CN108445011A - A kind of Defect Detection system and method based on deep learning - Google Patents
A kind of Defect Detection system and method based on deep learning Download PDFInfo
- Publication number
- CN108445011A CN108445011A CN201810200744.6A CN201810200744A CN108445011A CN 108445011 A CN108445011 A CN 108445011A CN 201810200744 A CN201810200744 A CN 201810200744A CN 108445011 A CN108445011 A CN 108445011A
- Authority
- CN
- China
- Prior art keywords
- image
- deep learning
- defect detection
- processing software
- picture
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The Defect Detection system based on deep learning that the invention discloses a kind of, including：Camera lens, PC hosts and the high in the clouds that belt transmission device, line are swept camera and be attached thereto；It is provided with image processing software in the PC hosts, Defect Detection and processing are carried out by image processing software, while handling result being shown to and being uploaded to high in the clouds, high in the clouds carries out big data analysis.The invention also discloses a kind of flaw detection methods based on deep learning, and present invention introduces the AI algorithms of deep learning, and efficient, high discrimination identification and extraction are carried out to unwanted visual characteristic in picture.Artificial less investment, image processing software detection performance maintenance cost are low.Compatibility is strong, in the case where facing model change, without separately developing in terms of image processing software algorithm, only needs product samples of collecting to carry out learning training more, just can meet the Defect Detection application of new product quickly.
Description
Technical field
The present invention relates to Defect Detection technical fields, particularly relate to a kind of Defect Detection system based on deep learning
And method.
Background technology
In industrial processes, Defect Detection is many product quality detections and its important step.Flaw is examined
It surveys device and the collected product surface image of industrial camera is subjected to defect identification process by image processing software, find out the flaw
Defect, while effectively classification and subsequent processing are carried out to flaw.There are several aspects in traditional image processing software：
One, image processing software opening parameter is more, need to put into great effort debugging and get to preferable detection performance.
Two, image processing software underlying algorithm versatility and functions expanding are weak, for new product and its client's new demand,
Personnel are needed to develop again.
Therefore, there is an urgent need for a kind of new technologies of design to improve its problem by the present inventor.
Invention content
In order to solve the above technical problems, the present invention provides a kind of Defect Detection system and method based on deep learning.
The technical scheme is that：
A kind of Defect Detection system based on deep learning, including：
Belt transmission device is used for transmission test product；
The camera lens that line is swept camera and is attached thereto, the two will be acquired for the test product on scanning belt transmitting device
To product surface image be sent at PC hosts；
It is provided with image processing software in the PC hosts, Defect Detection and processing are carried out by image processing software, together
When handling result is shown and is uploaded to high in the clouds；
The high in the clouds carries out big data analysis.
Preferably, described image processing software includes following module：
Preprocessing module, for being pre-processed to picture, the pretreatment includes but not limited to the greyscale transformation of image
With cutting for image；
Prediction module is predicted for passing through convolutional neural networks degree image, obtains prediction result；
Processing module obtains processing picture for handling prediction result；
Display module, for being shown to processing picture.
Preferably, further include a model library, connect with the prediction module, it is good that off-line training is provided in the model library
Convolutional neural networks model.
Preferably, the prediction result is single channel gray-scale map, and the image pixel value is in 0~255 distribution, each pixel value
Size indicates that the position is the score of flaw.
Preferably, in the processing module introduce two parameters of luminance threshold and area threshold to prediction result at
Reason.
A kind of flaw detection method based on deep learning, includes the following steps：
S1：Line sweeps the test product on camera scanning belt transmission device, and collected product surface image is sent to
At image processing software；
S2：Defect Detection and processing are carried out by image processing software, while handling result being shown and is uploaded to high in the clouds；
S3：High in the clouds carries out big data analysis.
Preferably, the step S2 is specifically included：
S21：Picture is pre-processed, the pretreatment includes but not limited to the sanction of the greyscale transformation and image of image
It cuts；
S22：It is predicted by convolutional neural networks degree image, obtains prediction result；
S23：Prediction result is handled, processing picture is obtained；
S24：Processing picture is shown.
