CN117115161A - Plastic defect inspection method - Google Patents
Plastic defect inspection method Download PDFInfo
- Publication number
- CN117115161A CN117115161A CN202311376360.7A CN202311376360A CN117115161A CN 117115161 A CN117115161 A CN 117115161A CN 202311376360 A CN202311376360 A CN 202311376360A CN 117115161 A CN117115161 A CN 117115161A
- Authority
- CN
- China
- Prior art keywords
- contour
- pixel
- sub
- plastic
- 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.)
- Granted
Links
- 229920003023 plastic Polymers 0.000 title claims abstract description 86
- 239000004033 plastic Substances 0.000 title claims abstract description 86
- 230000007547 defect Effects 0.000 title claims abstract description 78
- 238000007689 inspection Methods 0.000 title claims abstract description 40
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000009499 grossing Methods 0.000 claims description 19
- 230000008569 process Effects 0.000 claims description 9
- 230000003213 activating effect Effects 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 description 8
- 238000000605 extraction Methods 0.000 description 7
- 230000002159 abnormal effect Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 2
- 238000002955 isolation Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 1
- 238000006748 scratching Methods 0.000 description 1
- 230000002393 scratching effect 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
- G06T7/001—Industrial image inspection using an image reference approach
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20224—Image subtraction
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
-
- 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
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention provides a plastic defect inspection method, which belongs to the technical field of image processing and comprises the following steps: s1, acquiring an image of plastic to be inspected to obtain the image to be inspected; s2, subtracting the image to be inspected from the defect-free plastic sample image to obtain a difference image; s3, extracting contours of the difference images to obtain contours of the difference images; s4, inputting contour features of the difference image contours into a plastic defect inspection model to obtain plastic defect scores; the invention solves the problem of low defect inspection precision in the existing plastic defect inspection method.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a plastic defect inspection method.
Background
In the production process of plastics, the plastic is extremely easy to be influenced by the process or environment in the production process, thereby causing defects of the plastics. Common plastic defects include: shrinkage, air streaking, black spots, deformation, color leakage, scratching, scorching and the like. These defects affect the stability of the plastic structure while affecting the appearance of the plastic.
The existing plastic defect inspection method obtains gray level according to the gray level histogram by obtaining the gray level histogram of the plastic surface image, and obtains the gray level non-uniformity degree of the pixel points based on the gray level distribution, thereby evaluating whether the plastic surface is rough or not, and combining the rough surface images to find defects. However, the plastic surface is not smooth, and various rugged structures are included on the plastic surface, and a suspected defect area is determined according to the gray scale distribution, so that the defect inspection accuracy is not high.
Disclosure of Invention
The invention aims to provide a plastic defect inspection method which solves the problem that the existing plastic defect inspection method is low in defect inspection precision.
The embodiment of the invention is realized by the following technical scheme: a method of inspecting plastic defects, comprising the steps of:
s1, acquiring an image of plastic to be inspected to obtain the image to be inspected;
s2, subtracting the image to be inspected from the defect-free plastic sample image to obtain a difference image;
s3, extracting contours of the difference images to obtain contours of the difference images;
s4, inputting contour features of the difference image contours into a plastic defect inspection model to obtain plastic defect scores.
Further, the step S2 includes the following sub-steps:
s21, registering an image to be inspected and a defect-free plastic sample image through a preset positioning mark;
s25, subtracting the registered image to be inspected and the plastic sample image without defects according to the pixel point positions, and taking the absolute value of the pixel value difference value of the pixel point at the same position as the pixel value of the pixel point of the difference image at the same position.
The beneficial effects of the above further scheme are: according to the invention, the image to be inspected and the plastic sample image without defects can be registered by presetting the positioning mark, so that the alignment of the images is realized, and the pixel points at the same position are subtracted, so that the defect area is highlighted, and the defect area extraction is prevented from being influenced by the structure of the plastic.
