixels in the sequence to be determined for each row segment sequence,representation utilization->The function is normalized, and the function is added>Representing the merging degree of each line segment sequence and the sequence to be determined, wherein I represents an absolute value symbol;
combining the row segmentation sequences with the combination degree larger than a preset combination threshold value with sequences to be judged to obtain a defect region;
the independent defect area acquisition module is used for obtaining the adhesion degree of each defect area according to the regularity and the roundness of the outer edge of each defect area, obtaining an adhesion area and an independent defect area according to the adhesion degree of each defect area, obtaining a pit center area of each adhesion area according to the gray value distribution characteristic in each adhesion area, and dividing each adhesion area according to the pit center area to obtain the independent defect area of each adhesion area;
and the scribing early warning control module is used for carrying out laser scribing early warning control according to the area and the number of the independent defect areas.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311640281.2A CN117351034B (en) | 2023-12-04 | 2023-12-04 | Perovskite battery laser scribing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311640281.2A CN117351034B (en) | 2023-12-04 | 2023-12-04 | Perovskite battery laser scribing method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117351034A CN117351034A (en) | 2024-01-05 |
| CN117351034B true CN117351034B (en) | 2024-02-13 |
Family
ID=89365227
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311640281.2A Active CN117351034B (en) | 2023-12-04 | 2023-12-04 | Perovskite battery laser scribing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117351034B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019223068A1 (en) * | 2018-05-25 | 2019-11-28 | 平安科技（深圳）有限公司 | Iris image local enhancement method, device, equipment and storage medium |
| CN114862862A (en) * | 2022-07-11 | 2022-08-05 | 江苏大田阀门制造有限公司 | Pump body cold shut defect identification method and system based on image processing |
| CN115187602A (en) * | 2022-09-13 | 2022-10-14 | 江苏骏利精密制造科技有限公司 | Injection molding part defect detection method and system based on image processing |
| CN117036357A (en) * | 2023-10-10 | 2023-11-10 | 无锡市方顺型钢科技有限公司 | Profile steel surface defect detection method based on visual characteristics |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101540040B (en) * | 2008-03-21 | 2012-12-12 | 深圳迈瑞生物医疗电子股份有限公司 | Method and device for automatically detecting boundary of beam-limiting device |
-
2023
- 2023-12-04 CN CN202311640281.2A patent/CN117351034B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019223068A1 (en) * | 2018-05-25 | 2019-11-28 | 平安科技（深圳）有限公司 | Iris image local enhancement method, device, equipment and storage medium |
| CN114862862A (en) * | 2022-07-11 | 2022-08-05 | 江苏大田阀门制造有限公司 | Pump body cold shut defect identification method and system based on image processing |
| CN115187602A (en) * | 2022-09-13 | 2022-10-14 | 江苏骏利精密制造科技有限公司 | Injection molding part defect detection method and system based on image processing |
| CN117036357A (en) * | 2023-10-10 | 2023-11-10 | 无锡市方顺型钢科技有限公司 | Profile steel surface defect detection method based on visual characteristics |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117351034A (en) | 2024-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108711148B (en) | Tire defect intelligent detection method based on deep learning | |
| CN112233092A (en) | Deep learning method for intelligent defect detection of unmanned aerial vehicle power inspection | |
| CN111754498A (en) | Conveyor belt carrier roller detection method based on YOLOv3 | |
| CN111027539B (en) | License plate character segmentation method based on spatial position information | |
| CN114820625B (en) | Automobile top block defect detection method | |
| CN110544231A (en) | lithium battery electrode surface defect detection method based on background standardization and centralized compensation algorithm | |
| CN115019158B (en) | Image recognition-based marine pollution area recognition method and system | |
| CN113469951B (en) | Hub defect detection method based on cascade region convolutional neural network | |
| CN115456955A (en) | Method for detecting internal burr defect of ball cage dust cover | |
| CN115170518A (en) | Cell detection method and system based on deep learning and machine vision | |
| CN113240623A (en) | Pavement disease detection method and device | |
| CN115880571A (en) | Water level gauge reading identification method based on semantic segmentation | |
| CN116385374A (en) | Cell counting method based on convolutional neural network | |
| CN111951289A (en) | BA-Unet-based underwater sonar image data segmentation method | |
| CN114663393A (en) | Convolutional neural network-based detection method for defect of poor spacing between photovoltaic modules | |
| CN114782442A (en) | Photovoltaic cell panel intelligent inspection method and system based on artificial intelligence | |
| CN114926400A (en) | Fan blade defect detection method based on improved YOLOv5 | |
| CN117351034B (en) | Perovskite battery laser scribing method and system | |
| CN109685773B (en) | Method for determining synapse position points in immune cell image | |
| CN116052110B (en) | Intelligent positioning method and system for pavement marking defects | |
| CN116704387B (en) | Power line channel inspection system and method based on video structuring | |
| CN113554610A (en) | Photovoltaic module operation state detection method and application device thereof | |
| CN110765900B (en) | Automatic detection illegal building method and system based on DSSD | |
| CN115761613B (en) | Automatic tunnel crack detection method based on convolutional network | |
| US20240169510A1 (en) | Surface defect detection model training method, and surface defect detection method and system |
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 |