Preferably, the step S22 is specifically included：The good convolutional neural networks model of off-line training is loaded, to pretreatment
Image afterwards is predicted, test result is obtained.
Preferably, the prediction result is single channel gray-scale map, and the image pixel value is in 0~255 distribution, each pixel value
Size indicates that the position is the score of flaw.
Preferably, two parameters of luminance threshold and area threshold are introduced in the step S23 to handle prediction result.
Using above-mentioned technical proposal, the present invention includes at least following advantageous effect：
Defect Detection system and method for the present invention based on deep learning introduces the AI algorithms of deep learning, right
Unwanted visual characteristic carries out efficient, high discrimination identification and extraction in picture.Automatic collection and in real time uploading detection sort data and set
Standby status data carries out big data analysis to high in the clouds.Artificial less investment, image processing software detection performance maintenance cost are low.It is simultaneous
Appearance ability is strong, in the case where facing model change, without separately developing in terms of image processing software algorithm, only needs receive more
Collect product sample and carry out learning training, just can meet the Defect Detection application of new product quickly.
Description of the drawings
Fig. 1 is the structural schematic diagram of the Defect Detection system of the present invention based on deep learning；
Fig. 2 is the flow chart of the flaw detection method of the present invention based on deep learning.
Wherein：1- lines sweep camera, 2- camera lenses, 3- light sources, 4- measurement center lines, 5- belt transmission devices, 6- test products, 7-
PC hosts, the high in the clouds 8-.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Embodiment 1
As shown in Figure 1, to meet a kind of Defect Detection system based on deep learning of the present embodiment, including：
Belt transmission device 5 is used for transmission tested production (silicon chip)；
The camera lens 2 that line is swept camera 1 and is attached thereto, the two, will for the test product 6 on scanning belt transmitting device 5
Collected product surface image is sent at PC hosts 7；
Preferably, further include a light source 3, setting ensures adopting for image with 4 side of measurement center line for being camera light filling
Collect precision.
It is provided with image processing software in the PC hosts 7, Defect Detection and processing are carried out by image processing software, together
When handling result is shown and is uploaded to high in the clouds 8；
The high in the clouds 8 carries out big data analysis.
Preferably, described image processing software includes following module：
Preprocessing module, for being pre-processed to picture, the pretreatment includes but not limited to the greyscale transformation of image
With cutting for image；
Prediction module is predicted for passing through convolutional neural networks degree image, obtains prediction result；
Processing module obtains processing picture for handling prediction result；
Display module, for being shown to processing picture.
Preferably, further include a model library, connect with the prediction module, it is good that off-line training is provided in the model library
Convolutional neural networks model.
Preferably, the prediction result is single channel gray-scale map, and the image pixel value is in 0~255 distribution, each pixel value
Size indicates that the position is the score of flaw.
Preferably, in the processing module introduce two parameters of luminance threshold and area threshold to prediction result at
Reason.
In the present embodiment, image processing software gets line and sweeps 1 collected picture of camera, needs to locate picture in advance
Reason.Software image preprocessing module includes mainly the greyscale transformation of image and cutting for image.Software is directed to flaw on the image
Distribution situation take the different modes that cuts, reduce calculation amount, shorten predicted time.The first is that flaw is only distributed in image
Marginal position, second is that flaw is only distributed in a certain region of image.Flaw is distributed on general image, then without
It cuts.
Image can be cut into four small images by the first situation, software, this four small images are upper on image respectively
The image of edge, right hand edge, left hand edge and lower edge position, while affine transformation is carried out to specified size to this four small picture
Image.
The second situation, software can cut out the specific region in image, while the image to cutting out carries out
Image of the affine transformation to specified size.
Image processing software loads the good convolutional neural networks model of off-line training, predicts the image after cutting,
Prediction result is a single channel gray-scale map, which indicates the position in 0~255 distribution, each pixel value size
For the score of flaw.Pixel value is bigger, and score is bigger, show the position be flaw probability it is bigger.