Further, the step S3 includes the following sub-steps:
s31, any pixel point is taken from the difference image and used as a comparison point;
s32, calculating cosine similarity between 8 pixel points in a neighborhood range of the contrast point 3*3 and pixel values of the contrast point;
s33, marking the comparison points as non-contour points when the 8 cosine similarities are all larger than a similarity threshold value;
s34, any pixel point is taken in the neighborhood range of the comparison point to be used as the next comparison point, and the step S32 is skipped until the pixel point on the difference image is traversed;
s35, eliminating the pixel points marked as non-contour points from the difference image to obtain an initial contour;
s36, performing contour smoothing on the initial contour to obtain a gap image contour.
The beneficial effects of the above further scheme are: and taking any pixel point from the difference image as a comparison point, estimating the pixel value level of 9 pixel points through cosine similarity between the pixel point and 8 pixel points in the 3*3 neighborhood range, wherein if the pixel value levels are consistent, the 8 cosine similarity is larger than a similarity threshold, and if the pixel value levels are inconsistent, the cosine similarity is smaller than the similarity threshold. When the pixel values are consistent in level, the central pixel point is a non-contour point, all the non-contour points are removed after the difference image is traversed, and an initial contour is obtained, but an initial contour line is not smooth, and has individual abrupt abnormal points, so that the defect inspection accuracy is affected.
Further, the step S36 includes the following sub-steps:
s361, calculating a contour coefficient according to the number of pixel points existing in a 3*3 neighborhood range of each pixel point in the initial contour;
s362, performing smoothing processing on each pixel point according to the contour coefficient to obtain a difference image contour.
Further, the formula for calculating the contour coefficient in S361 is:
wherein,for the profile factor>3*3 neighborhood range memory for each pixelThe number of pixels in +.>Is a proportional coefficient->Is the pixel threshold.
The beneficial effects of the above further scheme are: in the invention, the quantity of the pixel points existing in the 3*3 neighborhood range of each pixel point is counted, thereby measuring the isolation condition of the pixel point, and the method is characterized in thatWhen the pixel is in the neighborhood, only 1 pixel point is in the neighborhood, the central pixel point is considered as a noise point, and is considered as an isolated pixel point, and the pixel point is in the range of +.>,/>When in use, then->The center pixel will be erased, in +.>,/>When the pixel is in the neighborhood range, only 2 pixels are in the neighborhood range, and the center pixel is considered as a noise point, < ->Also equal to 0, the center pixel will be erased, so the size of the pixel threshold in the present invention determines the degree of smoothing in the present invention, and in general, set +.>Or->。
Further, the formula for performing the smoothing process on each pixel in S362 is as follows:
wherein,pixel value for pixel point after smoothing,/-for the smoothing process>For the pixel value of the pixel point to be smoothed,/->For the average value of pixel values of pixel points existing in 3*3 neighborhood range of the pixel point to be smoothed, +.>Is a contour coefficient.
The beneficial effects of the above further scheme are: in the present inventionWhen (I)>Equal to 0, the pixel value of the pixel point after the smoothing is also 0, thereby removing the relatively isolated pixel point, avoiding affecting the extraction of the subsequent contour features, and ensuring that the pixel point is not affected by the extraction of the subsequent contour featuresIn this case, the number of pixels in the neighboring region of the pixel is large, so that the pixel value of the pixel is denoised by using the distribution condition of the pixel values in the neighboring region, and the abnormal pixel value is prevented from affecting the extraction of the subsequent contour features. The invention removes relatively isolated pixel points and abnormal pixel values, thereby enabling the contour lines to be smoother.
Further, the step S4 includes the following sub-steps:
s41, intercepting the outline of the difference image by adopting a window with a fixed size to obtain each outline sub-block;
s42, extracting the contour characteristics of each contour sub-block;
s43, inputting the outline characteristics into a plastic defect inspection model to obtain a plastic defect score.
The beneficial effects of the above further scheme are: according to the invention, a window with a fixed size is adopted to intercept the outline of the difference image, the outline of the difference image is intercepted into outline sub-blocks with the same size, and the plastic defect score is calculated according to the outline characteristics of each outline sub-block.
