• 대한전자공학회논문지SP
• /
• 제48권1호
• /
• pp.157-165
• /
• 2011

본 논문에서는 마르코프 랜덤 필드(Markov random fields: MRF) 기반으로 배경을 모델링하는 방식과 함께 관련 파라미터들을 추정하는 알고리즘을 제안한다. 화소 기반의 배경 모델링 기법은 인근 화소 간의 연관성을 고려하지 않고 화소 단위의 시간적 변화에 대한 통계적 특성에 주로 의존하므로 판정 오류를 줄이는데 한계가 있다. 제안 알고리즘은 화소 기반으로 배경 모델을 일차적으로 수행한 다음 MRF를 이용하여 시공간적으로 인근한 화소 간의 상호 의존성을 활용하여 배경모텔의 정확도를 향상시키는데 그 목적을 두고 있다. MRF는 기본적으로 파라미터의 크기에 매우 민감하므로 기존의 MRF 기반 알고리즘은 이미지에 따라 적절한 값을 사전에 구하여 적용하고 있다. 제안한 방식은 초기에 임의의 파라미터로 배경/전경 상태변수를 구한 후에 이의 통계적 특성을 이용하여 파라미터들을 추정하고 추정된 파라미터를 적용하여 상대변수를 재차 구하는 과정을 반복함으로써 최적의 파라미터에 적응적으로 수렴하도록 조정한다. 실내외의 다양한 환경에서 촬영한 비디오를 이용하여 제안한 방식 성능을 확인한다.

• Computers and Concrete
• /
• 제12권1호
• /
• pp.91-107
• /
• 2013

Corrosion of RC bridge decks eventually leads to delamination, severe cracking and spalling of the concrete cover. This is a prevalent deterioration mechanism and demands for the most costly repair interventions during the service life of bridges worldwide. On the other hand, decisions for repairs are usually made whenever the extent of a limit crack width, reported in routine visual inspections, exceeds an acceptable threshold level. In this paper, while random fields are applied to account for spatial variation of governing parameters of the corrosion process, an analytical model is used to simulate the corrosion induced crack width. However when dealing with random fields, the Monte Carlo simulation is apparently an inefficient and time consuming method, hence the utility of neural networks as a surrogate in simulation is investigated and found very promising. The proposed method can be regarded as an invaluable tool in decision making concerning maintenance of bridges.

• 인지과학
• /
• 제22권2호
• /
• pp.217-233
• /
• 2011

본 논문에서는 띄어쓰기가 전혀 되어 있지 않은 한국어 문장을 입력받아 자동으로 띄어쓰기를 해 주는 시스템을 제안한다. 띄어쓰기 문제는 주어진 문장의 각 음절에 대하여 띄어쓰기 여부를 나타내는 레이블을 부착하는 일종의 레이블링 문제이므로, 본 논문에서는 레이블링 문제 해결에 띄어난 성능을 보이는 것으로 알려진 CRF를 이용하여 자동 띄어쓰기를 시도하였다. 약 112만 음절 규모의 학습용 데이터로 학습을 하고, 2,114 문장(약 9.3만 음절)의 평가용 데이터로 띄어쓰기 정확도에 대한 평가를 하였다. 평가 결과 음절 단위의 정확도는 98.84%, 어절 단위의 정확도는 95.99%인 것으로 나타났다.

• Wind and Structures
• /
• 제29권6호
• /
• pp.389-403
• /
• 2019

In this study, two original spectral representations of stationary stochastic fields, say the continuous proper orthogonal decomposition (CPOD) and the frequency-wavenumber spectral representation (FWSR), are derived from the Fourier-Stieltjes integral at first. Meanwhile, the relations between the above two representations are discussed detailedly. However, the most widely used conventional Monte Carlo schemes associated with the two representations still leave two difficulties unsolved, say the high dimension of random variables and the incompleteness of probability with respect to the generated sample functions of the stochastic fields. In view of this, a dimension-reduction model involving merely one elementary random variable with the representative points set owing assigned probabilities is proposed, realizing the refined description of probability characteristics for the stochastic fields by generating just several hundred representative samples with assigned probabilities. In addition, for the purpose of overcoming the defects of simulation efficiency and accuracy in the FWSR, an improved scheme of non-uniform wavenumber intervals is suggested. Finally, the Fast Fourier Transform (FFT) algorithm is adopted to further enhance the simulation efficiency of the horizontal stochastic wind velocity fields. Numerical examplesfully reveal the validity and superiorityof the proposed methods.

