• 대한의용생체공학회:의공학회지
• /
• 제35권4호
• /
• pp.99-104
• /
• 2014

Generally, human eye blinks are closely associated with the cognitive state or visual attention such as attentional requirements on visual stimuli. These previous studies have reported that eye blinks are related to explicit visual attention using blink rate, pattern and blink timing across subjects. However, these results have been obtained in a well-controlled experimental settings. So, it would prove difficult to investigate human's natural response in a continuous and realistic situation. In our study, we measured the eye blink intervals while participants viewed a movie clip. And we analyzed the blink interval data for relationship between visual attention and eye blink intervals. 24 participants took part in two experimental sessions, first session to measure the IEBI while viewing the movie clip and second session to conduct a memory performance test using a self-questionnaire, which were spaced 3 weeks apart. The results indicate significantly higher memory performance at long IEBI period than short IEBI period while watching a movie clip(t = 3.257, df = 17, p < 0.005, 2-tailed). In addition, memory performance score significantly correlated with the IEBI value(spearman's rho = 0.40, N = 36, p < 0.01, 2-tailed). Our results suggest that IEBI is used to measure or assess visual attention while wiewing the movie that it is capable of simulating aspects of real-life experiences by visual attention. Thus, we expect IEBI to be used to measure or assess our visual attention, cognition, further emotion about not only movies, advertisements and other cultural contents but also cognitive science.

• 한국통신학회논문지
• /
• 제33권5B호
• /
• pp.318-325
• /
• 2008

HomePlug AV(HPAV)는 댁내에 깔려있는 전력선을 이용하여 데이터 전송뿐만 아니라 음향과 영상까지 전송을 목표로 HomePlug Alliance에서 제정한 표준이다. 이 표준은 TDMA와 CSMA/CA 기술을 혼합하여 사용하는 MAC 기술을 사용한다. HPAV의 CSMA/CA 프로토콜은 Contention Window(CW)와 Deferral Counter(DC)라는 두 가지 주요한 변수를 가지고 HPAV 네트워크를 제어한다. 본 논문에서는 CW와 DC가 성능에 미치는 영향을 여러 각도에서 조명하였고 이를 기반으로 HPAV MAC의 수율을 높이기 위한 적응적 CW 조절 방식을 제안한다. 여러 실험을 통해 얻은 결과 CW가 DC 보다 수율에 더 민감한 성질을 가지고 있어서 제안하는 방식은 DC는 표준에 정의된 기본값을 사용하고 CW만을 조정한다. 제안하는 방식은 네트워크가 과부하가 걸려있다면 CW를 두 배로 늘려주고 네트워크의 부하가 적으면 CW를 반으로 줄여주는 단순한 방식을 이용함으로 복잡도를 낮춰 구현이 가능하다는 장점이 있다. 모의실험과 수리분석을 통해 제안하는 방식이 여러 다양한 환경에서 좋은 성능을 보이는 것을 확인할 수 있다.

• 한국산학기술학회:학술대회논문집
• /
• 한국산학기술학회 2003년도 Proceeding
• /
• pp.223-230
• /
• 2003

Feature-based similarity retrieval has become an important research issue in multimedia database systems. The features of multimedia data are useful for discriminating between multimedia objects (e 'g', documents, images, video, music score, etc.). For example, images are represented by their color histograms, texture vectors, and shape descriptors, and are usually high-dimensional data. The performance of conventional multidimensional data structures(e'g', R- Tree family, K-D-B tree, grid file, TV-tree) tends to deteriorate as the number of dimensions of feature vectors increases. The R＊-tree is the most successful variant of the R-tree. In this paper, we propose a SOM-based R＊-tree as a new indexing method for high-dimensional feature vectors.The SOM-based R＊-tree combines SOM and R＊-tree to achieve search performance more scalable to high dimensionalities. Self-Organizing Maps (SOMs) provide mapping from high-dimensional feature vectors onto a two dimensional space. The mapping preserves the topology of the feature vectors. The map is called a topological of the feature map, and preserves the mutual relationship (similarity) in the feature spaces of input data, clustering mutually similar feature vectors in neighboring nodes. Each node of the topological feature map holds a codebook vector. A best-matching-image-list. (BMIL) holds similar images that are closest to each codebook vector. In a topological feature map, there are empty nodes in which no image is classified. When we build an R＊-tree, we use codebook vectors of topological feature map which eliminates the empty nodes that cause unnecessary disk access and degrade retrieval performance. We experimentally compare the retrieval time cost of a SOM-based R＊-tree with that of an SOM and an R＊-tree using color feature vectors extracted from 40, 000 images. The result show that the SOM-based R＊-tree outperforms both the SOM and R＊-tree due to the reduction of the number of nodes required to build R＊-tree and retrieval time cost.

