• Journal of KIISE
• /
• v.41 no.10
• /
• pp.847-858
• /
• 2014

Until now, a lot of image quality assessment techniques or tools for optimal human visual system(HVS)-awareness have been researched and SSIM(Structural SIMilarity) and its improved techniques are representative examples. However, they often cannot cope with various images and different distortion types robustly, and thus this can cause a large gap between their index values and HVS-awareness. In this paper, we conduct image quality assessment on SSIM and its variants intensively and analyze the causes of each component function's observed anomalies. Then, we propose a novel image quality assessment technique to compensate and improve such anomalies. Additionally, through extensive image assessment simulations, we show that the proposed technique can indicate HVS-awareness more robustly and consistently than SSIM and its variants for various images and different distortion types.

• Journal of the Korean Society for information Management
• /
• v.26 no.2
• /
• pp.127-147
• /
• 2009

The synonym issue is an inherent barrier in human-computer communication, and it is more challenging in a Web 2.0 application, especially in social tagging applications. In an effort to resolve the issue, the goal of this study is to test the feasibility of a Web 2.0 application as a potential source for synonyms. This study investigates a way of identifying similar tags from a popular collaborative tagging application, Delicious. Specifically, we propose an algorithm (FolkSim) for measuring the similarity of social tags from Delicious. We compared FolkSim to a cosine-based similarity method and observed that the top-ranked tags on the similar list generated by FolkSim tend to be among the best possible similar tags in given choices. Also, the lists appear to be relatively better than the ones created by CosSim. We also observed that tag folksonomy and similar list resemble each other to a certain degree so that it possibly serves as an alternative outcome, especially in case the FolkSim-based list is unavailable or infeasible.

• Computers and Concrete
• /
• v.22 no.1
• /
• pp.101-116
• /
• 2018

As one of the most important engineering structures, arch dams are huge constructions built with human hands and have strategical importance. Because of the fact that long construction duration, water supply, financial reasons, major loss of life and material since failure etc., the design of arch dams is very important problem and should be done by expert engineers to determine the structural behavior more accurately. Finite element analyses and non-destructive experimental measurements can be used to investigate the structural response, but there are some difficulties such as spending a long time while modelling, analysis and in-situ testing. Therefore, it is more useful to conduct the research on the laboratory conditions and to transform the obtained results into real constructions. Within the scope of this study, it is aimed to determine the structural behavior of arch dams considering experimentally validated prototype laboratory model using similitude and scaling laws. Type-1 arch dam, which is one of five arch dam types suggested at the "Arch Dams" Symposium in England in 1968 is selected as reference prototype model. The dam is built considering dam-reservoir-foundation interaction and ambient vibration tests are performed to validate the finite element results such as dynamic characteristics, displacements, principal stresses and strains. These results are considered as reference parameters and used to determine the real arch dam response with different scales factors such as 335, 400, 416.67 and 450. These values are selected by considering previously examined dam projects. Arch heights are calculated as 201 m, 240 m, 250 m and 270 m, respectively. The structural response is investigated between the model and prototype by using similarity requirements, field equations, scaling laws etc. To validate these results, finite element models are enlarged in the same scales and analyses are repeated to obtain the dynamic characteristics, displacements, principal stresses and strains. At the end of the study, it is seen that there is a good agreement between all results obtained by similarity requirements with scaling laws and enlarged finite element models.

• Journal of Information Processing Systems
• /
• v.13 no.3
• /
• pp.448-463
• /
• 2017

Multimedia is a ubiquitous and indispensable part of our daily life and learning such as audio, image, and video. Objective and subjective quality evaluations play an important role in various multimedia applications. Blind image quality assessment (BIQA) is used to indicate the perceptual quality of a distorted image, while its reference image is not considered and used. Blur is one of the common image distortions. In this paper, we propose a novel BIQA index for Gaussian blur distortion based on the fact that images with different blur degree will have different changes through the same blur. We describe this discrimination from three aspects: color, edge, and structure. For color, we adopt color histogram; for edge, we use edge intensity map, and saliency map is used as the weighting function to be consistent with human visual system (HVS); for structure, we use structure tensor and structural similarity (SSIM) index. Numerous experiments based on four benchmark databases show that our proposed index is highly consistent with the subjective quality assessment.