Image processing software introduces two parameters of luminance threshold and area threshold and handles prediction result.Luminance threshold
For inhibiting the lower region of score value.Less than the pixel value of this luminance threshold, then it is not considered as that there are flaws for the position.More than this
The pixel value of luminance threshold, then it is assumed that there are flaws for the position.The flaw of some small areas can be filtered out by area threshold
Region.Whether can be determined that prediction result figure with the presence of flaw by this operation, to judge the flake products be OK pieces or
NG pieces.For NG pieces, using radiation transformation matrix when cutting, position of the flaw before cutting i.e. on original image can be conversed
It sets.For OK pieces, software interface only shows artwork, and for NG pieces, software interface can mark flaw while showing artwork
Position and size.
High in the clouds 8 can be uploaded to per the testing results of flake products, high in the clouds 8 can integrate mass data, analyze slice, thin piece by the gross
NG and OK amounts carry out data analysis for software user of service.
Product surface Defect Detection is carried out in the AI algorithms to image processing software of the present embodiment introduction deep learning；Greatly
Amplitude improves the Defect Detection ability of image processing software, reduces false drop rate；Image processing software detection performance is easy to maintain,
Without debugging quantity of parameters, it is only necessary to personnel go safeguard one by depth learning technology train come model file；It improves
The versatility of image processing software, after product renewal, is developed again without labor intensive, and product samples of collecting only is needed to carry out more
Learning training updates model file, just can meet the Defect Detection application of new product quickly.
Embodiment 2
As shown in Fig. 2, on the basis of embodiment, a kind of Defect Detection side based on deep learning is present embodiments provided
Method includes the following steps：
S1：Line sweeps the test product 6 on 1 scanning belt transmitting device 5 of camera, and collected product surface image is sent
To image processing software；
S2：Defect Detection and processing are carried out by image processing software, while handling result being shown and is uploaded to high in the clouds
8；
S3：High in the clouds 8 carries out big data analysis.
Preferably, the step S2 is specifically included：
S21：Picture is pre-processed, the pretreatment includes but not limited to the sanction of the greyscale transformation and image of image
It cuts；
S22：It is predicted by convolutional neural networks degree image, obtains prediction result；
S23：Prediction result is handled, processing picture is obtained；
S24：Processing picture is shown.
Preferably, the step S22 is specifically included：The good convolutional neural networks model of off-line training is loaded, to pretreatment
Image afterwards is predicted, test result is obtained.
Preferably, the prediction result is single channel gray-scale map, and the image pixel value is in 0~255 distribution, each pixel value
Size indicates that the position is the score of flaw.
Preferably, two parameters of luminance threshold and area threshold are introduced in the step S23 to handle prediction result.
In the present embodiment, image processing software gets line and sweeps 1 collected picture of camera, needs to locate picture in advance
Reason.Software image preprocessing module includes mainly the greyscale transformation of image and cutting for image.Software is directed to flaw on the image
Distribution situation take the different modes that cuts, reduce calculation amount, shorten predicted time.The first is that flaw is only distributed in image
Marginal position, second is that flaw is only distributed in a certain region of image.Flaw is distributed on general image, then without
It cuts.
Image can be cut into four small images by the first situation, software, this four small images are upper on image respectively
The image of edge, right hand edge, left hand edge and lower edge position, while affine transformation is carried out to specified size to this four small picture
Image.
The second situation, software can cut out the specific region in image, while the image to cutting out carries out
Image of the affine transformation to specified size.
Image processing software loads the good convolutional neural networks model of off-line training, predicts the image after cutting,
Prediction result is a single channel gray-scale map, which indicates the position in 0~255 distribution, each pixel value size
For the score of flaw.Pixel value is bigger, and score is bigger, show the position be flaw probability it is bigger.
Image processing software introduces two parameters of luminance threshold and area threshold and handles prediction result.Luminance threshold
For inhibiting the lower region of score value.Less than the pixel value of this luminance threshold, then it is not considered as that there are flaws for the position.More than this
The pixel value of luminance threshold, then it is assumed that there are flaws for the position.The flaw of some small areas can be filtered out by area threshold
Region.Whether can be determined that prediction result figure with the presence of flaw by this operation, to judge the flake products be OK pieces or
NG pieces.For NG pieces, using radiation transformation matrix when cutting, position of the flaw before cutting i.e. on original image can be conversed
It sets.For OK pieces, software interface only shows artwork, and for NG pieces, software interface can mark flaw while showing artwork
Position and size.