• 의학교육논단
• /
• 제16권1호
• /
• pp.11-15
• /
• 2014

A number of medical schools in Korea have been using computer-based testing (CBT) for evaluating their students' scientific and/or clinical performance since the early 1990s. Introducing CBT to medical education would have several advantages: first, presenting figures and audio-video files of clinical content is simple with CBT, making it possible to evaluate medical students' competency with navigating more realistic clinical situations at minimum cost; second, CBT enables automatic item analysis and score reporting. To establish CBT, constructing an item bank with item parameters such as difficulty or discriminating parameters will be needed. To select more psychometrically sound items, analysis of the items according to item response theory is necessary. CBT has already been introduced in high stakes tests like the United States Medical Licensing Examination and the Medical Council of Canada Qualifying Examination. The National Health Personnel Examination Board in Korea is also planning to introduce a CBT-based version of the National Medical Examination soon. Thus all medical schools in Korea will need to introduce CBT and construct item banks to prepare their students for their licensing examinations and to measure the students' competency more accurately.

• Journal of Communications and Networks
• /
• 제14권3호
• /
• pp.319-335
• /
• 2012

In this study, we design an applicable model enabling internet protocol television (IPTV) service providers to use a virtual network (VN) for IPTV service delivery. The model addresses the guaranteed service delivery, cost effectiveness, flexible control, and scalable network infrastructure limitations of backbone or IP overlay-based content networks. There are two major challenges involved in this research: i) The design of an efficient, cost effective, and reliable virtual network topology (VNT) for IPTV service delivery and the handling of a VN allocation failure by infrastructure providers (InPs) and ii) the proper approach to reduce the cost of VNT recontruction and reallocation caused by VNT allocation failure. Therefore, in this study, we design a more reliable virtual network topology for solving a single virtual node, virtual link, or video server failure. We develop a novel optimization objective and an efficient VN construction algorithm for building the proposed topology. In addition, we address the VN allocation failure problem by proposing VNT decomposition and reconstruction algorithms. Various simulations are conducted to verify the effectiveness of the proposed VNT, as well as that of the associated construction, decomposition, and reconstruction algorithms in terms of reliability and efficiency. The simulation results are compared with the findings of existing works, and an improvement in performance is observed.

• International Journal of Contents
• /
• 제3권3호
• /
• pp.6-14
• /
• 2007

In this paper, we design and implement a XML Repository System(XRS) that exploits the advantages of DBMSs and IRSs. Our scheme uses BRS to support full text indexing and content-based queries efficiently, and ORACLE to store XML documents, multimedia data, DTD and structure information. We design databases to manage XML documents including audio, video, images as well as text. We employ the non-composition model when storing XML documents into ORACLE. We represent structured information as ETID(Element Type Id), SORD(Sibling ORDer) and SSORD(Same Sibling ORDer). ETID is a unique value assigned to each element of DTD. SORD and SSORD represent an order information between sibling nodes and an order information among the sibling nodes with the same element respectively. In order to show superiority of our XRS, we perform various experiments in terms of the document loading time, document extracting time and contents retrieval time. It is shown through experiments that our XRS outperforms the existing XML document management systems. We also show that it supports various types of queries through performance experiments.

• 한국통신학회논문지
• /
• 제38B권3호
• /
• pp.163-171
• /
• 2013

최근 HD급 동영상이나 3D 어플리케이션과 같은 이전보다 저사양, 모바일 단말에서는 구동하기 힘든 프로그램들에 대한 이용 요구가 확대되면서 처리해야 할 콘텐츠 데이터들이 고용량화 되고 있다. 클라우드 기반의 VDI(Virtual Desktop Infrastructure) 서비스는 이를 처리하기 위해 효율적인 데이터 처리 능력이 필요해졌으며 QoE(Quality of Experience) 보장을 위한 성능 개선 연구가 이슈가 되고 있다. 본 논문에서는 H/W 성능이 향상되어 CPU와 GPU를 탑재한 Thick Client기반의 3가지 Thick-Thin간 VDI 자원 공유 및 위임이 가능한 VDI 서비스에 대해 제안하며, VDI 서비스 성능의 개선을 위해 CPU와 GPU가 혼합된 Heterogeneous 멀티코어 환경에서 CPU와 GPU 병렬 처리 기법인 OpenMP와 CUDA를 활용하여 VDI 서비스 최적화 방안을 제안하고 기존의 VDI와 비교한 성능을 거론한다.

