CN115901789A - Cloth flaw detection system based on machine vision - Google Patents
Cloth flaw detection system based on machine vision Download PDFInfo
- Publication number
- CN115901789A CN115901789A CN202211700096.3A CN202211700096A CN115901789A CN 115901789 A CN115901789 A CN 115901789A CN 202211700096 A CN202211700096 A CN 202211700096A CN 115901789 A CN115901789 A CN 115901789A
- Authority
- CN
- China
- Prior art keywords
- cloth
- detection system
- machine vision
- prn
- 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.)
- Pending
Links
- 239000004744 fabric Substances 0.000 title claims abstract description 53
- 238000001514 detection method Methods 0.000 title claims abstract description 43
- 230000007547 defect Effects 0.000 claims abstract description 28
- 238000004364 calculation method Methods 0.000 claims abstract description 14
- 238000005286 illumination Methods 0.000 claims abstract description 10
- 238000006243 chemical reaction Methods 0.000 claims abstract description 8
- 230000008859 change Effects 0.000 claims abstract description 5
- 230000006870 function Effects 0.000 claims description 25
- 230000002950 deficient Effects 0.000 claims description 16
- 238000003705 background correction Methods 0.000 claims description 14
- 238000000034 method Methods 0.000 claims description 14
- 238000007781 pre-processing Methods 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 8
- 230000001502 supplementing effect Effects 0.000 claims description 4
- FFBHFFJDDLITSX-UHFFFAOYSA-N benzyl N-[2-hydroxy-4-(3-oxomorpholin-4-yl)phenyl]carbamate Chemical compound OC1=C(NC(=O)OCC2=CC=CC=C2)C=CC(=C1)N1CCOCC1=O FFBHFFJDDLITSX-UHFFFAOYSA-N 0.000 claims description 2
- 238000013527 convolutional neural network Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 4
- 230000004069 differentiation Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011089 mechanical engineering Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000012634 optical imaging Methods 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a cloth flaw detection system based on machine vision, which relates to the technical field and provides the following scheme comprising the following stepsThe system comprises the following components: an illumination system: spotlight spotlights are a plurality of; emitting light rays into the conical area at the light position for simulating a searchlight; a CCD camera: adopting a plurality of CCD color cameras; based on machine vision, an illumination system consisting of spot lights and a CCD camera are integrated to shoot the surface of the cloth, and the shot cloth surface picture is preprocessed by performing real-time shade correction, a gradual change filter, a spot filter and contrast conversion by arranging a cloth flaw detection system in an industrial computer; processing by FasterRCNN, adjusting PRN network, PRN loss function calculation, classification cross entropy loss calculation, regression SmoothL 1 And calculating to perform positive processing, and reducing the workload of the positive processing by matching the pretreatment with the positive processing, and in addition, the defects on the surface of the cloth can be more accurately detected by adding the pretreatment.
Description
Technical Field
The invention relates to the technical field of flaw detection, in particular to a cloth flaw detection system based on machine vision.
Background
Machine vision is a comprehensive technology, including image processing, mechanical engineering technology, control, electric light source illumination, optical imaging, sensors, analog and digital video technology, computer software and hardware technology; a typical machine vision application system comprises an image capture module, a light source system, an image digitization module, a digital image processing module, an intelligent judgment decision module and a mechanical control execution module. The machine vision system has the basic characteristics of improving the production flexibility and the automation degree, and the machine vision system is commonly used for replacing the artificial vision in dangerous working environments which are not suitable for manual operation or occasions where the requirements of the artificial vision are difficult to meet.
In the cloth manufacturing process, along with the increase of processing layers, the probability of generating defects is increased, and in the case of various defects, the defects on the surface of the cloth are difficult to compare and detect one by one through artificial vision; the machine vision can help to manually detect flaws of mass products; however, the conventional flaw detection method is usually an R-CNN algorithm, i.e. a convolutional neural network CNN is used to directly predict the types and positions of different targets, although the detection accuracy is higher than that of other algorithms, the speed is slow, and for large batches of cloth, the R-CNN algorithm is somewhat unwilling; therefore, in view of the drawbacks of the prior art, those skilled in the art propose a machine vision-based cloth defect detection system.
