CN117893467A - Textile defect type identification method - Google Patents
Textile defect type identification method Download PDFInfo
- Publication number
- CN117893467A CN117893467A CN202311665287.5A CN202311665287A CN117893467A CN 117893467 A CN117893467 A CN 117893467A CN 202311665287 A CN202311665287 A CN 202311665287A CN 117893467 A CN117893467 A CN 117893467A
- Authority
- CN
- China
- Prior art keywords
- textile
- defect
- image
- defects
- images
- 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
- 230000007547 defect Effects 0.000 title claims abstract description 139
- 239000004753 textile Substances 0.000 title claims abstract description 117
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000004458 analytical method Methods 0.000 claims abstract description 51
- 238000012545 processing Methods 0.000 claims abstract description 21
- 230000011218 segmentation Effects 0.000 claims abstract description 15
- 238000004364 calculation method Methods 0.000 claims abstract description 14
- 238000001914 filtration Methods 0.000 claims abstract description 12
- 238000007781 pre-processing Methods 0.000 claims abstract description 10
- 238000013135 deep learning Methods 0.000 claims abstract description 7
- 238000012549 training Methods 0.000 claims abstract description 4
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 230000002950 deficient Effects 0.000 description 7
- 238000001514 detection method Methods 0.000 description 7
- 239000012634 fragment Substances 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 241000239290 Araneae Species 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000013467 fragmentation Methods 0.000 description 1
- 238000006062 fragmentation reaction Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30124—Fabrics; Textile; Paper
-
- 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
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to the technical field of textile defect identification, and discloses a textile defect type identification method, which comprises the following steps: constructing a database, wherein a defect analysis model is built in the database, and the defect analysis model comprises a comparison module and an analysis module; the defect analysis model is used for classifying textile defects by training and learning known textile defect images and defect positions in a deep learning algorithm; step two: all images of the textile surface defects are acquired and saved to a storage unit of a database. According to the invention, three preprocessing including filtering processing, gray processing and segmentation processing are performed on the acquired textile image, noise of the target image can be suppressed under the condition that the detail characteristics of the image are reserved by the filtering processing, the gray processing can convert the color RGB image into the gray image, the contrast of the image can be enhanced, and the storage capacity and the calculation complexity are reduced.
Description
Technical Field
The invention relates to the technical field of textile defect identification, in particular to a textile defect type identification method.
Background
When defects appear on the surface of the textile, the appearance of the subsequent fabric is affected, and even quality problems are caused. Defect detection and identification of defect types are key links in textile industry. In the textile industry, there are 50 or more textile defects, most of which are caused by machine faults and yarn problems, and these defects can be classified into six types of defects, dirty yarn, spider web, broken warp, doubled weft, thin yarn and loose yarn. Textiles undergo various inspections and tests during the production process and before entering the market, and are an essential step for identifying defects on the textile surface.
At present, the detection of textile defects is mainly carried out by means of a manual or visual detection system, the efficiency of the detection system is low, the accuracy of detection results is not high and gradually eliminated by the market, and the accuracy of the detection results is high although the detection efficiency is high, but the defects still exist: the detected textile defects cannot be classified and counted, which is unfavorable for improving related processes, so that the yield of the textiles cannot be improved. To this end, we propose a method for identifying the type of textile defects to solve the above-mentioned problems.
Disclosure of Invention
Technical problem to be solved
Aiming at the defects of the prior art, the invention provides a method for identifying the defect type of the textile, which has the advantages of being capable of identifying the defects of the textile, carrying out classified statistical treatment on the defects of the textile, being beneficial to improving related processes and the like, and solves the technical problems.
Technical proposal
In order to achieve the above purpose, the present invention provides the following technical solutions: a method for identifying a type of textile defect, comprising the steps of:
step one: constructing a database, wherein a defect analysis model is built in the database, and the defect analysis model comprises a comparison module and an analysis module;
the defect analysis model is used for classifying textile defects by training and learning known textile defect images and defect positions in a deep learning algorithm;
step two: acquiring all images of the textile surface defects and saving the images to a storage unit of a database;
step three: acquiring a surface image of a textile, and preprocessing the image;
step four: inputting the pretreated textile surface image into a defect analysis model, and capturing key features of the textile surface image through the defect analysis model;
step five: inputting key features captured by the defect analysis model into a comparison module, comparing the key features with the textile surface defect images stored in the storage unit, and judging whether the textile has defects or not through the analysis module;
step six: if the textile is judged to have defects, the analysis module marks the positions of the defects, notes the types of the defects and counts the number of different types of the defects.
