• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.17 no.2
• /
• pp.15-23
• /
• 2017

The person Re-identification is the most challenging part of computer vision due to the significant changes in human pose and background clutter with occlusions. The picture from non-overlapping cameras enhance the difficulty to distinguish some person from the other. To reach a better performance match, most methods use feature selection and distance metrics separately to get discriminative representations and proper distance to describe the similarity between person and kind of ignoring some significant features. This situation has encouraged us to consider a novel method to deal with this problem. In this paper, we proposed an enhanced recurrent neural network with three-tier hierarchical network for person re-identification. Specifically, the proposed recurrent neural network (RNN) model contain an iterative expectation maximum (EM) algorithm and three-tier Hierarchical network to jointly learn both the discriminative features and metrics distance. The iterative EM algorithm can fully use of the feature extraction ability of convolutional neural network (CNN) which is in series before the RNN. By unsupervised learning, the EM framework can change the labels of the patches and train larger datasets. Through the three-tier hierarchical network, the convolutional neural network, recurrent network and pooling layer can jointly be a feature extractor to better train the network. The experimental result shows that comparing with other researchers' approaches in this field, this method also can get a competitive accuracy. The influence of different component of this method will be analyzed and evaluated in the future research.

• The Journal of the Korea Contents Association
• /
• v.7 no.10
• /
• pp.262-273
• /
• 2007

The current study sought to acknowledge the female sport fan as a sport marketing segment. In order to do so, female sport consumption motivation was identified and among motivation factors, factors effecting re-attend intention and word of mouth intention according to level of team identification was identified. In order to accomplish the purpose a focus group interview was conducted concerning female sport fans, and a survey took place in Korean Pro-sport event collecting 300 data. Among 300 data 248 data were used for the analysis. Using Lisrel 8.7 and SPSS 15.0 correlation analysis, descriptive analysis, confirmatory factor analysis, and multiple regression analysis was conducted. The results of study are as follows, First, 9 female sport consumption motivation was identified and among these factors, skill, achievement, drama, and escape had a positive effect on re-attend intension. Second, among 9 female sport consumption motivations skill, achievement, and social factor had a positive effect on word of mouth intention. Third, among the motivation factors, factors effecting re-attend intention had difference on the level of team identification. Forth, among the motivation factors, factors effecting word of mouth intention had difference on the level of team identification.

• The Journal of the Society of Stroke on Korean Medicine
• /
• v.10 no.1
• /
• pp.47-53
• /
• 2009

Objectives : This study was aim to investigate the relationship between NIHSS and distribution of pattern identification in stroke patients. Methods : 1471 participants from the patients hospitalized for stroke within 4 weeks from April 2007 to August 2009 were included. They were grouped according to the NIHSS score; group 1 for the participants whose NIHSS were less than six, group 2 for seven to fifteen, and group 3 for over than sixteen. And the patients were re-divided into two groups according to their post-onset interval. The difference of distribution of five pattern identification for each group were investigated. And five pattern identification were re-analyzed according to the deficiency-excess pattern identification. K-W test was used for statistical synthesis, and the result was regarded as significant one, if its p-value was below 0.05. Results : Dampness-phelegm pattern was the most frequent out of five patterns in total participants as well as all the subgroups. In group 3 with more serious neurological deficit, larger proportion of patients in early acute stage was diagnosed as excess pattern including Fire-Heat pattern. On the other side the proportion of Deficiency of Qi and Yin was larger in late convalescent stage of group 3 than in other groups. But nothing was statistically significant. Conclusions : Further study including patients with more variant classification with follow-up evaluation is needed to reflect the real characteristics of stroke population.

• The Journal of the Korea Contents Association
• /
• v.15 no.12
• /
• pp.459-470
• /
• 2015

In this study, we have developed a retirement coaching model for effective retirement planning by applying systematic conversation of the coaching technique. A draft of the retirement coaching model was complemented by the validity verification of experts. The retirement coaching model can be summarized as follows. The first step, rapport and empowerment(RE), consists of sharing the image of a happy retirement after establishing a rapport between the coach and the people wanting to prepare for retirement. The second step, target establishment(T), consists of setting the theme and goals of the conversation. The third step, identification(I), consists of developing a retirement plan for achieving the established goals. The fourth step, realization(R), consists of establishing a concrete schedule and a method of retirement planning. The fifth step, evaluation(E), consists of sharing impressions during coaching sessions, which facilitates future action plans.