Disclosure of Invention
In view of the above-mentioned drawbacks, the technical problem to be solved by the present invention is to provide a cloth defect detecting system based on machine vision;
the system comprises the following components:
an illumination system: the Spot light spotlights are a plurality of; emitting light rays into the conical area at the light position for simulating a searchlight;
a CCD camera: adopting a plurality of CCD color cameras;
an image acquisition card: the CCD camera is connected with a shooting unit of the CCD camera and is connected with an industrial computer;
an industrial computer;
the cloth defect detection system also comprises a preprocessing unit and a positive processing unit;
the preprocessing unit comprises four steps of real-time shading correction, a gradual change filter, a spot filter and contrast conversion;
the positive processing unit comprises target detection and identification, fasterRCNN processing, PRN network adjustment, PRN loss function calculation, classification cross entropy loss calculation and regression SmoothL 1 Six steps are calculated.
In the technical solution of the above system for detecting defects in cloth based on machine vision, it is preferable that the system further includes the following flow of detecting defects in cloth:
s1: the illumination system is used for supplementing light to the surface of the cloth, the CCD camera is used for shooting, recording and storing the image on the surface of the cloth into the image acquisition card, and the industrial computer is used for introducing the image in the image acquisition card into the flaw detection system;
s2: the industrial computer preprocesses the picture, carries out shading correction on the picture by adopting a general keyence shading correction algorithm, synchronously checks a gradient layer and a spot in the picture by a gradient filter and a spot filter, and finally carries out contrast conversion processing;
s3: after the picture is preprocessed, the picture is processed, the region with obvious difference of the picture in preprocessing is detected and identified, and the picture is processed by utilizing the Faster RCNN principleAdjusting the PRN network, calculating PRN loss function and classification cross entropy loss, and calculating regression Smoothl 1 And (4) loss of the function to obtain a detailed differentiated area in the picture.
In the technical solution of the cloth defect detecting system based on machine vision, preferably, the real-time shading correction in the preprocessing unit specifically adopts a shading correction algorithm of kirschner.
In the technical solution of the system for detecting defects in cloth based on machine vision, preferably, the adjusting PRN network in the positive processing specifically includes: the detection rate is improved by adjusting the size and the number of the candidate frames of the RPN in the fast RCNN, the anchor is moved, whether the candidate frames are defective areas is judged by utilizing a Softmax function, namely positive and negative are judged, two classification is realized, a detection target candidate area box is preliminarily extracted, after the defective areas are obtained, coordinate regression operation is carried out on the defective frames, more accurate spots of the defective areas are obtained, and the spots which are too small and exceed the boundaries are removed.
In the above technical solution of the cloth defect detecting system based on machine vision, preferably, the PRN loss function calculation in the process includes:
loss functions of the PRN network are categorical cross-entropy loss and regression SmoothL 1 The sum of losses, defined as:
in the above technical solution of the cloth defect detecting system based on machine vision, preferably, the classification cross entropy loss calculation in the forward processing is specifically defined as:
the classifier in the RPN network divides the candidate frame into foreground and background, labeled 1 and 0, respectively, N cls Representing the number of candidate boxes, P i Predicting an outline of a target for an in-frameThe ratio of the content to the content,
the probability of being an actual target is defined as:
in the above technical solution of the cloth defect detecting system based on machine vision, preferably, the regression SmoothL in the positive process 1 The calculation specifically comprises the following steps:
regression smoothL 1 The formula is defined as:
wherein, t i The predicted offset of the candidate box relative to the real box where the target is located,
is the actual offset of the candidate frame relative to the real frame where the target is located, t i And &>
Is defined as follows: />
t i ＝t xi +t yi +t wi +t hi
Wherein the content of the first and second substances,
L reg for regression loss, R is SmoothL 1 And the functions are defined as follows:
according to the technical scheme, compared with the prior art, the cloth flaw detection system based on the machine vision has the following beneficial effects:
based on machine vision, an illumination system consisting of spot lights and a CCD camera are integrated to shoot the surface of the cloth, and the shot cloth surface picture is preprocessed by performing real-time shade correction, a gradual change filter, a spot filter and contrast conversion by arranging a cloth flaw detection system in an industrial computer; processing by fast RCNN, adjusting PRN network, calculating PRN loss function, classifying cross entropy lossCalculating and regressing smoothL 1 And calculating to perform positive processing, and reducing the workload of the positive processing by matching the pretreatment with the positive processing, and in addition, the defects on the surface of the cloth can be more accurately detected by adding the pretreatment.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments of the present invention or the prior art will be briefly described and explained. It is obvious that the drawings in the following description are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.