In the third step, the high-definition industrial camera is adopted to acquire the textile surface image, the pretreatment of the image comprises segmentation processing, and the segmentation processing is used for segmenting the textile surface image into a plurality of parts by a segmentation algorithm based on areas.
As a preferred technical solution of the present invention, the preprocessing of the image in the third step further includes a filtering process and a gray scale process, and the enhancement of the gray scale image is performed, and the filtering process and the gray scale process are performed sequentially before the segmentation process.
As a preferred technical scheme of the invention, the comparison module is connected with the storage unit of the database, and the comparison module adopts a mean square error calculation formula to compare the preprocessed image with the pre-stored textile defect image, wherein the calculation formula is as follows:
where MSE is the mean square error value, n is the number of samples, yi is the true value of the ith sample, and y++i is the predicted value of the defect analysis model for the ith sample.
As a preferable technical scheme of the invention, the comparison module is connected with the analysis module, and the analysis module classifies the data calculated by the comparison module.
As a preferable technical scheme of the invention, the specific process of classification treatment is as follows:
determining the number of types of textile defects according to the number of known textile defect images, and marking the defects of different types by adopting a symbol;
comparing the pretreated textile image with each known textile defect image to obtain a mean square error value;
and respectively inputting the mean square error value calculated by the comparison module into an analysis module, selecting one data with the minimum mean square error value by the analysis module, judging which defect the data belongs to according to the symbol mark, and counting the symbol mark.
As a preferable technical scheme of the invention, the deep learning algorithm is specifically a convolutional neural network algorithm.
The preferable technical scheme of the invention is characterized in that: and step six, displaying the counted number of different defect types on a digital display screen.
Compared with the prior art, the invention provides a textile defect type identification method, which has the following beneficial effects:
1. according to the method, the known textile defect images are pre-stored in the storage unit of the database, when the images of the textiles are collected and then sent to the defect analysis model, the pre-processed images and the pre-stored known textile defect images are compared by using a mean square error method, the mean square error value between the two images is calculated through a mean square error method calculation formula, the larger the mean square error value is, the larger the difference between the two images is indicated, the larger the difference between the pre-processed images and the known textile defect images is, therefore, the textiles are judged to be good products, and the smaller the difference is, when the mean square error value is within a preset threshold range, the surface of the pre-processed images is very similar to the known textile defect images, the textiles are judged to be defective, and therefore whether the textiles are defective or not can be automatically detected.
2. According to the method, the number of the textile defect images is counted after the textile defect images are obtained, so that n types of the textile defect images are obtained, n types of obtained mean square error value marks are also obtained, if n mean square error values in the calculation result are all larger than a set threshold value, the fact that the textile is defective is indicated to be good is not found, otherwise, if one numerical value in the range of the set threshold value exists in the n types of the mean square error values, an analysis module finds out the marks corresponding to the numerical value, the defect is judged, the defect type is judged according to the marks, after the analysis module carries out classification statistics on the textile defects, the number of marks of each defect is displayed through a digital display screen so as to be convenient for a user to observe, and the process of relevant procedures for textile processing can be correspondingly adjusted through analyzing the number of different defects, so that the defect rate of the textile is reduced.
3. According to the invention, three preprocessing including filtering processing, gray processing and segmentation processing are performed on the acquired textile image, noise of a target image can be suppressed under the condition that the detail characteristics of the image are reserved by the filtering processing, the color RGB image can be converted into the gray image by the gray processing, the contrast of the image can be enhanced, the storage capacity and the computational complexity are reduced, the image fragmentation can be realized by the segmentation processing, the image fragments without characteristics are conveniently removed, the image fragments with the characteristics are reserved, and the computational complexity of subsequent comparison is further reduced.
Drawings
FIG. 1 is a schematic flow chart of the method of the present invention;
fig. 2 is a schematic diagram of the module structure of the present invention.
Detailed Description
Embodiments of the present invention are described in further detail below with reference to the accompanying drawings and examples. The following examples are illustrative of the invention but are not intended to limit the scope of the invention.
In the description of the present invention, unless otherwise indicated, the meaning of "a plurality" is two or more; the terms "upper," "lower," "left," "right," "inner," "outer," "front," "rear," "head," "tail," and the like are used as an orientation or positional relationship based on that shown in the drawings, merely to facilitate description of the invention and to simplify the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore should not be construed as limiting the invention. Furthermore, the terms "first," "second," "third," and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In the description of the present invention, it should be noted that, unless explicitly specified and limited otherwise, the terms "connected," "connected," and "connected" are to be construed broadly, and may be either fixedly connected, detachably connected, or integrally connected, for example; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.