Further, the profile feature in S42 includes: profile density, profile length, and smoothness;
the calculation formula of the contour density is as follows:
wherein,is->Profile density of individual profile sub-blocks, +.>Is->The number of pixels in each contour sub-block, < +.>Is the firstThe areas of the contour sub-blocks;
the calculation formula of the contour length is as follows:
wherein,is->Profile length of each profile sub-block, +.>Is->The abscissa of one of the furthest pair of pixels in a contour sub-block,/->Is->Ordinate of one of the most distant pair of pixels in each contour sub-block,/->Is->The abscissa of the other pixel of the furthest pair of pixels in the contour sub-block,is->An ordinate of another pixel point in the farthest pair of pixel points in the outline sub-block;
the calculation formula of the smoothness degree is as follows:
wherein,is->Smoothness of individual contour sub-blocks, +.>Is->The>Pixel value of each pixel, +.>Is->The number of pixels in each contour sub-block, the absolute value of +.>Numbering of contour sub-blocks +.>Is the number of the pixel point.
The beneficial effects of the above further scheme are: the contour density is set, the number of pixel points in one contour sub-block is expressed through the contour density, the length of the contour in one contour sub-block is represented through the distance between the farthest pair of pixel points, and the distribution condition of the pixel values in one contour sub-block is represented through the smoothness degree of the pixel values.
Further, the plastic defect inspection model in S43 is:
wherein,scoring for plastic defects->For the number of contour sub-blocks +.>Activating a function for sigmoid->Is a profile density coefficient>For the length of the profile, ->Is a smoothness coefficient>Is->The contour density of the individual contour sub-blocks,is->Profile length of each profile sub-block, +.>Is->The degree of smoothness of the individual contour sub-blocks.
The beneficial effects of the above further scheme are: in the invention, different coefficients are respectively set for the contour density, the contour length and the smoothness, so that different weights are distributed for the contour density, the contour length and the smoothness, and then the defect condition of the whole image to be inspected is estimated according to the contour density, the contour length and the smoothness in each contour sub-block.
The technical scheme of the embodiment of the invention has at least the following advantages and beneficial effects: the method comprises the steps of collecting images of plastics to be inspected, subtracting the images to be inspected from defect-free plastic sample images, constructing a new image, representing the difference between the two images, highlighting non-identical areas, extracting outlines of the difference images, highlighting defect characteristics, inputting outline characteristics of the outlines of the difference images into a plastic defect inspection model, obtaining a plastic defect score, and evaluating the plastic defect degree; in the invention, the influence of the same area on the defect inspection process is reduced by subtracting the two images, and then the data quantity is further reduced by extracting the outline features, the defect features are highlighted, the influence of the plastic structure on the plastic surface on the defect inspection is reduced as much as possible, and the defect inspection precision is improved.
Drawings
FIG. 1 is a flow chart of a method for inspecting defects in plastics.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
As shown in fig. 1, a plastic defect inspection method includes the steps of:
s1, acquiring an image of plastic to be inspected to obtain the image to be inspected;
s2, subtracting the image to be inspected from the defect-free plastic sample image to obtain a difference image;
the step S2 comprises the following sub-steps:
s21, registering an image to be inspected and a defect-free plastic sample image through a preset positioning mark;
s25, subtracting the registered image to be inspected and the plastic sample image without defects according to the pixel point positions, and taking the absolute value of the pixel value difference value of the pixel point at the same position as the pixel value of the pixel point of the difference image at the same position.
In S25, the pixel point where the pixel value is 0 is considered to be no pixel point, and the subsequent process is not participated.
According to the invention, the image to be inspected and the plastic sample image without defects can be registered by presetting the positioning mark, so that the alignment of the images is realized, and the pixel points at the same position are subtracted, so that the defect area is highlighted, and the defect area extraction is prevented from being influenced by the structure of the plastic.