• Lingua Humanitatis
• /
• v.5
• /
• pp.293-310
• /
• 2003

This study starts with a remark on the frequent usage of the term 'generation culture' in the explanations of cultural change. The term 'generation culture' itself does have neither a clear meaning nor any academic consensus for its operational definition yet. Depending on intuitions or common senses, in most cases, of everyday lift experience, the tends to designate the co-existence of different cultural layers as composing subsystems of a little broader Korean cultural trend. I tried here to analyze different positions and perspectives in the employment of this term, sometimes with strategic intentions of each social groups. Economic or political positions, for example, are intertwined with the quest of identification concerning 'we' and 'other' grouping dynamics in the Age of Globalization, which accelerates the speed of cultural re-territorialization. And the role of consumption activity as a kind of cultural indication has gained more weight in that process. This analysis will remind us, also, of the post-modem society's assumption about the space-and-time in transition, and its fluctuation.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.06a
• /
• pp.120-122
• /
• 2019

본 논문은 보행자 재 검출 알고리즘, 즉 person Re-Identification 알고리즘에 대하여 다루고 있다. 기존의 CNN 네트워크를 이용한 보행자 재 검출 알고리즘의 경우, 실제 감시 카메라 네트워크를 이용하여 보행자 재 검출을 할 경우 주변 환경 조건이 급격하게 변하는 경우 잘못 검출하는 경우가 발생하는 것을 확인할 수 있다. 이는 보행자 검출 후 해당 영역에 대하여 보행자 재 검출을 하는데 있어서 배경 부분의 변화에 영향을 받는다는 것을 의미한다. 따라서 본 논문에서는 배경 부분의 영향에 의한 효과를 줄이기 위하여, 보행자 영역 분리 알고리즘을 이용하여 보행자 영역을 분리한 후, 보행자 재 검출을 수행하는 연구를 진행한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.06a
• /
• pp.102-104
• /
• 2018

본 논문은 보행자 재 검출 알고리즘, 즉 person Re-Identification 알고리즘에 대하여 다루고 있다. 기존의 CNN 네트워크를 이용한 보행자 재 검출 알고리즘의 경우, 실제 감시 카메라 네트워크를 이용하여 보행자 재 검출을 할 경우 주변 환경 조건이 급격하게 변하는 경우 잘못 검출하는 경우가 발생하는 것을 확인할 수 있다. 이는 보행자 검출 후 해당 영역에 대하여 보행자 재 검출을 하는데 있어서 배경 부분의 변화에 영향을 받는다는 것을 의미한다. 따라서 본 논문에서는 배경 부분의 영향에 의한 효과를 줄이기 위하여, 보행자 영역 분리 알고리즘을 이용하여 보행자 영역을 분리한 후, 보행자 재 검출을 수행하는 연구를 진행한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.261-262
• /
• 2020

보행자 재 검출 알고리즘, 즉 person Re-Identification 알고리즘은 주어진 영상 내에 존재하는 보행자들 중 특정 보행자를 검출해내는 방법이다. 최근까지 보행자 재 검출 알고리즘에 대한 여러 연구가 진행되어 오고 있지만 기존의 CNN 네트워크를 이용한 보행자 재 검출 알고리즘의 경우, 실제 영상 데이터를 이용하여 보행자 재 검출을 할 경우 주변 환경 조건이나 조명의 조건, 보행자를 촬영한 방향 등에 따라 정확도가 떨어지는 현상이 발행한다. 이에 따라, 보행자 재 검출 알고리즘을 수행하는데 있어서, 조명 등의 조건에 구애 받지 않고 정확한 검출을 할 수 있도록 style transfer를 이용하여 영상을 변형하여 보행자 재검출을 수행하는 연구를 진행한다.

• Animal Systematics, Evolution and Diversity
• /
• v.29 no.4
• /
• pp.288-293
• /
• 2013

A single specimen of the genus Sillago, collected from Gwangyang, Korea, in May 2009, is characterized by XI first dorsal fin spines, 3 or 4 rows of melanophore pattern along the second dorsal fin membrane, and a darkish posterior margin of the caudal fin. Our specimen was identified as Sillago sinica reported as a new species; this identification is confirmed by mitochondrial DNA cytochrome oxidase subunit I sequences, which show that our specimen corresponds to S. sinica (d=0.000) and differs from the congeneric species Sillago parvisquamis (d=0.170). Comparisons of Korean specimens previously reported as S. parvisquamis with specimens of S. sinica show that the S. parvisquamis specimens are actually S. sinica. We propose the new Korean name "buk-bang-jeom-bo-ri-myeol" for S. sinica.

• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1166-1174
• /
• 2019

This paper propose and experiment advanced PersonNet, a human identification model, with advanced performance. We apply the inception layer to extract feature points, and increase the existing 32 feature points to 154. Also, we modify the CND method used by PersonNet to mitigate asymmetry, and apply weights to the feature map of pedestrian images in three parts, thereby making the features more distinct. Three databases were used for performance evaluation : CUHK01, CUHK03 and Market-1501. The experiment results showed 27-31% improvement in performance.