• Wind and Structures
• /
• 제16권5호
• /
• pp.433-455
• /
• 2013

A phase delay spectrum model towards the representation of spatial coherence of stochastic wind fields is proposed. Different from the classical coherence functions used in the spectral representation methods, the model is derived from the comprehensive description of coherence of fluctuating wind speeds and from the thorough analysis of physical accounts of random factors affecting phase delay, building up a consistent mapping between the simulated fluctuating wind speeds and the basic random variables. It thus includes complete probabilistic information of spatial stochastic wind fields. This treatment prompts a ready and succinct scheme for the simulation of fluctuating wind speeds, and provides a new perspective to the accurate assessment of dynamic reliability of wind-induced structures. Numerical investigations and comparative studies indicate that the developed model is of rationality and of applicability which matches well with the measured data at spatial points of wind fields, whereby the phase spectra at defined datum mark and objective point are feasibly obtained using the numerical scheme associated with the starting-time of phase evolution. In conjunction with the stochastic Fourier amplitude spectrum that we developed previously, the time history of fluctuating wind speeds at any spatial points of wind fields can be readily simulated.

• Wind and Structures
• /
• 제11권3호
• /
• pp.241-256
• /
• 2008

This paper adopts autoregressive (AR) model to simulate the wind velocity of spatial three-dimensional fields in accordance with the time and space dependent characteristics of the 3-D fields. Based on the built MATLAB programming, this paper discusses in detail the issues of the AR model deduced by matrix form in the simulation and proposes the corresponding solving methods: the over-relaxation iteration to solve the large sparse matrix equations produced by large number of degrees of freedom of structures; the improved Gauss formula to calculate the numerical integral equations which integral functions contain oscillating functions; the mixed congruence and central limit theorem of Lindberg-Levy to generate random numbers. This paper also develops a method of ascertaining the rank of the AR model. The numerical examples show that all those methods are stable and reliable, which can be used to simulate the wind velocity of all large span structures in civil engineering.

• Structural Engineering and Mechanics
• /
• 제5권6호
• /
• pp.839-851
• /
• 1997

Closed-form solutions are analytically derived for stochastic properties of earthquake ground motion fields, which are conditioned by an observed time series at certain observation sites and are characterized by spectra with uncertainties. The theoretical framework presented here can estimate not only the expectations of such simulated earthquake ground motions, but also the prediction errors which offer important information for the field of engineering. Before these derivations are made, the theory of conditional random fields is summarized for convenience in this study. Furthermore, a method for stochastic interpolation of power spectra is explained.

• 대한수학회지
• /
• 제31권4호
• /
• pp.679-689
• /
• 1994

Let ${X_\underline{j} : \underline{j} \in Z^d}$ be a random field on some probability space $(\Omega, F, P)$ with $EX_\underline{j} = 0, EX_\underline{j}^2 < \infty$.

• Journal of applied mathematics & informatics
• /
• 제33권5_6호
• /
• pp.517-530
• /
• 2015

In this paper, we establish a class of strong limit theorems, represented by inequalities, for the arbitrary random field with respect to the product binomial distributions indexed by the infinite tree on the generalized random selection system by constructing the consistent distri-bution and a nonnegative martingale with pure analytical methods. As corollaries, some limit properties for the Markov chain field with respect to the binomial distributions indexed by the infinite tree on the generalized random selection system are studied.

• 대한수학회지
• /
• 제38권3호
• /
• pp.577-594
• /
• 2001

In this paper we establish limit theorems on the large and small increments of a two-parameter Gaussian random process on rectangles in the Euclidean plane via estimating upper bounds of large deviation probabilities on suprema of the two-parameter Gaussian random process.