• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2019.05a
• /
• pp.442-444
• /
• 2019

Fabric defect classification is an important issue in fabric quality control. However, automated classification is difficult because it is hard to identify various types of defects in images. classification of fabric defects mostly rely on human ability. In this paper, to solve this problem we apply Convolutional Neural Networks (CNN) for fabric defect classification. To make training CNN easier, we propose a method that is manually attentive defect regions in images. we compare the proposed method with the original image and confirm that the proposed method is effective for learning.

• Smart Media Journal
• /
• v.11 no.11
• /
• pp.92-98
• /
• 2022

Identifying defects in textiles is a key procedure for quality control. This study attempted to create a model that detects defects by analyzing the images of the fabrics. The models used in the study were deep learning-based VGGNet and ResNet, and the defect detection performance of the two models was compared and evaluated. The accuracy of the VGGNet and the ResNet model was 0.859 and 0.893, respectively, which showed the higher accuracy of the ResNet. In addition, the region of attention of the model was derived by using the Grad-CAM algorithm, an eXplainable Artificial Intelligence (XAI) technique, to find out the location of the region that the deep learning model recognized as a defect in the fabric image. As a result, it was confirmed that the region recognized by the deep learning model as a defect in the fabric was actually defective even with the naked eyes. The results of this study are expected to reduce the time and cost incurred in the fabric production process by utilizing deep learning-based artificial intelligence in the defect detection of the textile industry.

• Proceedings of the Korean Fiber Society Conference
• /
• 1993.06a
• /
• pp.32-32
• /
• 1993

• Proceedings of the Korean Fiber Society Conference
• /
• 1998.10a
• /
• pp.430-434
• /
• 1998

편성 및 제직 공정의 고속화와 사 품질에 대한 소비자의 요구가 점점 고급화되면서 사품질에 대한 평가 기준도 변화하고 있다. 사 품질을 평가하는 여러 가지 요인 중에서 사 잔털은 편성 및 제직 공정에서 직물결함을 발생시키고, 작업효율을 저하시키므로 이에 대한 관심이 높아지고 있는 추세이다[1,2], 따라서, 사 잔털과 원면특성 및 공정조건의 관계에 대한 연구 결과가 보고되고 있다[3-9]. (중략)

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.4 no.1
• /
• pp.42-46
• /
• 2003

Liquid PVC, which is used widely as fiber coating, has brilliant non-luster effect, but it decreases flexibility of coated fiber surface. We used liquid silicone rubber in elastomer series as a coating material to alleviate this problem. We have conducted the former liquid PVC processing and used pressure of roller and preliminary hardening of processing. In this experiment, We measured 70 degree of hardness, 10.3 MPa of tensile strength and 200fs of tensile elongation of Liquid PVC-coated plastic fiber. We measured 40 degree of hardness, 5.1 MPa of tensile strength and 460% of tensile elongation of Liquid silicone PVC-coated plastic fiber. Therefore, Without the second process, Liquid silicone rubber coating increased non-luster effect and flexibility of plastic fiber surface more than Liquid PVC coating.

• Journal of Ocean Engineering and Technology
• /
• v.8 no.1
• /
• pp.96-104
• /
• 1994

The use of advanced composite materials has grown in recent years in aerospace and other structures. Out of various kinds of repairing methods the one selecteh for this study is an idealized case which simulates a situation where a damaged laminate has been repaired by drilling a hole and therefter plugging the hole with reinforcement. Two typesof reinforcement are investigated ;adhesively bonged plug reinforcement or snug-fit unbonded plug in the hole. For each case of reinforcement, four different sizes of hole diameter and three types of reinforcing material(steel, aluminum, plexiglass) are employed for investigation. The experiment are mainloy forced on the evaluation of ultimate strength of laminate with reinforced hole in comparison to its counterpart with the open hole.

• Journal of the Korean Institute of Navigation
• /
• v.20 no.1
• /
• pp.81-85
• /
• 1996

The defects occurred in knitted fabrics have several types due to some trouble sources. In particular, the defects caused by knitting machine troubles give a serious damage to the whole webs. It is, therefore, necessary to discriminate the kind of defects. The method to discriminate the type and size of defects has been proposed, which is used a pair of narrow band eliminating spatial filters. This method is based upon an isotropic signal processing in time domain. This paper is to confirm that the proposed method can be useful in the discrimination of defects, having analyzed in spatial frequency domain.

