http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
유비쿼터스 컴퓨팅 환경을 위한 RFID기반 소프트웨어 개발
김정년(Jungnyun Kim),박상성(Sangsung Park),장동식(Dongsik Jang) 한국정보과학회 2004 한국정보과학회 학술발표논문집 Vol.31 No.1B
유비쿼터스(Ubiquitous) 컴퓨팅이란 사용자가 다양한 형태의 컴퓨터를 원할 때 마다 접속할 수 있는 컴퓨팅 환경을 말한다. 현재 유비쿼터스 컴퓨팅은 차세대 정보기술의 핵심으로, 다양한 분야에서 하드웨어와 소프트웨어에 대한 개발이 진행되고 있다. 본 연구에서는 유비쿼터스 컴퓨팅 환경에서 사용할 수 있는 소프트웨어를 개발하여 실제로 적용시킴으로써, 유비쿼터스 컴퓨팅에서 사용될 수 있는 소프트웨어 모델을 제시하였다. 실험은 사무실 환경에서 실시하였고, 사용자는 개인의 정보를 담고 있는 RFID(Radio Frequency Identification)Tag를 사용함으로써 사용자등록과 인증을 받을 수 있다. 제시된 소프트웨어 모델은 사용자가 다른 컴퓨팅 환경으로 이동시에 RFID Tag와 RFID Reader를 이용하여 인증하고, 이전에사용하였던 컴퓨터의 작업환경을 자동으로 현재의 컴퓨터로 전송하는 시스템이다.
박상성(Sangsung Park),김정년(Jungnyun Kim),안동규(Dongkyu An),김재연(Jaeyeon Kim),장동식(Dongsik Jang) 한국정보과학회 2005 한국정보과학회 학술발표논문집 Vol.32 No.1
본 논문은 비전을 기반으로 한 감정인식 로봇에 관한 논문이다. 피부스킨칼라와 얼굴의 기하학적 정보를 이용한 얼굴검출과 감정인식 알고리즘을 제안하고, 개발한 로봇 시스템을 설명한다. 얼굴 검출은 RGB 칼라 공간을 CIElab칼라 공간으로 변환하여, 피부스킨 후보영역을 추출하고, Face Filter로 얼굴의 기하학적 상관관계를 통하여 얼굴을 검출한다. 기하학적인 특징을 이용하여 눈, 코, 입의 위치를 판별하여 표정 인식의 기본 데이터로 활용한다 눈썹과 입의 영역에 감정 인식 윈도우를 적용하여 윈도우 내에서의 픽셀값의 변화와 크기의 변화로 감정인식의 특징 값을 추출한다. 추출된 값은 실험에 의해서 미리 구해진 샘플과 비교를 통해 감정을 표현하고 표현된 감정은 Serial Communication을 통하여 로봇에 전달되고, 감정 데이터를 받은 얼굴에 장착되어 있는 모터를 통해 표정을 표현한다.
Methodology of improving illiteracy in P3 speller system with ICA blind detection
Jaehong Yoon,Mincheol Whang,Jungnyun Lee 한국HCI학회 2016 한국HCI학회 학술대회 Vol.2016 No.1
P3 speller is one of the most commonly used paradigm in the field of Brain Computer Interface (BCI). However, due to inherent low signal-to-noise, illiterates of this system remains unsolved. This study proposed a novel approach to overcome illiterates in P3 speller system. P3 speller paradigm with rapid serial visual presentation method was implemented to record single channel Event Related Potentials (ERP) at Fz. 8 laboratory researchers have been asked to count the number of target icon appeared when the icon prompted. Short-time Fourier transformation (STFT) was applied on recorded ERP. ICA blind detection discriminated independent components. The accuracy of system with and without ICA blind detection was compared. The average accuracy of ICA blind detection using Euclidean distance was highest but statistical difference was not found. The number of illiterates also decreased. The study has proven that ICA blind detection can effectively reduce illiterates of BCI system.
Bert와 K-means를 활용한 제조 서비스 연구 동향 분석
이기현(Gihyun Lee),이정년(Jungnyun Lee),김태훈(Taehoon Kim),이진명(Jinmyeong Lee),윤정민(Jungmin Yun) 대한전자공학회 2023 대한전자공학회 학술대회 Vol.2023 No.6
Manufacturing service, which provides added value to customers with products as a platform, is receiving attention. Research on manufacturing services is being conducted across various fields, but comprehensive analysis of research trends is still insufficient. This study analyzed the research trends of manufacturing and identified the major areas. A total of 7,061 related papers were collected through a Manufacturing service search on Web of Science, and were analyzed using BERT, K-means clustering, and TF-IDF. As a result of the analysis, the research fields of manufacturing service were classified into a total of 5 clusters, and cluster 0 (process control and supply chain management), 1 (cost-efficient design and quality control), 2 (product development and energy efficiency), and 4 (efficiency improvement in industrial product) increased or maintained research interest, while cluster 3 (quality control and cost management in raw material process) decreased research interest. This study identified the research trends of manufacturing service and can provide insights that can contribute to future research.
