CN110763705A - Deep learning identification method and system based on X-ray image and X-ray machine - Google Patents
Deep learning identification method and system based on X-ray image and X-ray machine Download PDFInfo
- Publication number
- CN110763705A CN110763705A CN201911043074.2A CN201911043074A CN110763705A CN 110763705 A CN110763705 A CN 110763705A CN 201911043074 A CN201911043074 A CN 201911043074A CN 110763705 A CN110763705 A CN 110763705A
- Authority
- CN
- China
- Prior art keywords
- defect
- ray image
- information data
- database
- ray
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N23/00—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Abstract
The invention relates to the technical field of detection, in particular to a deep learning identification method based on X-ray images, which comprises the steps of training a deep learning model according to the X-ray images of defective products, extracting defect information data in the X-ray images of the defective products, establishing a defect database, and storing the defect information data in the defect database; acquiring a reference value; acquiring an X-ray image of the workpiece, extracting the characteristics of the X-ray image, comparing the X-ray image with a defect database, and calculating whether the comparison degree is smaller than a reference value, if so, judging that the workpiece is qualified; if not, judging that the workpiece is unqualified, and storing the X-ray image with the comparison degree larger than the reference value into a defect database; according to the invention, through subsequent continuous workpiece detection, the defect information data of workpieces with different defects can be collected, more defect information data can be obtained, the identification speed and accuracy of defective products can be further ensured, and the identification effect is improved.
Description
Technical Field
The invention relates to the technical field of detection, in particular to a deep learning intelligent identification method based on X-ray images.
Background
In the industrial mass production process, people cannot look at the workpiece for a long time to see whether defects occur, and the workpiece is identified and checked by eyes with low quality efficiency and high error rate, and eye fatigue is easily generated. The detection precision, the production efficiency and the production automation degree can be greatly improved by using a machine detection method.
However, the existing machine needs human-computer cooperation and human eye identification, has a slow detection rate, needs a plurality of machines to detect simultaneously, and causes high cost, so that the intelligent identification capability and the automation degree of machine detection need to be improved.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: provided is a deep learning identification method based on X-ray images, which can improve the identification efficiency.
In order to solve the above technical problems, a first technical solution adopted by the present invention is:
a deep learning identification method based on X-ray images comprises the following steps
Step one, acquiring defect information data in an X-ray image of a defective product, establishing a defect database, and storing the defect information data into the defect database;
step two, acquiring a reference value;
step three, acquiring an X-ray image of the workpiece, extracting the characteristics of the X-ray image, comparing the X-ray image with a defect database, calculating whether the comparison degree is smaller than a reference value, and if so, judging that the workpiece is qualified; if not, determining that the workpiece is unqualified, and executing the next step;
and step four, storing the X-ray image with the comparison degree larger than the reference value into a defect database.
In order to solve the above technical problems, the second technical solution adopted by the present invention is:
a deep learning identification system based on X-ray images comprises:
the acquisition module is used for acquiring an X-ray image of the workpiece;
the characteristic extraction module is used for extracting the defect information data of the X-ray image of the workpiece;
the defect identification module is used for comparing the defect information data in the preset defect database with the defect information data of the workpiece and judging whether the workpiece is qualified or not;
and the result display module is used for displaying the judgment result of the defect identification module.
In order to solve the above technical problems, a third technical solution adopted by the present invention is:
an X-ray machine comprises a deep learning identification system based on X-ray images in the second technical scheme.
The invention has the beneficial effects that: through the defect database, a preliminary judgment mechanism can be established, and through subsequent continuous workpiece detection, the defect information data of workpieces with different defects can be collected to obtain more defect information data, so that the identification speed and the identification precision of defective products can be ensured, and the identification effect is improved; and by storing the X-ray image of the unqualified workpiece into the defect database, technicians can feed back the information in the defect database to the processing and manufacturing process of the workpiece, the processing process is improved, and the yield of the workpiece is improved.