Referring to fig. 1-2, a method for identifying a defect type of a textile comprises the following steps:
step one: constructing a database, wherein a defect analysis model is built in the database, and the defect analysis model comprises a comparison module and an analysis module;
the defect analysis model is used for classifying textile defects by training and learning known textile defect images and defect positions in a deep learning algorithm;
step two: acquiring all images of the textile surface defects and saving the images to a storage unit of a database;
step three: acquiring a surface image of a textile, and preprocessing the image;
step four: inputting the pretreated textile surface image into a defect analysis model, and capturing key features of the textile surface image through the defect analysis model;
step five: inputting key features captured by the defect analysis model into a comparison module, comparing the key features with the textile surface defect images stored in the storage unit, and judging whether the textile has defects or not through the analysis module;
step six: if the textile is judged to have defects, the analysis module marks the positions of the defects, notes the types of the defects and counts the number of different types of the defects.
In the third step, a high-definition industrial camera is adopted to acquire the textile surface image, the preprocessing of the image comprises segmentation processing, and the segmentation processing is that the textile surface image is segmented into a plurality of parts by a segmentation algorithm based on areas. The method has the advantages that the images on the outer surface of the textile are acquired through the high-definition industrial camera, the acquired images are sent to the database, the images to be acquired are divided into a plurality of parts after being subjected to filtering treatment and gray level treatment, the divided image fragments are convenient for comparing the follow-up images with the pre-stored known defective images of the textile, and the image fragments are more favorable for proposing characteristics after being fragmented, so that the image fragments without comparison value are removed, the calculated amount is reduced, and the speed of identifying the defects of the textile is improved.
Specifically, the preprocessing of the image in the third step further includes a filtering process and a gray scale process, and the enhancement of the gray scale image is performed, and the filtering process and the gray scale process are performed sequentially before the segmentation process. The method has the advantages that the filtering processing of the image is mainly carried out by carrying out convolution operation on the image, the pixel value is changed by utilizing the weighted average value of the local pixel value, the noise of the target image is suppressed under the condition of keeping the detail characteristic of the image as much as possible, and the original authenticity of the image is kept; the gray processing of the acquired textile image mainly converts the color RGB image into a gray image, so that the contrast of the image can be enhanced, and the storage capacity and the calculation complexity are reduced.
Specifically, the comparison module is connected with the storage unit of the database, and the comparison module adopts a mean square error calculation formula to compare the preprocessed image with the pre-stored textile defect image, wherein the calculation formula is as follows:
where MSE is the mean square error value, n is the number of samples, yi is the true value of the ith sample, and y++i is the predicted value of the defect analysis model for the ith sample. The method has the advantages that the mean square error method is used for comparing the preprocessed image with the pre-stored known textile defect image, the mean square error value between the two images is calculated through the calculation formula, the larger the mean square error value is, the larger the difference between the two images is, the larger the difference between the preprocessed image and the known textile defect image is, and therefore the textile is judged to be good, otherwise, the smaller the difference is, when the mean square error value is within a preset threshold range, the surface preprocessed image is very similar to the known textile defect image, and the textile is judged to be defective.
Specifically, the comparison module is connected with the analysis module, and the analysis module classifies the data calculated by the comparison module. The method has the advantages that when the mean square error value is calculated on the preprocessed image and the known textile defect image, the mean square error value of the preprocessed image and each different defect image is calculated once, the obtained mean square error values are marked as A1 and A2.
Specifically, the specific process of the classification treatment is as follows:
determining the number of types of textile defects according to the number of known textile defect images, and marking the defects of different types by adopting a symbol;
comparing the pretreated textile image with each known textile defect image to obtain a mean square error value;
and respectively inputting the mean square error value calculated by the comparison module into an analysis module, selecting one data with the minimum mean square error value by the analysis module, judging which defect the data belongs to according to the symbol mark, and counting the symbol mark. The method has the advantages that after the textile defect images are obtained, the number of the textile defect images is counted to obtain n types of textile defect, so that n types of obtained mean square error value marks are also included, if n mean square error values in the calculation result are all larger than a set threshold value, the fact that the textile is defective is indicated to be good, otherwise, if one numerical value in the range of the set threshold value exists in the n mean square error values, an analysis module finds out the mark corresponding to the numerical value, the defect of the textile is judged, and the defect type is judged according to the mark.