• Proceedings of the KIEE Conference
• /
• 1988.07a
• /
• pp.959-962
• /
• 1988

This paper describes an automatic visual inspection system for fabric defects based on pattern recognition techniques. The inspection for fabric defects can be separated into three sequences of operations which are the detection of fabric defects[1], the classification of figures of fabric defects, and the classification of fabric defects. Comparing projections of defect-detected images with the predefined complex, the classification accuracy of figures of fabric defects was found to be 95.3 percent. Employing the Bayes classifier using cluster shade in SGLDM and variance in decorrelation method as features, the classification accuracy of regional figure defects was found to be 82.4 percent. Finally, some experimental results for line and dispersed figures of fabric defects are included.

• The Journal of Information Systems
• /
• v.32 no.4
• /
• pp.149-162
• /
• 2023

Purpose In the textile industry, fabric defects significantly impact product quality and consumer satisfaction. This research seeks to enhance defect detection by developing a transformer-based deep learning image segmentation model for learning high-dimensional image features, overcoming the limitations of traditional image classification methods. Design/methodology/approach This study utilizes the ZJU-Leaper dataset to develop a model for detecting defects in fabrics. The ZJU-Leaper dataset includes defects such as presses, stains, warps, and scratches across various fabric patterns. The dataset was built using the defect labeling and image files from ZJU-Leaper, and experiments were conducted with deep learning image segmentation models including Deeplabv3, SegformerB0, SegformerB1, and Dinov2. Findings The experimental results of this study indicate that the SegformerB1 model achieved the highest performance with an mIOU of 83.61% and a Pixel F1 Score of 81.84%. The SegformerB1 model excelled in sensitivity for detecting fabric defect areas compared to other models. Detailed analysis of its inferences showed accurate predictions of diverse defects, such as stains and fine scratches, within intricated fabric designs.

• Journal of Sericultural and Entomological Science
• /
• no.11
• /
• pp.23-29
• /
• 1970

This experiment is to improve the wrinkle recovery (W.R.) of silk fabrics. The silk fabrics is creased very well, and the crease is the serious defection of it. This experiment is to improve the nature by use of formaldehyde on fabrics. The reagents used were HCl, CH$_3$COOH, CaC$_2$, HCHO, Na$_2$CO$_3$, NH$_4$OH, NaOH and NaHCO$_3$. The silk fabrics was treated, to compare 1 he influence of conditions, by varying the quantities of reagents and the temperature of solution, and the reaction time. The cotton fabrics and the viscose rayon were sunk with the silk at the same condition to be compared the influence. 1) Those of the most suitable temperature to improve for the better W.R. are 75$^{\circ}C$ for silk, 35-45$^{\circ}C$ for cotton, and no particular temperature under 75$^{\circ}C$ for viscose rayon. 2) The W.R. improvements after treated at the temperature of 1) were 11％ for silk and 33.4％ for cotton. 3) There are the best treating time for every fabrics. They were 60 to 90 min. for viscose rayon when HAC Ras used for solvent. It took, however, 60min. of the best time for silk, 120 min. for cotton, and 40 min. for viscose rayon when acetic anhydride instead of HAC was used. 4) It was possible to improve 16.6％ of W.R. for silk at the most suitable treating time, 25.0％ for cotton, and 13.3％ for viscose rayon. 5) Acetic anhydride was rather more effective to improve W.R. of both silk and viscose rayon than HAC. 6) Treating time was also shorter in case of using acetic anhydride than HAC. 7) The improvement of W.R. were 8.3％ for silk at the 10 to 14 ml. of HCHO the best volume, 21. 5％ for cotton at 18m!. of HCHO, and 70％ of for viscose rayon at 14 to 18ml. of HCHO. 8) The most effective quantity of HCI is 14 ml. for both silk and cotton. The W.R. improvement of silk was 22.2％, and that of cotton 19.5％. 9) The W.R. of 83.3％ the best for silk and 61. 6％ for cotton were gained when 4.2gr. of NaHCO$_3$ brings down the percent of W.R. for both silk and cotton. 10) The more NaOH and NH$_4$OH as neutralizing agents, the less effectivity of W.R. until the quantities of the reagents are reached to a special range which are 3. 3m!. for silk and 3.3-6.6 ml. for cotton, and then we can see the W.R. increasing as the quantities of reagents are increased. These facts were evident in case of silk and cotton. We can also see with this fact that the reminder of 〔OH$\^$-/〕 neutralizing 〔CH$\^$+/〕in solution makes it possible to treat formaldehyde on fabrics. 11) Low curing temperature was comparatively better for silk, and high temperature better for cotton. 12) The result of this experiment shows that the Improvement of W.R. for silk was possible to 94％ which means 22％ W.R. increase compared to the untreated silk. This effect also shows that the improvement to W '||'&'||' W (wash and wear) of silk will be possible.