CN116703894B - Lithium battery diaphragm quality detection system - Google Patents
Lithium battery diaphragm quality detection system Download PDFInfo
- Publication number
- CN116703894B CN116703894B CN202310960010.9A CN202310960010A CN116703894B CN 116703894 B CN116703894 B CN 116703894B CN 202310960010 A CN202310960010 A CN 202310960010A CN 116703894 B CN116703894 B CN 116703894B
- Authority
- CN
- China
- Prior art keywords
- coefficient
- diaphragm
- window area
- pixel point
- hole
- 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
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 title claims abstract description 39
- 229910052744 lithium Inorganic materials 0.000 title claims abstract description 39
- 238000001514 detection method Methods 0.000 title claims abstract description 19
- 230000008859 change Effects 0.000 claims abstract description 82
- 238000004458 analytical method Methods 0.000 claims abstract description 42
- 238000004364 calculation method Methods 0.000 claims abstract description 15
- 239000012528 membrane Substances 0.000 claims description 14
- 238000012935 Averaging Methods 0.000 claims description 4
- 238000003708 edge detection Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 2
- 230000007547 defect Effects 0.000 abstract description 13
- 238000012545 processing Methods 0.000 abstract description 5
- 238000000034 method Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000003064 k means clustering Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000003247 decreasing effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 229910052704 radon Inorganic materials 0.000 description 1
- SYUHGPGVQRZVTB-UHFFFAOYSA-N radon atom Chemical compound [Rn] SYUHGPGVQRZVTB-UHFFFAOYSA-N 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/10—Energy storage using batteries
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Educational Administration (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Strategic Management (AREA)
- Game Theory and Decision Science (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Image Analysis (AREA)
Abstract
The application relates to the field of image processing, and provides a lithium battery diaphragm quality detection system, which comprises the following components: the characteristic analysis module is used for calculating a diaphragm stretching gradual change rule coefficient of a window area where each pixel point in the image to be detected is located and a diaphragm darkness index of the window area; the characteristic calculation module is used for calculating a hole point gathering coefficient of each pixel point based on a diaphragm stretching gradual change rule coefficient of a window area where each pixel point is located in the image to be detected and a diaphragm darkness index of the window area; the clustering module clusters the pixel points to a first class cluster or a second class cluster based on the hole clustering coefficient of each pixel point, wherein the first class cluster represents a hole area, and the second class cluster represents a normal area. According to the scheme, whether the window area has stretching defects or not can be detected by calculating the diaphragm stretching gradual change rule coefficient of the window area, and whether the window area has hole defects or not can be detected by calculating the diaphragm darkness index of the window area, so that the accuracy of quality detection is improved.
Description
Technical Field
The application relates to the field of image processing, in particular to a lithium battery diaphragm quality detection system.
Background
The normal operation of the lithium battery needs to be guaranteed by the diaphragm, if flaws exist on the surface of the diaphragm, the product quality of the lithium battery can be greatly reduced, the performance of the lithium battery can be reduced, and safety problems such as electric leakage and the like can also occur. The surface of the lithium battery diaphragm has defects such as black spots, orange peel marks, holes and the like, wherein the most serious defects are the holes. If holes appear on the surface of the lithium battery diaphragm, the positive electrode and the negative electrode in the battery are in direct contact, short circuit of the battery is easy to be caused, and serious safety problems such as fire and explosion of the battery are possibly caused. And the quality of the lithium battery diaphragm can be detected by an image processing technology, so that the defect product can be effectively prevented from flowing into the market. The traditional image processing is to obtain a hole area by adopting an area growth method, and then evaluate and detect the quality of the lithium battery diaphragm by the detected hole flaws. However, the method does not identify the stretched area as a defective area, so that the accuracy of quality detection of the lithium battery separator is low.
Disclosure of Invention
The application provides a lithium battery diaphragm quality detection system, which can detect flaws in a broken hole area and also detect flaws in a stretching area, thereby improving accuracy of quality detection.
In a first aspect, the present application provides a lithium battery separator quality detection system, comprising:
the characteristic analysis module is used for calculating a diaphragm stretching gradual change rule coefficient of a window area where each pixel point in the image to be detected is located and a diaphragm darkness index of the window area; the diaphragm stretching gradual change rule coefficient represents whether a diaphragm stretching condition occurs in the window area, and the diaphragm darkness index represents whether the window area is a broken hole flaw area;
the characteristic calculation module is used for calculating a broken hole point gathering coefficient of each pixel point based on a diaphragm stretching gradual change rule coefficient of a window area where each pixel point is located in an image to be detected and a diaphragm darkness index of the window area;
the clustering module is used for clustering the pixel points into a first type cluster or a second type cluster based on the hole point coefficient of each pixel point, wherein the first type cluster represents a hole area, and the second type cluster represents a normal area.