Specifically, the deep learning algorithm is specifically a convolutional neural network algorithm. The convolution neural network comprises convolution calculation and a feedforward neural network with a depth structure, and mainly comprises a convolution layer, a pooling layer and a full-connection layer, the convolution operation is carried out on an input image through convolution check, local features of the image are extracted, the size of the feature image is reduced through downsampling operation on the feature image after convolution, main feature information is further extracted, the feature image after convolution and pooling is unfolded into a one-dimensional vector, features can be extracted on the acquired textile image through the convolution neural network, and the calculation amount of subsequent comparison is greatly reduced.
Specifically, step six also includes displaying the counted number of different defect types on the digital display screen. The method has the advantages that after the analysis module performs classification statistics on the textile defects, the marking quantity of each defect can be displayed through the digital display screen so as to be convenient for a user to observe, and the process of relevant procedures for textile processing can be correspondingly adjusted by analyzing the quantity of different defects, so that the defective rate of the textiles can be reduced.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (9)
1. A textile defect type identification method is characterized in that: the method comprises the following steps:
step one: constructing a database, wherein a defect analysis model is built in the database, and the defect analysis model comprises a comparison module and an analysis module;
the defect analysis model is used for classifying textile defects by training and learning known textile defect images and defect positions in a deep learning algorithm;
step two: acquiring all images of the textile surface defects and saving the images to a storage unit of a database;
step three: acquiring a surface image of a textile, and preprocessing the image;
step four: inputting the pretreated textile surface image into a defect analysis model, and capturing key features of the textile surface image through the defect analysis model;
step five: inputting key features captured by the defect analysis model into a comparison module, comparing the key features with the textile surface defect images stored in the storage unit, and judging whether the textile has defects or not through the analysis module;
step six: if the textile is judged to have defects, the analysis module marks the positions of the defects, notes the types of the defects and counts the number of different types of the defects.
2. A method of identifying a textile defect type according to claim 1, wherein: and thirdly, acquiring the textile surface image by adopting a camera under the high-definition industry, wherein the preprocessing of the image comprises segmentation processing, and the segmentation processing is to divide the textile surface image into a plurality of parts by a segmentation algorithm based on areas.
3. A method of identifying a textile defect type according to claim 1, wherein: the preprocessing of the image in the third step further comprises a filtering process and a gray scale process, and the enhancement of the gray scale image is performed, wherein the filtering process and the gray scale process are sequentially performed before the segmentation process.
4. A method of identifying a textile defect type according to claim 1, wherein: the comparison module is connected with the storage unit of the database, and compares the preprocessed image with the pre-stored textile defect image by adopting a mean square error calculation formula, wherein the calculation formula is as follows:
5. where MSE is the mean square error value, n is the number of samples, yi is the true value of the ith sample, and y++i is the predicted value of the defect analysis model for the ith sample.
6. A method of identifying a textile defect type according to claim 1, wherein: the comparison module is connected with the analysis module, and the analysis module classifies the data calculated by the comparison module.
7. A method of identifying a type of textile defect according to claim 5, wherein: the specific process of the classification treatment is as follows:
determining the number of types of textile defects according to the number of known textile defect images, and marking the defects of different types by adopting a symbol;
comparing the pretreated textile image with each known textile defect image to obtain a mean square error value;
and respectively inputting the mean square error value calculated by the comparison module into an analysis module, selecting one data with the minimum mean square error value by the analysis module, judging which defect the data belongs to according to the symbol mark, and counting the symbol mark.
8. A method of identifying a textile defect type according to claim 1, wherein: the deep learning algorithm is specifically a convolutional neural network algorithm.
