CN113506255A - Automatic clothing sewing line defect detection method based on high-precision digital image - Google Patents
Automatic clothing sewing line defect detection method based on high-precision digital image Download PDFInfo
- Publication number
- CN113506255A CN113506255A CN202110740826.1A CN202110740826A CN113506255A CN 113506255 A CN113506255 A CN 113506255A CN 202110740826 A CN202110740826 A CN 202110740826A CN 113506255 A CN113506255 A CN 113506255A
- Authority
- CN
- China
- Prior art keywords
- defects
- sewing
- clothing
- garment
- image
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- 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
-
- 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
Abstract
The invention discloses a clothing sewing line defect automatic detection method based on a high-precision digital image, which is used for dividing clothing parts according to the positions of clothing sewing lines and classifying the sewing line defects according to the clothing parts; acquiring a high-definition image of the garment to be measured by using an independently built image acquisition device provided with a high-precision industrial camera; preprocessing the acquired image, manually marking the garment part, and establishing a standard garment part data set; training an improved Faster R-CNN network by using the data set to realize automatic identification and division of clothing parts; extracting clothing part images, and extracting stitches by using a Gabor operator with the direction consistent with the direction of stitches of each clothing part; obtaining stitch segments with stitches as a unit by utilizing a Zhang-Suen algorithm, extracting the suture characteristics according to the stitch segments, identifying the defects of the garment sutures by using a classifier, and judging the categories of the defects.
Description
Technical Field
The invention belongs to the field of clothes quality detection, and particularly relates to a method for automatically detecting clothes sewing defects based on a high-precision digital image.
Background
The quality detection and control of the clothes are the key links of the clothes production. The quality inspection of the finished clothing is a process of comprehensively inspecting the finished clothing before delivery, mainly inspects problems occurring in the sewing and processing process and the after-finishing process of the clothing, and is the final link for controlling the quality of the clothing. The garment sewing defects are defects generated in the garment sewing processing process, are main factors influencing the garment grading standard, and directly influence the aesthetic property of the garment and the quality of the garment. Therefore, the detection of the sewing defects of the clothes is very important before the clothes are packed and delivered from a factory.
At present, clothing enterprises mainly detect the defects of clothing sewing threads in a manual visual inspection mode. In the production process of a large number of clothes, the traditional manual detection has the problems of high omission factor, low efficiency, high labor intensity and the like. With the rapid development of technologies in the fields of machine vision, digital image processing and the like, it becomes possible to solve the garment defect detection based on digital images. The method utilizes the high-precision digital image technology to detect the clothing sewing defects, not only can solve the problems in the manual detection process, but also improves the clothing defect detection precision, and ensures the clothing quality entering the sale market. Therefore, the quality detection industrialization degree of the clothing enterprises is effectively improved, the transformation and upgrading of the clothing industry are promoted, and the intellectualization pace of the clothing industry is accelerated.
Disclosure of Invention
The invention aims to provide a method for automatically detecting clothing sewing defects based on high-precision digital images, aiming at the defects in the prior art, the traditional image processing technology, the deep learning technology and the like are combined, and the automatic identification and division of clothing parts are realized through an improved Faster R-CNN network. Obtaining stitch segments with stitches as a unit by utilizing a Zhang-Suen algorithm, extracting the suture characteristics according to the stitch segments, identifying the defects of the garment sutures by using a classifier, and judging the categories of the defects. The invention solves the problems of low efficiency and missed inspection, optimizes the production flow and improves the quality of clothes.
In order to solve the technical problems, the following technical scheme is adopted:
a clothing sewing line defect automatic detection method based on a high-precision digital image is characterized by comprising the following steps:
(1) dividing garment parts according to positions of garment stitches, and classifying stitch defects according to the garment parts;
(2) acquiring a high-definition image of the garment to be measured by using an independently built image acquisition device provided with a high-precision industrial camera;
(3) preprocessing the image acquired in the step (2), manually marking the garment part, and establishing a standard garment part data set;
(4) training a Faster R-CNN network by using the garment part data set in the step (3) to realize automatic identification and division of garment parts;
(5) extracting the clothing part image automatically identified in the step (4), extracting the suture lines by using a Gabor operator with the direction consistent with the direction of the suture lines of each clothing part, and performing binarization processing on the extracted suture lines by using a threshold value;
(6) and (4) processing the binaryzation image of the sewing thread of the clothing part in the step (5) by utilizing a Zhang-Suen algorithm to obtain stitch segments with the stitch as a unit, extracting the sewing thread characteristics according to the stitch segments, and identifying the sewing defects of the clothing by using a classifier and judging the category of the sewing defects.