When the external environments are completely consistent, the pixel values of the same position points in the image to be inspected and the plastic sample image without defects are consistent, but in general, the external environments are changed at all times, so that the pixel values of the same position points in the image to be inspected and the plastic sample image without defects are different, and the pixel values of non-same areas of the difference image obtained by subtracting absolute values are larger.
S3, extracting contours of the difference images to obtain contours of the difference images;
the step S3 comprises the following substeps:
s31, any pixel point is taken from the difference image and used as a comparison point;
s32, calculating cosine similarity between 8 pixel points in a neighborhood range of the contrast point 3*3 and pixel values of the contrast point;
s33, marking the comparison points as non-contour points when the 8 cosine similarities are all larger than a similarity threshold value;
s34, any pixel point is taken in the neighborhood range of the comparison point to be used as the next comparison point, and the step S32 is skipped until the pixel point on the difference image is traversed;
s35, eliminating the pixel points marked as non-contour points from the difference image to obtain an initial contour;
s36, performing contour smoothing on the initial contour to obtain a gap image contour.
And taking any pixel point on the difference image as a comparison point, estimating the pixel value level of 9 pixel points through cosine similarity between the pixel point and 8 pixel points in the 3*3 neighborhood range, wherein if the pixel value levels are consistent, the 8 cosine similarity is larger than a similarity threshold, and if the pixel value levels are inconsistent, the cosine similarity is smaller than the similarity threshold. When the pixel values are consistent in level, the central pixel point is a non-contour point, all the non-contour points are removed after the difference image is traversed, and an initial contour is obtained, but an initial contour line is not smooth, and has individual abrupt abnormal points, so that the defect inspection accuracy is affected.
The step S36 includes the following sub-steps:
s361, calculating a contour coefficient according to the number of pixel points existing in a 3*3 neighborhood range of each pixel point in the initial contour;
s362, performing smoothing processing on each pixel point according to the contour coefficient to obtain a difference image contour.
The formula for calculating the contour coefficient in S361 is:
wherein,for the profile factor>For the number of pixels existing within 3*3 neighborhood of each pixel, +.>Is a proportional coefficient->Is the pixel threshold.
In the invention, the quantity of the pixel points existing in the 3*3 neighborhood range of each pixel point is counted, thereby measuring the isolation condition of the pixel point, and the method is characterized in thatWhen the pixel is in the neighborhood, only 1 pixel point is in the neighborhood, the central pixel point is considered as a noise point, and is considered as an isolated pixel point, and the pixel point is in the range of +.>,/>When in use, then->The center pixel will be erased, in +.>,/>When the pixel is in the neighborhood range, only 2 pixels are in the neighborhood range, and the center pixel is considered as a noise point, < ->Also equal to 0, the central pixel point will be erased, so the size of the pixel point threshold in the present invention determines the degree of smoothing in the present invention, and is typically setOr->。
The formula for performing the smoothing process on each pixel point in S362 is as follows:
wherein,pixel value for pixel point after smoothing,/-for the smoothing process>For the pixel value of the pixel point to be smoothed,/->For the average value of pixel values of pixel points existing in 3*3 neighborhood range of the pixel point to be smoothed, +.>Is a contour coefficient.
In the present inventionWhen (I)>When the pixel value is equal to 0, the pixel value of the pixel point after the smoothing is also 0, thereby removing the relatively isolated pixel point, avoiding affecting the extraction of the subsequent contour features, and being in +.>In this case, the number of pixels in the neighboring region of the pixel is large, so that the pixel value of the pixel is denoised by using the distribution condition of the pixel values in the neighboring region, and the abnormal pixel value is prevented from affecting the extraction of the subsequent contour features. The invention removes relatively isolated pixel points and abnormal pixel values, thereby enabling the contour lines to be smoother.
S4, inputting contour features of the difference image contours into a plastic defect inspection model to obtain plastic defect scores.
The step S4 comprises the following substeps:
s41, intercepting the outline of the difference image by adopting a window with a fixed size to obtain each outline sub-block;
s42, extracting the contour characteristics of each contour sub-block;
s43, inputting the outline characteristics into a plastic defect inspection model to obtain a plastic defect score.
