• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제7권5호
• /
• pp.458-466
• /
• 2001

본 논문은 HTML을 분석하여 추출된 내용을 음성으로 변환시키고, 웹 브라우저 제어 및 웹문서의 내용 지정을 위하여 음성인식을 사용하는 음성 웹 부라우저의 설계와 구현에 관한 것이다. 이 시스템의 특징은 보편적 설계기법을 사용하여 정상인과 시각장애인을 모두 고려하여 설계되었으며, 특히 시각장애인 인터페이스를 위하여 웹 문서의 구조를 쉽게 인식할 수 있는 트리 구조를 도입하였고, 웹 브라우저 상에 태그로 표현된 모든 요소를 추출하고, 각 요소에 따라 출력음성의 속성을 다르게 하여 별도의 요소 안내음성 출력이 없이 요소 특성을 인지할 수 있게 하였다. 이 방식은 음성 특징을 세부적으로 표현할 수 있는 청각 스타일 시트의 사용을 배제한 것으로써 웹 문서 작성자가 특별한 추가적 노력 없이, 사용자가 문서내의 구조와 요소들의 특징을 쉽게 인식할 수 있는 장점이 있다.

• 정보저장시스템학회:학술대회논문집
• /
• 정보저장시스템학회 2005년도 추계학술대회 논문집
• /
• pp.4-11
• /
• 2005

In this paper, we propose novel optics design for media inside near-field recording (NFR) using hemispherical solid immersion lens (HMS). To obtain strong advantage of data protection and high data capacity simultaneously, HMS based near field optics using aplanatic position of super hemisphere is designed. In this design, to improve small optical tolerance of this aplanatic position, additional aspheric lens surface is added on top of the HMS and it is combined with zoom optics which composed of two single lenses having low numerical aperture (NA). Also, to compensate chromatic aberration which happens seriously in optics using blue laser diode, diffractive optical element is used. Using zoom optics, additional aspheric lens surface, and diffractive optical element together, wavefront aberration and chromatic aberration are effectively reduced in broad range of cover layer thickness and wavelength variation. In addition, In this paper, effect of gap induced aberration is investigated by analyzing different behavior of each TM and TE wave for designed media inside dual-layer NFR optics.

• The International Journal of Costume Culture
• /
• 제2권1호
• /
• pp.62-76
• /
• 1999

The objectives of this study are to investigate consumer experiences with premiums, consumer preferences in premiums offered for clothing and the effect of premiums on clothing purchasing behaviors. A self-administered questionnaire survey was conducted to 610 men and women from ages 15 to 59 living in the Seoul metropolitan area from February 25 to March 9, 1999 ; 548 were used for the data analysis. Data was analyzed by descriptive statistics, t-test, ANOVA, chi-square analysis and Duncan's multiple range tests. Significant differences were found among selected demographic groups in information sources used for premium offers, experiences of receiving premiums, purchasing experiences due to premium offers, preferences between discount and premiums, additional purchase intentions because of premiums, and the premiums'influences. The groups with purchase experiences or brand and stores selecting experiences due to premium, showed significant differences in premiums'influences and the satisfaction levels with premiums. Regarding preferences between discount and premiums, significant differences were found in and the satisfaction levels with premiums.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제14권4호
• /
• pp.332-339
• /
• 2014

Face classification has wide applications in security and surveillance. However, this technique presents various challenges caused by pose, illumination, and expression changes. Face recognition with long-distance images involves additional challenges, owing to focusing problems and motion blurring. Multiple frames under varying spatial or temporal settings can acquire additional information, which can be used to achieve improved classification performance. This study investigates the effectiveness of multi-frame decision-level fusion with photon-counting linear discriminant analysis. Multiple frames generate multiple scores for each class. The fusion process comprises three stages: score normalization, score validation, and score combination. Candidate scores are selected during the score validation process, after the scores are normalized. The score validation process removes bad scores that can degrade the final output. The selected candidate scores are combined using one of the following fusion rules: maximum, averaging, and majority voting. Degraded facial images are employed to demonstrate the robustness of multi-frame decision-level fusion in harsh environments. Out-of-focus and motion blurring point-spread functions are applied to the test images, to simulate long-distance acquisition. Experimental results with three facial data sets indicate the efficiency of the proposed decision-level fusion scheme.

• 전자공학회논문지CI
• /
• 제43권5호
• /
• pp.8-16
• /
• 2006

본 논문에서는 VOD에서 스트리밍 데이터를 효율적으로 서비스할 수 있는 Consistency 알고리즘을 제시하고 개발한다. 미디어 데이터 서비스를 위해 하나의 미디어 데이터를 라운드 로빈 방식으로 스트라이핑하여 각 호스트의 저장 노드에 저장한다. 본 논문에서는 일반적인 계산 용도로 사용되는 매커니즘을 미디어 플레이 최소 단위인 (SH, GOP) 미디어 데이터에 서비스할 수 있도록 한다. VOD 시스템에서 RTP 패킷 데이터는 빠른 미디어 서비스 제공을 위하여 본 논문에서 제시한 consistency 알고리즘을 설계 및 구현하였다. 또한 이 알고리즘에 추가 기능으로 동기화에 따라 공유메모리에 존재하는 멀티미디어 데이터의 순서화된 서비스를 보장하는 것이다. 그리고 예상 페이징 기법을 적용하여 효율적인 버퍼링 기능을 보장하는 것이다. 그리고 본 논문에서 제시한 알고리즘에 대한 성능 평가를 실시하였다. 실험 결과 마지막 시점에 적용되는 베리어 알고리즘으로 인해 성능이 우수함을 알 수 있다.

