• International journal of advanced smart convergence
• /
• 제9권1호
• /
• pp.193-201
• /
• 2020

The main objective of this study is to investigate the impact of additional modalities on the performance of emotion recognition using speech, facial expression and physiological measurements. In order to compare different approaches, we designed a feature-based recognition system as a benchmark which carries out linear supervised classification followed by the leave-one-out cross-validation. For the classification of four emotions, it turned out that bimodal fusion in our experiment improves recognition accuracy of unimodal approach, while the performance of trimodal fusion varies strongly depending on the individual. Furthermore, we experienced extremely high disparity between single class recognition rates, while we could not observe a best performing single modality in our experiment. Based on these observations, we developed a novel fusion method, called parametric decision fusion (PDF), which lies in building emotion-specific classifiers and exploits advantage of a parametrized decision process. By using the PDF scheme we achieved 16% improvement in accuracy of subject-dependent recognition and 10% for subject-independent recognition compared to the best unimodal results.

• 한국지진공학회:학술대회논문집
• /
• 한국지진공학회 2005년도 학술발표회 논문집
• /
• pp.177-184
• /
• 2005

Even in moderate to low seismic regions like Korean peninsular where wind loading usually governs the structural design of a tall building, the probable structural impact of the design basis earthquake or the maximum credible earthquake on the selected structural system should be considered at least in finalizing the design. In this study, by using response spectrum analysis and time history analysis method, seismic performance evaluation was conducted for wind-designed concentrically braced steel highrise buildings. Input ensemble was normalized to be compatible with expected peak ground acceleration. The analysis results showed that wind-designed concentrically braced steel highrise buildings possess significantly increased elastic seismic capacity due to the system overstrength resulting from the wind-serviceability criterion and the width-to-thickness ratio limits on steel members. The time history analysis tended to significantly underestimated the seismic response as compared to response spectrum analysis. Further detailed studies regarding selection and scaling scheme of input ground motions is needed.

• Earthquakes and Structures
• /
• 제1권3호
• /
• pp.307-326
• /
• 2010

This paper is concerned with the optimal seismic design of added viscous dampers in yielding plane frames. The total added damping is minimized for allowable values of local performance indices under the excitation of an ensemble of ground motions in both regular and irregular structures. The local performance indices are taken as the maximal inter-story drift of each story and/or the normalized hysteretic energy dissipated at each of the plastic hinges. Gradients of the constraints with respect to the design variables (damping coefficients) are derived, via optimal control theory, to enable an efficient first order optimization scheme to be used for the solution of the problem. An example of a ten story three bay frame is presented. This example reveals the following 'fully stressed characteristics' of the optimal solution: damping is assigned only to stories for which the local performance index has reached the allowable value. This may enable the application of efficient and practical analysis/redesign type methods for the optimal design of viscous dampers in yielding plane frames.

• 대한전기학회논문지
• /
• 제43권6호
• /
• pp.1027-1035
• /
• 1994

This paper aims to make an artificial arm control strategy. For this, we propose a new feature extraction method and design artificial neural network for the functional classification of myoelectric signal(MES). We first transform the two channel myoelectric signals (MES) for biceps and triceps into frequency domain using fast Fourier transform (FFT). And features were obtained by comparing the magnitudes of ensemble spectrum data and used as inputs to the three-layer neural network for the learning. By changing the number of units in hidden layer of neural network we observed the improvement of classification performance. To observe the effeciency of the proposed scheme we performed experiments for classification of six arm functions to the three subjects. And we obtained on average 94[%] the ratio of classification.

• JSTS:Journal of Semiconductor Technology and Science
• /
• 제14권1호
• /
• pp.23-28
• /
• 2014

In this paper, we discuss on the optimal design scheme of the bilayer OLED (Organic Light Emitting Diodes) with micro-cavity structure. We carried out the optical simulation on the OLED device and calculated optimal scale of devices with taking the micro-cavity effect into account. Our emission model is based upon an ensemble of radiating dipole antennas. Consequently, we applied Maxwell's equation to this sequence, followed by the analysis on the electrical behaviors of OLED device using Poisson's equation. It contains carrier injection and transportation mechanism. In this process, we found out the thickness of each layer can affect the recombination rate at the emission layer. Therefore, we optimized the thickness of each layer to improve the efficiency of the device.

• 천문학회보
• /
• 제46권1호
• /
• pp.61.2-61.2
• /
• 2021

SkyMapper is the largest-aperture optical wide-field telescope in Australia and can be used for transient detection in the Southern sky. Reference images from its Southern Survey cover the sky at δ <+10 deg to a depth of I ~ 20 mag. It has been used for surveys of extragalactic transients such as supernovae, optical counterparts to gravitational-wave (GW) and fast radio bursts. We adopt an ensemble-based machine learning technique and further filtering scheme that provides high completeness ~98% and purity ~91% across a wide magnitude range. Here we present an important use-case of our robotic transient search, which is the follow-up of GW event triggers from LIGO/Virgo. We discuss the facility's performance in the case of the second binary neutron star merger GW190425. In time for the LIGO/Virgo O4 run, we will have deeper reference images for galaxies within out to ~200 Mpc distance, allowing rapid transient detection to i ~ 21 mag.

• 한국수자원학회논문집
• /
• 제39권3호
• /
• pp.275-288
• /
• 2006

본 연구에서는 GDAPS(T213) 중기 기상 수치예보 자료를 활용한 ESP (Ensemble Streamflow Prediction) 기법을 개발하여 미래에 발생할 수 있는 댐 유입량의 중장기적 확률예측을 위해 초과 확률구간별 댐 유입량을 예측하고 RPSS 검증기법으로 예측결과의 정확도를 분석하였다. 개발된 ESP시스템을 적용한 결과 일단위 개념의 확률예보는 높은 불확실성을 내포할 수 있고, 중장기 확률예보에 초점을 맞추어 1, 3, 7일 등의 예측시간 해상도에 대한 ESP정확도의 민감도를 분석한 결과 예측시간 해상도 간격이 증가할수록 예측결과의 불확실성이 감소하면서 그 정확도가 전반적으로 증가함을 살펴볼 수 있었다. 이러한 결과를 바탕으로 GDAPS 자료를 활용한 1주 단위의 한달(28일)예보를 수행한 ESP 결과는 각 초과 확률구간 분포의 적절한 증가 및 감소로 인하여 그 시간적 변동성이 안정적으로 예측되고 예측결과의 불확실성을 감소시킬 수 있어 그 활용가치가 높은 것으로 나타났다. 이러한 관점에서 본 연구의 ESP 시스템은 중장기적 측면에서 GDAPS 자료의 활용가치를 높일 수 있고, 기존 ESP 결과보다 향상된 정확도로 댐 유입량을 예측할 수 있으므로 실시간 댐 유입량 예측에 적용한다면 수자원 관리 차원에서 유용한 수단이 될 수 있을 것이다.

• 한국컴퓨터정보학회논문지
• /
• 제19권3호
• /
• pp.45-54
• /
• 2014

본 논문에서는 기존 기법들의 제한점을 개선하기 위해 휴리스틱 규칙 및 기계학습 분석 결과를 이용한 두 단계의 P2P 트래픽 분류 기법을 제안한다. 첫 번째 단계는 패킷 레벨의 시그니처 기반 분류기이고, 두 번째 단계는 플로우 레벨에서 수행되는 패턴 휴리스틱 규칙 및 통계 기반 분류기이다. 제안된 패턴 휴리스틱 규칙은 분류의 정확도를 높이고 통계 기반 분류기가 처리할 트래픽의 양을 줄일 수 있다. 다양한 의사 결정 트리 알고리즘의 분석을 기반으로 통계 기반 분류기는 가장 효율적인 REPTree로 구현하고, 앙상블 알고리즘을 통해 통계 기반 분류기의 성능을 개선한다. 실제 환경의 데이터 집합을 이용한 검증 분석을 통해, 본 제안 기법이 기존 기법에 비해 높은 정확도와 낮은 과부하를 제공함을 제시한다.

• 대기
• /
• 제27권1호
• /
• pp.105-118
• /
• 2017

This study assesses the performance of the Weather Research and Forecasting (WRF) model in reproducing regional climate over CORDEX-East Asia Phase 2 domain with different cumulus parameterization schemes [Kain-Fritch (KF), Betts-Miller-Janjic (BM), and Grell-Devenyi-Ensemble (GD)]. The model is integrated for 27 months from January 1979 to March 1981 and the initial and boundary conditions are derived from European Centre for Medium-Range Weather Forecast Interim Reanalysis (ERA-Interim). The WRF model reasonably reproduces the temperature and precipitation characteristics over East Asia, but the regional scale responses are very sensitive to cumulus parameterization schemes. In terms of mean bias, WRF model with BM scheme shows the best performance in terms of summer/winter mean precipitation as well as summer mean temperature throughout the North East Asia. In contrast, the seasonal mean precipitation is generally overestimated (underestimated) by KF (GD) scheme. In addition, the seasonal variation of the temperature and precipitation is well simulated by WRF model, but with an overestimation in summer precipitation derived from KF experiment and with an underestimation in wet season precipitation from BM and GD schemes. Also, the frequency distribution of daily precipitation derived from KF and BM experiments (GD experiment) is well reproduced, except for the overestimation (underestimation) in the intensity range above (less) then $2.5mm\;d^{-1}$. In the case of the amount of daily precipitation, all experiments tend to underestimate (overestimate) the amount of daily precipitation in the low-intensity range < $4mm\;d^{-1}$ (high-intensity range > $12mm\;d^{-1}$). This type of error is largest in the KF experiment.

• 한국산업정보학회논문지
• /
• 제29권2호
• /
• pp.15-25
• /
• 2024

효과적인 동물 생태계 분석을 위해서는 동물 서식 현황을 자동으로 파악할 수 있는 동물 관제 기술이 중요하다. 특히 울음소리로 종을 판별하는 동물 소리 분류 기술은 영상을 통한 판별이 어려운 환경에서 큰 주목을 받고 있다. 기존 연구들은 단일 딥러닝 모델을 사용하여 동물 소리를 분류하였으나, 야외 환경에서 수집된 동물 소리는 많은 배경 잡음을 포함하여 단일 모델의 판별력을 악화시키며, 종에 따른 데이터 불균형으로 인해 모델의 편향된 학습을 야기한다. 이에, 본 논문에서는 클래스의 데이터 수를 고려하여 페널티를 부여하는 Focal Loss를 사용한 여러 분류 모델의 예측결과를 앙상블을 통해 결합하여 잡음이 많은 동물 소리를 효과적으로 분류할 수 있는 기법을 제안한다. 공개 데이터 셋을 사용한 실험에서, 제안된 기법은 단일 모델의 평균 성능에 비해 Recall 기준으로 최대 22.6%의 성능 개선을 달성하였다.