• 말소리와 음성과학
• /
• 제7권1호
• /
• pp.3-10
• /
• 2015

In this paper, different frequency scales in cepstral feature extraction are evaluated for the text-independent speaker recognition. To this end, mel-frequency cepstral coefficients (MFCCs), linear frequency cepstral coefficients (LFCCs), and bilinear warped frequency cepstral coefficients (BWFCCs) are applied to the speaker recognition experiment. In addition, the spectro-temporal features extracted by the cepstral-time matrix (CTM) are examined as an alternative to the delta and delta-delta features. Experiments on the NIST speaker recognition evaluation (SRE) 2004 task are carried out using the Gaussian mixture model-universal background model (GMM-UBM) method and the joint factor analysis (JFA) method, both based on the ALIZE 3.0 toolkit. Experimental results using both the methods show that BWFCC with appropriate warping factor yields better performance than MFCC and LFCC. It is also shown that the feature set including the spectro-temporal information based on the CTM outperforms the conventional feature set including the delta and delta-delta features.

• 천문학논총
• /
• 제30권2호
• /
• pp.309-313
• /
• 2015

If the present universe is slightly open then pre-inflation curvature would appear as a cosmic dark-flow component of the CMB dipole moment. We summarize current cosmological constraints on this cosmic dark flow and analyze the possible constraints on parameters characterizing the pre-inflating universe in an inflation model with a present-day very slightly open ${\Lambda}CDM$ cosmology. We employ an analytic model to show that for a broad class of inflation-generating effective potentials, the simple requirement that the observed dipole moment represents the pre-inflation curvature as it enters the horizon allows one to set upper and lower limits on the magnitude and wavelength scale of pre-inflation fluctuations in the inflaton field and the curvature parameter of the pre-inflation universe, as a function of the fraction of the total initial energy density in the inflaton field. We estimate that if the current CMB dipole is a universal dark flow (or if it is near the upper limit set by the Planck Collaboration) then the present constraints on ${\Lambda}CDM$ cosmological parameters imply rather small curvature ${\Omega}_k{\sim}0.1$ for the pre-inflating universe for a broad range of the fraction of the total energy in the inflaton field at the onset of inflation. Such small pre-inflation curvature might be indicative of open-inflation models in which there are two epochs of inflation.

• 대한인간공학회지
• /
• 제30권1호
• /
• pp.265-273
• /
• 2011

Objective: The aim of this study is to investigate the attributing factors influencing usability of the mobile phones specifically designed for the elderly users. Efforts to identify factors that cause usability problems for the elderly users in using mobile phones can provide the groundwork for changes aimed at usability enhancement and design of interfaces of mobile phones. Background: It became important to understand the behaviors and tendencies of the elderly in communication as the society became older. The problems in digital divide is contributed to mainly lack of understanding in terms of the use pattern of the elderly and lack of consideration of their characteristics in designing user interfaces of most ICT devices. Method: A total of 30 elderly users who were over 65 years in age participated in usability evaluation test experiment and performed seven different tasks using a widely accepted model of universally designed mobile phone. Their performance was compared with that of contrast group that consisted of 10 younger participants who were on their 20s. Results: It was found that the elderly users had hard times in using mobile phones, especially in keypad manipulation among search, understanding, and manipulation subtasks. Conclusion: Older users seemed to have difficulty in all the subtasks of search, recognition, and manipulation. It was suggested that designers of mobile phones need to give careful consideration into designing visual interfaces for search tasks and keypads for easier control and input for the elderly users. Application: The study is expected to provide guidelines for the universal design of mobile phones and their interfaces for enhancing usability of the mobile phone for elderly users.