In an alternative embodiment, the feature analysis module includes: a first feature analysis module;
the first feature analysis module is used for carrying out edge detection on the image to be detected to obtain edge binary images of all areas in the image to be detected; and constructing a window area with a preset size by taking each pixel point as a center, and determining the diaphragm dark index of the window area based on the gray value of the pixel point of the window area and the edge line in the edge binary image.
In an alternative embodiment, the first feature analysis module is further configured to:
determining the number of segmentation areas formed by the edges of the window area and edge lines in the edge binary image in the window area;
if the number of the divided areas is 1, taking the gray average value of the pixel points in the window area as the diaphragm dark index of the window area;
if the number of the divided areas is larger than 1, calculating gray average values corresponding to all the divided areas, and taking the ratio of the gray average value of the smallest divided area to the difference value of the maximum gray value and the minimum gray value in the window area as the diaphragm dark channel index of the window area.
In an alternative embodiment, the feature analysis module includes: a second feature analysis module;
the second feature analysis module is used for calculating a gray rule change coefficient of a projection value of each pixel point in each angle direction by adopting a pulling transformation algorithm based on the gray value of the pixel point in each angle direction of the window area, so as to obtain a projection value sequence in each angle direction; the diaphragm stretching gradual change rule coefficient of the window area is determined based on the projection value sequence in each angle direction and the gray value of the pixel point in the angle direction.
In an alternative embodiment, the second feature analysis module is further configured to:
calculating the gray value difference value between two adjacent pixel points in the longitudinal axis direction of each pixel point in the angle direction, summing the gray value difference values calculated by all pixel points in the longitudinal axis direction, and averaging to obtain the gray rule change coefficient of the projection value of each pixel point in the angle direction, wherein the gray rule change coefficient of the projection value of all pixel points in the angle direction forms a projection value sequence in the angle direction;
calculating a gray scale rule change coefficient mean value in the angle direction based on all gray scale rule change coefficients in the projection value sequence in the angle direction;
based on the difference between the average value of the gray scale rule change coefficients in the angle direction and the gray scale rule change coefficient of the projection value of each pixel point, the diaphragm stretching and gradual change rule coefficients in the angle direction are obtained, and the diaphragm stretching and gradual change rule coefficients in all the angle directions are used as the diaphragm stretching and gradual change rule coefficients of the window area.
In an alternative embodiment, the second feature analysis module is further configured to:
calculating the diaphragm stretching gradual change rule coefficient of each angle direction by using the following formula:
wherein,a membrane stretch gradient law coefficient representing angular direction, +.>Represents the number of pixels in the angular direction, < +.>Representing the number of pixels in the vertical axis direction of the pixels in the angular direction, +.>Representing the gray value of a pixel point (i, j) of which the coordinate point (i) is adjacent to the pixel point (i) of which the abscissa is in the vertical axis direction in the angle direction,/>The pixel point with the abscissa of i is in the vertical axis direction in the angular directionGray value of pixel point with (i, j+1) as coordinate point adjacent upward,/>The average value of the change coefficient of the gray scale rule in the angle direction is represented.
In an alternative embodiment, the feature calculation module is configured to:
determining the maximum coefficient and the minimum coefficient of the diaphragm stretching gradual change rule coefficient in all the angle directions of the window area; and determining a minimum index of diaphragm dark indexes for all window areas;
if the diaphragm darkness index of the current window area is not equal to the minimum index, determining that a hole area or diaphragm stretching condition exists in the current window area, and calculating a hole convergence point coefficient of a central pixel point of the current window area based on the minimum index, the diaphragm darkness index of the current window area, the maximum coefficient and the minimum coefficient;
if the diaphragm darkness index of the current window area is equal to the minimum index, determining that the current window area has no hole area or diaphragm stretching condition, and calculating the hole gathering point coefficient of the central pixel point of the current window area based on the diaphragm darkness index of the current window area.
In an alternative embodiment, if the diaphragm dark index of the current window area is not equal to the minimum index, the hole convergence point coefficient of the central pixel point of the current window area is calculated by:
wherein,a hole convergence point coefficient representing a center pixel of the current window region, +.>The maximum coefficient is represented by a value of,representing the minimum coefficient, +.>Diaphragm dark index representing current window area, +.>Representing the minimum index.
In an alternative embodiment, the clustering module is configured to:
calculating the difference value of the hole point coefficient of the current pixel point and the hole point coefficient of the clustering center pixel points of the first class cluster and the second class cluster to obtain the difference value of the hole point coefficient;
calculating Euclidean distance between the current pixel point and the clustering center pixel points of the first type cluster and the second type cluster;
calculating the product between the broken hole point coefficient difference and the Euclidean distance obtained by calculation to obtain the clustering distance between the current pixel point and the first class cluster and the second class cluster;
and clustering the current pixel points into the first type cluster or the second type cluster based on the clustering distance between the current pixel points and the first type cluster and the second type cluster.
In an alternative embodiment, if the clustering distance between the current pixel point and the first cluster is smaller than the clustering distance between the current pixel point and the second cluster, the current pixel point is clustered to the first cluster.
