CN113870299A - 3D printing fault detection method based on edge detection and morphological image processing - Google Patents
3D printing fault detection method based on edge detection and morphological image processing Download PDFInfo
- Publication number
- CN113870299A CN113870299A CN202111044933.7A CN202111044933A CN113870299A CN 113870299 A CN113870299 A CN 113870299A CN 202111044933 A CN202111044933 A CN 202111044933A CN 113870299 A CN113870299 A CN 113870299A
- Authority
- CN
- China
- Prior art keywords
- image
- method based
- detection method
- fault detection
- edge detection
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 30
- 238000010146 3D printing Methods 0.000 title claims abstract description 26
- 238000003708 edge detection Methods 0.000 title claims abstract description 21
- 230000000877 morphologic effect Effects 0.000 title claims abstract description 19
- 238000000034 method Methods 0.000 claims abstract description 26
- 238000011144 upstream manufacturing Methods 0.000 claims abstract description 3
- 238000005260 corrosion Methods 0.000 claims description 10
- 230000007797 corrosion Effects 0.000 claims description 10
- 230000008961 swelling Effects 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 6
- 230000003628 erosive effect Effects 0.000 claims description 3
- 238000007639 printing Methods 0.000 claims 1
- 230000003044 adaptive effect Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 230000018109 developmental process Effects 0.000 description 3
- 238000007689 inspection Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000011218 segmentation Effects 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/73—Deblurring; Sharpening
-
- 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
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/155—Segmentation; Edge detection involving morphological operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- 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
- G06T2207/20036—Morphological image processing
- G06T2207/20041—Distance transform
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Geometry (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a 3D printing fault detection method based on edge detection and morphological image processing, which comprises the following steps: s1, acquiring a 3D printed object image A under a preset working condition through an industrial camera; step S2: sharpening the image A by using a Prewitt operator to obtain an image B; step S3: converting the image B into a binary image C by using an Otsu algorithm; step S4: performing closed operation on the image C, and closing the small cracks to obtain an image D; step S5: performing opening operation on the image D, and removing isolated dots, burrs and bridges to obtain an image E; step S6: acquiring a 4-connected region of the image E by adopting a bwporim algorithm, and extracting the outline of the binary image to obtain an image G; step S7: reading the upstream contour position and the downstream contour position of the image, and subtracting the upper line position and the lower line position on the same column to obtain the line width of the position of the object; step S8: and counting the number of positions with the continuous line width of 0 to obtain the number of broken points. The 3D printing fault detection method based on edge detection and morphological image processing can simply and quickly detect and process the breakpoint fault existing in the 3D printing process.
Description
Technical Field
The invention belongs to the field of image processing and 3D printers, and particularly relates to a 3D printing fault detection method based on edge detection and morphological image processing.
Background
As an important component in the world economy, China strongly supports the development of the 3D printing industry from multiple levels such as development targets, industry standards, financial support, major project establishment and the like. Defects generated in the 3D printing process not only affect the appearance and the service performance of parts, cause the waste of time and materials, but also severely restrict the further development of the 3D printing process. For a long time, quality inspection is judged by depending on experience manually, and deviation occurs in the generated quality judgment result. The on-site inspection has fast pace, does not form complete quality data record accumulation, and loses the bonus that big data can bring to management.
The artificial intelligence technology can make up the difficulties of difficult finding of inspectors, uneven technical level, standard scale understanding and control difference and the like. The current mature fault detection method based on image processing comprises the following steps: edge detection based methods, image based color information methods, mathematical morphology based methods.
The method based on edge detection comprises the steps of firstly carrying out edge enhancement on an image, then carrying out binarization on the image, and finally carrying out matching between the image and a template. However, the method has high requirements for selecting the template and is easily influenced by the size, color and light of the template, the traditional edge detection method is to extract an image edge feature image, and the edge extraction method used for 3D printing fault detection only needs to reserve the outermost edge of the target, so the method needs to be improved.
The color information method based on the image firstly finds out the ground color area of the 3D printing object from the image to determine the threshold value of each related component of the color of the target object in the space, and then uses the color space distance and the similarity calculation to carry out the image color segmentation. However, when the color of the 3D printing material is close to the color of the nearby area, the method may cause false fault detection under the interference of the copy background.