• 보건행정학회지
• /
• 제33권3호
• /
• pp.273-294
• /
• 2023

Background: Most developed countries are working to improve their universal health coverage systems. This study investigates regional disparities in unmet healthcare needs and their causes in South Korea. Additionally, it compares the unmet healthcare needs rate in South Korea with that of 33 European countries. Methods: The analysis incorporates information from 13,359 adults aged 19 or older, using data from the Korea Health Panel. The dependent variables encompass the experience of unmet healthcare needs and the three causes of occurrence: "burden of medical expenses," "time constraints," and "lack of care." The primary variable of interest is the region of residence, while control variables encompass 14 socio-demographic, health, and functional characteristics. Multivariable binary logistic regression analysis, accounting for the sampling design, is conducted. Results: The rate of unmet healthcare needs in Korea is 11.7% (95% confidence interval [CI], 11.0%-13.3%), which is approximately 30 times higher than that of Austria (0.4%). The causes of unmet healthcare needs, ranked in descending order, are "lack of care," "time constraints," and "burden of medical expenses." Predictive probabilities for experiencing unmet healthcare needs and each cause differ significantly between regions. For instance, the probability of experiencing unmet healthcare needs due to "lack of care" is approximately 10 times higher in Gangwon-do (13.5%; 95% CI, 13.0%-14.1%) than in Busan (1.3%; 95% CI, 1.3%-1.4%). The probability due to "burden of medical expenses" is approximately 14 times higher in Seoul (4.1%; 95% CI, 3.6%-4.6%) compared to Jeollanam-do (0.3%; 95% CI, 0.2%-0.4%). Conclusion: Amid rapid sociodemographic transitions, South Korea must make significant efforts to alleviate unmet healthcare needs and the associated regional disparities. To effectively achieve this, it is recommended that South Korea involves the National Assembly in healthcare policy-making, while maintaining a centralized financing model and delegating healthcare planning and implementation to regional authorities for their local residents-similar to the approaches of the United Kingdom and France.

• 보건행정학회지
• /
• 제30권2호
• /
• pp.253-262
• /
• 2020

Background: The public health care system of a country is shaped and driven by its historical background as well as social, economic, and cultural structures. This study sheds light on the unique features, strengths, and weaknesses of the health insurance systems of South Korea (Korea) and Peru. Methods: The capacity mapping tool was used to explore the Korean and Peruvian population and geographical structures; health insurance laws, regulations, and policies; payment systems; eligibility and contribution collection; and long-term care insurance. Results: The study found that the Korean government took the lead in integrating multiple insurers into a single-payer system in an effort to reinforce and stabilize its health insurance system in 2000. Peru has been developed mixed model such based on taxes and contributions, to address a gap between different social classes. Peruvian government developed a two-axis system, one for low-income earners, financed by taxes, and another financed by contributions paid by workers and government officials in the formal sector. Peru has introduced many variations to its fee payment and insurer systems, target population, and coverage scope, and maintains its health insurance system accordingly to this day. Conclusion: The current study provides observation of the Health Insurance System in two different countries and helps to understand possible ways to improve the health insurance system in both countries. Based on this study, Peru will be able to see how its system differs from Korea's and benefit from the related policy implications.

• 스마트미디어저널
• /
• 제6권3호
• /
• pp.64-67
• /
• 2017

문학작품은 연극, 영화, 만화, 게임을 비롯한 다양한 문화 콘텐츠의 소재로 활용되고 있다. 원천자료에서 추출된 요소를 활용하여 게임 스토리텔링을 진행하여도 개연성과 보편성이 확보된 게임 스토리 창작이 가능하다는 기존 연구 결과에 따라, 앞으로 문학작품이 변용된 콘텐츠들의 분석에 대한 활용이 확대될 것으로 기대되어 문학작품이 변용된 콘텐츠들의 분석이 필요할 것으로 보고 원전의 변용양상을 분석하고자 한다. 본 논문에서는 국내 온라인 게임 중 문학작품을 활용한 "마비노기"의 시나리오를 그레마스의 행위소 모형을 이용하여 분석하였다. 분석된 결과는 원천 소스인"햄릿"과 비교하여 구조, 시대, 배경, 주인공, 목표, 주제를 분석하였다. 이를 통해 문학을 콘텐츠화할 때 서사 변화를 확인하였으며, 해당 분석은 문학작품을 기반으로 한 게임 시나리오의 변용양상을 분석한 데에 의의가 있다. 이후 문학작품의 다양한 소재들을 다양하게 녹여낸 스토리텔링 방법을 연구하는데 기여할 것으로 기대한다.

• 대기
• /
• 제23권4호
• /
• pp.453-470
• /
• 2013

As a part of the KIAPS Observation Processing System (KOPS), we have developed the modules of satellite radiance data pre-processing and quality control, which include observation operators to interpolate model state variables into radiances in observation space. AMSU-A (Advanced Microwave Sounding Unit-A) level-1d radiance data have been extracted using the BUFR (Binary Universal Form for the Representation of meteorological data) decoder and a first guess has been calculated with RTTOV (Radiative Transfer for TIROS Operational Vertical Sounder) version 10.2. For initial quality checks, the pixels contaminated by large amounts of cloud liquid water, heavy precipitation, and sea ice have been removed. Channels for assimilation, rejection, or monitoring have been respectively selected for different surface types since the errors from the skin temperature are caused by inaccurate surface emissivity. Correcting the bias caused by errors in the instruments and radiative transfer model is crucial in radiance data pre-processing. We have developed bias correction modules in two steps based on 30-day innovation statistics (observed radiance minus background; O-B). The scan bias correction has been calculated individually for each channel, satellite, and scan position. Then a multiple linear regression of the scan-bias-corrected innovations with several predictors has been employed to correct the airmass bias.

• 한국실내디자인학회논문집
• /
• 제3호
• /
• pp.26-32
• /
• 1994

The aim of this study is to provide practical guide for exterior design in apartment and to propose a device to improve the quality of residential environment by the color planning. Color is a language of context and used as a visual aid to recognize the objects and spaces in the world around us. In this sense, each color perception is unique to its own setting and not subject to any universal rules of thumbs, Therefore, when designing with color, each setting will demand its own specific set of rules. A balanced color scheme has to make the environment more pleasant for people to live in . Fro the exterior in housing the use of color has to be considered not just in its wider context of the architecture or surrounding but as an element of the life of the inhabitant. So it is considered on the type of people who will live in and visual condition of site. In this study , the objective model of systematic design process for environmental color planning is proposed on the basis of theroetical background and references investigated, By the existing thesis, resident's preference of the exterior color and foumular of composition in apartment was invastigated . In general theories, based on the theory Birren's color harmony for color coordination an design principles, a device of exterior color design in apartment was suggested by the course of reasonable design process. Color in environment plays many different roles. It integrates the architecture with the site. If color is use well, it can soften architecture that is too harsh, it can give a sense of identity to different part of site and improve a city's image. The environmental color is effective media to enhance the aesthetic value of architecture and environment.

• 대한전자공학회논문지SP
• /
• 제39권2호
• /
• pp.69-78
• /
• 2002

본 논문에서는 복잡한 배경에서의 얼굴 추출 방법을 제안한다. 제안된 알고리즘은 적응 퍼지 색 분할기법을 사용하여 얼굴색과 머리색을 분할시킨다. 얼굴색 분포는 Y,Cb,Cr 색 공간내에서 유도되어지고, 조명값에 적응적인 퍼지 시스템을 사용하여 얼굴색을 구분해낸다. 머리색은 RGB 색 공간내에서 구분되어진다. 전처리 과정을 거쳐 추출되어진 얼굴색과 머리색 영역에 컨벡스 헐을 적용하여 그들의 관계를 통해 최종적인 얼굴 영역이 추출되어진다. 제안된 방법은 기존의 패턴 매칭 방법에 비해 효율적인 성능을 나타낸다. 제안된 알고리즘의 유효성을 실험을 통해 증명하며, 색 영역에서의 제한 조건 없이 성공적으로 얼굴 영역을 추출해 냄을 알 수 있다.

• 정보보호학회논문지
• /
• 제27권3호
• /
• pp.449-457
• /
• 2017

본 논문에서는 양자난수발생기 Quantis의 후처리 과정에 대하여 암호학의 관점에서 실험을 통하여 안전성과 성능을 분석하였다. Quantis의 후처리 과정은 수학적 이론에 근거한 이진행렬-벡터 곱 연산을 통해 풀엔트로피(full-entropy)를 출력하도록 설계되었고, NIST SP 800-90B의 최소엔트로피(min-entropy) 추정 테스트를 이용하여 이를 검증하였다. 이진행렬-벡터 곱 연산에 최적화 기법을 사용함으로써 난수 생성 속도에 미치는 영향을 최소화하였음을 확인하고, NIST SP 800-90B에서 제시한 검증된 Conditioning과의 난수 출력 성능을 비교하였다. 또한, 미국 NIST와 독일 BSI의 난수발생기 표준 모델과 Quantis의 부합되는 요소와 아닌 요소를 구분하였다. Quantis를 암호학의 용도로 사용하고자할 경우, CMVP 기준에 적합하게 이용하기 위해 Quantis의 출력 데이터를 승인된 의사 난수발생기의 씨드로 사용하여 출력한 난수를 사용하는 것이 적절하다고 판단된다.