The application has the beneficial effects that the lithium battery diaphragm quality detection system is different from the prior art, and comprises: the characteristic analysis module is used for calculating a diaphragm stretching gradual change rule coefficient of a window area where each pixel point in the image to be detected is located and a diaphragm darkness index of the window area; the diaphragm stretching gradual change rule coefficient represents whether a diaphragm stretching condition occurs in the window area, and the diaphragm darkness index represents whether the window area is a broken hole flaw area; the characteristic calculation module is used for calculating a hole point gathering coefficient of each pixel point based on a diaphragm stretching gradual change rule coefficient of a window area where each pixel point is located in an image to be detected and a diaphragm darkness index of the window area; and the clustering module clusters the pixel points to a first type cluster or a second type cluster based on the hole point coefficient of each pixel point, wherein the first type cluster represents a hole area, and the second type cluster represents a normal area. According to the scheme, whether the window area has stretching defects or not can be detected by calculating the diaphragm stretching gradual change rule coefficient of the window area, and whether the window area has hole defects or not can be detected by calculating the diaphragm darkness index of the window area, so that the accuracy of quality detection is improved.
Drawings
FIG. 1 is a schematic diagram of a system for detecting the quality of a lithium battery separator according to an embodiment of the present application;
fig. 2 is a schematic view of a pixel point in the vertical axis direction of the pixel point in an angle direction according to the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The traditional method for detecting the quality of the lithium battery diaphragm can only identify the flaws in the hole area and cannot identify the flaws in the stretching area, so that the quality detection of the diaphragm is inaccurate, and the diaphragm with low stretching quality flows into the market. According to the lithium battery diaphragm quality detection system, whether the window area has stretching defects can be detected by calculating the diaphragm stretching gradual change rule coefficient of the window area, and whether the window area has hole defects can be detected by calculating the diaphragm darkness index of the window area, so that the accuracy of quality detection is improved. The present application will be described in detail with reference to the accompanying drawings and examples.
Referring to fig. 1, fig. 1 is a schematic structural diagram of an embodiment of a lithium battery separator quality detection system according to the present application, which specifically includes: a feature analysis module 11, a feature calculation module 12 and a clustering module 13.
The characteristic analysis module 11 is used for calculating a diaphragm stretching gradual change rule coefficient of a window area where each pixel point in the image to be detected is located and a diaphragm darkness index of the window area; the diaphragm stretching gradual change rule coefficient represents whether the window area has diaphragm stretching conditions, and the diaphragm darkness index represents whether the window area is a hole flaw area.
In a specific embodiment, the feature analysis module 11 is configured to obtain an image to be detected. Specifically, the CCD camera is erected above a production line, the back surface of the lithium battery diaphragm is polished by using the white LED light source, good color reduction degree can be provided, the diaphragm defect is more clearly displayed, the CCD camera is used for acquiring color RGB images of the lithium battery diaphragm, and the RGB images are converted into gray images. Since noise occurs in images taken on industrial production lines, median filtering techniques are used to denoise the images in order to eliminate these noise, enhancing the accuracy of subsequent analysis. And taking the denoised gray image as an image to be detected.
In the production process of lithium battery separators, the surface of the separator is generally smooth and flat, and has a certain stretchability. If a lithium battery separator breaks during production, the separator appears as a black, irregular circle in the image. The hole is darker than the white diaphragm background, if a point is in the hole area, the gray values of the pixels surrounding the pointSmaller. Meanwhile, if there are hole boundaries around the point, these boundaries are generally not smooth. The application calculates the diaphragm darkness index of the characterization hole flaw area through the characteristic analysis module 11. In particular. The feature analysis module 11 calculates a diaphragm dark index of a window area where each pixel point in the image to be detected is located, wherein the diaphragm dark index represents whether the window area is a hole flaw area or not.
In a specific embodiment, the feature analysis module 11 includes a first feature analysis module 111, where the first feature analysis module 111 is configured to perform edge detection on the image to be detected, so as to obtain edge binary images of all areas in the image to be detected. Specifically, the first feature analysis module 111 obtains edge binary images of all areas in the to-be-detected image by adopting Canny operator on the gray level image of the lithium battery diaphragm, i.e. the to-be-detected image. And constructing a window area with a preset size, such as W multiplied by W, by taking each pixel point as a center, and analyzing the distribution condition of gray values in the window area of W multiplied by W so as to determine the diaphragm darkness index of the window area. Wherein the value of W can be set by the practitioner, in this embodiment, the value of W is 7.
The first feature analysis module 111 determines a diaphragm dark index for the window region based on gray values of pixels of the window region and edge lines in the edge binary image. In an embodiment, the first feature analysis module 111 determines the number V of segmented regions in the window region, where the segmented regions are formed by edges of the window region and edge lines in the edge binary image. If the number V of the divided regions is 1, i.e., v=1, the gray average value of the pixel points in the window region is used as the diaphragm dark index a of the window region, i.e.，/>And representing the gray average value of the pixel points in the window area. It can be understood that the gray average value of the pixel points in the window area represents the brightness degree of the color of the window area, and the diaphragm darkness index +.>The smaller the window area, the darker the color, the more likely the window area is a hole defect area.