Preferably, the step (1) can divide the garment into six parts, namely a collar part, a front fly, a hem, a split part of the garment body and the sleeves and the cuffs, according to the distribution position of the sewing threads in the garment.
Preferably, the garment stitch defects can be divided into the following parts according to garment parts: collar defects including broken threads, jumpers, heavy threads and thread ends; defects at the front fly, including broken thread, jumper, heavy thread and thread end; the defects at the lower hem comprise broken lines, jumper wires, heavy lines and thread ends; the defects of the splicing part of the clothes body and the sleeves comprise sewing folds, sewing holes and thread ends; the defects at the cuffs comprise broken lines, jumpers, heavy lines and thread ends.
Preferably, the acquisition equipment in the step (2) comprises a Mars4072S-24uc industrial camera, a stepping motor, a high-performance computer, a set of conveyor belts and an LED lamp strip, wherein the lens of the Mars 4072-4072S-24 uc industrial camera is a VT-LEM1614CBMPB (16mm) lens.
Preferably, the step (4) of preprocessing the data set of the garment part is to perform size normalization processing on the acquired image, expand the data set through rotation and translation, and mark different parts of the garment with rectangular frames of different colors and name the garment.
Preferably, the Faster R-CNN network backbone network in the step (4) adopts a Resnet network, removes the full connection layer, only leaves the convolution layer, and outputs the feature map after down sampling; adjusting the sizes of the first three layers of convolution layers in the Resnet network, placing the convolution layers in front of the RPN convolution layers, and enhancing shallow layer characteristics such as clothing texture, gradient and the like to enable the generated frame to be detected to be more accurate; the region of interest Pooling (RoI Pooling) in the Faster R-CNN network is improved to be the region of interest calibration RoI Align, so that the quantization error is reduced, and a target rectangular frame in the garment part detection is more accurate.
Preferably, the step (5) extracts the clothing part image according to the automatically identified rectangular frame coordinates of the clothing part, sets the Gabor operator direction according to the sewing direction of the clothing part, and extracts the sewing line by using the Gabor operator with the direction consistent with the sewing direction.
Preferably, the step (6) utilizes a Zhang-Suen algorithm to extract a central axis with a single-pixel width from a binary image of the suture to obtain a stitch segment with a stitch as a unit; taking the central axis end point as a starting point, recording all connected stitch segments, and obtaining five groups of characteristic parameters of the end point position, the length, the central position, the distance between the end point position and the central position of the adjacent segment and the straightness of each stitch segment; the feature vectors with the dimension of 1 × 5 are input into a Support Vector Machine (SVM) classifier which takes a Radial Basis Function (RBF) as a kernel function to obtain a classification result.
Due to the adoption of the technical scheme, the method has the following beneficial effects:
compared with the prior art, the method divides the garment parts according to the positions of the garment stitches, and classifies the stitch defects according to the garment parts; acquiring a high-definition image of the garment to be measured by using an independently built image acquisition device provided with a high-precision industrial camera; preprocessing the acquired image, manually marking the garment part, and establishing a standard garment part data set; training an improved Faster R-CNN network by using the data set to realize automatic identification and division of clothing parts; extracting clothing part images, and extracting stitches by using a Gabor operator with the direction consistent with the direction of stitches of each clothing part; obtaining stitch segments with stitches as a unit by utilizing a Zhang-Suen algorithm, extracting the suture characteristics according to the stitch segments, identifying the defects of the garment sutures by using a classifier, and judging the categories of the defects. The sewing problems of the clothes produced in each batch are filed, the results are fed back to the production link, the production flow is optimized, the problems of low efficiency and missing detection of manual detection are solved, and the development of the clothes industry chain is effectively promoted.