FIG. 1 is a schematic diagram of a frame of a machine vision based cloth defect detection system according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the drawings in the embodiments of the present invention, and it is obvious that the embodiments described below are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In order to make the technical solution and implementation of the present invention more clearly explained and illustrated, several preferred embodiments for implementing the technical solution of the present invention are described below.
It will be understood that when an element is referred to as being "fixed" or "disposed" on another element, it can be directly on the other element or be indirectly disposed on the other element; when an element is referred to as being "connected to" another element, it can be directly connected to the other element or be indirectly connected to the other element.
Additionally, the terms herein: the references to "inner, outer", "front, rear", "left, right", "vertical, horizontal", "top, bottom", etc. are to be construed as references to orientations or positional relationships based on those shown in the drawings, merely for convenience of describing the application and to simplify the description, and are not intended to indicate or imply that the referenced device or element must have a particular orientation, be constructed and operated in a particular orientation, and are not to be construed as limiting the application.
The terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.
Example 1.
Cloth flaw detection process:
the illumination system is used for supplementing light to the surface of the cloth, the CCD camera is used for shooting, recording and storing images on the surface of the cloth into the image acquisition card, and the industrial computer is used for guiding the images in the image acquisition card into the flaw detection system; the industrial computer preprocesses the picture, carries out shading correction on the picture by adopting a general kirschner shading correction algorithm, synchronously checks a gradient layer and a spot in the picture by a gradient filter and a spot filter, and finally carries out contrast conversion processing.
After the picture is subjected to a preprocessing program, processing is started, areas with obvious differences of the picture in preprocessing are detected and identified, and the picture is processed by utilizing the fast RCNN principle; and meanwhile, the PRN network is adjusted, the detection rate is improved by adjusting the size and the number of candidate frames of the RPN network in the Faster RCNN, the anchor is moved, whether the detection target is a defective area or not is judged by utilizing a Softmax function, namely, positive and negative are judged, secondary classification is realized, a detection target candidate area box is preliminarily extracted, after the defective area is obtained, coordinate regression operation is carried out on a defective frame, more accurate propulses of the defective area are obtained, and meanwhile, the propulses which are too small and exceed the boundary are removed.
Calculating PRN loss functions and categorical cross-entropy losses, the loss functions of the PRN network being categorical cross-entropy losses and regression SmoothL 1 The sum of losses, defined as:
in the above technical solution of the cloth defect detecting system based on machine vision, preferably, the classification cross entropy loss calculation in the forward processing is specifically defined as:
the classifier in the RPN network divides the candidate frame into foreground and background, labeled 1 and 0 respectively cls Representing the number of candidate boxes, P i The probability of predicting the target in the frame is,
the probability of being an actual target is defined as:
finally, the regression smoothL is calculated 1 Loss of function, regression SmoothL 1 The formula is defined as:
wherein, t i Is the real frame of the candidate frame relative to the targetThe predicted offset of (a) is predicted,
is the actual offset of the candidate frame relative to the real frame where the target is located, t i And &>
Is defined as follows:
t i ＝t xi +t yi +t wi +t hi
wherein the content of the first and second substances,
L reg for regression loss, R is SmoothL 1 And the definition of the function are respectively as follows:
and finally, obtaining a detailed differentiation area in the picture in a cloth flaw detection system of the industrial computer.