• 전자공학회논문지CI
• /
• 제44권4호통권316호
• /
• pp.45-49
• /
• 2007

송신 시스템과 수신 시스템에 동일한 주파수를 사용하는 pulse-modulated 방식의 RFID (Radio Frequency IDentification) 시스템에서 송신 시스템과 수신 시스템 사이의 isolation특성은 중요한 문제점이다. 제안된 방법은 송신 시스템에 스위치와 믹서 오실레이터를 사용하여 RFID 시스템의 송신 시스템과 수신 시스템 사이의 isolation 특성을 개선하였다. 제안된 시스템은 송신 시스템에서 up-convert된 출력 신호와 수신 신호의 주파수를 다르게 하여 isolation 특성을 개선한 것이다. 이 방법을 사용하여 20 dB의 isolation 특성 개선 효과를 얻었다.

• International Journal of CAD/CAM
• /
• 제8권1호
• /
• pp.21-28
• /
• 2009

In the present work, an attempt has been made to construct branching surface from 2-D contours, which are given at different layers and may have branches. If a layer having more than one contour and corresponds to contour at adjacent layers, then it is termed as branching problem and approximated by adding additional points in between the layers. Firstly, the branching problem is converted to single contour case in which there is no branching at any layer and the final branching surface is obtained by skinning. Contours are constructed from the given input points at different layers by energy-based B-Spline approximation. 3-D curves are constructed after adding additional points into the contour points for all the layers having branching problem by using energy-based B-Spline formulation. Final 3-D surface is obtained by skinning 3-D curves and 2-D contours. There are three types of branching problems: (a) One-to-one, (b) One-to-many and (c) Many-to-many. Oneto-one problem has been done by plethora of researchers based on minimizations of twist and curvature and different tiling techniques. One-to-many problem is the one in which at least one plane must have more than one contour and have correspondence with the contour at adjacent layers. Many-to-many problem is stated as m contours at i-th layer and n contours at (i+1)th layer. This problem can be solved by combining one-to-many branching methodology. Branching problem is very important in CAD, medical imaging and geographical information system(GIS).

• 패션비즈니스
• /
• 제17권4호
• /
• pp.177-190
• /
• 2013

The purpose of this study is 1) to investigate the types of upscale fashion department stores' Facebook page contents, 2) to compare the types of Facebook page contents with the department stores, and 3) to explore the dimensions of the Facebook page components and their relations. For the study, three preeminent department stores in social media marketing were chosen: Bergdorf Goodman, Barneys New York, and Saks Fifth Avenue. Three hundred sixty five contents of these department stores' pages were investigated, which were uploaded from February 1st to March 31st of 2013. Content analysis, correspondence analysis, and categorical principal component analysis were used for the research. The result showed that there are four important types of contents in pages: product-related contents, fashion-related contents, department stores-related contents, and the contents of communicating with users. And these components of contents were related with department stores distinctively. The two dimensions of the page components were revealed: the basic components (contents, 'like', 'share', and 'comments') and the additional components (links and photos). Among contents, the introduction of products was appealed but news and events were not liked by users; the contents without a photo were not linked to additional information either.

• 한국해양공학회지
• /
• 제19권5호
• /
• pp.71-77
• /
• 2005

This paper presents an integrated underwater navigational algorithm for unmanned underwater vehicles, using additional two-range transducers. This paper proposes a measurement model, using two range measurements, to improve the performance of an IMU-DVL (inertial measurement unit - Doppler velocity log) navigation system for long-time operation of underwater vehicles, excluding DVL measurement. Extended Kalman filter was adopted to propagate the error covariance, to update the measurement errors, and to correct the state equation when the external measurements are available. Simulation was conducted with the 6-d.o.f nonlinear numerical model of an AUV in lawn-mowing survey mode, at current flaw, where the velocity information is unavailable. Simulations illustrate the effectiveness of the integrated navigation system, assisted by the additional range measurements without DVL sensing.

• 정보처리학회논문지C
• /
• 제9C권6호
• /
• pp.945-952
• /
• 2002

For a successful inter-system handoff, several important issues must be handled and additional new features must be considered. This paper focuses on the cellular structure of small cells which are required for the high density of population and a handoff scheme designed between two heterogenous networks. Incase of inter-system handoff (ISHO), the time required to complete the handoff can vary and depends on the structure of networks. And also the transmission of additional signals can increase the probability of failure for ISHO. Here we propose the sub-boundary cell base station (Sub-BBS) to alleviate the role of the BBS. The Sub-BBS is adjacent to BBS in the same regional mobile network. 쪼en the mobile terminal enters Sub-BBS, the network starts finding a new route and after entering BBS, it initiates the transformation process. The proposed scheme significantly reduces the ISHO failure rate compared to the existing one which is the most recent and known as efficient.