Drawings
The invention will be further described with reference to the accompanying drawings in which:
FIG. 1 is a flow chart of the present invention for automatically detecting defects in clothing stitches;
FIG. 2 is a classification chart of sewing defects of the garment according to the garment parts;
FIG. 3 is a schematic view of an image acquisition device;
FIG. 4 is an illustration of the garment portion identified by Faster R-CNN.
Detailed Description
For explaining the purposes, technical schemes and advantages of the invention, the blouse is selected as the garment in the invention and is explained in detail with the accompanying drawings. It should be understood that the specific examples described herein are intended to be illustrative only and are not intended to be limiting.
FIG. 1 is a flow chart of the method for automatically detecting defects of clothing stitches based on high-precision digital images, which specifically comprises the following steps:
(1) the clothing parts are divided according to the positions of clothing stitches, and the stitch defects are classified according to the clothing parts. As shown in figure 2, the clothes can be divided into six parts, namely a collar part, a front fly, a lap, a splicing part of a clothes body and sleeves and a sleeve opening according to the distribution position of the sewing thread in the clothes. The sewing defects of the clothes can be divided into the following parts according to the parts of the clothes: collar defects including broken threads, jumpers, heavy threads and thread ends; defects at the front fly, including broken thread, jumper, heavy thread and thread end; the defects at the lower hem comprise broken lines, jumper wires, heavy lines and thread ends; the defects of the splicing part of the clothes body and the sleeves comprise sewing folds, sewing holes and thread ends; the defects at the cuffs comprise broken lines, jumpers, heavy lines and thread ends.
(2) And acquiring a high-definition image of the garment to be detected by using image acquisition equipment provided with a high-precision industrial camera. As shown in fig. 3, the acquisition equipment comprises Mars4072S-24uc industrial cameras using VT-LEM1614CBMPB (16mm) lenses, one stepper motor, one high-performance computer, one set of conveyor belts, and LED strips. Wherein the stepping motor and the conveyor belt form a conveying device for conveying the clothes to be detected. The LED lamp area sets up and waits to examine clothing all around, makes and waits to examine clothing illumination even, improves and gathers the picture quality. The camera is controlled by the computer to take a picture, and the acquired clothing image information is stored in the computer.
(3) Establishing a standard clothing part data set, collecting a large number of clothing sample images of various styles, preprocessing the obtained images, normalizing the sizes of the images, expanding the data set through rotation and translation, facilitating the more accurate identification of the clothing parts by the model, and then labeling different parts of the clothing with rectangular frames of different colors and naming.
(4) The improved Faster R-CNN network backbone network adopts the Resnet network, removes the full connection layer therein, only leaves the convolution layer, and outputs the characteristic diagram after down sampling; adjusting the sizes of the first three layers of convolution layers in the Resnet network, placing the convolution layers in front of the RPN convolution layers, and enhancing shallow layer characteristics such as clothing texture, gradient and the like to enable the generated frame to be detected to be more accurate; region of interest Pooling (RoI Pooling) in the fast R-CNN network is improved to RoI alignment. The RoI Align is to obtain an image numerical value on a pixel point with the coordinate as a floating point number by using a bilinear interpolation method, and change the whole regional characteristic gathering process into a continuous process, so that a target rectangular frame in the clothing part detection is more accurate. The improved Faster R-CNN network is trained by using the garment part data set, automatic identification and division of garment parts are realized as shown in figure 4, and the accuracy can reach 96%.
(5) Extracting a clothing part image according to the automatically identified rectangular frame coordinates of the clothing part, setting the Gabor operator direction according to the sewing direction of the clothing part, extracting sewing threads by using a horizontal 0-degree Gabor operator for a collar, a vertical 90-degree Gabor operator for a front fly, a horizontal 0-degree Gabor operator for a lower hem, a 45-degree Gabor operator for the splicing position of a clothing body and a right sleeve, a 135-degree Gabor operator for the splicing position of the clothing body and a left sleeve, a 45-degree Gabor operator for a right sleeve and a 135-degree Gabor operator for a left sleeve. And carrying out binarization on the extracted suture by using a threshold value, separating the suture from the background, and keeping the suture information.