The method based on mathematical morphology comprises the steps of firstly carrying out graying processing on an image, then carrying out filtering processing, and finally carrying out analysis processing on the image by using expansion, decaying candle, opening and closing operations and the like of the mathematical morphology. However, this method has a poor detection effect when there is an artifact in the image.
In order to overcome the problem that the conventional 3D printing fault detection method is easily influenced by color, illumination conditions and interference under a complex background, a 3D printing fault detection method based on edge detection and morphological image processing is provided. Firstly, sharpening an image A by using a Prewitt operator, and in order to solve the problem that an image seriously polluted by artifacts, light rays and colors is difficult to carry out edge detection processing, the invention carries out mathematical morphological expansion, candle corrosion, opening and closing operations and the like on the result of binary image processing to analyze and process the image, closes small cracks, removes isolated dots, burrs and bridges, thins the area of a target object to obtain the image, and finally carries out edge detection on the image by adopting a bwwperm algorithm to extract the outline information of a printed object. Reading the upper outline position and the lower outline position of the image, and subtracting the upper line position and the lower line position on the same column to obtain the line width of the position of the object. And counting the position number of the continuous line width of 0 to obtain the broken point number of the printed line segment.
Disclosure of Invention
The invention mainly solves the technical problem of intelligently monitoring a 3D printing automatic production line by using a machine vision method, and mainly aims to provide a 3D printing fault detection method based on edge detection and morphological image processing, which is applied to an appearance detection link of the 3D printing automatic production line, detects the number, position and unreasonable line width of breakpoints by using a machine vision system, and automatically eliminates defective 3D products according to a detection result.
In order to solve the technical problems, the invention adopts a technical scheme that: provided is a 3D printing fault detection method based on edge detection and morphological image processing, comprising the following steps:
s1, acquiring a 3D printed object image A under a preset working condition through an industrial camera;
step S2: sharpening the image A by using a Prewitt operator to obtain an image B;
step S3: converting the image B into a binary image C by using an Otsu algorithm;
step S4: performing closed operation on the image C, and closing the small cracks to obtain an image D;
step S5: performing opening operation on the image D, and removing isolated dots, burrs and bridges to obtain an image E;
step S6: acquiring a 4-connected region of the image E by adopting a bwporim algorithm, and extracting the outline of the binary image to obtain an image G;
step S7: reading the upstream contour position and the downstream contour position of the image, and subtracting the upper line position and the lower line position on the same column to obtain the line width of the position of the object;
step S8: and counting the number of positions with the continuous line width of 0 to obtain the number of broken points.
Further, in the step S1, the industrial camera acquires the object image a printed in 3D under the preset working condition, wherein the industrial camera in the step S1 is an area-array camera or a line-array camera.
Further, in the step S2, the Prewitt operator is used to sharpen the image a, specifically:
the Prewitt operator comprises two sets of 3 x 3 operator templates, GxIndicating detection of horizontal edge transverse template, GyRepresenting the vertical edge longitudinal template;
calculating the gradient value of the image by taking the sum of the transverse gradient and the longitudinal gradient mode of each pixel point, and expressing the gradient value by the following formula:
G=|Gx|+|Gy|
further, in step S3, the binarization processing is performed on the image by using an Otsu algorithm, which specifically is as follows:
the number of pixels in the image having a gray value less than the threshold T is denoted N1The number of pixels having a pixel gray level greater than the threshold T is denoted by N2The single-channel gray scale M is 256, and the total number of pixels Sum, then
(4) The proportion of the number of background pixels in the whole image is as follows:
(5) the proportion of the number of foreground pixels in the whole image is as follows:
(6) average gray value of background:
(4) average gray value of foreground:
(5) gradation integrated value in 0 to M gradation sections:
μ=μ1*ω1+μ2*ω2
(6) the between-class variance:
g=ω1*(μ-μ1)2+ω2*(μ-μ2)2
(7) substituting the formulas (3), (4) and (5) into the formula (6) can obtain the final simplified formula
g=ω1*ω2*(μ1-μ2)2
And obtaining the threshold T which enables the inter-class variance to be maximum by adopting a traversal method, namely obtaining the threshold T.
Further, the step S4 performs a close operation on the image to close the small crack, specifically: expansion followed by corrosion. Where Θ and ≦ indicate erosion and swelling, respectively.
A●B=(A⊕B)ΘB
Further, the step S5 performs an opening operation on the image to remove isolated dots, burrs, and bridges, specifically: corrosion first and then swelling. Where the sum indicates corrosion and swelling, respectively.
