• IEMEK Journal of Embedded Systems and Applications
• /
• v.19 no.1
• /
• pp.47-55
• /
• 2024

We spend a lot of time in indoor space, and the space has a huge impact on our lives. Interior design plays a significant role to make an indoor space attractive and functional. However, it should consider a lot of complex elements such as color, pattern, and material etc. With the increasing demand for interior design, there is a growing need for technologies that analyze these design elements accurately and efficiently. To address this need, this study suggests a deep learning-based design analysis system. The proposed system consists of a semantic segmentation model that classifies spatial components and an image classification model that classifies attributes such as color, pattern, and material from the segmented components. Semantic segmentation model was trained using a dataset of 30000 personal indoor interior images collected for research, and during inference, the model separate the input image pixel into 34 categories. And experiments were conducted with various backbones in order to obtain the optimal performance of the deep learning model for the collected interior dataset. Finally, the model achieved good performance of 89.05% and 0.5768 in terms of accuracy and mean intersection over union (mIoU). In classification part convolutional neural network (CNN) model which has recorded high performance in other image recognition tasks was used. To improve the performance of the classification model we suggests an approach that how to handle data that has data imbalance and vulnerable to light intensity. Using our methods, we achieve satisfactory results in classifying interior design component attributes. In this paper, we propose indoor space design analysis system that automatically analyzes and classifies the attributes of indoor images using a deep learning-based model. This analysis system, used as a core module in the A.I interior recommendation service, can help users pursuing self-interior design to complete their designs more easily and efficiently.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.33 no.10
• /
• pp.1541-1551
• /
• 2009

This study gives guidelines to pattern designing by supplying various images of the transformed shape of jeans with the location of the waistline and the pants silhouette. For this study, 9 kinds of sample cloths (100% cotton denim) were designed considering the laying measurement of the jeans with the location of the waistline and the pants silhouette. The images of each sample were evaluated after the measurement of the completed samples. Clothing and textiles specialists used a semantic differential scale as the evaluation method of the images. For the statistical analysis of the data, one way Anova and Duncan test were adopted using the SPSS 12.0 program. The results of this study are as follows: 1. The visual image by the location of waistline and the change of the pants silhouette is composed of 3 factors (attraction, fashion and comfort factors) of which the attraction factor is the most important factor. 2. The visual image is positive when the location of the waistline is in the low waist position. It is attractive, fashionable, and comfortable. The visual image is negative when the location of the waistline is in the position of the natural waist. 3. The visual image is attractive when the pants silhouette is a boot-cut and fashionable when the pants silhouette are skinny. The straight-cut is comfortable but the visual image is negative. 4. There is no correlation between visual image by the location of the waistline and the change of the pants silhouette.

• Journal of the Korean Society of Costume
• /
• v.53 no.2
• /
• pp.35-55
• /
• 2003

The purpose of the study is to get some pieces of prior information to eventually develop competitive swimming suit designs with Korean traditional images. The study has been accomplished in following steps. First, as theoretical research, the history of swimming suit design and the requirements in designing competitive swimming suits were studied. Second, as practical research, at first, the trend of the competitive swimming suit designs in the national swimming suit market was examined. And then. the questionnaire surveys of both professional and amateur swimmers about their design preferences with purchasing and fabric-related tendencies were conducted for understanding professionals' characters more exactly. This data were analyzed through cross-analysis and multi-response analysis and x 2 was used. The results of this study can be summarized as follows : First, athletes' preferential design tendency for the swimming suit designs were examined in the aspects of style, color, pattern, logo, and accessary. In style, female athletes preferred an athletic one-piece style and male ones preferred a brief style. In color. neutral, mostly black was preferred most and cold color group like dark blue and navy blue next, regardless of sex. In addition, they preferred similar color coordination. In pattern, though solid fabrics were mostly preferred in both sexes, especially Taeguk motive were considered the most appropriate pattern to show Korean image. Besides, logo was considered importantly, and swimming caps and goggles of accessaries were generally used. Second, athletes' swimming suit purchasing tendency was as a following. the number of swimming suits possessed was more than 4 pieces. the durable period was less than 3 months. and the most important point considered in purchasing was an easy fitting for men and a design with an easy fitting for women. And most of swimming athletes preferred foreign products than domestic ones, which was attributed to excellent quality, easy fitting, and good design. and so on. Third, in fabric-related tendency, food touch, easy fitting. and opaqueness were considered importantly. and durability to chlorine, elasticity, color fastness, easy draining, lightness, and so on were demanded forward.

• Korean Journal of Human Ecology
• /
• v.4 no.2
• /
• pp.123-140
• /
• 2001

The purpose of this study is to find the suitable suit design of mid-aged men by examining visual effects in Physical and Psychological side aspects. I used convenience sampling by considering sex and age distribution. These methods were used to analyze data that were frequency analysis, variate analysis, centesimal scale, primary factor analysis, three primary variate analysis, duality variate analysis, equation variate analysis and Cronbach's multiplex range verification 1. Image perception of men's suit was analyzed with arrangement, numbers of button, pattern and the shape of one's body and was divided into charming, youthfulness, modernism. affection and manly. 2. 1) There are partly resemblance difference in physical visual effect by the shape of body and suit design. 2) There are partly resemblance differences in psychological visual effect by one's shape and suit design. In conclusion arrangement, number of button and Pattern are very important to give an effect on whole visual effect man's suit. Also, these were felt differently in each types of body Therefore it is confirmed that the type of body is important fluent to make people perceive.

• Korean Institute of Interior Design Journal
• /
• v.19 no.3
• /
• pp.86-94
• /
• 2010

Art-Wall is the term for a surface wall representing artistic expression by using different materials or images in interior spaces. Usually this is the focal point in an interior space and sometimes it called as an image wall. In the new millenium Art-walls began to use in living spaces. While in the early stages of the development, Art-Wall was limited in a living room, in contemporary it spread into overall living spaces using various patterns and materials. Therefore, the research on Art-Walls in contemporary living spaces is essential to understand this development. The purpose of this study is to find out characteristics of patterns on Art-Walls as an imagery element in contemporary interior spaces. To analyze the characteristic, the analysis table for Art-Walls are constructed based on classifications of wallpapers and textiles. Based on this analysis table, 89 examples of Art-Walls from four major periodicals from 2008 to 2009 were examined. As a result of the analysis, in contemporary interior spaces solid-texture pattern is most frequently appeared. Then natural motifs are the next most popular using oversized or irregular layouts. The result shows that naturalistic expression is significant in interior space in 2008 to 2009. Therefore, Art-Walls that adapted diverse patterns are an essential element in contemporary interiors to represent the major concept of the space.

• Journal of the Korean Society of Costume
• /
• v.61 no.10
• /
• pp.32-41
• /
• 2011

The purpose of this study is to prove an importance of systemic analysis paradigm to interpret fashion make-up. This research employed 'paradigm for interpretation of fashion make-up signs' theory to analyze data. This theory consists of three steps to interpret the meaning of a work: (1) recognition of form, (2) analysis of connotation meanings, and (3) analysis of symbolic meanings and communication. The results of this study were as follows. Jean Paul Gaultier's eye make-up uses a square pattern that mostly consists of the color black. He presents a Glam Rock image by using rimmed make-up that portrays an avant-garde and a nonmainstream culture. Jean Paul Gaultier's make-up was showing black color and square eye pattern and was presenting Glam Rock image and was a rimmed make-up meaning a nonmainstream culture and Avant-garde symbolically. Chanel's make-up emphasizes the eyehole with the rimmed technique by applying the make-up of Marchesa Casati and including non-popularity, Limited Marketing symbolically. Christian Dior's make-up applies the representative make-up of Joan of Arc and Siouxsie sioux and also uses rimmed make-up to portray the meaning of religion persecution, the dark-ages militarism and including Humanism symbolically.

• Journal of the Semiconductor & Display Technology
• /
• v.8 no.2
• /
• pp.23-29
• /
• 2009

To increase the data capacity of one-dimensional symbology, 2D barcodes have been proposed a decade ago. In this paper, a new 2D barcode detection algorithm based on Local Binary Pattern is presented. To locate 2D barcode symbols, a texture analysis scheme based on the Local Binary Pattern is adopted, and a gray-scale projection with sub-pixel operation is utilized to separate the symbol precisely from the input image. Finally, the segmented symbol is normalized using the inverse perspective transformation for the decoding process. The proposed method ensures high performances under various lighting/printing conditions and strong perspective deformations. Experiments show that our method is very robust and efficient in detecting the symbol area for the various types of 2D barcodes.

• Journal of Industrial Technology
• /
• v.17
• /
• pp.331-338
• /
• 1997

In this paper, we propose a new neural pattern recognition from wavelet transform. We first analysis in BFT(Binary Field Transform) in character image. The proposed neural network and wavelet transform is able to improve learning time and scaling. The ability and effectiveness of identifying image using the proposed wavelet transform will be demonstrated by computer simulation.

• Journal of the Optical Society of Korea
• /
• v.18 no.5
• /
• pp.472-476
• /
• 2014

The Joint Transform Correlator (JTC) has been the most suitable technique for real time optical pattern recognition and target tracking applications. This paper proposes a new application of the JTC system for an analysis of the blurring effect of the optical images caused by a defocused lens. We present the relation between the correlation peak, optical transfer function (OTF), and the amount of blurring caused by focusing error. Moreover, we show a possibility of calibrating the blurred image by simply measuring the correlation peak.

• Journal of Korea Multimedia Society
• /
• v.20 no.10
• /
• pp.1655-1661
• /
• 2017

A high-speed directional interpolation algorithm based on the pattern of a $2{\times}2$ pixel block is proposed in this paper. The basic concept of the proposed algorithm is started from UDWT(un-decimated discrete wavelet transform), but there are no transform operations. In order to detect the direction of the edge, 4-pixel differences of two pairs in the $2{\times}2$ block are compared. The $2{\times}2$ block patterns are grouped into total 8 classes, and thereafter the directional interpolation is executed according to the type of the pattern. Since the calculation of the proposed algorithm is very simple and needs a few additions on integer data type, the computation time is almost same as that of bilinear interpolation algorithm. However, experimental results show that the output quality of the proposed one is better than those of the conventional interpolation ones in the objective quality and the computation time.