• Journal of the Korean Data and Information Science Society
• /
• v.14 no.3
• /
• pp.535-543
• /
• 2003

Cluster analysis has been widely used in many applications, such as pattern analysis or recognition, data analysis, image processing, market research on on-line or off-line and so on. Clustering can identify dense and sparse regions among data attributes or object attributes. But it requires many hours to get clusters that we want, because clustering is more primitive, explorative and we make many data an object of cluster analysis. In this paper we propose a new method of clustering using sample based on grid. It is more fast than any traditional clustering method and maintains its accuracy.

• 한국데이터정보과학회:학술대회논문집
• /
• 2003.05a
• /
• pp.97-108
• /
• 2003

Cluster analysis has been widely used in many applications, such that pattern analysis or recognition, data analysis, image processing, market research on on-line or off-line and so on. Clustering can identify dense and sparse regions among data attributes or object attributes. But it requires many hours to get clusters that we want, because of clustering is more primitive, explorative and we make many data an object of cluster analysis. In this paper we propose a new method of clustering using sample based on grid. It is more fast than any traditional clustering method and maintains its accuracy. It reduces running time by using grid-based sample. And other clustering applications can be more effective by using this methods with its original methods.

• 한국데이터정보과학회:학술대회논문집
• /
• 2004.04a
• /
• pp.51-60
• /
• 2004

K-means clustering is an iterative algorithm in which items are moved among sets of clusters until the desired set is reached. K-means clustering has been widely used in many applications, such as market research, pattern analysis or recognition, image processing, etc. It can identify dense and sparse regions among data attributes or object attributes. But k-means algorithm requires many hours to get k clusters that we want, because it is more primitive, explorative. In this paper we propose a new method of k-means clustering using a center of gravity for grid-based sample. It is more fast than any traditional clustering method and maintains its accuracy.

• Journal of Korea Multimedia Society
• /
• v.20 no.2
• /
• pp.126-133
• /
• 2017

The conventional hand-crafted features used to track objects have limitations in object representation. Convolutional neural networks, which show good performance results in various areas of computer vision, are emerging as new ways to break through the limitations of feature extraction. CNN extracts the features of the image through layers of multiple layers, and learns the kernel used for feature extraction by itself. In this paper, we use the feature map extracted from the convolution layer of the convolution neural network to create an outline model of the object and use it for tracking. We propose a method to adaptively update the outline model to cope with various environment change factors affecting the tracking performance. The proposed algorithm evaluated the validity test based on the 11 environmental change attributes of the CVPR2013 tracking benchmark and showed excellent results in six attributes.

• Archives of design research
• /
• v.4 no.1
• /
• pp.31-37
• /
• 1991

Process of thinking on the form in product design includes the process in which unclear requirements are transformed into defined concepts, and the process in which the design as a form conforming to the concepts in conc1uded.(The former is called fconceptualizationJ which deals with abstract matters, and the latter "manipulation of the formJ which deals with more embodied matters) In conceptualization process the operation which transform the diverse requirements in the initial phase of de\ulcornersign to related linguistic expressions or to an image that directly suggests the form is executed. In the process of manipulation of the form the transformation to design as the description of relationships between the formal elements of the design solution resulted from the transformed requirements. It is the job of selecting the most suitable formal elements, balancing the relationships between them, out 01 the possibilities that reside in the individual attributes of formal elements, and finally leading to the design as the end solution. This writing focuses down to the wilfulness and ambiguity of design, and the diversity of design solution after the requirements, formal elements, and the process of relating the formal attributes to each other is built up the two phases.ases.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.4
• /
• pp.25-31
• /
• 2018

In recent years the privacy issue on social media is often being discussed. The purpose of this study is to explore the relationships among user gender, user group according to user activity level (highly active vs less active) and self-disclosure in social media. We collected a total of 180 million tweets issued by 13 million twitter users for 12 months and investigated attributes of tweet (user's profile, profile image, description, geographic information, URL) which are related to self-disclosure and boundary impermeability. The results show there are significant (p<0.001) interactions between user gender, user group and each attribute of tweet that are related to self-disclosure and show that the patterns of self-disclosure are different across attributes. The results also show that the mean self-disclosure scores and boundary impermeability of top 10% highly active users are significantly higher than other less active users for all genders.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.6
• /
• pp.2115-2127
• /
• 2021

Although significant progress has been made in synthesizing visually realistic face images by Generative Adversarial Networks (GANs), there still lacks effective approaches to provide fine-grained control over the generation process for semantic facial attribute editing. In this work, we propose a novel cross channel self-attention based generative adversarial network (CCA-GAN), which weights the importance of multiple channels of features and archives pixel-level feature alignment and conversion, to reduce the impact on irrelevant attributes while editing the target attributes. Evaluation results show that CCA-GAN outperforms state-of-the-art models on the CelebA dataset, reducing Fréchet Inception Distance (FID) and Kernel Inception Distance (KID) by 15~28% and 25~100%, respectively. Furthermore, visualization of generated samples confirms the effect of disentanglement of the proposed model.

• Korean Institute of Interior Design Journal
• /
• v.26 no.4
• /
• pp.93-102
• /
• 2017

The study of space through image evaluation has reached the stage of measuring the emotion with the development of IT technology and the advent of VR AR era. Many researches have been trying to evaluate the space and measure the emotion in objective form by providing various forms of spatial images such as photographs, sketches, and CG. In order for these studies to be used as objective data with logical relevance, it is necessary to describe in detail the method of collecting the data used in the experiment, the characteristics of the test procedure, and the method of analysis. This study is a basic study for constructing a database by systematically organizing the attributes, experimental methods and experimental procedures of subjects. From the viewpoint of securing the objectivity and reproducibility of the paper, we analyzed the spatial image evaluation process focusing on 1) evaluation subject, 2) experimental method, and 3) analysis standard. It is necessary to examine whether the object of evaluation and the form of image and the method of providing it meet the purpose of the study. In addition, the size and order of the image, the viewing time and the interval (break time) should be different according to the gender and group experiment and the individual test method.

• Asian Journal of Business Environment
• /
• v.9 no.3
• /
• pp.21-26
• /
• 2019

Purpose - This study aims to investigate the effect of brand image and trust of a home meal replacement (HMR) industry on customer satisfaction and repurchase during online shopping. Research design, data, and methodology - With 217 questionnaires, this study was conducted by AMOS 20.0, and the Structural Equation Model (SEM) as statistical method was used for examining the hypotheses in this study. Factors such as brand image and brand trust in customer shopping for HMR products online were tested, and relationships between satisfaction and repurchase were studied. Results - Brand image and brand trust in terms of online shopping for HMR were found to affect satisfaction significantly; in addition, the path where satisfaction leads to repurchase was found to be significant. However, brand image and brand trust for HMR in online shopping differed depending on customer age groups. The path-coefficients from brand image of HMR in online shopping to satisfaction were more significant in the older age group; meanwhile, the path-coefficient from brand trust to satisfaction was significant in the younger age group. Conclusions - Results of the study suggested the importance of the attributes for buying HMR products online and provided meaningful implications of difference between age groups when they choose the products.

• The Research Journal of the Costume Culture
• /
• v.21 no.1
• /
• pp.42-56
• /
• 2013

Nowadays the advantages of multichannel retailing strategy in fashion business have been widely discussed, but empirical research on fashion retail has been limited. The purpose of this research is to provide some ideas on multichannel retailing strategy to fashion retailers through the case of UNIQLO. The online survey was conducted on each 100 female customers in their 20s, 30s, 40s living in seoul among UNIQLO customers. The survey was consisted of measurement items for UNIQLO's online store image and offline store image, customer satisfaction, purchase intention, and demographic attributes. The online survey was found that 30.3% of UNIQLO's multichannel customers bought a product from offline store using online shopping mall as a search channel, on the other hand, 20.7% of UNIQLO's multichannel customers bought a product from online store using offline store as a search channel. Factors of the online shopping mall image were consisted of shopping convenience, product information, price policy, trust. And factors of the offline store image were consisted of trust and store, product information, service. Some factors of online store and offline store image had impact on multichannel customer satisfaction. And, customer satisfaction also had impact on purchase intention of UNIQLO product. Some suggestion for the future of multichannel research in fashion retailing was given.