If the number V of the divided areas is larger than 1, namely V is larger than 1, calculating the gray average value corresponding to all the divided areas, and enabling the gray average value of the divided areas to be the smallestAnd a maximum gray value +.>And minimum gray valueAs the membrane dark channel index A of said window area, i.e. +.>. Minimum gray-scale mean value of the divided regions->Smaller indicates that the color is darker in the window area, i.e., has a hole flaw color characteristic. By calculating the difference between the maximum gray value and the minimum gray value in the window area>Obtaining the degree index of the window area with possible hole area, if the difference is larger, the window area is located at the position close to the boundary of the hole, namely the defect area with the hole in one window area and the normal diaphragm surface, the diaphragm darkness index->The smaller the darkest segmented region within the window region, the more likely the membrane hole defect region.
When a hole appears in the process of producing the lithium battery diaphragm, the state around the hole can be changed correspondingly. The quality of the diaphragm around the hole is also deteriorated due to the stretching effect, and the thickness of the diaphragm around the hole is reduced relative to the diaphragm with better quality, and the effect in the image is that the gray value is gradually decreased as the diaphragm approaches the hole, which is also required as a feature for judging the quality of the diaphragm. For those separators with better quality and stronger stretchability, it is not easy for the surface of the separator around such a hole to be stretched over a large area. If such a regular decrease in surrounding gray values occurs in the image, then it is also necessary to cluster these regions into the hole regions to help make further decisions on detecting diaphragm quality next. Based on this, in the present application, the feature analysis module 11 is further configured to calculate a rule coefficient of stretching and gradual change of the diaphragm in the window area where each pixel point is located in the image to be detected, where the rule coefficient of stretching and gradual change of the diaphragm characterizes whether the window area has stretching of the diaphragm.
In a specific embodiment of the present application, the feature analysis module 11 further includes: a second feature analysis module 112. The second feature analysis module 112 is configured to calculate a gray scale rule change coefficient of a projection value of each pixel point in each angle direction based on the gray scale value of the pixel point in each angle direction of the window area by using a pull transformation algorithm, so as to obtain a projection value sequence in each angle direction. Specifically, the second feature analysis module 112 calculates, according to the distribution of gray values in each pixel window area in the image to be detected, a Radon projection rotation window for each window area, calculates the average value of the sum of gray value differences between two adjacent pixels in the longitudinal axis direction of each pixel in the window area every 5 degrees, sums the gray value differences calculated by all the pixels in the longitudinal axis direction, averages the sum of gray value differences calculated by all the pixels in the longitudinal axis direction, and obtains the gray rule change coefficient of the projection value of each pixel in the angle direction, where the gray rule change coefficient of the projection value of all the pixels in the angle direction constitutes the projection value sequence in the angle direction. Specifically, referring to fig. 2, the second feature analysis module 112 is further configured to: and calculating the gray value difference value between two adjacent pixel points in the longitudinal axis direction y of each pixel point in the angle direction alpha, summing the gray value difference values calculated by all the pixel points in the longitudinal axis direction, and averaging to obtain the projection value of the abscissa corresponding to each pixel point in the angle direction. For example, the gray-scale difference between every two adjacent pixels, such as the pixel (i, j) and the pixel (i, j+1), in all pixels in the same column as the pixel (i, j) in fig. 2 is calculated, and the calculated gray-scale differences are summed and averaged to obtain the projection of the abscissa i in the angle direction αThe gray scale law change coefficient of the value.
The second feature analysis module 112 calculates a mean value of the gray scale rule change coefficients in the angle direction based on all the gray scale rule change coefficients in the projection value sequence in the angle direction; based on the difference between the average value of the gray scale rule change coefficients in the angle direction and the gray scale rule change coefficient of the projection value of each pixel point, the diaphragm stretching and gradual change rule coefficients in the angle direction are obtained, and the diaphragm stretching and gradual change rule coefficients in all the angle directions are used as the diaphragm stretching and gradual change rule coefficients of the window area.