9. A method of identifying a textile defect type according to claim 1, wherein: and step six, displaying the counted number of different defect types on a digital display screen.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311665287.5A CN117893467B (en) | 2023-12-06 | 2023-12-06 | Textile defect type identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311665287.5A CN117893467B (en) | 2023-12-06 | 2023-12-06 | Textile defect type identification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117893467A true CN117893467A (en) | 2024-04-16 |
| CN117893467B CN117893467B (en) | 2024-09-06 |
Family
ID=90649706
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311665287.5A Active CN117893467B (en) | 2023-12-06 | 2023-12-06 | Textile defect type identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117893467B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107341499A (en) * | 2017-05-26 | 2017-11-10 | 昆明理工大学 | It is a kind of based on non-formaldehyde finishing and ELM fabric defect detection and sorting technique |
| CN108133472A (en) * | 2017-11-28 | 2018-06-08 | 长乐万达纺织机械有限公司 | Textile fabric defect inspection method and textile fabric defect detecting device |
| CN112907519A (en) * | 2021-01-29 | 2021-06-04 | 广州信邦智能装备股份有限公司 | Metal curved surface defect analysis system and method based on deep learning |
| CN114463314A (en) * | 2022-02-11 | 2022-05-10 | 磐柔(厦门)工业智能有限公司 | Wafer defect detection method and system based on color difference shadow model |
| CN114708235A (en) * | 2022-04-12 | 2022-07-05 | 嘉兴炬星数字科技有限公司 | Textile detection system based on artificial intelligence discernment AI |
| CN116385426A (en) * | 2023-04-28 | 2023-07-04 | 中南大学 | Textile surface defect detection method and related equipment |
| CN116823817A (en) * | 2023-08-28 | 2023-09-29 | 江苏州际数码印花有限公司 | Textile jacquard defect detection method and system based on deep learning |
| CN117115147A (en) * | 2023-10-19 | 2023-11-24 | 山东华盛创新纺织科技有限公司 | Textile detection method and system based on machine vision |
-
2023
- 2023-12-06 CN CN202311665287.5A patent/CN117893467B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107341499A (en) * | 2017-05-26 | 2017-11-10 | 昆明理工大学 | It is a kind of based on non-formaldehyde finishing and ELM fabric defect detection and sorting technique |
| CN108133472A (en) * | 2017-11-28 | 2018-06-08 | 长乐万达纺织机械有限公司 | Textile fabric defect inspection method and textile fabric defect detecting device |
| CN112907519A (en) * | 2021-01-29 | 2021-06-04 | 广州信邦智能装备股份有限公司 | Metal curved surface defect analysis system and method based on deep learning |
| CN114463314A (en) * | 2022-02-11 | 2022-05-10 | 磐柔(厦门)工业智能有限公司 | Wafer defect detection method and system based on color difference shadow model |
| CN114708235A (en) * | 2022-04-12 | 2022-07-05 | 嘉兴炬星数字科技有限公司 | Textile detection system based on artificial intelligence discernment AI |
| CN116385426A (en) * | 2023-04-28 | 2023-07-04 | 中南大学 | Textile surface defect detection method and related equipment |
| CN116823817A (en) * | 2023-08-28 | 2023-09-29 | 江苏州际数码印花有限公司 | Textile jacquard defect detection method and system based on deep learning |
| CN117115147A (en) * | 2023-10-19 | 2023-11-24 | 山东华盛创新纺织科技有限公司 | Textile detection method and system based on machine vision |
Non-Patent Citations (1)
| Title |
|---|
| 杨学志;沈晶;: "采用小波框架的纺织品缺陷分类方法", 工程图学学报, no. 05, 15 October 2009 (2009-10-15), pages 108 - 112 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117893467B (en) | 2024-09-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114549522B (en) | Textile quality detection method based on target detection | |
| CN114757900B (en) | Artificial intelligence-based textile defect type identification method | |
| CN111223093A (en) | AOI defect detection method | |
| CN111402226A (en) | Surface defect detection method based on cascade convolution neural network | |
| CN113658131B (en) | Machine vision-based tour ring spinning broken yarn detection method | |
| CN108562589A (en) | A method of magnetic circuit material surface defect is detected | |
| CN111047655B (en) | High-definition camera cloth defect detection method based on convolutional neural network | |
| CN108364291A (en) | Grey cloth rapid detection method based on computer vision technique | |
| CN109886954B (en) | Printed matter defect detection method | |
| CN111667475B (en) | Machine vision-based Chinese date grading detection method | |
| CN115131348B (en) | Method and system for detecting textile surface defects | |
| CN116990323B (en) | High-precision printing plate visual detection system | |
| CN115266732B (en) | Carbon fiber tow defect detection method based on machine vision | |
| CN115100206A (en) | Printing defect identification method for textile with periodic pattern | |
| CN113936001B (en) | Textile surface flaw detection method based on image processing technology | |
| CN115205209A (en) | Monochrome cloth flaw detection method based on weak supervised learning | |
| CN115841731A (en) | Infrared-monitoring park fire early warning method | |
| CN114565607A (en) | Fabric defect image segmentation method based on neural network | |
| CN117893467B (en) | Textile defect type identification method | |
| CN114820597B (en) | Smelting product defect detection method, device and system based on artificial intelligence | |
| CN116309493A (en) | Method and system for detecting defects of textile products | |
| CN114820453A (en) | Method for detecting surface flaws of packaged filament based on deep learning | |
| CN114511851A (en) | Hairspring algae cell statistical method based on microscope image | |
| CN114092441A (en) | Product surface defect detection method and system based on dual neural network | |
| 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |