• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 2002.05a
• /
• pp.295-299
• /
• 2002

본 연구에서는 피륙의 물리화학적 특성에 의해 결정되는 촉감, 태 이외에도 색채, 무의 등 여러 요소들의 영향을 받아 복합적으로 표현되는 의류소재의 총체적인 개념인 의류소재 이미지는 어떤 것들이 있으며 이러한 이미지들은 어떻게 분류될 수 있는지를 알아보기 위하여 의류소재 이미지의 평가를 위한 축을 개발해 보았다. 1995년부터 2000년까지의 Texjournal과 인터패션플래닝에서 발간되는 98/99FW부터 0255까지 트렌드 북에서 소재를 설명하는 형용사를 조사하여 유사한 형용사를 통합 처리하여 87개의 형용사를 최종 추출하여 형용사쌍을 만들고 소재 자극 없이 형용사쌍이 주는 소재이미지만을 가지고 쌍비교법을 통해 유사성을 7점 척도로 표시하도록 하였다. 얻어진 결과를 다차원척도법을 이용하여 분석하여 87개의 형용사의 평가차원을 살펴보았다. 의류소재 이미지를 평가하는 축을 다차원 척도법을 이용하여 개발한 결과 '남성적-여성적', '새로운-낡은 듯한', '캐주얼-클래식', '모호한-정돈된'의 4가지 차원의 8개축이 개발되었다.

• Smart Media Journal
• /
• v.12 no.7
• /
• pp.43-51
• /
• 2023

As online transactions increase, the image of clothing has a great influence on consumer purchasing decisions. The importance of image information for clothing materials has been emphasized, and it is important for the fashion industry to analyze clothing images and grasp the materials used. Textile materials used for clothing are difficult to identify with the naked eye, and much time and cost are consumed in sorting. This study aims to classify the materials of textiles from clothing images based on deep learning algorithms. Classifying materials can help reduce clothing production costs, increase the efficiency of the manufacturing process, and contribute to the service of recommending products of specific materials to consumers. We used machine vision-based deep learning algorithms ResNet and Vision Transformer to classify clothing images. A total of 760,949 images were collected and preprocessed to detect abnormal images. Finally, a total of 167,299 clothing images, 19 textile labels and 20 fabric labels were used. We used ResNet and Vision Transformer to classify clothing materials and compared the performance of the algorithms with the Top-k Accuracy Score metric. As a result of comparing the performance, the Vision Transformer algorithm outperforms ResNet.

• Smart Media Journal
• /
• v.13 no.4
• /
• pp.57-64
• /
• 2024

Attributes such as material, color, and fit in fashion images are important factors for consumers to purchase clothing. However, the process of classifying clothing attributes requires a large amount of manpower and is inconsistent because it relies on the subjective judgment of human operators. To alleviate this problem, there is a need for research that utilizes artificial intelligence to classify clothing attributes in fashion images. Previous studies have mainly focused on classifying clothing attributes for either tops or bottoms, so there is a limitation that the attributes of both tops and bottoms cannot be identified simultaneously in the case of full-body fashion images. In this study, we propose a deep learning model that can distinguish between tops and bottoms in fashion images and classify the category of each item and the attributes of the clothing material. The deep learning models ResNet and EfficientNet were used in this study, and the dataset used for training was 1,002,718 fashion images and 125 labels including clothing categories and material properties. Based on the weighted F1-Score, ResNet is 0.800 and EfficientNet is 0.781, with ResNet showing better performance.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.27 no.5
• /
• pp.533-544
• /
• 2003

The purpose of this study was to provide practical information to design woolen fabrics in terms of structural and surface characteristics, which produce texture images of fabrics. The relationship among structural, surface characteristics and texture images, and preference and purchase intention were analyzed. To evaluate the texture images of the fabrics subjectively, 7 rank＇s semantic differential scale questionnaires were developed with thirty adjective pairs. Blind and non-blind test were performed with 320 female subjects who were in their 20-30＇s. Commercially available 48 woolen fabrics were used as specimens. Results showed that five factors were obtained: classic, elegance, warmth, natural and casual. These factors were closely related to fiber type, weave type, fabric counts, and finishes.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.26 no.11
• /
• pp.1638-1648
• /
• 2002

In this study, we classified adjectives to represent the image of clothing materials as the fundamental process for evaluation of the images on various fabrics and reviewed hierarchy and evaluation dimensions regarding the image of clothing materials. The adjectives to express the image of clothing materials were extracted from Fashion Magazine and Fashion Trend Book The similarity among adjectives was measured by pair-wise comparison without showing fabrics. From the result of the cluster analysis, 87 adjectives were finally extracted through the integrated processing of the adjectives with similar meaning and a close distance. Through the cluster analysis, the hierarchy of the clothing material images was examined. The clothing material images were classified into 12 primary sub-clusters such as ‘feminine', ‘warm', ‘neat', ‘classical', ‘pastoral.' ‘casual', ‘modern'. ‘ambiguous', ‘primitive', masculine', ‘abundant', and ‘arranged'. The dimensions evaluating the clothing material images were also developed using the multi-dimensional scaling method. A 4-dimensions and 8-axes system was established, which is composed of ‘masculine-feminine', ‘new-old', ‘casual-classical', and ‘ambiguous-arranged' images.