Drawings
FIG. 1 is a schematic flow chart of an intelligent deep learning identification method based on X-ray images according to the present invention;
Detailed Description
In order to explain technical contents, achieved objects, and effects of the present invention in detail, the following description is made with reference to the accompanying drawings in combination with the embodiments.
Referring to fig. 1, a deep learning identification method based on X-ray images includes the following steps
Training a deep learning model according to an X-ray image of a defective product, extracting defect information data in the X-ray image of the defective product, establishing a defect database, and storing the defect information data into the defect database;
step two, acquiring a reference value; the reference value can be set to different reference values according to the requirements of different defects;
step three, acquiring an X-ray image of the workpiece, extracting the characteristics of the X-ray image, comparing the X-ray image with a defect database, calculating whether the comparison degree is smaller than a reference value, and if so, judging that the workpiece is qualified; if not, determining that the workpiece is unqualified, and executing the next step;
and step four, storing the X-ray image with the comparison degree larger than the reference value into a defect database.
From the above description, a preliminary judgment mechanism can be established through an initial deep learning model, and through subsequent continuous workpiece detection, the defect information data of workpieces with different defects can be collected to obtain more defect information data, so that the identification speed and accuracy of defective products can be ensured, and the identification effect is improved; and by storing the X-ray image of the unqualified workpiece into the defect database, technicians can feed back the information in the defect database to the processing and manufacturing process of the workpiece, the processing process is improved, and the yield of the workpiece is improved.
Further, the first step further includes:
training a deep learning model according to the X-ray image of the defective product, wherein the deep learning model is fast R-CNN, YOLOv3 or SSD and the like; and extracting defect information data in the X-ray image of the defective product, establishing a defect database, and classifying the defect information data according to the difference of the defects and storing the defect information data into the defect database.
From the above description, by classifying the defect information data, data comparison can be facilitated, and the recognition speed can be improved.
Further, an image preprocessing step is also included before the characteristic extraction of the X-ray image;
the image preprocessing step comprises the following steps: and (3) denoising and enhancing the X-ray image by using a mean filtering algorithm or a high-pass filtering algorithm, distinguishing and listing according to the defect types presented by the X-ray image, and selecting the classification of the defect database corresponding to the defect types.
From the above description, it can be known that, through the image preprocessing step, various defects of the X-ray image can be arranged and sorted, and correspond to the classification information in the defect database, and further, when performing the next comparison, the information in the entire defect database does not need to be compared.
Furthermore, the number of the reference values is several;
and if all the comparison degrees are smaller than the reference value, judging that the workpiece is qualified.
As is apparent from the above description, since there are more than one type of defect and more than one type of defect of one workpiece by several reference values, it is possible to set a plurality of different reference values according to the requirements.
Further, a defect acquisition step is required between the image preprocessing and the feature extraction:
all defects in the X-ray image are selected one by one.
Further, the method for extracting the features and calculating the comparison degree further comprises the following defect sorting steps:
different defects are mapped to defect information data within the classification of the defect database.
Further, the defect information data after feature extraction is compared with the defect information data in the classification of the defect database, and the comparison degree is calculated.
Further, the fourth step further includes:
and classifying and storing the X-ray images with the comparison degrees larger than the reference value into a defect database.
Further, the X-ray image is composed of a plurality of basic X-ray images.
From the above description, the X-ray image can be synthesized into a three-dimensional X-ray image from a plurality of basic X-ray images at different angles, so as to improve the recognition accuracy; the X-ray image synthesis can be used for pseudo-color or different-color filtering and distinguishing industries (or workpieces), and a color separation three-dimensional imaging technology or a color difference three-dimensional imaging technology is used for generating an image for three-dimensional color separation distinguishing display of articles with different densities in the image.
A deep learning identification method based on X-ray images comprises the following steps
Step one, acquiring defect information data in an X-ray image of a defective product, establishing a defect database, and storing the defect information data into the defect database;
step two, acquiring a reference value;
step three, acquiring an X-ray image of the workpiece, extracting the characteristics of the X-ray image, comparing the X-ray image with a defect database, calculating whether the comparison degree is smaller than a reference value, and if so, judging that the workpiece is qualified; if not, determining that the workpiece is unqualified, and executing the next step;
and step four, storing the X-ray image with the comparison degree larger than the reference value into a defect database.
