• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 1998년도 가을 학술발표논문집 Vol.25 No.2 (2)
• /
• pp.467-468
• /
• 1998

Impulsive noise appears as black and/or white spots in an image. It is usually caused by errors during the image acquisition or transmission through communication channels. This paper presents a study on the impulsive noise reduction filter of digital image. A much more effective method for removing impulse noise is weighted median filtering. But it loses some information by changing center value with no condition. We propose some new technique to change center value with some conditions. In this paper, the performance of conditional weighted median filter is compared to the commonly used median filter, mean filter, max/min filter, and weighted median filter. A quantitative comparison is performed on MSE (Mean Square Error), RMSE (Root Mean Square Error), and SNR (Signal to Noise Ratio). Proposed conditional weighted median filter can yield better performance than regular filters.

• 대한수학회지
• /
• 제53권6호
• /
• pp.1275-1292
• /
• 2016

The Marcinkiewicz-Zygmund strong law of large numbers for conditionally independent and conditionally identically distributed random variables is an existing, but merely qualitative result. In this paper, for the more general cases where the conditional order of moment belongs to (0, ${\infty}$) instead of (0, 2), we derive results on convergence rates which are quantitative ones in the sense that they tell us how fast convergence is obtained. Furthermore, some conditional probability inequalities are of independent interest.

• 한국지진공학회논문집
• /
• 제23권6호
• /
• pp.301-309
• /
• 2019

A methodology to assess seismic fragility of a nuclear power plant (NPP) using a conditional mean spectrum is proposed as an alternative to using a uniform hazard response spectrum. Rather than the single-scenario conditional mean spectrum, which is the conventional conditional mean spectrum based on a single scenario, a multi-scenario conditional mean spectrum is proposed for the case in which no single scenario is dominant. The multi-scenario conditional mean spectrum is defined as the weighted average of different conditional mean spectra, each one of which corresponds to an individual scenario. The weighting factors for scenarios are obtained from a deaggregation of seismic hazards. As a validation example, a seismic fragility assessment of an NPP containment structure is performed using a uniform hazard response spectrum and different single-scenario conditional mean spectra and multi-scenario conditional mean spectra. In the example, the number of scenarios primarily influences the median capacity of the evaluated structure. Meanwhile, the control frequency, a key parameter of a conditional mean spectrum, plays an important role in reducing logarithmic standard deviation of the corresponding fragility curves and corresponding high confidence of low probability of failure (HCLPF) capacity.

• Communications for Statistical Applications and Methods
• /
• 제1권1호
• /
• pp.33-40
• /
• 1994

The conditional expectation of a random variable in a multivariate normal random vector is a multiple linear regression on its predecessors. Using this fact, the least median of squares estimation method developed in a multiple linear regression is adapted to a multivariate data to identify influential observations. The resulting method clearly detect outliers and it avoids the masking effect.

• 대한교통학회지
• /
• 제19권5호
• /
• pp.109-117
• /
• 2001

본 논문은 특정 도로조건에 대하여 어떤 형태의 중앙선침범 예방시설물을 설치할 것인가에 대한 비용-효율적인 기준 마련을 목표로 삼고, 그 첫 단계로 중앙선침범사고 예측모델을 개발하였다. 예측모델은 중앙선침범 예측과 침범후 대향차량과의 충돌 확률로 이루어진다. 중앙선침범 예측은 교통량에 관계되며, 이는 Hutchinson과 Kennedy(1966)의 연구에서 인용하였다. 중앙선을 침범한 후 대향차량과의 충돌 확률은 중앙선을 침범하는 차량의 침범각의 분포에 따라 주행거리와 주행시간을 계산하고, 음지수분포로 가정한 대향차량의 출현에 간격수락이론을 적용하였다. 제시된 중앙선침범사고 예측모델을 통하여 예측된 사고건수는 사고감소에 대한 편익을 계산할 수 있게 하며, 이는 중앙선침범예방시설 설치비용 계산과 함께 경제성분석을 가능하게 하여 비용-효율적인 중앙선침범예방시설 설치기준마련에 도움을 줄 것이다.

• Communications for Statistical Applications and Methods
• /
• 제17권1호
• /
• pp.27-37
• /
• 2010

This article concerns the forecasting in binomial AR(p) models which is proposed by Wei$\ss$ (2009b) for time series of binomial counts. Our method extends to binomial AR(p) models a recent result by Jung and Tremayne (2006) for integer-valued autoregressive model of second order, INAR(2), with simple Poisson innovations. Forecasts are produced by conditional median which gives 'coherent' forecasts, and we estimate the forecast distributions of future values of binomial AR(p) models by means of a Monte Carlo method allowing for parameter uncertainty. Model parameters are estimated by the method of moments and estimated standard errors are calculated by means of block of block bootstrap. The method is fitted to log data set used in Wei$\ss$ (2009b).