Specifically, the projection value sequenceFor example, calculate +.>Dividing the sum by n to obtain the average value of the gray rule change coefficients in the angle direction ++>. Based on the difference between the average value of the gray scale rule change coefficients in the angle direction and the gray scale rule change coefficient of the projection value of each pixel point, the diaphragm stretching and gradual change rule coefficients in the angle direction are obtained, and the diaphragm stretching and gradual change rule coefficients in all the angle directions are used as the diaphragm stretching and gradual change rule coefficients of the window area. Specifically, the second feature analysis module 112 is further configured to: calculating the diaphragm stretching gradual change rule coefficient of each angle direction by using the following formula:
wherein,each projection value in the projection value sequence represents the gray rule change coefficient in the vertical axis direction of the corresponding pixel point in the angle direction and is used for representing the vertical axis where the abscissa is locatedGradation degree between pixel points in the axis direction. The gray scale rule change coefficient is obtained by calculating the absolute value of the difference value of gray scale values between adjacent pixel points in m pixel points in the longitudinal axis direction of the corresponding pixel point, and then averaging. />A membrane stretch gradient law coefficient representing angular direction, +.>Represents the number of pixels in the angular direction, < +.>The number of pixels in the vertical axis direction of the pixel in the angular direction is represented, and taking the pixel (i, j) in fig. 2 as an example, m represents the number of pixels in the same column as the pixel (i, j). />Representing the gray value of a pixel point (i, j) of which the coordinate point (i) is adjacent to the pixel point (i) of which the abscissa is in the vertical axis direction in the angle direction,/>Representing the gray value of a pixel point (i, j+1) with a coordinate point adjacent to a pixel point (i) with an abscissa in the vertical axis direction in the angular direction, +.>The average value of the change coefficient of the gray scale rule in the angle direction is represented. The membrane stretching gradual change rule coefficient of window area +.>The smaller the window area, the more pronounced the regularity of the tensile gradation of the membrane occurs. Calculate->The difference between the gray scale law change coefficient of each projection point, namely +.>And obtaining a diaphragm stretching gradual change rule coefficient C in the angle direction, wherein the coefficient C is used for representing whether the stretching degree of each projection point is consistent.
The feature calculation module 12 of the present application is configured to calculate a hole point gathering coefficient of each pixel based on a diaphragm stretching gradient rule coefficient of a window area where each pixel is located in an image to be detected and a diaphragm darkness index of the window area. Specifically, the membrane stretching gradual change rule coefficient of the window areaThe smaller the window area, the more pronounced the regularity of the tensile gradation of the membrane occurs. Thereby obtaining the minimum diaphragm stretching gradual change rule coefficient in one projection angle direction in the window areaI.e. the hole area in the window area if the darkest area exists in the angular direction. Some window areas are uniform in color, no gradual change rule appears, and only the coefficient of the diaphragm stretching gradual change rule of each window area is calculated, so that whether the window area is actually the area with the gradual change rule of gray scale caused by stretching generated by diaphragm broken holes cannot be accurately judged. According to the characteristic of consistent color in the window area, the maximum diaphragm stretching gradual change rule coefficient is +.>Should be in accordance with the minimum membrane stretch gradient law coefficient +.>The phase difference is not great. Thus, the feature calculation module 12 is configured to: determining the maximum coefficient of the membrane stretching gradual change rule coefficient in all angle directions of the window area +.>And minimum coefficient->. And the feature calculation module 12 is also used to determine diaphragm darkness for all window areasMinimum index>。
If the diaphragm dark index A of the current window area is not equal to the minimum indexI.e. +.>Determining that a hole area or a diaphragm stretching condition exists in the current window area based on the minimum index +.>A diaphragm dark index A of the current window area, the maximum coefficient +.>And said minimum coefficient->Calculating a hole point coefficient of a central pixel point of the current window area +.>Specifically, the->. Wherein (1)>A hole convergence point coefficient representing a center pixel of the current window region, +.>Representing the maximum coefficient>Representing the minimum coefficient, +.>Diaphragm dark index representing current window area, +.>Representing the minimum index.
If the diaphragm dark index A of the current window area is equal to the minimum indexI.e. +.>Determining that no hole area or diaphragm stretching condition exists in the current window area, and calculating a hole point gathering coefficient of a central pixel point of the current window area based on a diaphragm darkness index A of the current window area>Specifically, the->。
When the following is performedAt the time, a stretching diaphragm area with gradual change of gray values in and around the hole exists in the window area, and the stretching diaphragm area is marked according to the rule coefficient of the minimum diaphragm stretching gradual change>Calculated->Molecule->The method is to emphasize that the window area is a gray stretching gradual change rule degree instead of a uniform color area, and the larger the value is, the more the window area is a stretching area around the hole; molecule->The gray level decreasing rule degree used for representing the window area is that the smaller the value is, the more rule of the gradient rule is, i.e. the window area is more likely to be stretched around the broken holeA region; />Is provided with the minimum membrane stretching gradual change rule coefficient +.>Membrane dark index of window region +.>The smaller the value, the darker the window area, and the darker the color of the stretched area around the hole. According to->Obtaining the dark color characteristic of the hole area in the window area, namely the +.>，/>The smaller the color of the hole area within the window area, the darker the color, i.e., the greater the degree to which the point is a hole convergence point. When->When the window area is not provided with a membrane area with gradually-changed stretching around the hole, the +.>It is necessary to determine whether the darkest window area is the only hole area in the window area based on the magnitude of this value,/o>The smaller, i.e.)>The larger the darkness of the region, the greater the degree of color that the darkness of the region can represent the hole dark region, indicating that the center point should be clustered as a hole region point.