Further, the first step further includes:
training the deep learning model according to the X-ray image of the defective product, extracting defect information data in the X-ray image of the defective product, establishing a defect database, and classifying the defect information data according to the difference of the defects and storing the defect information data into the defect database.
Further, an image preprocessing step is also included before the characteristic extraction of the X-ray image;
the image preprocessing step comprises the following steps: distinguishing and listing according to the defect types presented by the X-ray images, and selecting the classification of a defect database corresponding to the defect types;
the method also comprises the following defect sorting steps between the feature extraction and the comparison calculation:
different defects correspond to defect information data in the classification of the defect database;
the types of the defect information data after the characteristic extraction correspond to the classification of a defect database;
the fourth step further comprises:
and classifying and storing the X-ray images with the comparison degrees larger than the reference value into a defect database.
Further, the X-ray image is composed of a plurality of basic X-ray images.
A deep learning identification system based on X-ray images comprises:
the acquisition module is used for acquiring an X-ray image of the workpiece;
the characteristic extraction module is used for extracting the defect information data of the X-ray image of the workpiece;
the defect identification module is used for comparing the defect information data in the preset defect database with the defect information data of the workpiece and judging whether the workpiece is qualified or not;
and the result display module is used for displaying the judgment result of the defect identification module.
Further, the system also comprises an image preprocessing module which is used for distinguishing and listing according to the defect types presented by the X-ray images and selecting the classification of the defect database corresponding to the defect types;
the system also comprises a defect sorting module, and the types of the defect information data after the characteristic extraction correspond to the classification of the defect database.
Further, the system also comprises a writing module which is used for writing the defect information data of the unqualified workpiece into the defect database of the corresponding classification.
Furthermore, the X-ray images are synthesized by a plurality of basic X-ray images through a synthesis module.
An X-ray machine comprises the deep learning identification system based on X-ray images.
Example one
The defects of the existing defective product comprise:
welding seams: desoldering, cold bonding, unfused, lack of penetration, etc.;
casting: slag inclusion, loosening, sand holes, shrinkage holes, slag holes, air holes, cracks, cold shut and the like;
chain: empty rings, connecting ring cracks, parent metal cracks, depressions, openings, duckbills, welding line cracks, deformation, dislocation and the like;
a composite article: voids, faults, dislocations, foreign bodies, impurities, fiber uniformity, wrinkles, cracks, and the like.
A deep learning identification method based on X-ray images comprises the following steps:
training a fast R-CNN deep learning model according to an X-ray image of a defective product, extracting defect information data in the X-ray image of the defective product, generating and establishing a defect database by adopting a manual calibration mode for the defect data, and classifying the defect information data according to different defects and storing the defect information data into the defect database;
acquiring different defect information data reference values;
step three, obtaining an X-ray image of the workpiece,
using a mean filtering algorithm or a high-pass filtering algorithm to carry out denoising and enhancement on the X-ray image, distinguishing and listing according to the defect types presented by the X-ray image, selecting the classification of a defect database corresponding to the defect types,
selecting all defects in the X-ray image one by one,
extracting the characteristics of each defect of the X-ray image,
different defects are mapped to defect information data within the classification of the defect database,
comparing the defect information data after the characteristic extraction with defect information data in the classification of the defect database and calculating comparison degrees, and if all different comparison degrees are smaller than corresponding reference values, judging that the workpiece is qualified; if not, determining that the workpiece is unqualified, and executing the next step;
and step four, storing the X-ray images with the comparison degrees larger than the reference value into a defect database in a classified manner.
