CN114998187A - Glass fiber separator defect detection method and system - Google Patents
Glass fiber separator defect detection method and system Download PDFInfo
- Publication number
- CN114998187A CN114998187A CN202210345162.3A CN202210345162A CN114998187A CN 114998187 A CN114998187 A CN 114998187A CN 202210345162 A CN202210345162 A CN 202210345162A CN 114998187 A CN114998187 A CN 114998187A
- Authority
- CN
- China
- Prior art keywords
- gradient
- glass fiber
- obtaining
- fiber separator
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000003365 glass fiber Substances 0.000 title claims abstract description 116
- 230000007547 defect Effects 0.000 title claims abstract description 63
- 238000001514 detection method Methods 0.000 title claims description 13
- 238000005192 partition Methods 0.000 claims abstract description 41
- 230000002159 abnormal effect Effects 0.000 claims abstract description 30
- 238000000034 method Methods 0.000 claims abstract description 29
- 238000012545 processing Methods 0.000 claims abstract description 10
- 230000032798 delamination Effects 0.000 claims description 17
- 238000005520 cutting process Methods 0.000 claims description 15
- 230000002950 deficient Effects 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 6
- 238000004891 communication Methods 0.000 claims description 2
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000012216 screening Methods 0.000 abstract description 2
- 239000011152 fibreglass Substances 0.000 description 5
- 239000010410 layer Substances 0.000 description 4
- 230000011218 segmentation Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 239000002994 raw material Substances 0.000 description 3
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 238000013517 stratification Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 239000002253 acid Substances 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910002092 carbon dioxide Inorganic materials 0.000 description 1
- 239000001569 carbon dioxide Substances 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000002356 single layer Substances 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P70/00—Climate change mitigation technologies in the production process for final industrial or consumer products
- Y02P70/50—Manufacturing or production processes characterised by the final manufactured product
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to the technical field of artificial intelligence, in particular to a method and a system for detecting defects of a glass fiber partition plate. The method comprises the steps of obtaining defect probability through the difference of pixel information of a surface image of a glass fiber partition plate and a standard average pixel value of the surface image of a standard glass fiber partition plate, and screening an abnormal partition plate image according to the defect probability. And obtaining gradient information of the abnormal partition plate, and determining whether the defect type is completely layered or partially layered according to the magnitude of the gradient amplitude. If the partial layering is performed, classifying according to the gradient amplitude to obtain a plurality of gradient categories, obtaining a gradient edge according to coordinate information of pixel points in the gradient categories, and obtaining a layering range according to the gradient edge and the gradient direction. The invention can accurately identify the layering defects according to the pixel characteristics and the gradient characteristics of the image, and can realize targeted subsequent processing on the product according to the layering range.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a method and a system for detecting defects of a glass fiber partition plate.
Background
AGM is a superfine glass fiber separator, is an important part in the production of storage batteries, the quality of the AGM directly influences the discharge capacity and the charge-discharge cycle service life of the storage batteries, and a multilayer composite separator (mostly in a double-layer or three-layer structure) has stronger capillary action and better liquid absorption capacity than a single-layer separator and can eliminate the layering phenomenon of acid liquor. And the battery assembled with the multi-layer separator has better large-current discharge performance under low-temperature conditions.