Spatial and Time Domain Feature of ERP Speller System Extracted via Convolutional Neural Network
Yoon, Jaehong,Lee, Jungnyun,Whang, Mincheol Hindawi 2018 Computational intelligence and neuroscience Vol.2018 No.-
<P>Feature of event-related potential (ERP) has not been completely understood and illiteracy problem remains unsolved. To this end, P300 peak has been used as the feature of ERP in most brain–computer interface applications, but subjects who do not show such peak are common. Recent development of convolutional neural network provides a way to analyze spatial and temporal features of ERP. Here, we train the convolutional neural network with 2 convolutional layers whose feature maps represented spatial and temporal features of event-related potential. We have found that nonilliterate subjects' ERP show high correlation between occipital lobe and parietal lobe, whereas illiterate subjects only show correlation between neural activities from frontal lobe and central lobe. The nonilliterates showed peaks in P300, P500, and P700, whereas illiterates mostly showed peaks in around P700. P700 was strong in both subjects. We found that P700 peak may be the key feature of ERP as it appears in both illiterate and nonilliterate subjects.</P>
문성철(Sungchul Mun),황민철(Mincheol Whang),이정년(Jungnyun Lee),김종화(Jonghwa Kim),김지혜(Jihye Kim) 대한인간공학회 2011 大韓人間工學會誌 Vol.30 No.1
Objective: The aim of this study was to investigate the association between conceptual metaphors of affect and physical dimensions. Background: The evidence that affect is unconsciously linked to physical dimensions(e.g., position and color) has been obtained in western cultures. However, in eastern cultures, especially, in Korea, it has been seldom tested. It is indispensable to examine whether or not cultural differences exist for better understanding of the conceptual metaphors. Method: 50 positive and 50 negative words were adopted from previous studies in order to determine the relationship between physical domains and abstract affect. Three hypotheses were tested in this study: (1) Good is up. (2) Good is bright. (3) Good is right(for right-handers). The automatic association between valence(positive vs. negative) and vertical position (up vs. down) was investigated in experiment 1, in which 100 words were randomly presented on either the top or the bottom of a LCD screen and the evaluation time of the presented word was recorded. The unconscious association between valence and color(white and black) was examined in experiment 2, where each word was randomly presented in either white or black. Lastly, we tested whether right-handers tend to associate rightward space with positive ideas and leftward space with negative ideas. Results: Reaction time was faster when words were presented in metaphor congruent domains(e.g., 'positive words on the top, in bright color, and on right space vs. negative words on the bottom, in dark color, and on left region) than when they were shown in metaphor incongruent domains. Conclusion: The results might give the groundwork for various directions in further understanding perception, emotion, and behavior. Application: This study may provide useful points to GUI designers and service developers if further studies enhanced in validity and reliability are followed.
ERP 를 이용한 Word 와 Icon 의 인지적 차이 분석
김송이(Songyi Kim),황민철(Mincheol Whang),이정년(Jungnyun Lee),차예술(Yesool Cha),김종화(Jonghwa Kim) 한국HCI학회 2012 한국HCI학회 학술대회 Vol.2012 No.1
본 연구에서는 Word 와 Icon 의 인지적 차이를 확인하여, BCI(Brain-Computer Interface)에서 사용될 자극 콘텐츠를 선정하고자 한다. 실험은 20 대 대학생 남자 8 명, 여자 7 명을 피실험자로 하였고, ISI(Inter-Stimulus Interval) time 150ms 일 때, Word, Icon 자극에 대하여 각각 ERP 요소 P3, N1 의 Latency 와 Amplitude 를 분석하였다. 이 때, Target 자극 제시 방법은 독립변수의 통제변인을 같게 하기 위하여 청각자극을 사용하였다. EEG (Electroencephalography) 신호는 국제 10-20 전극 배치 법에 의거하여 F3, F4, P3, P4, O1, O2 의 6 개의 채널을 측정하였으며, Reference 와 Ground 는 각각 Cz 와 Fpz 지점에 부착하여 측정하였다. 실험 결과는 Word 가 Icon 보다 P3 Latency 가 빠르게 나타났으며, P3 Amplitude 는 높게 나타났다. 이는 자극에 대하여 Word 가 인지 반응이 더 빠르고 집중이 더 잘 되었음을 확인할 수 있었다. 따라서 본 연구를 통해 BCI 를 이용한 Menu Selection 및 Manipulation 을 할 때에 사용될 자극 콘텐츠 설계 시, 기초 연구로써 활용될 수 있을 것으로 기대된다.