High in the clouds 8 can be uploaded to per the testing results of flake products, high in the clouds 8 can integrate mass data, analyze slice, thin piece by the gross
NG and OK amounts carry out data analysis for software user of service.
Product surface Defect Detection is carried out in the AI algorithms to image processing software of the present embodiment introduction deep learning；Greatly
Amplitude improves the Defect Detection ability of image processing software, reduces false drop rate；Image processing software detection performance is easy to maintain,
Without debugging quantity of parameters, it is only necessary to personnel go safeguard one by depth learning technology train come model file；It improves
The versatility of image processing software, after product renewal, is developed again without labor intensive, and product samples of collecting only is needed to carry out more
Learning training updates model file, just can meet the Defect Detection application of new product quickly.
The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present invention.
Various modifications to these embodiments will be apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, of the invention
It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one
The widest range caused.
Claims (10)
1. a kind of Defect Detection system based on deep learning, which is characterized in that including：
Belt transmission device is used for transmission test product；
The camera lens that line is swept camera and is attached thereto, the two, will be collected for the test product on scanning belt transmitting device
Product surface image is sent at PC hosts；
It is provided with image processing software in the PC hosts, Defect Detection and processing are carried out by image processing software, simultaneously will
Handling result shows and is uploaded to high in the clouds；
The high in the clouds carries out big data analysis.
2. the Defect Detection system based on deep learning as described in claim 1, which is characterized in that described image processing software
Including following module：
Preprocessing module, for being pre-processed to picture, the pretreatment includes but not limited to greyscale transformation and the figure of image
Picture is cut；
Prediction module is predicted for passing through convolutional neural networks degree image, obtains prediction result；
Processing module obtains processing picture for handling prediction result；
Display module, for being shown to processing picture.
3. the Defect Detection system based on deep learning as claimed in claim 1 or 2, it is characterised in that：It further include a model
Library is connect with the prediction module, and the good convolutional neural networks model of off-line training is provided in the model library.
4. the Defect Detection system based on deep learning as claimed in claim 2, it is characterised in that：The prediction result is single
Channel gray-scale map, for the image pixel value in 0~255 distribution, each pixel value size indicates that the position is the score of flaw.
5. the Defect Detection system based on deep learning as claimed in claim 2, it is characterised in that：Draw in the processing module
Enter two parameters of luminance threshold and area threshold to handle prediction result.
6. a kind of flaw detection method based on deep learning, which is characterized in that include the following steps：
S1：Line sweeps the test product on camera scanning belt transmission device, and collected product surface image is sent to image
At processing software；
S2：Defect Detection and processing are carried out by image processing software, while handling result being shown and is uploaded to high in the clouds；
S3：High in the clouds carries out big data analysis.
7. the flaw detection method based on deep learning as claimed in claim 6, which is characterized in that the step S2 is specifically wrapped
It includes：
S21：Picture is pre-processed, the pretreatment includes but not limited to the greyscale transformation of image and cutting for image；
S22：It is predicted by convolutional neural networks degree image, obtains prediction result；
S23：Prediction result is handled, processing picture is obtained；
S24：Processing picture is shown.
8. the flaw detection method based on deep learning as claimed in claim 6, which is characterized in that the step S22 is specific
Including：The good convolutional neural networks model of off-line training is loaded, pretreated image is predicted, obtains test result.
9. the flaw detection method based on deep learning as claimed in claim 7 or 8, it is characterised in that：The prediction result
For single channel gray-scale map, for the image pixel value in 0~255 distribution, each pixel value size indicates that the position is the score of flaw.