When the raw materials and the pressing temperature are abnormal, the surface of the multilayer partition plate can be layered, and the product quality is seriously influenced. However, due to the influence of illumination and the fact that the layering phenomenon is located inside the glass fiber partition plate, the gradient difference between the scoring layer area and the normal area is small, and the detection effect is poor when the traditional threshold segmentation technology or the edge detection technology is used for detection.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a method and a system for detecting defects of a glass fiber separator, and the adopted technical scheme is as follows:
the invention provides a method for detecting defects of a glass fiber separator, which comprises the following steps:
obtaining continuous multiframe glass fiber separator surface images; obtaining the pixel value category in the surface image of the glass fiber separator; obtaining a standard average pixel value of a standard glass fiber separator surface image; obtaining a defect probability according to the difference between the pixel value category and the standard average pixel value; if the defect probability is larger than a preset probability threshold value, the corresponding surface image of the glass fiber separator is an abnormal separator image;
obtaining gradient information at each pixel point position of the abnormal partition plate image; the gradient information comprises a gradient amplitude and a gradient direction; if the gradient amplitudes are all zero, the defect type of the abnormal partition plate image is considered to be complete delamination; otherwise, the defect type of the abnormal partition plate image is considered to be partial layering;
obtaining a plurality of gradient categories according to the gradient amplitude classification; obtaining a plurality of gradient edges according to the coordinate information of the pixel points in the gradient category; and determining a layering range according to the gradient edge and the gradient direction.
Further, the obtaining of the continuous multiframe glass fiber separator surface images comprises:
collecting a scene image of a glass fiber partition plate in a cutting processing scene; removing background information in the scene image to obtain an initial glass fiber clapboard surface image;
obtaining a maximum connected domain in the initial glass fiber separator surface image; and taking the image in the maximum communication domain as the surface image of the glass fiber separator.
Further, the obtaining of the pixel value category in the image of the surface of the glass fiber separator comprises:
obtaining a first normalized gray level histogram of the surface image of the glass fiber separator; the vertical axis of the first normalized gray level histogram is gray level proportion, and the horizontal axis of the first normalized gray level histogram is gray level; taking a plurality of gray levels with a gray level ratio different from zero in the first normalized gray histogram as the pixel value category;
obtaining a standard average pixel value for a standard glass fiber separator surface image includes:
obtaining a second normalized gray level histogram of the surface image of the standard glass fiber separator; and obtaining the standard average pixel value according to the gray level ratio and the gray level size in the second normalized gray level histogram.
Further, the obtaining the defect probability according to the difference between the pixel value category and the standard average pixel value comprises:
obtaining the defect probability according to a defect probability formula, wherein the defect probability formula comprises:
wherein P is the defect probability,
is the average pixel value of the image of the surface of the glass fiber separator,
is the standard average pixel value, H i Is the ith said pixel value class, and N is the number of said pixel value classes.
Further, the obtaining a plurality of gradient classes according to the gradient magnitude classification includes:
and classifying by using a mean shift clustering algorithm according to the gradient amplitude to obtain a plurality of gradient categories.
Further, the obtaining a plurality of gradient edges according to the coordinate information of the pixel points in the gradient category includes:
classifying by adopting a density clustering algorithm according to the coordinate information of the pixel points in the gradient category to obtain a plurality of clustering clusters; and calculating the inter-class variance of each cluster, removing the cluster corresponding to the largest inter-class variance, and taking each remaining cluster as one gradient edge.
Further, the determining a stratification range according to the gradient edge and the gradient direction comprises:
obtaining a plurality of regions according to the gradient edge, and taking a region pointed by the gradient direction as an initial layering region; and taking the range of the maximum abscissa and the minimum abscissa of the initial layering area as the layering range.
Further, after determining the hierarchical scope, the method further includes:
longitudinally cutting according to the layering range to obtain a defective glass fiber separator and a normal glass fiber separator; and removing the defective glass fiber separator and reserving the normal glass fiber separator.
The invention also provides a glass fiber separator defect detection system, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to realize any one of the steps of the glass fiber separator defect detection method.
The invention has the following beneficial effects:
according to the embodiment of the invention, the pixel characteristics of the acquired surface image of the glass fiber separator are compared with the pixel characteristics of the surface image of the qualified glass fiber separator, so that the defect probability of the current glass fiber separator is determined. And primarily screening the current glass fiber partition plate according to the defect probability to obtain an abnormal partition plate image. And further judging the layering condition of the current glass fiber separator according to the gradient information in the abnormal separator image, if partial layering occurs, further determining the layering range according to the gradient direction, and performing targeted processing on the glass fiber separator according to the layering condition. And the accurate detection of the layering defect of the glass fiber separator is realized.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart of a method for detecting defects in a glass fiber separator according to an embodiment of the present invention;
FIG. 2 is a schematic view of a glass fiber separator processing scenario according to an embodiment of the present invention;
FIG. 3 is a schematic view of a glass fiber separator including delamination defects according to an embodiment of the present invention.