Example two
A deep learning identification method based on X-ray images comprises the following steps:
inputting a notch or a sunken X-ray image of a chain lacking into a preset YOLOv3 deep learning model, extracting defect information data in the X-ray image of a defective product, establishing a defect database, and classifying the defect information data according to different defects and storing the defect information data into the defect database;
step two, acquiring the reference value of 0.5 mm;
step three, obtaining an X-ray image of the chain, wherein the X-ray image is synthesized by a plurality of basic X-ray images through a chromatic aberration type stereo imaging technology;
using a mean filtering algorithm or a high-pass filtering algorithm to carry out denoising and enhancement on the X-ray image, distinguishing and listing according to the defect types presented by the X-ray image, selecting the classification of a defect database corresponding to the defect types,
selecting all defects in the X-ray image one by one,
extracting the characteristics of each defect of the X-ray image,
different defects are mapped to defect information data within the classification of the defect database,
comparing the defect information data after the characteristic extraction with defect information data in the classification of the defect database and calculating the comparison degree (tolerance), and if the tolerance of the notch or the recess and the defect information data in the classification of the defect database is less than 0.5mm, judging that the workpiece is qualified; if not, determining that the workpiece is unqualified, and executing the next step;
and step four, storing the X-ray images with the comparison degrees larger than the reference value into a defect database in a classified manner.
EXAMPLE III
A deep learning identification method based on X-ray images comprises the following steps:
inputting an X-ray image of the fiber uniformity of a composite material product into a preset SSD deep learning model, extracting defect information data in the X-ray image of a defective product, establishing a defect database, and classifying the defect information data according to different defects and storing the defect information data into the defect database;
step two, acquiring the reference value of 30 percent;
step three, obtaining an X-ray image of the chain, wherein the X-ray image is synthesized by a plurality of basic X-ray images through a color separation stereo imaging technology;
using a mean filtering algorithm or a high-pass filtering algorithm to carry out denoising and enhancement on the X-ray image, distinguishing and listing according to the defect types presented by the X-ray image, selecting the classification of a defect database corresponding to the defect types,
selecting all defects in the X-ray image one by one,
extracting the characteristics of each defect of the X-ray image,
different defects are mapped to defect information data within the classification of the defect database,
comparing the defect information data after the characteristic extraction with defect information data in the classification of the defect database and calculating the comparison degree (similarity), and if the fiber uniformity and the similarity of the defect information data in the classification of the defect database are both less than 30%, judging that the workpiece is qualified; if not, determining that the workpiece is unqualified, and executing the next step;
and step four, classifying and storing the X-ray images with the similarity larger than the reference value into a defect database.
Example four
A deep learning identification system based on X-ray images comprises:
the acquisition module is used for acquiring an X-ray image of the workpiece;
the image preprocessing module is used for carrying out denoising and enhancement on the X-ray image by using a mean filtering algorithm or a high-pass filtering algorithm, distinguishing and listing according to the defect types presented by the X-ray image and selecting the classification of a defect database corresponding to the defect types;
the characteristic extraction module is used for extracting the defect information data of the X-ray image of the workpiece;
the defect sorting module corresponds the types of the defect information data after the characteristic extraction to the classification of the defect database;
the defect identification module is used for comparing the defect information data in the preset defect database with the defect information data of the workpiece and judging whether the workpiece is qualified or not;
the result display module is used for displaying the judgment result of the defect identification module;
and the writing module is used for writing the defect information data of the unqualified workpiece into the defect database classified correspondingly.
EXAMPLE five
A deep learning identification system based on X-ray images comprises:
the acquisition module is used for acquiring an X-ray image of the workpiece; the X-ray image is synthesized by a plurality of basic X-ray images through a synthesis module;
the image preprocessing module is used for carrying out denoising and enhancement on the X-ray image by using a mean filtering algorithm or a high-pass filtering algorithm, distinguishing and listing according to the defect types presented by the X-ray image and selecting the classification of a defect database corresponding to the defect types;
the characteristic extraction module is used for extracting the defect information data of the X-ray image of the workpiece;
the defect sorting module corresponds the types of the defect information data after the characteristic extraction to the classification of the defect database;
the defect identification module is used for comparing the defect information data in the preset defect database with the defect information data of the workpiece and judging whether the workpiece is qualified or not;
the result display module is used for displaying the judgment result of the defect identification module;
and the writing module is used for writing the defect information data of the unqualified workpiece into the defect database classified correspondingly.