Example 2.
The cloth flaw detection system needs system support, popular and general algorithms can be divided into two types, one type is an R-CNN algorithm, the other type is a Yolo one-stage algorithm such as SSD, only one convolutional neural network CNN is used for directly predicting the types and positions of different targets, the first type method is high in accuracy but low in speed, and the second type algorithm is high in speed but low in accuracy. The selection may be based on the number of detections and the speed.
On the basis, a database (both a local mode and a cloud server mode) can be arranged in the cloth defect detection system to provide a large number of defect samples for the system, and the cloth defect detection system can realize higher speed of a preprocessing link or select direct processing through a large number of samples provided by the large database and deep learning in a certain period.
Cloth flaw detection process:
the illumination system is used for supplementing light to the surface of the cloth, the CCD camera is used for shooting, recording and storing images on the surface of the cloth into the image acquisition card, and the industrial computer is used for guiding the images in the image acquisition card into the flaw detection system; the industrial computer preprocesses the picture, carries out shading correction on the picture by adopting a general keyence shading correction algorithm, synchronously checks a shading layer and a spot in the picture by a shading filter and a spot filter, and finally carries out contrast conversion processing.
After the picture is subjected to a preprocessing program, processing is started, areas with obvious differences of the picture in preprocessing are detected and identified, and the picture is processed by utilizing the fast RCNN principle; and adjusting the PRN network, improving the detection rate by adjusting the size and the number of candidate frames of the RPN network in the fast RCNN, moving the anchor, judging whether the detected region is a defective region by utilizing a Softmax function, namely judging positive and negative, realizing secondary classification, preliminarily extracting a detection target candidate region box, performing coordinate regression operation on a defective frame after the defective region is obtained, obtaining more accurate spots of the defective region, and simultaneously removing the spots which are too small and exceed the boundary.
Calculating PRN loss functions and categorical cross-entropy losses, the loss functions of the PRN network being categorical cross-entropy losses and regression SmoothL 1 The sum of losses, defined as:
in the above technical solution of the cloth defect detecting system based on machine vision, preferably, the classification cross entropy loss calculation in the forward processing is specifically defined as:
the classifier in the RPN network divides the candidate frame into foreground and background, labeled 1 and 0 respectively cls Representing the number of candidate boxes, P i The probability of predicting the target in the frame is,
the probability of being an actual target is defined as:
finally, the regression smoothL is calculated 1 Loss of function, regression SmoothL 1 The formula is defined as:
wherein, t i The predicted offset of the candidate box relative to the real box where the target is located,
is the actual offset of the candidate frame relative to the real frame where the target is located, t i And &>
Is defined as follows:
wherein, the first and the second end of the pipe are connected with each other,
L reg for regression loss, R is SmoothL 1 And the functions are defined as follows:
and finally, obtaining a detailed differentiation area in the picture in a cloth flaw detection system of the industrial computer.
Finally, it should be noted that the structures, ratios, sizes, and the like shown in the drawings are only used for matching the disclosure of the specification, so as to be understood and read by those skilled in the art, and are not used to limit the conditions that the present application can implement, so that the present application has no technical essence, and any structural modification, ratio relationship change, or size adjustment should still fall within the scope of the present application without affecting the efficacy and the achievable purpose of the present application.
As used herein, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one of 8230" does not exclude the presence of additional identical elements in the process, method, article, or apparatus that comprises the element.
The present invention is not limited to the above preferred embodiments, and any structural changes made under the teaching of the present invention shall fall within the protection scope of the present invention.
Claims (7)
1. Cloth flaw detection system based on machine vision, its characterized in that includes following system components:
an illumination system: spotlight spotlights; emitting light rays into the conical area at the light position for simulating a searchlight;
a CCD camera: adopting a plurality of CCD color cameras;
an image acquisition card: the CCD camera is connected with a shooting unit of the CCD camera and is connected with an industrial computer;
an industrial computer;
the cloth flaw detection system also comprises a preprocessing unit and a positive processing unit;
the preprocessing unit comprises four steps of real-time shading correction, a gradual change filter, a spot filter and contrast conversion;
the positive processing unit comprises target detection and identification, fasterRCNN processing, PRN network adjustment, PRN loss function calculation, classification cross entropy loss calculation and regression SmoothL 1 And six steps are calculated.
2. The machine vision based cloth defect detecting system of claim 1, further comprising a cloth defect detecting process:
s1: the illumination system is used for supplementing light to the surface of the cloth, the CCD camera is used for shooting, recording and storing images on the surface of the cloth into the image acquisition card, and the industrial computer is used for guiding the images in the image acquisition card into the flaw detection system;
s2: the industrial computer preprocesses the picture, carries out shading correction on the picture by adopting a general keyence shading correction algorithm, synchronously checks a gradient layer and a spot in the picture by a gradient filter and a spot filter, and finally carries out contrast conversion processing;
s3: after a picture is preprocessed, positive processing is started, areas with obvious differences of the picture in the preprocessing are detected and identified, the picture is processed by utilizing a FasterRCNN principle, a PRN network is adjusted, a PRN loss function and a classification cross entropy loss are calculated, and finally a regression SmoothL is calculated 1 And (4) loss of the function to obtain a detailed differentiated area in the picture.
3. The machine-vision-based cloth defect detection system of claim 1, wherein the real-time shading correction in the preprocessing unit specifically employs a shading correction algorithm of kirschner.
4. The machine vision based cloth defect detection system of claim 1, wherein the adjusting PRN network in the forward process is specifically: the detection rate is improved by adjusting the size and the number of the candidate frames of the RPN in the FasterRCNN, the anchor is moved, whether the candidate frames are defective areas is judged by utilizing a Softmax function, namely positive and negative are judged, two classification is realized, a detection target candidate area box is preliminarily extracted, after the defective areas are obtained, coordinate regression operation is carried out on the defective frames, more accurate spots of the defective areas are obtained, and small spots and spots exceeding the boundary are removed.
5. The machine vision based cloth defect detection system of claim 1, wherein the PRN penalty function calculation in-process is specifically:
loss functions of PRN networks are categorical cross-entropy losses and regression SmoothL 1 The sum of losses, defined as:
6. the machine vision-based cloth defect detection system of claim 1, wherein said in-process classification cross entropy loss calculation is specifically defined as:
the classifier in the RPN network divides the candidate frame into foreground and background, labeled 1 and 0, respectively, N cls Represents the number of candidate boxes, P i The probability of predicting the target in the frame is,
the probability of being an actual target is defined as:
7. the machine vision-based cloth defect detection system of claim 1, wherein the regression SmoothL in the forward process 1 The calculation specifically comprises the following steps:
regression smoothL 1 The formula is defined as:
wherein, t i The predicted offset of the candidate box relative to the real box where the target is located,
is the actual offset of the candidate frame relative to the real frame where the target is located, t i And &>
Is defined as follows:
t i ＝t xi +t yi +t wi +t hi
wherein, the first and the second end of the pipe are connected with each other,
L reg for regression loss, R is SmoothL 1 And the functions are defined as follows:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211700096.3A CN115901789A (en) | 2022-12-28 | 2022-12-28 | Cloth flaw detection system based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211700096.3A CN115901789A (en) | 2022-12-28 | 2022-12-28 | Cloth flaw detection system based on machine vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115901789A true CN115901789A (en) | 2023-04-04 |