• Journal of Korea Multimedia Society
• /
• v.21 no.5
• /
• pp.571-582
• /
• 2018

Recently, development of deep learning has shown better image classification result than human. According to recent research, a hidden layer of deep learning is deeper, and a preservation of extracted features shows good results. However, in the case of general images, the extracted features are clear and easy to sort. This study aims to classify plant leaf images. This plant leaf image has high similarity in each image. Since plant leaf images have high similarity not only between images of different species but also within the same species, classification accuracy is not increased by simply extending the hidden layer or connecting the layers. Therefore, in this paper, we tried to improve the hidden layer of the algorithm called DenseNet which shows the recent excellent classification results, and compare the results of several different modified layers. The proposed method makes it possible to classify plant leaf images collected in a natural environment more easily and accurately than conventional methods. This results in good classification of plant leaf image data including unnecessary noise obtained in a natural environment.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.37 no.7
• /
• pp.907-921
• /
• 2013

This study shows the importance of human factors in relationship marketing by revealing the influence of Hanbok salesperson's attributes on relationship quality and long-term relationship orientation at the time when a culture is newly understood. A questionnaire surveyed 376 women living around the Seoul and Daejeon Metropolitan areas from February to March 2013. SPSS WIN 20.0 and AMOS 20.0 programs analyzed the gathered data. We review the influence of Hanbok salesperson's attributes (expertise, ethics, communication skill, customer orientation, similarity, and likeability) on relationship quality and behavioral intention. The research indicates that only customer orientation and Hanbok salesperson's expertise attributes influence relationship quality. All the attributes had a positive influence; in addition, the relationship quality had a significant influence on customer loyalty. However, communication skill, similarity, and likeability did not influence relationship quality. Customer orientation, which provides a customized service based on the recognition of individual customer trends and expertise in developing a relationship with customers, are important factors to form relationship quality and loyalty.

• Archives of design research
• /
• v.15 no.4
• /
• pp.337-346
• /
• 2002

Among many design processes such as learning, storing, retrieving and applying, the process that learns design knowledge is very important for producing creative results that solve design purposes in design computations. The computer should have the ability similar to human in learning design knowledge. It should recognize not only physical properties but also high level design knowledge constructed from the first level physical properties. The high level design knowledge are recognised in terms of isometric translation relationships. This paper explains properties of isometric translation and methods how the computer can recognize high level shape design knowledge using shape pattern representation.

• Journal of the Korea Computer Industry Society
• /
• v.3 no.9
• /
• pp.1269-1278
• /
• 2002

The more high growth of knowledge, the more need personal identity technique. Fingerprint or iris of the eye identity techniques are already commercialized and used various field. Using human face recognition or authentication are not high performance yet. But application for an organism or face recognition are expected getting important. We propose a user recognition system by verifying similarity comparison of eye and lip component images which are splitted, calculated characteristic rate of each facial components and added weight to special formula. Through test proposed methods and analysis the result, we got a high recognition rate.

• Fisheries and Aquatic Sciences
• /
• v.10 no.1
• /
• pp.16-23
• /
• 2007

Myostatin (GDF8) is a growth factor that limits muscle tissue growth and development in vertebrates. We isolated a myostatin-like gene (Py-MSTN) from the marine invertebrate, the scallop Patinopecten yessoensis. Py-MSTN was highly expressed in the adductor muscle and in the gill unexpectedly. Amino acid analysis showed that Py-MSTN has 49% amino acid sequence identity and 64% similarity to human myostatin (Hs-MSTN), and 42% identity and 61% similarity to myoglianin, the only invertebrate homolog. These results indicated that Py-MSTN may be functionally similar to the vertebrate MSTN than the invertebrate homolog. Phylogenetic analysis suggested that Py-MSTN is an ancestral form of vertebrate MSTN and GDF11 and does not belong to other $TGF-{\beta}$ family members. Molecular modeling showed that Py-MSTN exhibits a similar tertiary structure to mammalian BMP7, a member of $TGF-{\beta}$ family. In addition, the amino acid residues which contact extracellular domain of the receptor were relavively conserved. Given these results, we propose that Py-MSTN is a functionally active member of the $TGF-{\beta}$ family and is involved In muscle growth and regulation.

• Journal of the Korean Society of Costume
• /
• v.58 no.9
• /
• pp.38-50
• /
• 2008

In the late Joseon period, as Korea suffered from two battles such as Byeongjahoran and Imjinweran, the people wanted to find comfortable shelters, which caused Bigi to prevail. That is, as an alternative to find a shelter which is physically prosperous and mentally comfortable and to get out of turbulent days, the people desired the utopian world of Sipseungji. The Sipseungji of 'Namsako' was deeply rooted in the society, which had a great impact on it. However, it is very surprising to find that the Sipseungji is metaphored in the process to make the shapes of Korean $J{\check{o}}gori$ which are a represented product of our culture. In other words, the ideal world we desired is a clothe itself which wraps human bodies, not any mysterious place that can not be found. They wanted to deliver the assumption that the ideal world is in humans themselves through clothes. Though a shaman sign of Sip (meaning ten), the Sipseungjiseol of Namsako became rooted in the living of the common people, which caused social confusion, but the Sipseungji suggests that as humans are the very lucky place, the utopian world is in themselves. Therefore, it should not be overlooked that the shapes of Korean $J{\check{o}}gori$ have instructional values that can not be found in those of other foreign cultures.