According to the invention, a window with a fixed size is adopted to intercept the outline of the difference image, the outline of the difference image is intercepted into outline sub-blocks with the same size, and the plastic defect score is calculated according to the outline characteristics of each outline sub-block.
The profile features in S42 include: profile density, profile length, and smoothness;
the calculation formula of the contour density is as follows:
wherein,is->Profile density of individual profile sub-blocks, +.>Is->The number of pixels in each contour sub-block, < +.>Is the firstThe areas of the contour sub-blocks;
in the invention, the area of the outline sub-block is defined by the number of pixel points distributed.
The calculation formula of the contour length is as follows:
wherein,is->Profile length of each profile sub-block, +.>Is->The abscissa of one of the furthest pair of pixels in a contour sub-block,/->Is->Ordinate of one of the most distant pair of pixels in each contour sub-block,/->Is->The abscissa of the other pixel of the furthest pair of pixels in the contour sub-block,is->An ordinate of another pixel point in the farthest pair of pixel points in the outline sub-block;
the calculation formula of the smoothness degree is as follows:
wherein,is->Smoothness of individual contour sub-blocks, +.>Is->The>Pixel value of each pixel, +.>Is->The number of pixels in each contour sub-block, the absolute value of +.>Numbering of contour sub-blocks +.>Is the number of the pixel point.
The contour density is set, the number of pixel points in one contour sub-block is expressed through the contour density, the length of the contour in one contour sub-block is represented through the distance between the farthest pair of pixel points, and the distribution condition of the pixel values in one contour sub-block is represented through the smoothness degree of the pixel values.
The plastic defect inspection model in S43 is:
wherein,scoring for plastic defects->For the number of contour sub-blocks +.>Activating a function for sigmoid->Is a profile density coefficient>For the length of the profile, ->Is a smoothness coefficient>Is->The contour density of the individual contour sub-blocks,is->Profile length of each profile sub-block, +.>Is->The degree of smoothness of the individual contour sub-blocks.
In the invention, different coefficients are respectively set for the contour density, the contour length and the smoothness, so that different weights are distributed for the contour density, the contour length and the smoothness, and then the defect condition of the whole image to be inspected is estimated according to the contour density, the contour length and the smoothness in each contour sub-block.
The method comprises the steps of collecting images of plastics to be inspected, subtracting the images to be inspected from defect-free plastic sample images, constructing a new image, representing the difference between the two images, highlighting non-identical areas, extracting outlines of the difference images, highlighting defect characteristics, inputting outline characteristics of the outlines of the difference images into a plastic defect inspection model, obtaining a plastic defect score, and evaluating the plastic defect degree; in the invention, the influence of the same area on the defect inspection process is reduced by subtracting the two images, and then the data quantity is further reduced by extracting the outline features, the defect features are highlighted, the influence of the plastic structure on the plastic surface on the defect inspection is reduced as much as possible, and the defect inspection precision is improved.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (9)
1. A method for inspecting defects in plastics, comprising the steps of:
s1, acquiring an image of plastic to be inspected to obtain the image to be inspected;
s2, subtracting the image to be inspected from the defect-free plastic sample image to obtain a difference image;
s3, extracting contours of the difference images to obtain contours of the difference images;
s4, inputting contour features of the difference image contours into a plastic defect inspection model to obtain plastic defect scores.
2. The plastic defect inspection method of claim 1, wherein S2 comprises the sub-steps of:
s21, registering an image to be inspected and a defect-free plastic sample image through a preset positioning mark;
s25, subtracting the registered image to be inspected and the plastic sample image without defects according to the pixel point positions, and taking the absolute value of the pixel value difference value of the pixel point at the same position as the pixel value of the pixel point of the difference image at the same position.