• Proceedings of the Costume Culture Conference
• /
• 2001.04a
• /
• pp.47-48
• /
• 2001

• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 2009.05a
• /
• pp.135-138
• /
• 2009

포토닉 의류는 의류에 다양한 광원과 광전달 소재를 적용하여 빛을 발현하는 디지털 기술을 적용하여 의류의 색채를 제어함으로써 착용자의 감성을 시각화할 수 있도록 하는 스마트 의류의 일종이다. 앞서 개발한 포토닉 의류의 중 LED램프와 광섬유의 조합으로 색광을 발현하는 기능은 특별히 RGB 색조합을 통해 의도하는 색채를 다양하게 구현할 수 있다는 특징이 있다. 현재 색채 발현 방법은 Red, Green, Blue의 3가지 색상의 LED의 적절한 조합을 통하여 색채를 발현하는 방식을 가지고 있으나 색체계 상의 기준 색상과 실제 발현되어 육안으로 확인되는 색채의 차이가 있어 의도와는 다른 색채가 나타나 디자이너의 색채 기획의도를 포토닉 의류에 적용하기 힘들며, 기존의 색체계에 의해 생산된 다른 텍스타일 색채와의 컬러 조합에 있어서 필요한 정확한 데이터를 얻기 어렵다. 따라서, 포토닉 의류에 의도하는 색채를 발현하기 위해서는 기존 색상과의 차이점 보정과 효과적인 데이터 베이스 구축이 필요하다. 본 연구에서는 포토닉 의류의 색채에 관한 보다 체계적인 연구를 통해 웹 컬러와 광섬유를 이용한 포토닉 컬러에서의 색채 이미지를 비교, 분석하고 RGB값과 색차값을 통한 차이점 보정을 참고하여 광섬유를 사용하여 제작되는 스마트 포토닉 의류의 제작 시 색채선정 기획에 있어서 기초적인 자료로 활용될 수 있도록 제시하고자 한다.

• Science of Emotion and Sensibility
• /
• v.8 no.2
• /
• pp.117-128
• /
• 2005

The purpose of this study was to clarify the effects of constituent characteristics and the mechanical properties by KES-FB system on the visual texture images and preference of men's suit fabrics. Twenty different kinds of black fabrics, which were used mainly for the spring and summer men's suits, were selected and the subjective evaluation of the visual texture images was tested. Sixty experts in department of clothing and apparel industries participated in the subjective evaluation. Factor analysis showed visual texture images were classified into 4 categories : 'bulkiness', $'extensibility{\cdot}rapability'$, 'flexibility' and 'smoothness'. All of relationships were established between the mechanical properties and the visual texture images except the relation between 'flexibility' and bending properties. The significant factors affecting preference were 'smoothness', compression energy(WC), fabric count and 'flexibility', As the fabrics had higher value for 'smoothness' and 'flexibility' , and lower ones for compression energy and fabric count, they tended to be more preferred.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.25 no.8
• /
• pp.1408-1419
• /
• 2001

The purpose of this study was to investigate the fashion trends of last three years and how the trends were imaged by the structural characteristics of the fabrics. The characteristics for 897 fabrics were analyzed from four kinds. eleven volumes of fashion trend magazines. From these magazines, three panels categorized by their frequencies as nine images such as natural, innocent, ethnic, childish, casual, classic, modern and technical, Sub-images of each image such as rustic, irregular, decorative, etc were also categorized. For the each image, fiber contents and structural characteristics of weave type, weight, density, yarn size, twist and fabric finishes were investigated in terms of frequency, range and mean, Results showed that chiffon and organza seemed to have very specific images and used to represent specifically the romantic or ethnic images; whereas voile and jersey was used to represent various images. For S/S seasons, most popular fiber type was cotton. The weave type was not the important factor to give variations in images; plan weave exclusively used irrespective of image. For the romantic, ethnic and innocent images, rather light fabrics were used. For the childish and natural, medium weights, and for the technical, modern and classic images heavy weight fabrics were used. Vaious finishes were employed to represent specific images.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.1166-1170
• /
• 2009

Colors convey emotions and feelings. This study investigated human's emotional responses on both single colors and changes of colors in clothing. From experiment 1, we found that the important possibility that color emotion also can apply on photonic clothing and it can play a significant role in expressing emotions. We also found there are differences in emotional dimensions between web colors and photonic colors.