The hole clustering coefficient of each pixel point in the image to be detected is obtained through the calculation, and in the application, the clustering module 13 is further utilized to cluster the pixel points to a first type cluster or a second type cluster based on the hole clustering coefficient of each pixel point, wherein the first type cluster represents a hole area, and the second type cluster represents a normal area.
Because the surface of the lithium battery diaphragm is white, the hole flaws on the surface of the diaphragm are expected to be identified under the background, a K-means clustering algorithm can be used for gathering the images of the lithium battery diaphragm into two types, wherein one type is the hole flaws on the surface of the diaphragm, and the other type is the diaphragm background, so that the hole flaws on the surface of the lithium battery diaphragm are identified, and the quality of the lithium battery diaphragm is conveniently detected.
The K-means clustering algorithm is calculated based on Euclidean distance between the distance center points, but hole clustering obtained by means of distance calculation is not accurate, so that characteristics are needed to be built on hole flaws to optimize the clustering distance. Calculating the diaphragm image of the lithium batteryThe clustering effect is that the group is a hole cluster and the group is a normal cluster. Hole flaw feature indices will be constructed to optimize cluster distance.
And mapping the gray level image onto a gray level histogram, taking any one pixel point in gray levels of two maximum peaks as a center point of two types of clustering, and calculating the correlation between the pixel points and the two center points according to the rest points so as to finish the clustering process.
The clustering module 13 is used for: and calculating the difference value of the hole point coefficient of the current pixel point and the hole point coefficient of the clustering center pixel points of the first class cluster and the second class cluster to obtain the difference value of the hole point coefficient. And calculating the Euclidean distance between the current pixel point and the clustering center pixel points of the first type cluster and the second type cluster. And calculating the product of the difference of the broken hole point coefficient and the Euclidean distance to obtain the clustering distance between the current pixel point and the first class cluster and the second class cluster. And clustering the current pixel points into the first type cluster or the second type cluster based on the clustering distance between the current pixel points and the first type cluster and the second type cluster.
In an embodiment, the clustering distance between the current pixel point and the first class cluster and the second class cluster is calculated by:
wherein,represents->The position coordinates of the pixel points, c, are 1 and 2, and respectively represent the clustering center pixel points of the first type of clusters, namely the hole clusters, and the second type of clusters, namely the normal clusters,/the clustering center pixel points of the first type of clusters, namely the hole clusters>Representing the position coordinates of the pixel points of the cluster center, < + >>Represents->Hole convergence point coefficient of each pixel, < ->Representing the hole point coefficient of the pixel point in the center of the cluster. />For the +.>Euclidean distance between each pixel point and clustering center pixel point, < ->For the +.>The difference of the hole clustering point coefficients between each pixel point and the clustering center pixel point is calculated by the formula to obtain the clustering between the current pixel point and the first class cluster and the second class clusterDistance->. Comparing cluster distance->The smaller cluster of the first and second clusters is the first +.>The addition of a pixel to the cluster indicates the +.>The distance between each pixel point and the cluster and the hole feature have similarity.
In the application, a hole cluster area in the lithium battery diaphragm is obtained by clustering images, and the ratio of the area of the hole cluster in the images to the total area of the lithium battery diaphragm is larger than a threshold valueAnd when the quality of the lithium battery diaphragm is judged to be poor, a worker is prompted to further detect or repair. />The value of (2) is set by the value implementer, the application sets the threshold value +.>Set to 0.005.
The traditional image processing can not completely identify the stretched area around the lithium battery diaphragm hole, so that the accuracy of the quality detection of the lithium battery diaphragm is lower. According to the application, by adopting a K-means clustering algorithm, according to the characteristics of the hole flaws on the surface of the lithium battery diaphragm and the characteristics of gray level gradual change of the stretched diaphragm around the hole, the hole flaw index is constructed, so that the flaws in the hole area can be detected, the flaws in the stretched area can be detected, the clustering distance in the clustering process is optimized, and the accuracy of the algorithm is improved.
The foregoing is only the embodiments of the present application, and therefore, the patent scope of the application is not limited thereto, and all equivalent structures or equivalent processes using the descriptions of the present application and the accompanying drawings, or direct or indirect application in other related technical fields, are included in the scope of the application.