Detailed Description
To further illustrate the technical means and effects of the present invention adopted to achieve the predetermined objects, the following detailed description will be given for the method and system for detecting defects of a glass fiber separator according to the present invention, and the detailed description, structure, features and effects thereof with reference to the accompanying drawings and preferred embodiments. In the following description, the different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following describes a specific scheme of a method and a system for detecting defects of a glass fiber separator provided by the invention in detail with reference to the accompanying drawings.
Referring to fig. 1, a flow chart of a method for detecting defects of a glass fiber separator according to an embodiment of the present invention is shown, the method including:
step S1: obtaining continuous multiframe glass fiber separator surface images; obtaining the pixel value category in the surface image of the glass fiber separator; obtaining a standard average pixel value of a standard glass fiber separator surface image; obtaining the defect probability according to the difference between the pixel value category and the standard average pixel value; and if the defect probability is greater than a preset probability threshold value, the corresponding surface image of the glass fiber separator is an abnormal separator image.
After the glass fiber separator is manufactured, the large glass fiber separator is usually cut into small pieces with the same or different specifications, so that the glass fiber separator is convenient to transport and use. Referring to fig. 2, a schematic diagram of a processing scene of a glass fiber partition board provided by an embodiment of the invention is shown, in which white is the glass fiber partition board, the glass fiber partition board is fixed on a cutting machine by two press rolls, wherein the outermost press roll is provided with a cutting blade, the cutting of the glass fiber partition board is realized by moving the glass fiber partition board, and the innermost press roll is used for fixing and conveying the glass fiber partition board.
In order to completely and clearly obtain image information of the glass fiber partition plate, a gray camera is arranged right above the cutting device, the height of the camera is moderate, the camera view can completely contain the area before the glass fiber partition plate is cut, and the sampling rate of the camera is adjusted to be matched with the moving rate of the glass fiber cover plate, so that the camera can acquire continuous, multiple frames of clear and complete scene images.
The method comprises the steps that a set camera is used for collecting scene images of glass fiber partition plates in a cutting processing scene, and a large amount of background information such as local information of the ground, a press roller or a cutting machine is contained in the scene images. For the convenience of subsequent detection, background information needs to be removed to obtain an image of the surface of the initial glass fiber separator. Two areas of the glass fiber separator information, namely the area to be cut and the area outside the innermost press roll, may be included in the initial glass fiber separator surface image. The glass fiber separator aimed at by the embodiment of the invention is a separator in a region to be cut, so that glass fiber separators in other regions need to be removed. And obtaining the maximum connected domain in the initial glass fiber separator surface image by a connected domain analysis method. The image in the maximum connected domain is taken as the surface image of the glass fiber separator for the process analysis.
In the embodiment of the invention, because the surface of the glass fiber partition plate is white and has a higher gray value compared with other pixel information of the surrounding environment, the scene image can be segmented by using a threshold segmentation method to obtain the initial glass fiber partition plate. The segmentation threshold for the threshold segmentation may use an average pixel value within the scene image.
In the embodiment of the invention, the connected component analysis algorithm adopts a seed filling method.
Referring to fig. 3, a schematic diagram of a glass fiber separator including a delamination defect according to an embodiment of the invention is shown. In fig. 3, the difference between the pixel values of the two regions is not large, and the layered region cannot be accurately divided by using a fixed threshold, but the pixel information of the layered region can still indicate the abnormal condition of the current glass fiber separator by using the pixel information of the normal region as a standard.
A first normalized grayscale histogram of a glass fiber separator surface image is obtained. The vertical axis of the first normalized gray level histogram is gray level ratio, and the horizontal axis is gray level. And taking a plurality of gray levels with the gray level proportion not being zero in the first normalized gray level histogram as pixel value categories. Obtaining a standard average pixel value of a standard glass fiber separator surface image in the same way: and obtaining a second normalized gray level histogram of the surface image of the standard glass fiber separator. And obtaining a standard average pixel value according to the gray level occupation ratio and the gray level size in the second normalized gray level histogram.
The standard glass fiber separator surface image does not have a delamination defect, and therefore the standard average pixel value can be considered as the pixel value of the qualified area of the glass fiber separator. The method for obtaining the defect probability of the current glass fiber separator surface image according to the difference between the pixel value category and the standard average pixel value specifically comprises the following steps:
obtaining the defect probability according to a defect probability formula, wherein the defect probability formula comprises:
wherein, P is the defect probability,
is the average pixel value of the image of the surface of the glass fiber separator,
is a standard average pixel value, H i Is the ith pixel value class and N is the number of pixel value classes.
In the defect probability formula, use
Representing a first degree of offset of the current fiberglass separator surface image from the standard fiberglass separator surface image,
the closer to 1, the closer the two are. It should be noted that if the glass fiber separator is delaminated, the pixel value of the delamination area is larger than that of the normal area, and thus, the delamination area is larger than that of the normal area
The amount of the carbon dioxide gas to be mixed is more than or equal to 1,
the larger the delamination degree, the larger the defect probability.
In the defect probability formula，
And a second degree of deviation of each gray level from the standard average pixel value, wherein the first degree of deviation and the second degree of deviation jointly reflect the overall degree of deviation of the current glass fiber separator surface image and the standard glass fiber separator surface image.
If the defect probability is larger than the preset probability threshold value, the fact that the glass fiber separator has abnormal information is shown, and the corresponding surface image of the glass fiber separator is an abnormal separator image. In the embodiment of the present invention, the probability threshold is set to 0.5.
Step S2: obtaining gradient information at each pixel point position of the abnormal partition plate image; the gradient information comprises gradient amplitude and gradient direction; if the gradient amplitudes are all zero, the defect type of the abnormal partition plate image is considered to be complete delamination; otherwise, the defect type of the abnormal partition image is considered to be partial delamination.
The abnormal partition image shows that the pixel abnormality occurs in the current glass fiber partition, and may include two layering situations, namely complete layering and partial layering. Complete delamination indicates that the current glass fiber separator is totally abnormal and has no normal area. Partial delamination indicates the presence of one or more delamination areas in the current fiberglass separator. Gradient edges exist between layered regions, so that the gradient information can be used for determining the abnormal condition of the current glass fiber separator.
In the embodiment of the invention, after the abnormal partition plate image is subjected to Gaussian blur processing, a sobel operator is used for carrying out gradient detection, and gradient information at the position of each pixel point is obtained. The gradient information includes gradient magnitude and gradient direction.
If the gradient amplitudes are all 0, it is indicated that no edge information exists in the current abnormal partition image, that is, the pixel values in the abnormal partition image are uniformly distributed. Therefore, the defect type corresponding to the abnormal partition image is considered to be completely layered. In order to ensure the product quality, the glass fiber partition plates which are completely layered need to be cut off and removed.
If the gradient amplitude is not 0, the abnormal partition image is indicated to have edge information, and a layered region is formed. I.e. the defect type at this time is partially layered.
Step S3: obtaining a plurality of gradient classes according to the gradient amplitude classification; obtaining a plurality of gradient edges according to the coordinate information of the pixel points in the gradient category; the stratification range is determined from the gradient edges and the gradient direction.
Because the gradient amplitudes on the edges between the same layered regions have certain similarity, a plurality of gradient categories can be obtained by classifying according to the gradient amplitudes, and the method specifically comprises the following steps:
and classifying by using a mean shift clustering algorithm according to the gradient amplitude to obtain a plurality of gradient classes.
The continuous pixel points in each gradient category can form a gradient edge, so that a plurality of gradient edges are obtained according to the coordinate information of the pixel points in the gradient categories, and the method specifically comprises the following steps:
and classifying by adopting a density clustering algorithm according to the coordinate information of the pixel points in the gradient category to obtain a plurality of clustering clusters. In a plurality of clustering clusters, discrete points such as noise and the like can be classified into one class, so that in order to prevent the influence of the noise on detection, the inter-class variance of each clustering cluster is calculated, the clustering cluster corresponding to the largest inter-class variance is removed, and the rest clustering clusters are each a gradient edge.
The gradient edge may enclose one or more regions. Because the layering degree is complex, the layering area may not be in a regular shape, and the cutting equipment cannot perform targeted cutting, the layering range needs to be determined according to the gradient edge and the gradient direction in order to facilitate the production process, so that the subsequent cutting equipment can perform effective cutting according to the layering range, and qualified areas are reserved as much as possible. The obtained layering ranges specifically include:
a plurality of regions are obtained according to the gradient edges, and the region pointed in the gradient direction is used as an initial layered region. The range of the maximum abscissa and the minimum abscissa of the initial delamination area is taken as the delamination range.
And after the layering range of each layering region is determined, the cutting equipment can be controlled to cut longitudinally according to the layering range, and the defective glass fiber separator and the normal glass fiber separator are obtained. The defective fiberglass separator contains most of the hierarchical region information thereon, and the normal fiberglass separator contains only the normal region information. Therefore, the defective glass fiber separator is removed, and only the normal glass fiber separator is reserved, so that the product quality is ensured. Waste recovery or product repair can be performed on the defective glass fiber separator, so that the production efficiency is maximized.
In summary, in the embodiment of the present invention, the defect probability is obtained through the difference between the pixel information of the surface image of the glass fiber separator and the standard average pixel value of the surface image of the standard glass fiber separator, and the abnormal separator image is screened out according to the defect probability. And obtaining gradient information of the abnormal partition plate, and determining whether the defect type is completely layered or partially layered according to the magnitude of the gradient amplitude. If the partial layering is performed, classifying according to the gradient amplitude to obtain a plurality of gradient categories, obtaining a gradient edge according to coordinate information of pixel points in the gradient categories, and obtaining a layering range according to the gradient edge and the gradient direction. According to the embodiment of the invention, accurate layered defect identification is carried out according to the pixel characteristics and the gradient characteristics of the image, and targeted subsequent processing on the product can be realized according to the layered range.
The invention also provides a glass fiber separator defect detection system, which comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein when the processor executes the computer program, any one of the steps of the glass fiber separator defect detection method is realized.
It should be noted that: the precedence order of the above embodiments of the present invention is only for description, and does not represent the merits of the embodiments. The processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (9)
1. A method of detecting defects in a glass fiber separator, the method comprising:
obtaining continuous multiframe glass fiber separator surface images; obtaining the pixel value category in the surface image of the glass fiber separator; obtaining a standard average pixel value of a standard glass fiber separator surface image; obtaining a defect probability according to the difference between the pixel value category and the standard average pixel value; if the defect probability is larger than a preset probability threshold value, the corresponding surface image of the glass fiber separator is an abnormal separator image;
obtaining gradient information at each pixel point position of the abnormal partition plate image; the gradient information comprises a gradient amplitude and a gradient direction; if the gradient amplitudes are all zero, the defect type of the abnormal partition plate image is considered to be complete delamination; otherwise, the defect type of the abnormal partition plate image is considered to be partial layering;
obtaining a plurality of gradient categories according to the gradient amplitude classification; obtaining a plurality of gradient edges according to the coordinate information of the pixel points in the gradient category; and determining a layering range according to the gradient edge and the gradient direction.
2. The method as claimed in claim 1, wherein said obtaining a plurality of consecutive frames of images of the surface of the glass fiber separator comprises:
collecting a scene image of a glass fiber partition plate in a cutting processing scene; removing background information in the scene image to obtain an initial glass fiber clapboard surface image;
obtaining a maximum connected domain in the initial glass fiber separator surface image; and taking the image in the maximum communication domain as the surface image of the glass fiber separator.
3. The method as claimed in claim 1, wherein the obtaining of the pixel value category in the image of the surface of the glass fiber separator comprises:
obtaining a first normalized gray level histogram of the surface image of the glass fiber separator; the vertical axis of the first normalized gray level histogram is gray level proportion, and the horizontal axis of the first normalized gray level histogram is gray level; taking a plurality of gray levels with the gray level proportion not being zero in the first normalized gray level histogram as the pixel value category;
obtaining a standard average pixel value for a standard glass fiber separator surface image includes:
obtaining a second normalized gray level histogram of the surface image of the standard glass fiber separator; and obtaining the standard average pixel value according to the gray level ratio and the gray level size in the second normalized gray level histogram.
4. A method as claimed in claim 3, wherein said obtaining a defect probability based on the difference between the pixel value class and the standard average pixel value comprises:
obtaining the defect probability according to a defect probability formula, wherein the defect probability formula comprises:
wherein P is the defect probability,
is the average pixel value of the image of the surface of the glass fiber separator,
is the standard average pixel value, H i Is the ith said pixel value class, N is said imageNumber of prime value classes.
5. A method as claimed in claim 1, wherein said classifying according to said gradient magnitude to obtain a plurality of gradient classes comprises:
and classifying by using a mean shift clustering algorithm according to the gradient amplitude to obtain a plurality of gradient classes.
6. The method of claim 1, wherein the obtaining a plurality of gradient edges according to the coordinate information of the pixels in the gradient category comprises:
classifying by adopting a density clustering algorithm according to the coordinate information of the pixel points in the gradient category to obtain a plurality of clustering clusters; and calculating the inter-class variance of each cluster, removing the cluster corresponding to the maximum inter-class variance, and taking each residual cluster as one gradient edge.
7. A method as claimed in claim 1, wherein said determining a delamination range based on said gradient edge and said gradient direction comprises:
obtaining a plurality of regions according to the gradient edge, and taking a region pointed by the gradient direction as an initial layering region; and taking the range of the maximum abscissa and the minimum abscissa of the initial layering area as the layering range.
8. The method as claimed in claim 7, wherein the step of determining the delamination range further comprises:
longitudinally cutting according to the layering range to obtain a defective glass fiber separator and a normal glass fiber separator; and removing the defective glass fiber separator and reserving the normal glass fiber separator.
9. A glass fibre separator defect detection system comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor when executing the computer program implements the steps of the method of any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210345162.3A CN114998187A (en) | 2022-03-31 | 2022-03-31 | Glass fiber separator defect detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210345162.3A CN114998187A (en) | 2022-03-31 | 2022-03-31 | Glass fiber separator defect detection method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114998187A true CN114998187A (en) | 2022-09-02 |
Family
ID=83024368
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210345162.3A Pending CN114998187A (en) | 2022-03-31 | 2022-03-31 | Glass fiber separator defect detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114998187A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115311284A (en) * | 2022-10-12 | 2022-11-08 | 南通嘉鹏家居用品有限公司 | Production quality detection method for glass fiber mat |
| CN117218117A (en) * | 2023-11-07 | 2023-12-12 | 常熟市东宇绝缘复合材料有限公司 | Glass fiber yarn detection method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6167150A (en) * | 1998-07-24 | 2000-12-26 | Cognex Corporation | Method and apparatus for detecting extended defects in an object |
| JP2012184942A (en) * | 2011-03-03 | 2012-09-27 | Central Glass Co Ltd | Method for detecting surface defect of glass fiber product |
| CN113781482A (en) * | 2021-11-11 | 2021-12-10 | 山东精良海纬机械有限公司 | Method and system for detecting crack defects of mechanical parts in complex environment |
| EP3965056A1 (en) * | 2020-09-04 | 2022-03-09 | General Electric Company | Systems and methods for generating a single observation image to analyze coating defects |
| CN114240888A (en) * | 2021-12-17 | 2022-03-25 | 沭阳安宏木制品厂 | Furniture assembly paint spraying defect repairing method and system based on image processing |
-
2022
- 2022-03-31 CN CN202210345162.3A patent/CN114998187A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6167150A (en) * | 1998-07-24 | 2000-12-26 | Cognex Corporation | Method and apparatus for detecting extended defects in an object |
| JP2012184942A (en) * | 2011-03-03 | 2012-09-27 | Central Glass Co Ltd | Method for detecting surface defect of glass fiber product |
| EP3965056A1 (en) * | 2020-09-04 | 2022-03-09 | General Electric Company | Systems and methods for generating a single observation image to analyze coating defects |
| CN113781482A (en) * | 2021-11-11 | 2021-12-10 | 山东精良海纬机械有限公司 | Method and system for detecting crack defects of mechanical parts in complex environment |
| CN114240888A (en) * | 2021-12-17 | 2022-03-25 | 沭阳安宏木制品厂 | Furniture assembly paint spraying defect repairing method and system based on image processing |
Non-Patent Citations (1)
| Title |
|---|
| 余方林 等: "弱磁检测技术在玻璃纤维中的应用研究", 《失效分析与预防》, vol. 14, no. 4, 31 December 2019 (2019-12-31), pages 232 - 239 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115311284A (en) * | 2022-10-12 | 2022-11-08 | 南通嘉鹏家居用品有限公司 | Production quality detection method for glass fiber mat |
| CN117218117A (en) * | 2023-11-07 | 2023-12-12 | 常熟市东宇绝缘复合材料有限公司 | Glass fiber yarn detection method and system |
| CN117218117B (en) * | 2023-11-07 | 2024-01-26 | 常熟市东宇绝缘复合材料有限公司 | Glass fiber yarn detection method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114998187A (en) | Glass fiber separator defect detection method and system | |
| CN102937595B (en) | Method, device and system for detecting printed circuit board (PCB) | |
| CN111080691A (en) | Infrared hot spot detection method and device for photovoltaic module | |
| CN115294140A (en) | Hardware part defect detection method and system | |
| CN113205063A (en) | Visual identification and positioning method for defects of power transmission conductor | |
| CN115272321B (en) | Textile defect detection method based on machine vision | |
| CN116166960B (en) | Big data characteristic cleaning method and system for neural network training | |
| CN114581692B (en) | Vibration damper fault detection method and system based on intelligent pattern recognition | |
| CN114170140A (en) | Membrane defect identification method based on Yolov4 | |
| CN115953398B (en) | Defect identification method for strip steel surface | |
| CN113298797B (en) | Solar cell surface defect detection method | |
| CN114581446B (en) | Battery core abnormity detection method and system of laminated battery | |
| CN112270659A (en) | Rapid detection method and system for surface defects of pole piece of power battery | |
| CN112258459A (en) | Lithium battery defect detection method based on integrated deep learning | |
| CN116523923B (en) | Battery case defect identification method | |
| CN115456955A (en) | Method for detecting internal burr defect of ball cage dust cover | |
| CN115184364A (en) | Tire bubble defect detection method and system based on 3D point cloud | |
| CN114170168A (en) | Display module defect detection method, system and computer readable storage medium | |
| CN115330768B (en) | Quality grading method for solar cell panel | |
| CN106204889B (en) | A kind of localization method and device of bank note fluorescence area | |
| CN115330766B (en) | Method for identifying formation quality defects of capacitor anode aluminum foil | |
| CN117788469B (en) | Corrugated board surface defect intelligent identification method based on simulation image | |
| CN111192264A (en) | Automatic paper defect detection method based on improved non-local mean value fast-RCNN algorithm | |
| US20240221150A1 (en) | Electrode plate wrinkling detection method and system, terminal, and storage medium | |
| CN118657774B (en) | Intelligent identification method for technological defects of negative electrode material graphite box plate |
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 |