(6) Extracting a central axis with a single-pixel width from a binary image of the suture by using a Zhang-Suen algorithm to obtain a stitch segment with a stitch as a unit; taking the central axis end point as a starting point, recording all connected stitch segments, and obtaining five groups of characteristic parameters of the end point position, the length, the central position, the distance between the end point position and the central position of the adjacent segment and the straightness of each stitch segment; the feature vectors with the dimension of 1 × 5 are input into a Support Vector Machine (SVM) classifier which takes a Radial Basis Function (RBF) as a kernel function to obtain a classification result.
The method and the device realize automatic detection of the stitch defects generated in the sewing process of the clothes, the detection precision can reach 3-5 mm, and the method and the device have the characteristics of objectivity, high efficiency and accuracy compared with the traditional manual detection. And moreover, the sewing problems of the clothes produced in each batch can be filed, the results are fed back to the production link, the production flow is optimized, and the production efficiency is improved. The quality detection industrialization degree of the clothing enterprises is effectively improved, the clothing quality of the sales market is guaranteed, the transformation and upgrading of the clothing industry are promoted, and the intelligent development of the clothing industry is promoted.
The above is only a specific embodiment of the present invention, but the technical features of the present invention are not limited thereto. Any simple changes, equivalent substitutions or modifications made on the basis of the present invention to solve the same technical problems and achieve the same technical effects are all covered in the protection scope of the present invention.
Claims (8)
1. A clothing sewing line defect automatic detection method based on a high-precision digital image is characterized by comprising the following steps:
(1) dividing garment parts according to positions of garment stitches, and classifying stitch defects according to the garment parts;
(2) acquiring a high-definition image of the garment to be measured by using an independently built image acquisition device provided with a high-precision industrial camera;
(3) preprocessing the image acquired in the step (2), manually marking the garment part, and establishing a standard garment part data set;
(4) training a Faster R-CNN network by using the garment part data set in the step (3) to realize automatic identification and division of garment parts;
(5) extracting the clothing part image automatically identified in the step (4), extracting the suture lines by using a Gabor operator with the direction consistent with the direction of the suture lines of each clothing part, and performing binarization processing on the extracted suture lines by using a threshold value;
(6) and (4) processing the binaryzation image of the sewing thread of the clothing part in the step (5) by utilizing a Zhang-Suen algorithm to obtain stitch segments with the stitch as a unit, extracting the sewing thread characteristics according to the stitch segments, and identifying the sewing defects of the clothing by using a classifier and judging the category of the sewing defects.
2. The method for automatically detecting the defects of the sewing threads based on the high-precision digital images as claimed in claim 1, wherein the method comprises the following steps: the step (1) can divide the clothes into six parts, namely a collar part, a front fly, a lap, a split part of the clothes body and the sleeves and the sleeve openings according to the distribution position of the sewing thread in the clothes.
3. The method for automatically detecting the clothing sewing defects based on the high-precision digital images as claimed in claim 1 or 2, wherein the method comprises the following steps: the sewing defects of the clothes can be divided into the following parts according to the parts of the clothes: collar defects including broken threads, jumpers, heavy threads and thread ends; defects at the front fly, including broken thread, jumper, heavy thread and thread end; the defects at the lower hem comprise broken lines, jumper wires, heavy lines and thread ends; the defects of the splicing part of the clothes body and the sleeves comprise sewing folds, sewing holes and thread ends; the defects at the cuffs comprise broken lines, jumpers, heavy lines and thread ends.
4. The method for automatically detecting the defects of the sewing threads based on the high-precision digital images as claimed in claim 1, wherein the method comprises the following steps: the acquisition equipment in the step (2) comprises a Mars4072S-24uc industrial camera, a stepping motor, a high-performance computer, a set of conveyor belts and an LED lamp belt, wherein a lens of the Mars4072S-24uc industrial camera is a VT-LEM1614CBMPB 16mm lens.
5. The method for automatically detecting the defects of the sewing threads based on the high-precision digital images as claimed in claim 1, wherein the method comprises the following steps: and (4) preprocessing the garment part data set in the step (4) is to perform size normalization processing on the acquired image, expand the data set through rotation and translation, mark different parts of the garment with rectangular frames of different colors and name the garment.