EXAMPLE six
An X-ray machine comprises the fifth embodiment.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all equivalent changes made by using the contents of the present specification and the drawings, or applied directly or indirectly to the related technical fields, are included in the scope of the present invention.
Claims (9)
1. A deep learning identification method based on X-ray images is characterized by comprising the following steps
Step one, acquiring defect information data in an X-ray image of a defective product, establishing a defect database, and storing the defect information data into the defect database;
step two, acquiring a reference value;
step three, acquiring an X-ray image of the workpiece, extracting the characteristics of the X-ray image, comparing the X-ray image with a defect database, calculating whether the comparison degree is smaller than a reference value, and if so, judging that the workpiece is qualified; if not, determining that the workpiece is unqualified, and executing the next step;
and step four, storing the X-ray image with the comparison degree larger than the reference value into a defect database.
2. The method of claim 1, wherein the first step further comprises:
training the deep learning model according to the X-ray image of the defective product, extracting defect information data in the X-ray image of the defective product, establishing a defect database, and classifying the defect information data according to the difference of the defects and storing the defect information data into the defect database.
3. The method according to claim 2, further comprising an image preprocessing step before the feature extraction of the X-ray image;
the image preprocessing step comprises the following steps: distinguishing and listing according to the defect types presented by the X-ray images, and selecting the classification of a defect database corresponding to the defect types;
the method also comprises the following defect sorting steps between the feature extraction and the comparison calculation:
different defects correspond to defect information data in the classification of the defect database;
the types of the defect information data after the characteristic extraction correspond to the classification of a defect database;
the fourth step further comprises:
and classifying and storing the X-ray images with the comparison degrees larger than the reference value into a defect database.
4. The method of claim 2, wherein the X-ray image is composed of a plurality of basic X-ray images.
5. A deep learning identification system based on X-ray images is characterized by comprising:
the acquisition module is used for acquiring an X-ray image of the workpiece;
the characteristic extraction module is used for extracting the defect information data of the X-ray image of the workpiece;
the defect identification module is used for comparing the defect information data in the preset defect database with the defect information data of the workpiece and judging whether the workpiece is qualified or not;
and the result display module is used for displaying the judgment result of the defect identification module.
6. The X-ray image based deep learning identification system according to claim 5,
the system also comprises an image preprocessing module which is used for distinguishing and listing according to the defect types presented by the X-ray images and selecting the classification of the defect database corresponding to the defect types;
the system also comprises a defect sorting module, and the types of the defect information data after the characteristic extraction correspond to the classification of the defect database.
7. The X-ray image based deep learning identification system of claim 6, further comprising a writing module for writing defect information data of unqualified workpieces into a defect database of corresponding classification.
8. The system of claim 5, wherein the X-ray image is composed of a plurality of basic X-ray images by a composition module.