• 한국응급구조학회지
• /
• 제28권2호
• /
• pp.77-98
• /
• 2024

Purpose: This study aimed to derive detailed insights into emergency medical technicians' (EMTs) perceptions regarding continuing education. Methods: We conducted a descriptive survey. The interview participants were 20 EMTs, divided into four groups: six Fire Department EMTs, five Hospital EMTs, five Clinical EMTs (excluding the ER and Fire Department), and four Non-Clinical EMTs. Data were collected through focus group interviews (FGIs) and semistructured questionnaires. The interview periods were March 25, 2022, and March 27, 2022. Results: Continuing education was primarily field-oriented, leading to lower satisfaction among non-field EMTs. To enhance satisfaction, training should be tailored to the specific roles and experiences of the EMTs. Additionally, quality management of lectures, instructor selection, and video quality require reconsideration. Key factors that enhance satisfaction include a wider variety of courses, up-to-date information, and relevant course content, regardless of job role or experience. Conclusion: Incorporating feedback into EMTs' perceptions of continuing education can lead to the development of more effective curricula and improved quality management. This approach is expected to enhance both educational satisfaction and job performance.

• Journal of Platform Technology
• /
• 제12권1호
• /
• pp.128-140
• /
• 2024

딥페이크(Deepfake)란 다양한 인공지능 기술을 활용해 진짜와 같은 가짜를 만드는 영상 합성기술로, 가짜 뉴스 생성, 사기, 악의적인 도용 등에 활용되어 개인과 사회에게 심각한 혼란을 유발시키고 있다. 사회적 문제방지를 위해, 딥페이크로 생성된 이미지를 정교하게 분석하고 탐지하는 방법이 필요하다. 따라서, 본 논문에서는 딥페이크로 생성된 가짜 이미지와 진짜 이미지에서 Saliency 특징을 각각 추출하고 분석하여 합성 후보 영역을 검출하며, 추출된 특징들을 중점으로 학습하여 최종적으로 딥페이크 이미지 탐지 모델을 구축하였다. 제안된 Saliency 기반의 딥페이크 탐지 모델은 합성된 이미지, 동영상 등의 딥페이크 검출 상황에서 공통적으로 사용될 수 있으며, 다양한 비교실험을 통해 본 논문의 제안 방법이 효과적임을 보였다.

• 대한전자공학회논문지TC
• /
• 제49권3호
• /
• pp.47-61
• /
• 2012

현재 HFC(Hybrid Fiber Coaxial) 망에서 UHD(Ultra High Definition) 콘텐츠와 같은 대용량 방송콘텐츠 전송을 위한 대표적인 전송기술로, 다수의 채널을 결합시켜 논리적인 하나의 광대역 채널로 변경하여 데이터를 병렬 전송하는 방법이 대두되고 있다. 하지만 채널 결합을 통해 콘텐츠를 전송하는 시스템의 경우 하나의 콘텐츠 전송을 위해서 다수의 채널을 결합하기 때문에 채널 자원의 부족이 예상된다. 따라서 결합된 채널을 효율적으로 사용할 수 있는 기술이 필요하다. 본 논문에서는 결합된 채널의 효율적인 대역폭 사용을 위하여 세 가지 방식의 VBR(Variable Bit Rate) 대역폭 추정 알고리즘을 분석하였다. 세 가지의 유효 대역폭 추정 방식은 Guerin이 제안한 가우시안 근사를 통한 유효 대역폭 추정, Lee가 제안한 비디오 프레임 특성을 기반으로 한 유효 대역폭 추정과 Nagarajan이 제안한 가우시안 트래픽을 기반으로 한 유효 대역폭 추정 알고리즘이다. 또한, 이들을 분석하여 대용량 방송 시스템으로의 적용가능성을 평가하였다. 성능 분석을 위한 시뮬레이션은 OPNET 시뮬레이터를 사용하였고, 성능 분석의 정확성을 위해 실제 HD 방송 트래픽을 분석하여 대용량 방송 트래픽을 생성하였다.