6. The method for automatically detecting the defects of the sewing threads based on the high-precision digital images as claimed in claim 1, wherein the method comprises the following steps: the Faster R-CNN network backbone network in the step (4) adopts a Resnet network, removes all connection layers, only leaves a convolution layer, and outputs a feature map after down sampling; adjusting the sizes of the first three layers of convolution layers in the Resnet network, and placing the convolution layers in front of the RPN convolution layer; region of interest Pooling RoI Pooling in FasterR-CNN networks was improved to region of interest alignment RoI Align.
7. The method for automatically detecting the defects of the sewing threads based on the high-precision digital images as claimed in claim 1, wherein the method comprises the following steps: and (5) extracting a clothing part image according to the automatically identified rectangular frame coordinates of the clothing part, setting the Gabor operator direction according to the sewing line direction of the clothing part, and extracting the sewing line by using the Gabor operator with the direction consistent with the sewing line direction.
8. The method for automatically detecting the defects of the sewing threads based on the high-precision digital images as claimed in claim 1, wherein the method comprises the following steps: extracting a central axis with a single-pixel width from a binary image of the suture by using a Zhang-Suen algorithm to obtain a stitch segment with a stitch as a unit; taking the central axis end point as a starting point, recording all connected stitch segments, and obtaining five groups of characteristic parameters of the end point position, the length, the central position, the distance between the end point position and the central position of the adjacent segment and the straightness of each stitch segment; and inputting the feature vector with the parameter synthesis dimension of 1 multiplied by 5 into a Support Vector Machine (SVM) classifier taking a Radial Basis Function (RBF) as a kernel function to obtain a classification result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110740826.1A CN113506255B (en) | 2021-07-01 | 2021-07-01 | Automatic detection method for clothing suture defects based on high-precision digital image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110740826.1A CN113506255B (en) | 2021-07-01 | 2021-07-01 | Automatic detection method for clothing suture defects based on high-precision digital image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113506255A true CN113506255A (en) | 2021-10-15 |
| CN113506255B CN113506255B (en) | 2023-09-19 |
Family
ID=78009511
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110740826.1A Active CN113506255B (en) | 2021-07-01 | 2021-07-01 | Automatic detection method for clothing suture defects based on high-precision digital image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113506255B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115624227A (en) * | 2022-06-16 | 2023-01-20 | 旭日商贸(中国)有限公司 | Garment processing intelligent transmission process flow method and system based on AOI |
| CN116536906A (en) * | 2023-05-11 | 2023-08-04 | 杭州探观科技有限公司 | Three-dimensional cutting method for underpants |
| CN117663720A (en) * | 2024-01-29 | 2024-03-08 | 石狮市飞轮线带织造有限公司 | Drying process in polyester sewing thread preparation process |
| CN117663720B (en) * | 2024-01-29 | 2024-04-30 | 石狮市飞轮线带织造有限公司 | Drying process in polyester sewing thread preparation process |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100305909A1 (en) * | 2009-05-26 | 2010-12-02 | MettleWorks, Inc. | Garment digitization system and method |
| CN109174694A (en) * | 2018-08-22 | 2019-01-11 | 杰克缝纫机股份有限公司 | Cut-parts Defect Detection system, method, electric terminal and storage medium |
| CN110264310A (en) * | 2019-05-30 | 2019-09-20 | 肖伯祥 | A kind of clothing pattern making method based on human body big data |
| KR102178961B1 (en) * | 2020-04-21 | 2020-11-13 | 주식회사 스타일쉐어 | Artificial neural network apparatus and method for recommending fashion item using user clustering |
| CN112102224A (en) * | 2019-06-18 | 2020-12-18 | 浙江工业大学 | Cloth defect identification method based on deep convolutional neural network |
| US20200401114A1 (en) * | 2019-06-24 | 2020-12-24 | Amazon Technologies, Inc. | Digital file structure for garment production |
| CN112214630A (en) * | 2020-09-10 | 2021-01-12 | 武汉纺织大学 | Clothing image retrieval system and method based on expansion convolution residual error network |