Claims (3)
1. A lithium battery separator quality detection system, comprising:
the characteristic analysis module is used for calculating a diaphragm stretching gradual change rule coefficient of a window area where each pixel point in the image to be detected is located and a diaphragm darkness index of the window area; the diaphragm stretching gradual change rule coefficient represents whether a diaphragm stretching condition occurs in the window area, and the diaphragm darkness index represents whether the window area is a broken hole flaw area;
the characteristic calculation module is used for calculating a broken hole point gathering coefficient of each pixel point based on a diaphragm stretching gradual change rule coefficient of a window area where each pixel point is located in an image to be detected and a diaphragm darkness index of the window area;
the clustering module is used for clustering the pixel points to a first type cluster or a second type cluster based on the hole point coefficient of each pixel point, wherein the first type cluster represents a hole area, and the second type cluster represents a normal area;
the feature analysis module comprises: a first feature analysis module;
the first feature analysis module is used for carrying out edge detection on the image to be detected to obtain edge binary images of all areas in the image to be detected; constructing a window area with a preset size by taking each pixel point as a center, and determining a diaphragm dark index of the window area based on the gray value of the pixel point of the window area and an edge line in the edge binary image;
the first feature analysis module is further to:
determining the number of segmentation areas formed by the edges of the window area and edge lines in the edge binary image in the window area;
if the number of the divided areas is 1, taking the gray average value of the pixel points in the window area as the diaphragm dark index of the window area;
if the number of the divided areas is greater than 1, calculating gray average values corresponding to all the divided areas, and taking the ratio of the gray average value of the minimum of the divided areas to the difference value of the maximum gray value and the minimum gray value in the window area as the diaphragm dark index of the window area;
the feature analysis module comprises: a second feature analysis module;
the second feature analysis module is used for calculating a gray rule change coefficient of a projection value of each pixel point in each angle direction by adopting a pulling transformation algorithm based on the gray value of the pixel point in each angle direction of the window area, so as to obtain a projection value sequence in each angle direction; determining the diaphragm stretching gradual change rule coefficient of the window area based on a projection value sequence in each angle direction and gray values of pixel points in the angle direction;
the second feature analysis module is further to:
calculating the gray value difference value between two adjacent pixel points in the longitudinal axis direction of each pixel point in the angle direction, summing the gray value difference values calculated by all pixel points in the longitudinal axis direction, and averaging to obtain the gray rule change coefficient of the projection value of each pixel point in the angle direction, wherein the gray rule change coefficient of the projection value of all pixel points in the angle direction forms a projection value sequence in the angle direction;
calculating a gray scale rule change coefficient mean value in the angle direction based on all gray scale rule change coefficients in the projection value sequence in the angle direction;
based on the difference between the average value of the gray scale rule change coefficients in the angle direction and the gray scale rule change coefficient of the projection value of each pixel point, further obtaining the diaphragm stretching and gradual change rule coefficient in the angle direction, and taking the diaphragm stretching and gradual change rule coefficients in all the angle directions as the diaphragm stretching and gradual change rule coefficient of the window area;
the second feature analysis module is further configured to:
calculating the diaphragm stretching gradual change rule coefficient of each angle direction by using the following formula:
wherein,a membrane stretch gradient law coefficient representing angular direction, +.>Represents the number of pixels in the angular direction, < +.>Representing the number of pixels in the vertical axis direction of the pixels in the angular direction, +.>Representing the gray value of a pixel point (i, j) of which the coordinate point (i) is adjacent to the pixel point (i) of which the abscissa is in the vertical axis direction in the angle direction,/>Representing the gray value of a pixel point (i, j+1) with a coordinate point adjacent to a pixel point (i) with an abscissa in the vertical axis direction in the angular direction, +.>The average value of the change coefficient of the gray rule in the angle direction is represented;
the feature calculation module is used for:
determining the maximum coefficient and the minimum coefficient of the diaphragm stretching gradual change rule coefficient in all the angle directions of the window area; and determining a minimum index of diaphragm dark indexes for all window areas;
if the diaphragm darkness index of the current window area is not equal to the minimum index, determining that a hole area or diaphragm stretching condition exists in the current window area, and calculating a hole convergence point coefficient of a central pixel point of the current window area based on the minimum index, the diaphragm darkness index of the current window area, the maximum coefficient and the minimum coefficient;
if the diaphragm darkness index of the current window area is equal to the minimum index, determining that a hole area or diaphragm stretching condition does not exist in the current window area, and calculating a hole gathering point coefficient of a central pixel point of the current window area based on the diaphragm darkness index of the current window area;
if the diaphragm dark index of the current window area is not equal to the minimum index, the calculation mode of the hole gathering point coefficient of the central pixel point of the current window area is as follows:
wherein,a hole convergence point coefficient representing a center pixel of the current window region, +.>Representing the maximum coefficient>Representing the minimum coefficient, +.>Diaphragm dark index representing current window area, +.>Representing the minimum index.
2. The lithium battery separator quality detection system of claim 1, wherein the clustering module is configured to:
calculating the difference value of the hole point coefficient of the current pixel point and the hole point coefficient of the clustering center pixel points of the first class cluster and the second class cluster to obtain the difference value of the hole point coefficient;
calculating Euclidean distance between the current pixel point and the clustering center pixel points of the first type cluster and the second type cluster;
calculating the product between the broken hole point coefficient difference and the Euclidean distance obtained by calculation to obtain the clustering distance between the current pixel point and the first class cluster and the second class cluster;
and clustering the current pixel points into the first type cluster or the second type cluster based on the clustering distance between the current pixel points and the first type cluster and the second type cluster.