10. the flaw detection method based on deep learning as claimed in claim 7, it is characterised in that：Draw in the step S23
Enter two parameters of luminance threshold and area threshold to handle prediction result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810200744.6A CN108445011B (en) | 2018-03-12 | 2018-03-12 | Defect detection system and method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810200744.6A CN108445011B (en) | 2018-03-12 | 2018-03-12 | Defect detection system and method based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108445011A true CN108445011A (en) | 2018-08-24 |
| CN108445011B CN108445011B (en) | 2021-06-11 |
Family
ID=63194102
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810200744.6A Active CN108445011B (en) | 2018-03-12 | 2018-03-12 | Defect detection system and method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108445011B (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109738595A (en) * | 2019-03-07 | 2019-05-10 | 福建工程学院 | A kind of system and method based on Internet of Things detection transformer fault |
| CN109903267A (en) * | 2019-01-22 | 2019-06-18 | 江苏恒力化纤股份有限公司 | A method of based on image processing techniques test network wire network degree |
| CN109975308A (en) * | 2019-03-15 | 2019-07-05 | 维库（厦门）信息技术有限公司 | A kind of surface inspecting method based on deep learning |
| CN110070521A (en) * | 2019-03-19 | 2019-07-30 | 广东工业大学 | A kind of 3D printing model flaw anticipation system and method for view-based access control model nerve study |
| CN110349108A (en) * | 2019-07-10 | 2019-10-18 | 北京字节跳动网络技术有限公司 | Handle method, apparatus, electronic equipment and the storage medium of image |
| CN111340798A (en) * | 2020-03-16 | 2020-06-26 | 浙江一木智能科技有限公司 | Application of deep learning in product appearance flaw detection |
| CN111598827A (en) * | 2019-02-19 | 2020-08-28 | 富泰华精密电子（郑州）有限公司 | Appearance flaw detection method, electronic device and storage medium |
| CN111784644A (en) * | 2020-06-11 | 2020-10-16 | 上海布眼人工智能科技有限公司 | Printing defect detection method and system based on deep learning |
| KR102259519B1 (en) * | 2020-08-28 | 2021-06-03 | 주식회사농심 | cutting error detecting system for pack unti |
| CN113189109A (en) * | 2021-01-15 | 2021-07-30 | 深圳锦绣创视科技有限公司 | Flaw judgment system and flaw judgment method based on artificial intelligence |
| CN113358020A (en) * | 2020-03-05 | 2021-09-07 | 青岛海尔工业智能研究院有限公司 | Machine vision detection system and method |
| KR102670982B1 (en) * | 2023-12-15 | 2024-05-30 | 재단법인 포항산업과학연구원 | Defect detecting method, defect detecting device about color steel plate, and computer program |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103411974A (en) * | 2013-07-10 | 2013-11-27 | 杭州赤霄科技有限公司 | Cloud big data-based planar material detection remote system and cloud big data-based planar material detection method |
| CN106290388A (en) * | 2016-08-03 | 2017-01-04 | 国网山东省电力公司电力科学研究院 | A kind of insulator breakdown automatic testing method |
| CN107123114A (en) * | 2017-04-21 | 2017-09-01 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of cloth defect inspection method and device based on machine learning |
-
2018
- 2018-03-12 CN CN201810200744.6A patent/CN108445011B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103411974A (en) * | 2013-07-10 | 2013-11-27 | 杭州赤霄科技有限公司 | Cloud big data-based planar material detection remote system and cloud big data-based planar material detection method |
| CN106290388A (en) * | 2016-08-03 | 2017-01-04 | 国网山东省电力公司电力科学研究院 | A kind of insulator breakdown automatic testing method |
| CN107123114A (en) * | 2017-04-21 | 2017-09-01 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of cloth defect inspection method and device based on machine learning |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109903267A (en) * | 2019-01-22 | 2019-06-18 | 江苏恒力化纤股份有限公司 | A method of based on image processing techniques test network wire network degree |
| CN109903267B (en) * | 2019-01-22 | 2021-01-05 | 江苏恒力化纤股份有限公司 | Method for testing network wire network degree based on image processing technology |
| CN111598827A (en) * | 2019-02-19 | 2020-08-28 | 富泰华精密电子（郑州）有限公司 | Appearance flaw detection method, electronic device and storage medium |