A○B=(AΘB)⊕B
Further, in step S6, a bwperemm algorithm is used to obtain a 4-pass region of the image E and perform contour extraction of the binary image, specifically: contour extraction was performed using the bwporim (E,4) function of Matlab.
Compared with the prior art, the invention has the beneficial effects that:
(a) the binarization processing of the 3D printed object image adopts an adaptive threshold value binarization Otsu algorithm, the adaptive threshold value does not need to be fixed, but can be adaptively set according to the local characteristics of the image according to a corresponding adaptive method, and the binarization processing is carried out.
(b) Performing morphological closing operation and opening operation on the binarized 3D printed object image by using mathematical morphology, acquiring a 4-connection region of the image by using a bwpherim algorithm, extracting the outline of the binary image, judging whether a breakpoint exists according to whether the edge outline distance of the object is greater than 0, and determining the breakpoint as the fault point. Therefore, the fault detection method is not influenced by the size ratio of the 3D printed object, does not need to use a plurality of standard template images according to different objects to be detected, has the advantages of rapidness and high detection precision compared with the traditional method based on template comparison, and is not easily influenced by the manufacturing of the 3D printed object and image acquisition errors.
Drawings
FIG. 1 is a schematic flow diagram of an embodiment of the present invention.
Fig. 2 is an object image a which is acquired by acquiring 3D printing under a preset working condition through an industrial camera according to the invention.
FIG. 3 is a diagram of an image B obtained by sharpening the image A with the Prewitt operator according to the present invention.
FIG. 4 is a binary image C obtained by binarizing the image B according to the Otsu algorithm
FIG. 5 is a diagram of an image D obtained by performing a close operation on the image C to close the small cracks according to the present invention;
FIG. 6 is an image E obtained by performing an open operation on an image D according to the present invention to remove isolated dots, burrs and bridges;
FIG. 7 is an image G obtained by acquiring a 4-connected region of an image E by a bwpherim algorithm and extracting a contour of a binary image;
FIG. 8 is a diagram of the line width of an object at the position of the upper line and the lower line of the image read by the present invention, obtained by subtracting the upper and lower lines of the same column;
fig. 9 shows the position number and the broken point number of the statistical continuous line width of 0 according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. The described embodiments are only some embodiments of the invention, not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without any inventive step, are within the scope of the present invention.
Referring to fig. 1 to 9, an embodiment of the present invention includes: in particular to a 3D printing fault detection method based on edge detection and morphological image processing, which comprises the following steps:
step S1: the method comprises the steps of collecting a 3D printed object image A under a preset working condition through an industrial camera, wherein the industrial camera adopts an area-array camera or a linear array camera. As shown in fig. 2;
step S2: sharpening the image A by using a Prewitt operator to obtain an image B, as shown in FIG. 3;
sharpening the image A by using a Prewitt operator, specifically:
the Prewitt operator comprises two sets of 3 x 3 operator templates, GxIndicating detection of horizontal edge transverse template, GyRepresenting the vertical edge longitudinal template;
calculating the gradient value of the image by taking the sum of the transverse gradient and the longitudinal gradient mode of each pixel point, and expressing the gradient value by the following formula:
G=|Gx|+|Gy|
step S3: converting the image B into a binary image C by using an Otsu algorithm, as shown in the attached figure 4;
carrying out binarization processing on the image by using an Otsu algorithm, which specifically comprises the following steps:
the number of pixels in the image having a gray value less than the threshold T is denoted N1The number of pixels having a pixel gray level greater than the threshold T is denoted by N2The single-channel gray scale M is 256, and the total number of pixels Sum, then
(1) The proportion of the number of background pixels in the whole image is as follows:
(2) the proportion of the number of foreground pixels in the whole image is as follows:
(3) average gray value of background:
(4) average gray value of foreground:
(5) gradation integrated value in 0 to M gradation sections:
μ=μ1*ω1+μ2*ω2
(6) the between-class variance:
g=ω1*(μ-μ1)2+ω2*(μ-μ2)2
(7) substituting the formulas (3), (4) and (5) into the formula (6) can obtain the final simplified formula
g=ω1*ω2*(μ1-μ2)2
And obtaining the threshold T which enables the inter-class variance to be maximum by adopting a traversal method, namely obtaining the threshold T.
Step S4: performing a closing operation on the image C, and closing the small cracks to obtain an image D, as shown in FIG. 5;
performing closed operation on the image to close the small cracks, specifically: expansion followed by corrosion. Where Θ and ≦ indicate erosion and swelling, respectively.
A●B=(A⊕B)ΘB
Step S5: performing an opening operation on the image D, and removing isolated dots, burrs and bridges to obtain an image E, as shown in FIG. 6;
performing open operation on the image to remove isolated dots, burrs and bridges, specifically: corrosion first and then swelling. Where the sum indicates corrosion and swelling, respectively.
A○B=(AΘB)⊕B
Step S6: acquiring a 4-connected region of the image E by adopting a bwpherem algorithm, and extracting a contour of a binary image to obtain an image G, as shown in the attached figure 7;
acquiring a 4-connected region of the image E by adopting a bwporim algorithm and extracting a contour of a binary image, wherein the method specifically comprises the following steps: contour extraction was performed using the bwporim (E,4) function of Matlab.
Step S7: reading the up-line outline position and the down-line outline position of the image, and subtracting the upper line position and the lower line position on the same column to obtain the line width of the position of the object, as shown in the attached figure 8;
step S8: counting the number of positions of the continuous line width of 0, the number of broken points is shown in fig. 9.
The binarization processing of the 3D printed object image adopts an adaptive threshold value binarization Otsu algorithm, the adaptive threshold value does not need to be fixed, but can be adaptively set according to the local characteristics of the image according to a corresponding adaptive method, and the binarization processing is carried out.
Performing morphological closing operation and opening operation on the binarized 3D printed object image by using mathematical morphology, acquiring a 4-connection region of the image by using a bwpherim algorithm, extracting the outline of the binary image, judging whether a breakpoint exists according to whether the edge outline distance of the object is greater than 0, and determining the breakpoint as the fault point. Therefore, the fault detection method is not influenced by the size ratio of the 3D printed object, does not need to use a plurality of standard template images according to different objects to be detected, has the advantages of rapidness and high detection precision compared with the traditional method based on template comparison, and is not easily influenced by the manufacturing of the 3D printed object and image acquisition errors.
Claims (7)
1. A3D printing fault detection method based on edge detection and morphological image processing is characterized by comprising the following steps:
step S1: acquiring a 3D printed object image A under a preset working condition through an industrial camera;
step S2: sharpening the image A by using a Prewitt operator to obtain an image B;
step S3: converting the image B into a binary image C by using an Otsu algorithm;
step S4: performing closed operation on the image C, and closing the small cracks to obtain an image D;
step S5: performing opening operation on the image D, and removing isolated dots, burrs and bridges to obtain an image E;
step S6: acquiring a 4-connected region of the image E by adopting a bwporim algorithm, and extracting the outline of the binary image to obtain an image G;
step S7: reading the upstream contour position and the downstream contour position of the image, and subtracting the upper line position and the lower line position on the same column to obtain the line width of the position of the object;
step S8: and counting the number of positions with the continuous line width of 0 to obtain the number of broken points.
2. The 3D printing fault detection method based on edge detection and morphological image processing as claimed in claim 1, wherein the industrial camera in the step S1 is an area-array camera or a line-array camera.
3. The 3D printing fault detection method based on edge detection and morphological image processing according to claim 1, wherein the step S2 uses Prewitt operator to sharpen the image a, specifically:
the Prewitt operator comprises two sets of 3 x 3 operator templates, GxIndicating detection of horizontal edge transverse template, GyRepresenting the vertical edge longitudinal template;
calculating the gradient value of the image by taking the sum of the transverse gradient and the longitudinal gradient mode of each pixel point, and expressing the gradient value by the following formula:
G=|Gx|+|Gy|
4. the 3D printing fault detection method based on edge detection and morphological image processing as claimed in claim 1, wherein the step S3 is to perform binarization processing on the image by using Otsu algorithm, specifically:
the number of pixels in the image having a gray value less than the threshold T is denoted N1The number of pixels having a pixel gray level greater than the threshold T is denoted by N2The single-channel gray scale M is 256, and the total number of pixels Sum, then
(1) The proportion of the number of background pixels in the whole image is as follows:
(2) the proportion of the number of foreground pixels in the whole image is as follows:
(3) average gray value of background:
(4) average gray value of foreground:
(5) gradation integrated value in 0 to M gradation sections:
μ=μ1*ω1+μ2*ω2
(6) the between-class variance:
g=ω1*(μ-μ1)2+ω2*(μ-μ2)2
(7) substituting the formulas (3), (4) and (5) into the formula (6) can obtain the final simplified formula
g=ω1*ω2*(μ1-μ2)2
And obtaining the threshold T which enables the inter-class variance to be maximum by adopting a traversal method, namely obtaining the threshold T.
5. The 3D printing fault detection method based on edge detection and morphological image processing as claimed in claim 1, wherein the image is subjected to a close operation in step S4 to close small cracks, specifically swelling before corrosion. Where Θ and ≦ indicate erosion and swelling, respectively.
A●B=(A⊕B)ΘB
6. The 3D printing fault detection method based on edge detection and morphological image processing as claimed in claim 1, wherein the step S5 is to perform an opening operation on the image to remove isolated dots, burrs and bridges, specifically corrosion first and then swelling. Where the sum indicates corrosion and swelling, respectively.
A○B=(AΘB)⊕B
7. The 3D printing fault detection method based on edge detection and morphological image processing as claimed in claim 1, wherein the step S6 adopts a bwphererm algorithm to obtain a 4-connected region of the image E and perform contour extraction of a binary image, specifically, to perform contour extraction by using bwpherem (E,4) function of Matlab. .
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111044933.7A CN113870299A (en) | 2021-09-07 | 2021-09-07 | 3D printing fault detection method based on edge detection and morphological image processing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111044933.7A CN113870299A (en) | 2021-09-07 | 2021-09-07 | 3D printing fault detection method based on edge detection and morphological image processing |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113870299A true CN113870299A (en) | 2021-12-31 |
Family
ID=78994635
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111044933.7A Pending CN113870299A (en) | 2021-09-07 | 2021-09-07 | 3D printing fault detection method based on edge detection and morphological image processing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113870299A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114862793A (en) * | 2022-05-06 | 2022-08-05 | 深圳市微特智能系统有限公司 | Method and system for comparing Printed Circuit Board (PCB) ink-jet printing images |
-
2021
- 2021-09-07 CN CN202111044933.7A patent/CN113870299A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114862793A (en) * | 2022-05-06 | 2022-08-05 | 深圳市微特智能系统有限公司 | Method and system for comparing Printed Circuit Board (PCB) ink-jet printing images |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106780486B (en) | Steel plate surface defect image extraction method | |
CN109191459B (en) | Automatic identification and rating method for continuous casting billet macrostructure center segregation defect | |
CN115345885B (en) | Appearance quality detection method for metal fitness equipment | |
CN107437243B (en) | Tire impurity detection method and device based on X-ray image | |
CN107742286B (en) | Method for detecting EL test crack defects of polycrystalline silicon solar cell | |
CN113592861B (en) | Bridge crack detection method based on dynamic threshold | |
CN111260616A (en) | Insulator crack detection method based on Canny operator two-dimensional threshold segmentation optimization | |
CN108765402B (en) | Non-woven fabric defect detection and classification method | |
CN109886954B (en) | Printed matter defect detection method | |
CN115063430B (en) | Electric pipeline crack detection method based on image processing | |
CN111738256B (en) | Composite CT image segmentation method based on improved watershed algorithm | |
CN107490583A (en) | A kind of intermediate plate defect inspection method based on machine vision | |
CN112734729B (en) | Water gauge water level line image detection method and device suitable for night light supplement condition and storage medium | |
CN111681213A (en) | Light guide plate line scratch defect detection method based on deep learning | |
CN113870299A (en) | 3D printing fault detection method based on edge detection and morphological image processing | |
CN114494318A (en) | Method for extracting cornea contour from cornea dynamic deformation video based on Otsu algorithm | |
CN102313740B (en) | Solar panel crack detection method | |
CN112200053B (en) | Form identification method integrating local features | |
CN117036259A (en) | Metal plate surface defect detection method based on deep learning | |
CN109886912A (en) | A kind of thrust bearing retainer detection method of surface flaw | |
CN112581433B (en) | Geometric information extraction method for thermal barrier coating cracks | |
CN111127450B (en) | Bridge crack detection method and system based on image | |
CN113538418A (en) | Tire X-ray image defect extraction model construction method based on morphological analysis | |
CN110322437B (en) | Fabric defect detection method based on automatic encoder and BP neural network | |
CN110458042B (en) | Method for detecting number of probes in fluorescent CTC |
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 |