3. The lithium battery separator quality detection system according to claim 2, wherein the current pixel point is clustered to a first cluster if the clustering distance of the current pixel point to the first cluster is smaller than the clustering distance of the current pixel point to a second cluster.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310960010.9A CN116703894B (en) | 2023-08-02 | 2023-08-02 | Lithium battery diaphragm quality detection system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310960010.9A CN116703894B (en) | 2023-08-02 | 2023-08-02 | Lithium battery diaphragm quality detection system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116703894A CN116703894A (en) | 2023-09-05 |
| CN116703894B true CN116703894B (en) | 2023-11-17 |
Family
ID=87826087
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310960010.9A Active CN116703894B (en) | 2023-08-02 | 2023-08-02 | Lithium battery diaphragm quality detection system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116703894B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117333489B (en) * | 2023-12-01 | 2024-02-02 | 苏州普洛泰科精密工业有限公司 | Film damage detection device and detection system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103698334A (en) * | 2013-10-25 | 2014-04-02 | 明基材料有限公司 | Pin hole flaw detection system for isolated film and detection method thereof |
| WO2022052480A1 (en) * | 2020-09-11 | 2022-03-17 | 广东奥普特科技股份有限公司 | Method and system for detecting and processing defect in lithium battery electrode plate in real time |
| CN115588022A (en) * | 2022-11-10 | 2023-01-10 | 合肥惠强新能源材料科技有限公司 | Lithium battery isolation film quality detection system based on process index data |
| CN115641329A (en) * | 2022-11-15 | 2023-01-24 | 武汉惠强新能源材料科技有限公司 | Lithium battery diaphragm defect detection method and system |
| CN115937204A (en) * | 2023-01-09 | 2023-04-07 | 江苏惠汕新能源集团有限公司 | Welded pipe production quality detection method |
| WO2023124585A1 (en) * | 2021-12-31 | 2023-07-06 | 广东利元亨智能装备股份有限公司 | Detection method and system based on battery defect detection system, and storage medium |
-
2023
- 2023-08-02 CN CN202310960010.9A patent/CN116703894B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103698334A (en) * | 2013-10-25 | 2014-04-02 | 明基材料有限公司 | Pin hole flaw detection system for isolated film and detection method thereof |
| WO2022052480A1 (en) * | 2020-09-11 | 2022-03-17 | 广东奥普特科技股份有限公司 | Method and system for detecting and processing defect in lithium battery electrode plate in real time |
| WO2023124585A1 (en) * | 2021-12-31 | 2023-07-06 | 广东利元亨智能装备股份有限公司 | Detection method and system based on battery defect detection system, and storage medium |
| CN115588022A (en) * | 2022-11-10 | 2023-01-10 | 合肥惠强新能源材料科技有限公司 | Lithium battery isolation film quality detection system based on process index data |
| CN115641329A (en) * | 2022-11-15 | 2023-01-24 | 武汉惠强新能源材料科技有限公司 | Lithium battery diaphragm defect detection method and system |
| CN115937204A (en) * | 2023-01-09 | 2023-04-07 | 江苏惠汕新能源集团有限公司 | Welded pipe production quality detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116703894A (en) | 2023-09-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109035195B (en) | Fabric defect detection method | |
| CN115841493B (en) | Cable detection method based on image processing | |
| CN116703894B (en) | Lithium battery diaphragm quality detection system | |
| CN115249246B (en) | Optical glass surface defect detection method | |
| CN113313677B (en) | Quality detection method for X-ray image of wound lithium battery | |
| CN105158258A (en) | Computer vision-based bamboo strip surface defect detection method | |
| CN115619793B (en) | Power adapter appearance quality detection method based on computer vision | |
| CN116721107B (en) | Intelligent monitoring system for cable production quality | |
| CN115294116B (en) | Method, device and system for evaluating dyeing quality of textile material based on artificial intelligence | |
| CN114494256A (en) | Electric wire production defect detection method based on image processing | |
| CN116188468B (en) | HDMI cable transmission letter sorting intelligent control system | |
| CN110648330B (en) | Defect detection method for camera glass | |
| CN116883408B (en) | Integrating instrument shell defect detection method based on artificial intelligence | |
| CN116245880A (en) | Electric vehicle charging pile fire risk detection method based on infrared identification | |
| CN116703890A (en) | Method and system for detecting tab defects | |
| CN116152242A (en) | Visual detection system of natural leather defect for basketball | |
| CN112435235A (en) | Seed cotton impurity content detection method based on image analysis | |
| CN117252872B (en) | Mobile phone screen defect visual detection method and system | |
| CN116630332B (en) | PVC plastic pipe orifice defect detection method based on image processing | |
| CN116543238B (en) | Image detection method for cable insulating layer | |
| CN108965749A (en) | Defect pixel detection and means for correcting and method based on texture recognition | |
| US20200408697A1 (en) | Belt inspection system, belt inspection method, and recording medium for belt inspection program | |
| CN116229438B (en) | Spinning quality visual identification system | |
| CN110223253B (en) | Defogging method based on image enhancement | |
| CN114998346B (en) | Waterproof cloth quality data processing and identifying 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 |