| CN109738595A (en) * | 2019-03-07 | 2019-05-10 | 福建工程学院 | A kind of system and method based on Internet of Things detection transformer fault |
| CN109975308A (en) * | 2019-03-15 | 2019-07-05 | 维库（厦门）信息技术有限公司 | A kind of surface inspecting method based on deep learning |
| CN110070521A (en) * | 2019-03-19 | 2019-07-30 | 广东工业大学 | A kind of 3D printing model flaw anticipation system and method for view-based access control model nerve study |
| CN110349108A (en) * | 2019-07-10 | 2019-10-18 | 北京字节跳动网络技术有限公司 | Handle method, apparatus, electronic equipment and the storage medium of image |
| CN110349108B (en) * | 2019-07-10 | 2022-07-26 | 北京字节跳动网络技术有限公司 | Method, apparatus, electronic device, and storage medium for processing image |
| CN113358020A (en) * | 2020-03-05 | 2021-09-07 | 青岛海尔工业智能研究院有限公司 | Machine vision detection system and method |
| CN111340798A (en) * | 2020-03-16 | 2020-06-26 | 浙江一木智能科技有限公司 | Application of deep learning in product appearance flaw detection |
| CN111784644A (en) * | 2020-06-11 | 2020-10-16 | 上海布眼人工智能科技有限公司 | Printing defect detection method and system based on deep learning |
| KR102259519B1 (en) * | 2020-08-28 | 2021-06-03 | 주식회사농심 | cutting error detecting system for pack unti |
| CN113189109A (en) * | 2021-01-15 | 2021-07-30 | 深圳锦绣创视科技有限公司 | Flaw judgment system and flaw judgment method based on artificial intelligence |
| KR102670982B1 (en) * | 2023-12-15 | 2024-05-30 | 재단법인 포항산업과학연구원 | Defect detecting method, defect detecting device about color steel plate, and computer program |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108445011B (en) | 2021-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108445011A (en) | A kind of Defect Detection system and method based on deep learning | |
| CN103913468B (en) | Many defects of vision checkout equipment and the method for large-scale LCD glass substrate on production line | |
| CN104256882B (en) | Based on reconstituted tobacco ratio measuring method in the pipe tobacco of computer vision | |
| CN104764744B (en) | Visual inspection device and method for inspecting freshness of poultry eggs | |
| CN110473178A (en) | A kind of open defect detection method and system based on multiple light courcess fusion | |
| WO2004070653A3 (en) | Image analysis system and method | |
| CN104198325B (en) | Stem ratio measuring method in pipe tobacco based on computer vision | |
| CN109307675A (en) | A kind of product appearance detection method and system | |
| CN106442525B (en) | Online detection method for walnut internal shriveling defect | |
| CN114638797A (en) | Method and device for detecting copper surface defects based on linear array camera | |
| CN114062366B (en) | Online detection method and system for winnowing stem-removing quality in tobacco shred production | |
| CN106326632A (en) | Object analyzing method and object analyzing system | |
| CN110298830A (en) | Fresh concrete based on convolutional neural networks vibrates apparent mass recognition methods | |
| US6683975B2 (en) | Apparatus and method for determining the dispersibility of a product in particulate form | |
| EP1360481A2 (en) | System for automated and remote histological analysis and new drug assessment | |
| CN111344555B (en) | Method and system for assessing the health status of a lactating mammal | |
| CN115661057A (en) | Industrial nondestructive testing system and method based on cloud edge cooperation and deep learning | |
| CN109490491A (en) | A kind of food safety integrated analysis device | |
| CN104198491B (en) | Based on expansive cut tobacco ratio measuring method in the tobacco shred of computer vision | |
| Niskanen et al. | Experiments with SOM based inspection of wood | |
| CN117115636B (en) | Algae and plankton analysis method, analyzer, medium and equipment | |
| CN113418919A (en) | Textile fiber component qualitative and quantitative online analysis system and method | |
| CN109632799A (en) | The mobile detection stage division of rice leaf nitrogen content based on machine vision, system and computer readable storage medium | |
| Liu et al. | A Novel Apple Size and Surface Quality Detection and Grading System | |
| CN118735873A (en) | Automatic detection method for body surface defects of chicken carcasses |
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 |