3. The plastic defect inspection method of claim 1, wherein S3 comprises the sub-steps of:
s31, any pixel point is taken from the difference image and used as a comparison point;
s32, calculating cosine similarity between 8 pixel points in a neighborhood range of the contrast point 3*3 and pixel values of the contrast point;
s33, marking the comparison points as non-contour points when the 8 cosine similarities are all larger than a similarity threshold value;
s34, any pixel point is taken in the neighborhood range of the comparison point to be used as the next comparison point, and the step S32 is skipped until the pixel point on the difference image is traversed;
s35, eliminating the pixel points marked as non-contour points from the difference image to obtain an initial contour;
s36, performing contour smoothing on the initial contour to obtain a gap image contour.
4. A plastic defect inspection method according to claim 3, wherein S36 comprises the sub-steps of:
s361, calculating a contour coefficient according to the number of pixel points existing in a 3*3 neighborhood range of each pixel point in the initial contour;
s362, performing smoothing processing on each pixel point according to the contour coefficient to obtain a difference image contour.
5. The plastic defect inspection method of claim 4, wherein the formula for calculating the profile factor in S361 is:
wherein,for the profile factor>For the number of pixels existing within 3*3 neighborhood of each pixel, +.>Is a proportional coefficient->Is the pixel threshold.
6. The plastic defect inspection method according to claim 5, wherein the formula for smoothing each pixel in S362 is:
wherein,pixel value for pixel point after smoothing,/-for the smoothing process>For the pixel value of the pixel point to be smoothed,/->For the average value of pixel values of pixel points existing in 3*3 neighborhood range of the pixel point to be smoothed, +.>Is a contour coefficient.
7. The plastic defect inspection method of claim 1, wherein S4 comprises the substeps of:
s41, intercepting the outline of the difference image by adopting a window with a fixed size to obtain each outline sub-block;
s42, extracting the contour characteristics of each contour sub-block;
s43, inputting the outline characteristics into a plastic defect inspection model to obtain a plastic defect score.
8. The plastic defect inspection method of claim 7, wherein the profile features in S42 comprise: profile density, profile length, and smoothness;
the calculation formula of the contour density is as follows:
wherein,is->Profile density of individual profile sub-blocks, +.>Is->The number of pixels in each contour sub-block, < +.>Is->The areas of the contour sub-blocks;
the calculation formula of the contour length is as follows:
wherein,is->Profile length of each profile sub-block, +.>Is->The abscissa of one of the furthest pair of pixels in a contour sub-block,/->Is->Ordinate of one of the most distant pair of pixels in each contour sub-block,/->Is->The abscissa of the other pixel point of the furthest pair of pixel points in the outline sub-blocks, +.>Is->An ordinate of another pixel point in the farthest pair of pixel points in the outline sub-block;
the calculation formula of the smoothness degree is as follows:
wherein,is->Smoothness of individual contour sub-blocks, +.>Is->The>The pixel values of the individual pixel points,is->Each contour isThe number of pixels in the block, ||is absolute value, |is +.>Numbering of contour sub-blocks +.>Is the number of the pixel point.