Family
ID=86483904
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211700096.3A Pending CN115901789A (en) | 2022-12-28 | 2022-12-28 | Cloth flaw detection system based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115901789A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110827277A (en) * | 2019-11-26 | 2020-02-21 | 山东浪潮人工智能研究院有限公司 | Cloth flaw detection method based on yolo3 network |
| CN111047655A (en) * | 2020-01-10 | 2020-04-21 | 北京盛开互动科技有限公司 | High-definition camera cloth defect detection method based on convolutional neural network |
| CN111260614A (en) * | 2020-01-13 | 2020-06-09 | 华南理工大学 | Convolutional neural network cloth flaw detection method based on extreme learning machine |
| CN111398292A (en) * | 2020-04-07 | 2020-07-10 | 苏州哈工吉乐优智能装备科技有限公司 | Gabor filtering and CNN-based cloth defect detection method, system and equipment |
| WO2021164168A1 (en) * | 2020-02-20 | 2021-08-26 | 苏州浪潮智能科技有限公司 | Object detection method for image data and related device |
| CN115393265A (en) * | 2022-07-06 | 2022-11-25 | 北京理工大学 | Cross-cloth flaw accurate detection method based on visual field adaptive learning |
-
2022
- 2022-12-28 CN CN202211700096.3A patent/CN115901789A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110827277A (en) * | 2019-11-26 | 2020-02-21 | 山东浪潮人工智能研究院有限公司 | Cloth flaw detection method based on yolo3 network |
| CN111047655A (en) * | 2020-01-10 | 2020-04-21 | 北京盛开互动科技有限公司 | High-definition camera cloth defect detection method based on convolutional neural network |
| CN111260614A (en) * | 2020-01-13 | 2020-06-09 | 华南理工大学 | Convolutional neural network cloth flaw detection method based on extreme learning machine |
| WO2021164168A1 (en) * | 2020-02-20 | 2021-08-26 | 苏州浪潮智能科技有限公司 | Object detection method for image data and related device |
| CN111398292A (en) * | 2020-04-07 | 2020-07-10 | 苏州哈工吉乐优智能装备科技有限公司 | Gabor filtering and CNN-based cloth defect detection method, system and equipment |
| CN115393265A (en) * | 2022-07-06 | 2022-11-25 | 北京理工大学 | Cross-cloth flaw accurate detection method based on visual field adaptive learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Computer vision based system for apple surface defect detection | |
| US6061086A (en) | Apparatus and method for automated visual inspection of objects | |
| CN111650210A (en) | Burr detection method and detection system for high-speed high-precision lithium ion battery pole piece | |
| CN111007073A (en) | Method and system for online detection of part defects in additive manufacturing process | |
| CN108760747A (en) | A kind of 3D printing model surface defect visible detection method | |
| Liu et al. | Automatic detection technology of surface defects on plastic products based on machine vision | |
| CN104034638B (en) | The diamond wire online quality detecting method of granule based on machine vision | |
| CN111558542A (en) | Ceramic tile surface quality online detection sorting system and method | |
| CN113920090B (en) | Prefabricated rod appearance defect automatic detection method based on deep learning | |
| CN113177924A (en) | Industrial production line product flaw detection method | |
| CN115184359A (en) | Surface defect detection system and method capable of automatically adjusting parameters | |
| CN104483320A (en) | Digitized defect detection device and detection method of industrial denitration catalyst | |
| CN113418933B (en) | Flying shooting visual imaging detection system and method for detecting large-size object | |
| CN106709529B (en) | Visual detection method for photovoltaic cell color difference classification | |
| CN110763700A (en) | Method and equipment for detecting defects of semiconductor component | |
| CN110554052A (en) | artificial board surface defect detection method and system | |
| CN114136975A (en) | Intelligent detection system and method for surface defects of microwave bare chip | |
| CN106529510A (en) | Wrinkle recognition method and apparatus for capacitor thin film | |
| CN110567968A (en) | part defect detection method and device | |
| CN113781458A (en) | Artificial intelligence based identification method | |
| CN115901789A (en) | Cloth flaw detection system based on machine vision | |
| CN107833223B (en) | Fruit hyperspectral image segmentation method based on spectral information | |
| CN206114545U (en) | Singly take photograph wide visual field of camera vision thread detection device | |
| CN210775265U (en) | Artificial board surface defect detection system | |
| CN109013388B (en) | Automatic sorting system and method for ceramic balls |
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 |