| CN112529768A (en) * | 2020-12-04 | 2021-03-19 | 中山大学 | Garment editing and generating method based on generation countermeasure network |
-
2021
- 2021-07-01 CN CN202110740826.1A patent/CN113506255B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100305909A1 (en) * | 2009-05-26 | 2010-12-02 | MettleWorks, Inc. | Garment digitization system and method |
| CN109174694A (en) * | 2018-08-22 | 2019-01-11 | 杰克缝纫机股份有限公司 | Cut-parts Defect Detection system, method, electric terminal and storage medium |
| CN110264310A (en) * | 2019-05-30 | 2019-09-20 | 肖伯祥 | A kind of clothing pattern making method based on human body big data |
| CN112102224A (en) * | 2019-06-18 | 2020-12-18 | 浙江工业大学 | Cloth defect identification method based on deep convolutional neural network |
| US20200401114A1 (en) * | 2019-06-24 | 2020-12-24 | Amazon Technologies, Inc. | Digital file structure for garment production |
| KR102178961B1 (en) * | 2020-04-21 | 2020-11-13 | 주식회사 스타일쉐어 | Artificial neural network apparatus and method for recommending fashion item using user clustering |
| CN112214630A (en) * | 2020-09-10 | 2021-01-12 | 武汉纺织大学 | Clothing image retrieval system and method based on expansion convolution residual error network |
| CN112529768A (en) * | 2020-12-04 | 2021-03-19 | 中山大学 | Garment editing and generating method based on generation countermeasure network |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115624227A (en) * | 2022-06-16 | 2023-01-20 | 旭日商贸(中国)有限公司 | Garment processing intelligent transmission process flow method and system based on AOI |
| CN116536906A (en) * | 2023-05-11 | 2023-08-04 | 杭州探观科技有限公司 | Three-dimensional cutting method for underpants |
| CN116536906B (en) * | 2023-05-11 | 2023-09-22 | 杭州探观科技有限公司 | Forming quality detection method of three-dimensional cutting underpants |
| CN117663720A (en) * | 2024-01-29 | 2024-03-08 | 石狮市飞轮线带织造有限公司 | Drying process in polyester sewing thread preparation process |
| CN117663720B (en) * | 2024-01-29 | 2024-04-30 | 石狮市飞轮线带织造有限公司 | Drying process in polyester sewing thread preparation process |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113506255B (en) | 2023-09-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106127780B (en) | A kind of curved surface defect automatic testing method and its device | |
| CN109239076B (en) | Sewing stitch defect detection method based on machine vision | |
| CN113506255B (en) | Automatic detection method for clothing suture defects based on high-precision digital image | |
| CN103868935A (en) | Cigarette appearance quality detection method based on computer vision | |
| US9527115B2 (en) | Computer vision and machine learning software for grading and sorting plants | |
| CN107895362B (en) | Machine vision method for detecting quality of miniature wiring terminal | |
| CN108181316B (en) | Bamboo strip defect detection method based on machine vision | |
| CN108629319B (en) | Image detection method and system | |
| WO2022236876A1 (en) | Cellophane defect recognition method, system and apparatus, and storage medium | |
| CN110403232A (en) | A kind of cigarette quality detection method based on second level algorithm | |
| CN106501272B (en) | Machine vision soldering tin positioning detection system | |
| CN109324056B (en) | Sewing thread trace measuring method based on machine vision | |
| CN109003271A (en) | A kind of Wiring harness connector winding displacement quality determining method based on deep learning YOLO algorithm | |
| CN106248680A (en) | A kind of engine commutator quality detecting system based on machine vision and detection method | |
| CN107891012B (en) | Pearl size and circularity sorting device based on equivalent algorithm | |
| CN114004826A (en) | Visual sense-based method for detecting appearance defects of medical injection molding part | |
| CN111060518A (en) | Stamping part defect identification method based on instance segmentation | |
| CN111739003B (en) | Machine vision method for appearance detection | |
| CN108088848A (en) | A kind of online test method based on phython machine vision defects | |
| CN105699386B (en) | A kind of automatic cloth inspection labeling method using contact-type image sensor | |
| Huang et al. | Mango surface defect detection based on HALCON | |
| CN116091506A (en) | Machine vision defect quality inspection method based on YOLOV5 | |
| CN206470210U (en) | Machine vision scolding tin position detecting system | |
| CN109345519A (en) | Wiring harness connector based on deep learning YOLO algorithm processes model detection method | |
| CN114494142A (en) | Mobile terminal middle frame defect detection method and device based on deep learning |
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 |