9. The plastic defect inspection method according to claim 8, wherein the plastic defect inspection model in S43 is:
wherein,scoring for plastic defects->For the number of contour sub-blocks +.>Activating a function for sigmoid->Is a profile density coefficient>For the length of the profile, ->Is a smoothness coefficient>Is->Of sub-blocks of contoursProfile density->Is->Profile length of each profile sub-block, +.>Is->The degree of smoothness of the individual contour sub-blocks.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311376360.7A CN117115161B (en) | 2023-10-24 | 2023-10-24 | Plastic defect inspection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311376360.7A CN117115161B (en) | 2023-10-24 | 2023-10-24 | Plastic defect inspection method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117115161A true CN117115161A (en) | 2023-11-24 |
CN117115161B CN117115161B (en) | 2024-01-02 |
Family
ID=88798749
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311376360.7A Active CN117115161B (en) | 2023-10-24 | 2023-10-24 | Plastic defect inspection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117115161B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117649566A (en) * | 2024-01-30 | 2024-03-05 | 四川省机场集团有限公司成都天府国际机场分公司 | Airport luggage size classification method based on image processing |
Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2002063887A1 (en) * | 2001-02-06 | 2002-08-15 | Koninklijke Philips Electronics N.V. | Contour filter for image sensor output signals |
CN102509300A (en) * | 2011-11-18 | 2012-06-20 | 深圳市宝捷信科技有限公司 | Defect detection method and system |
US20130307963A1 (en) * | 2005-01-14 | 2013-11-21 | Hitachi High-Technologies Corporation | Method and apparatus for inspecting patterns formed on a substrate |
CN104697476A (en) * | 2015-03-19 | 2015-06-10 | 北京时代之峰科技有限公司 | Automatic detection method and device for roughness light-incision profile curve |
CN107014822A (en) * | 2017-02-28 | 2017-08-04 | 深圳市维图视技术有限公司 | A kind of non-woven fabrics gauze mask defective vision detecting system and method |
CN112508903A (en) * | 2020-12-01 | 2021-03-16 | 电子科技大学 | Method for detecting surface defect contour of lens of satellite telescope |
CN112508826A (en) * | 2020-11-16 | 2021-03-16 | 哈尔滨工业大学(深圳) | Printed matter defect detection method based on feature registration and gradient shape matching fusion |
CN113920096A (en) * | 2021-10-14 | 2022-01-11 | 广东工业大学 | Method for detecting metal packaging defects of integrated circuit |
CN114387329A (en) * | 2022-01-16 | 2022-04-22 | 四川轻化工大学 | Building contour progressive regularization method based on high-resolution remote sensing image |
CN116137036A (en) * | 2023-04-19 | 2023-05-19 | 吉林省英华恒瑞生物科技有限公司 | Gene detection data intelligent processing system based on machine learning |
CN116703903A (en) * | 2023-08-04 | 2023-09-05 | 山东济宁运河煤矿有限责任公司 | Anchor rod repairing quality detection method based on machine vision |
CN116779465A (en) * | 2023-08-25 | 2023-09-19 | 青岛天仁微纳科技有限责任公司 | Nano-imprinting wafer defect detection method |
CN116823755A (en) * | 2023-06-20 | 2023-09-29 | 浙江大学 | Flexible circuit board defect detection method based on skeleton generation and fusion configuration |
-
2023
- 2023-10-24 CN CN202311376360.7A patent/CN117115161B/en active Active
Patent Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2002063887A1 (en) * | 2001-02-06 | 2002-08-15 | Koninklijke Philips Electronics N.V. | Contour filter for image sensor output signals |
US20130307963A1 (en) * | 2005-01-14 | 2013-11-21 | Hitachi High-Technologies Corporation | Method and apparatus for inspecting patterns formed on a substrate |
CN102509300A (en) * | 2011-11-18 | 2012-06-20 | 深圳市宝捷信科技有限公司 | Defect detection method and system |
CN104697476A (en) * | 2015-03-19 | 2015-06-10 | 北京时代之峰科技有限公司 | Automatic detection method and device for roughness light-incision profile curve |
CN107014822A (en) * | 2017-02-28 | 2017-08-04 | 深圳市维图视技术有限公司 | A kind of non-woven fabrics gauze mask defective vision detecting system and method |
CN112508826A (en) * | 2020-11-16 | 2021-03-16 | 哈尔滨工业大学(深圳) | Printed matter defect detection method based on feature registration and gradient shape matching fusion |
CN112508903A (en) * | 2020-12-01 | 2021-03-16 | 电子科技大学 | Method for detecting surface defect contour of lens of satellite telescope |
CN113920096A (en) * | 2021-10-14 | 2022-01-11 | 广东工业大学 | Method for detecting metal packaging defects of integrated circuit |
CN114387329A (en) * | 2022-01-16 | 2022-04-22 | 四川轻化工大学 | Building contour progressive regularization method based on high-resolution remote sensing image |
CN116137036A (en) * | 2023-04-19 | 2023-05-19 | 吉林省英华恒瑞生物科技有限公司 | Gene detection data intelligent processing system based on machine learning |
CN116823755A (en) * | 2023-06-20 | 2023-09-29 | 浙江大学 | Flexible circuit board defect detection method based on skeleton generation and fusion configuration |
CN116703903A (en) * | 2023-08-04 | 2023-09-05 | 山东济宁运河煤矿有限责任公司 | Anchor rod repairing quality detection method based on machine vision |
CN116779465A (en) * | 2023-08-25 | 2023-09-19 | 青岛天仁微纳科技有限责任公司 | Nano-imprinting wafer defect detection method |
Non-Patent Citations (4)
Title |
---|
FACHRUDDIN FACHRUDDIN等: "Extraction of object image features with gradation contour", 《TELKOMNIKA TELECOMMUNICATION, COMPUTING, ELECTRONICS AND CONTROL》, vol. 19, no. 6, pages 1913 - 1923 * |
JINGFAN HANG等: "Sanitary Ceramic Surface Defect Detection Method Based on Neighborhood Pixel Gray Information", 《IECON 2020 THE 46TH ANNUAL CONFERENCE OF THE IEEE INDUSTRIAL ELECTRONICS SOCIETY》, pages 5337 - 5342 * |
朱璐: "基于调色板的自然图像重着色及其应用研究", 《中国优秀硕士学位论文全文数据库_信息科技辑》, pages 138 - 2610 * |
王伟梁: "基于机器视觉算法的PCB裸板缺陷检测算法及应用研究", 《万方数据知识服务平台》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117649566A (en) * | 2024-01-30 | 2024-03-05 | 四川省机场集团有限公司成都天府国际机场分公司 | Airport luggage size classification method based on image processing |
CN117649566B (en) * | 2024-01-30 | 2024-04-09 | 四川省机场集团有限公司成都天府国际机场分公司 | Airport luggage size classification method based on image processing |
Also Published As
Publication number | Publication date |
---|---|
CN117115161B (en) | 2024-01-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108764229B (en) | Water gauge image automatic identification method based on computer vision technology | |
CN117115161B (en) | Plastic defect inspection method | |
CN109886939B (en) | Bridge crack detection method based on tensor voting | |
CN115100221B (en) | Glass defect segmentation method | |
CN111415376B (en) | Automobile glass subpixel contour extraction method and automobile glass detection method | |
CN116758077B (en) | Online detection method and system for surface flatness of surfboard | |
CN116580021B (en) | Environment-friendly concrete carbon reduction product production and quality detection method | |
JP2014228357A (en) | Crack detecting method | |
CN117237368B (en) | Bridge crack detection method and system | |
CN111161260A (en) | Hot-rolled strip steel surface defect detection method and device based on deep learning | |
CN116993731B (en) | Shield tunneling machine tool bit defect detection method based on image | |
CN111325164A (en) | Pointer indication number identification method and device and electronic equipment | |
CN114965483A (en) | Quantitative evaluation method for various complex defects of spacecraft | |
CN113705564B (en) | Pointer type instrument identification reading method | |
CN108764343B (en) | Method for positioning tracking target frame in tracking algorithm | |
CN111415378B (en) | Image registration method for automobile glass detection and automobile glass detection method | |
CN111815625A (en) | Early lung cancer prediction method based on deep learning | |
Nagase et al. | Automatic calculation and visualization of nuclear density in whole slide images of hepatic histological sections | |
CN113570524B (en) | Repairing method for high reflection noise depth image | |
CN108010076B (en) | End face appearance modeling method for intensive industrial bar image detection | |
CN112750119B (en) | Detection and measurement method for weak defects on surface of white glass cover plate | |
CN115082447A (en) | Method and device for measuring uniformity of mortar, readable storage medium and electronic equipment | |
CN114926360A (en) | Image noise reduction processing working method based on noise estimation | |
CN114529803A (en) | Pointer instrument identification enhancement detection method | |
CN113763279A (en) | Accurate correction processing method for image with rectangular frame |
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 |