9. An X-ray machine comprising the X-ray image based deep learning identification system according to any one of claims 5 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911043074.2A CN110763705A (en) | 2019-10-30 | 2019-10-30 | Deep learning identification method and system based on X-ray image and X-ray machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911043074.2A CN110763705A (en) | 2019-10-30 | 2019-10-30 | Deep learning identification method and system based on X-ray image and X-ray machine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110763705A true CN110763705A (en) | 2020-02-07 |
Family
ID=69334633
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911043074.2A Pending CN110763705A (en) | 2019-10-30 | 2019-10-30 | Deep learning identification method and system based on X-ray image and X-ray machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110763705A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113470018A (en) * | 2021-09-01 | 2021-10-01 | 深圳市信润富联数字科技有限公司 | Hub defect identification method, electronic device, device and readable storage medium |
| CN113758932A (en) * | 2021-09-07 | 2021-12-07 | 广东奥普特科技股份有限公司 | Lithium battery diaphragm defect vision system based on deep learning |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103984951A (en) * | 2014-04-25 | 2014-08-13 | 西南科技大学 | Automatic defect recognition method and system for magnetic particle testing |
| CN108711148A (en) * | 2018-05-11 | 2018-10-26 | 沈阳理工大学 | A kind of wheel tyre defect intelligent detecting method based on deep learning |
| CN109001230A (en) * | 2018-05-28 | 2018-12-14 | 中兵国铁（广东）科技有限公司 | Welding point defect detection method based on machine vision |
| WO2019176720A1 (en) * | 2018-03-13 | 2019-09-19 | オムロン株式会社 | Discrimination system, discrimination device, learning device, discrimination method and program |
-
2019
- 2019-10-30 CN CN201911043074.2A patent/CN110763705A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103984951A (en) * | 2014-04-25 | 2014-08-13 | 西南科技大学 | Automatic defect recognition method and system for magnetic particle testing |
| WO2019176720A1 (en) * | 2018-03-13 | 2019-09-19 | オムロン株式会社 | Discrimination system, discrimination device, learning device, discrimination method and program |
| CN108711148A (en) * | 2018-05-11 | 2018-10-26 | 沈阳理工大学 | A kind of wheel tyre defect intelligent detecting method based on deep learning |
| CN109001230A (en) * | 2018-05-28 | 2018-12-14 | 中兵国铁（广东）科技有限公司 | Welding point defect detection method based on machine vision |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113470018A (en) * | 2021-09-01 | 2021-10-01 | 深圳市信润富联数字科技有限公司 | Hub defect identification method, electronic device, device and readable storage medium |
| CN113758932A (en) * | 2021-09-07 | 2021-12-07 | 广东奥普特科技股份有限公司 | Lithium battery diaphragm defect vision system based on deep learning |
| CN113758932B (en) * | 2021-09-07 | 2022-11-25 | 广东奥普特科技股份有限公司 | Deep learning-based visual detection method for defects of lithium battery diaphragm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10391692B2 (en) | Injection molding system | |
| CN102253048B (en) | Machine vision detection method and system for detection of various products | |
| CN107607554A (en) | A kind of Defect Detection and sorting technique of the zinc-plated stamping parts based on full convolutional neural networks | |
| CN109685760B (en) | MATLAB-based SLM powder bed powder laying image convex hull depression defect detection method | |
| CN104914111A (en) | Strip steel surface defect on-line intelligent identification and detection system and detection method | |
| CN110992349A (en) | Underground pipeline abnormity automatic positioning and identification method based on deep learning | |
| CN111982916A (en) | Welding seam surface defect detection method and system based on machine vision | |
| CN112232399B (en) | Automobile seat defect detection method based on multi-feature fusion machine learning | |
| US20230306577A1 (en) | Cross-scale defect detection method based on deep learning | |
| CN111815555A (en) | Metal additive manufacturing image detection method and device combining anti-neural network with local binary | |
| CN112037219A (en) | Metal surface defect detection method based on two-stage convolution neural network | |
| CN110763705A (en) | Deep learning identification method and system based on X-ray image and X-ray machine | |
| CN110108712A (en) | Multifunctional visual sense defect detecting system | |
| CN110308151A (en) | A kind of surface of workpiece defect identification method and device based on machine vision | |
| CN110349125A (en) | A kind of LED chip open defect detection method and system based on machine vision | |
| CN110096980A (en) | Character machining identifying system | |
| CN112288741A (en) | Product surface defect detection method and system based on semantic segmentation | |
| CN115035092A (en) | Image-based bottle detection method, device, equipment and storage medium | |
| CN116678826A (en) | Appearance defect detection system and method based on rapid three-dimensional reconstruction | |
| CN114565314A (en) | Hot rolled steel coil end face quality control system and method based on digital twinning | |
| Ekambaram et al. | Identification of defects in casting products by using a convolutional neural network | |
| CN116978834A (en) | Intelligent monitoring and early warning system for wafer production | |
| CN107516309B (en) | Visual inspection method for defects of printing panel | |
| CN114119562A (en) | Brake disc outer surface defect detection method and system based on deep learning | |
| CN111899221A (en) | Appearance defect detection-oriented self-migration learning method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |