• 한국음향학회:학술대회논문집
• /
• 한국음향학회 2004년도 춘계학술발표대회 논문집 제23권 1호
• /
• pp.65-68
• /
• 2004

본 논문에서는 반복학습으로 음소모델을 강건하게 하여 음소기반 핵심어 검출 시스템의 성능을 개선하고자 하였다. 가변어휘 핵심어 검출 시스템은 인식 대상 핵심어의 추가와 변경이 용이하도록 모노폰 단위로 핵심어 모델과 필러 모델을 구성하였다. 핵심어 모델과 필러 모델은 동일한 음소모델을 이용하므로 각각의 음소 모델의 분별력 향상은 핵심어 검출 성능과 밀접한 관계에 있다. 따라서 본 논문에서는 음소 HMM(Hidden Markov Model)의 학습시에 반복 학습을 통하여 음소 모델을 강건하게 만든 후 핵심어 검출 실험을 수행하였다. 그 결과, 10회의 반복학습을 통하여 얻어진 음소 HMM을 이용한 핵심어 검출의 성능은 반복학습을 하지 않은 경우보다 핵심어 검출의 CA-CR 평균 성능이 $4\%$ 향상됨을 확인할 수 있었다.

• 한국농공학회지
• /
• 제35권3호
• /
• pp.117-129
• /
• 1993

This study was conducted to get reasonable and abundant hydrological time series of monthly flows simulated by a best fitting stochastic simulation model for the establishment of rational design and the rationalization of management for agricultural hydraulic structures including reservoirs. Comparative analysis carried out for both statistical characteristics and synthetic monthly flows simulated by the multi-season first order Markov model based on Gamma distribution which is confirmed as good one in the first report of this study and by Harmonic synthetic model analyzed in this report for the six watersheds of Yeong San and Seom Jin river systems. 1.Arithmetic mean values of synthetic monthly flows simulated by Gamma distribution are much closer to the results of the observed data than those of Harmonic synthetic model in the applied watersheds. 2.In comparison with the coefficients of variation, index of fluctuation for monthly flows simulated by two kinds of synthetic models, those based on Gamma distribution are appeared closer to the observed data than those of Harmonic synthetic model both in Yeong San and Seom Jin river systems. 3.It was found that synthetic monthly flows based on Gamma distribution are considered to give better results than those of Harmonic synthetic model in the applied watersheds. 4.Continuation studies by comparison with other simulation techniques are to be desired for getting reasonable generation technique of synthetic monthly flows for the various river systems in Korea.

• 경영과학
• /
• 제31권2호
• /
• pp.1-13
• /
• 2014

In almost all of the organizations, the cost for acquiring and maintaining the inventory takes a considerable portion of the management budget, and thus a certain constraint is set upon the budget itself. The previous studies on inventory control for each item that aimed to improve the fill rate, backorder, and the expenditure on inventory are fitting for the commercially-operated SCM, but show some discrepancies when they are applied to the spare parts for repairing disabled systems. Therefore, many studies on systematic approach concept considering spare parts of various kinds simultaneously have been conducted to achieve effective performance for the inventory control at a lower cost, and primarily, METRIC series models can be named. However, the past studies were limited when dealing with the probability distributions for representing the situation on demand and transportation of the parts, with the (S-1, S) inventory control policy, and so on. To address these shortcomings, the Continuous Time Markov Chain (CTMC) model, which considers the phase-type distributions and the (s, Q) inventory control policies to best describe the real-world situations inclusively, is presented in this study. Additionally, by considering the cost versus the system availability, the optimization of the inventory level, based on this model, is also covered.

• 한국정보처리학회논문지
• /
• 제5권10호
• /
• pp.2713-2724
• /
• 1998

본 논문에서는 칼라 텍스쳐 영상의 MRF모델에서 새로운 파라미터 추정 방법을 제안한다. MRF모델은 RGB 칼라 면 내부의 상호작용뿐만 아니라 칼라 면들 사이의 상호작용도 고려한다. 모델에서의 파리미터들은 공간적 상호작용의 정도를 나타내며 균질한 영역들을 구별하기 위해 사용된다. 그러나 MRF모델을 기반으로 한 칼라 텍스쳐 영상 모델링은 추정해야할 파라미터 수가 너무 많다는 문제를 안고 있다. 본 논문에서는 이러한 계산상의 문제점을 해결하기 위해서 유전자 알고리즘을 사용한다. 제안한 방법의 유효성을 검증하기 위한 실험에서 칼라 자연 영상을 크기에 제한 없이 안정되게 영역 분할하였다.

• Communications for Statistical Applications and Methods
• /
• 제25권2호
• /
• pp.131-157
• /
• 2018

We propose a robust event date model to estimate the date of a target event by a combination of individual dates obtained from archaeological artifacts assumed to be contemporaneous. These dates are affected by errors of different types: laboratory and calibration curve errors, irreducible errors related to contaminations, and taphonomic disturbances, hence the possible presence of outliers. Modeling based on a hierarchical Bayesian statistical approach provides a simple way to automatically penalize outlying data without having to remove them from the dataset. Prior information on individual irreducible errors is introduced using a uniform shrinkage density with minimal assumptions about Bayesian parameters. We show that the event date model is more robust than models implemented in BCal or OxCal, although it generally yields less precise credibility intervals. The model is extended in the case of stratigraphic sequences that involve several events with temporal order constraints (relative dating), or with duration, hiatus constraints. Calculations are based on Markov chain Monte Carlo (MCMC) numerical techniques and can be performed using ChronoModel software which is freeware, open source and cross-platform. Features of the software are presented in Vibet et al. (ChronoModel v1.5 user's manual, 2016). We finally compare our prior on event dates implemented in the ChronoModel with the prior in BCal and OxCal which involves supplementary parameters defined as boundaries to phases or sequences.

• 한국HCI학회:학술대회논문집
• /
• 한국HCI학회 2007년도 학술대회 1부
• /
• pp.1057-1062
• /
• 2007

얼굴표정 인식은 심리학 연구, 얼굴 애니메이션 합성, 로봇공학, HCI(Human Computer Interaction) 등 다양한 분야에서 중요성이 증가하고 있다. 얼굴표정은 사람의 감정 표현, 관심의 정도와 같은 사회적 상호작용에 있어서 중요한 정보를 제공한다. 얼굴표정 인식은 크게 정지영상을 이용한 방법과 동영상을 이용한 방법으로 나눌 수 있다. 정지영상을 이용할 경우에는 처리량이 적어 속도가 빠르다는 장점이 있지만 얼굴의 변화가 클 경우 매칭, 정합에 의한 인식이 어렵다는 단점이 있다. 동영상을 이용한 얼굴표정 인식 방법은 신경망, Optical Flow, HMM(Hidden Markov Models) 등의 방법을 이용하여 사용자의 표정 변화를 연속적으로 처리할 수 있어 실시간으로 컴퓨터와의 상호작용에 유용하다. 그러나 정지영상에 비해 처리량이 많고 학습이나 데이터베이스 구축을 위한 많은 데이터가 필요하다는 단점이 있다. 본 논문에서 제안하는 실시간 얼굴표정 인식 시스템은 얼굴영역 검출, 얼굴 특징 검출, 얼굴표정 분류, 아바타 제어의 네 가지 과정으로 구성된다. 웹캠을 통하여 입력된 얼굴영상에 대하여 정확한 얼굴영역을 검출하기 위하여 히스토그램 평활화와 참조 화이트(Reference White) 기법을 적용, HT 컬러모델과 PCA(Principle Component Analysis) 변환을 이용하여 얼굴영역을 검출한다. 검출된 얼굴영역에서 얼굴의 기하학적 정보를 이용하여 얼굴의 특징요소의 후보영역을 결정하고 각 특징점들에 대한 템플릿 매칭과 에지를 검출하여 얼굴표정 인식에 필요한 특징을 추출한다. 각각의 검출된 특징점들에 대하여 Optical Flow알고리즘을 적용한 움직임 정보로부터 특징 벡터를 획득한다. 이렇게 획득한 특징 벡터를 SVM(Support Vector Machine)을 이용하여 얼굴표정을 분류하였으며 추출된 얼굴의 특징에 의하여 인식된 얼굴표정을 아바타로 표현하였다.

• 천문학회보
• /
• 제41권2호
• /
• pp.62.2-62.2
• /
• 2016

We present a newly developed algorithm based on a Bayesian method for 2D tilted-ring analysis of disk galaxies which operates on velocity fields. Compared to the conventional ones based on a chi-squared minimisation procedure, this new Bayesian-based algorithm less suffers from local minima of the model parameters even with high multi-modality of their posterior distributions. Moreover, the Bayesian analysis implemented via Markov Chain Monte Carlo (MCMC) sampling only requires broad ranges of posterior distributions of the parameters, which makes the fitting procedure fully automated. This feature is essential for performing kinematic analysis of an unprecedented number of resolved galaxies from the upcoming Square Kilometre Array (SKA) pathfinders' galaxy surveys. A standalone code, the so-called '2D Bayesian Automated Tilted-ring fitter' (2DBAT) that implements the Bayesian fits of 2D tilted-ring models is developed for deriving rotation curves of galaxies that are at least marginally resolved (> 3 beams across the semi-major axis) and moderately inclined (20 < i < 70 degree). The main layout of 2DBAT and its performance test are discussed using sample galaxies from Australia Telescope Compact Array (ATCA) observations as well as artificial data cubes built based on representative rotation curves of intermediate-mass and massive spiral galaxies.

• Journal of the Korean Data and Information Science Society
• /
• 제26권1호
• /
• pp.77-87
• /
• 2015

파론도 역설은 두 개의 지는 게임이 결합하여 이기게 되거나, 두 개의 이기는 게임이 결합하여 지게 되는 현상을 말한다. 본 논문에서는 여러 명으로 구성된 집단에서 임의로 한 명을 선택하여 본인의 과거 실적에 의해 승패 확률이 정해지는 과거의존 파론도 게임을 실시하거나 또는 단순히 상금을 임의로 선택한 또 다른 사람에게 전달만 하는 게임을 진행하는 경우를 살펴본다. 각 게임은 지거나 공정한 게임인 반면에 두 게임을 임의로 결합한 혼합게임은 이기게 되는 파론도 효과가 존재함을 확인한다. 또한 각 게임은 이기거나 공정한 게임인데 임의로 결합한 혼합게임은 지게 되는 역 파론도 효과가 존재하는 확률 모수의 범위도 완성한다.

• ETRI Journal
• /
• 제39권3호
• /
• pp.398-405
• /
• 2017

In this paper, an artificial stereo extension method that creates stereophonic sound from a mono sound source is proposed. The proposed method first trains deep neural networks (DNNs) that model the nonlinear relationship between the dominant and residual signals of the stereo channel. In the training stage, the band-wise log spectral magnitude and unwrapped phase of both the dominant and residual signals are utilized to model the nonlinearities of each sub-band through deep architecture. From that point, stereo extension is conducted by estimating the residual signal that corresponds to the input mono channel signal with the trained DNN model in a sub-band domain. The performance of the proposed method was evaluated using a log spectral distortion (LSD) measure and multiple stimuli with a hidden reference and anchor (MUSHRA) test. The results showed that the proposed method provided a lower LSD and higher MUSHRA score than conventional methods that use hidden Markov models and DNN with full-band processing.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권12호
• /
• pp.5905-5926
• /
• 2017

Sentiments can profoundly affect individual behavior as well as decision-making. Confronted with the ever-increasing amount of review information available online, it is desirable to provide an effective sentiment model to both detect and organize the available information to improve understanding, and to present the information in a more constructive way for consumers. This study developed a unified phrase-based topic and sentiment detection model, combined with a tracking model using incremental hierarchical dirichlet allocation (PTSM_IHDP). This model was proposed to discover the evolutionary trend of topic-based sentiments from online reviews. PTSM_IHDP model firstly assumed that each review document has been composed by a series of independent phrases, which can be represented as both topic information and sentiment information. PTSM_IHDP model secondly depended on an improved time-dependency non-parametric Bayesian model, integrating incremental hierarchical dirichlet allocation, to estimate the optimal number of topics by incrementally building an up-to-date model. To evaluate the effectiveness of our model, we tested our model on a collected dataset, and compared the result with the predictions of traditional models. The results demonstrate the effectiveness and advantages of our model compared to several state-of-the-art methods.