• Communications for Statistical Applications and Methods
• /
• 제30권5호
• /
• pp.453-465
• /
• 2023

The prediction problem of univariate records, though not addressed in multivariate records, has been discussed by many authors based on records values. There are various definitions for multivariate records among which depth-based records have been selected for the aim of this paper. In this paper, by means of the maximum likelihood and conditional median methods, point and interval predictions of depth values which are related to the future depth-based multivariate records are considered on the basis of the observed ones. The observations derived from some elements of the elliptical distributions are the main reason of studying this problem. Finally, the satisfactory performance of the prediction methods is illustrated via some simulation studies and a real dataset about Kermanshah city drought.

• 한국정보처리학회논문지
• /
• 제6권12호
• /
• pp.3718-3726
• /
• 1999

칼라 영상은 단색조의 영상에 비해 인간의 시각을 크게 향상시킨다. 따라서, 칼라 영상 처리에 관한 연구는 매우 중요하다. 칼라 영상은 센서 잡음이나 채널 전송 에러에 의해 생기는 잡음에 의해 자주 오염되어진다. 이러한 칼라 잡음을 제거하기 위해 벡터 미디안, 평균 필터, 벡터 $\alpha-trimmed$ 평균 필터 등 여러 형태의 필터들이 개발되어왔는데 특히, 혼합된 칼라 잡음의 조건에서 벡터 $\alpha-trimmed$ 평균 필터는 우수한 성능을 보여왔다. 그러나, 벡터 $\alpha-trimmed$ 평균 필터는 필터링 과정이 영상의 전 영역에 걸쳐 동일한 가중치로 균일하게 적용되어지기 때문에 스텝 윤곽선 이동이 일어나고, 이에 따라 blurring 현상이 나타나는 문제점이 있다. 이러한 문제점을 개선하기 위해 본 논문에서는 윤곽선 영역과 평탄 영역을 구분한 뒤 각 영역에 적합한 선택적인 필터링을 하는 조건적인 퍼지 클러스터 필터를 제안하였고, 제안된 조건적인 퍼지 클러스터 필터는 혼합된 잡음의 조건에서 기존의 벡터 $\alpha-trimmed$ 평균 필터에 비해 NCD척도 및 사람의 시각에 의한 평가에 의해 우수한 성능을 보였다.

• 대한지역사회영양학회지
• /
• 제27권4호
• /
• pp.309-320
• /
• 2022

Objectives: This study aimed to analyze the consumption values of fast foods among American consumers and compare the consumption values according to the levels of health consciousness. Methods: An online survey using a self-administered questionnaire was conducted on 351 American consumers. Based on the median health consciousness score (3.83 out of 5 points), the subjects were classified into the low health-conscious group (Low group) and the high health-conscious group (High group). Factor analysis was used to extract factors for the five consumption values (functional, social, emotional, conditional, and epistemic values). The differences in the consumption values between the two groups were analyzed. Results: A total of 14 factors were extracted for the five consumption values and 9 factors among them (convenience, healthiness, and taste in functional values; health-conscious people, young, busy, obese people, low class, and budget restricted people in social values; guilt in emotional values; accidental situations in conditional values) showed significant differences between the two groups. The Low group had a higher perception of the factor of healthiness (P < 0.001) than the High group. The High group had a relatively higher perception of the factors of convenience (P < 0.001), taste (P < 0.001), and guilt (P < 0.001). In addition, the High group perceived the social values of fast foods more negatively. The High group consumed fast foods less frequently than the Low group and perceived their health status and healthiness of eating habits more positively. Conclusions: The results reveal that the health consciousness level significantly influences consumption value perceptions about fast foods in American consumers. Policymakers and marketers can develop effective strategies based on the results of this study.

• 한국수자원학회논문집
• /
• 제45권8호
• /
• pp.815-826
• /
• 2012

기존 Ordinary Regression (OR) 방법을 이용한 경향성 분석은 경향성을 과소평가하는 문제점을 나타낸다. 이러한 점에서 본 연구에서는 자료의 정규분포 가정과 평균을 중심으로 경향성 평가가 이루어지는 기존 Ordinary Regression (OR) 방법을 개선한 Quantile Regression (QR) 방법을 제안하였다. 본 연구에서는 64개 강우 관측지점의 연 최대 극대강수량 자료에 대하여 QR 방법과 OR 방법에 대하여 통계적 성능을 평가하였다. QR 방법의경향성 분석결과 47개 지점에서 5% 오차수준 내에서 t-검정을 통과한 반면 OR 방법에서는 13개 지점 만이 통계적 유의성을 가지는 것으로 나타났다. 이는 OR 방법이 자료의 평균을 중심으로 경향성을 평가하는 기법인데 반해 QR은 자료의 다양한 분위에서 경향성을 평가함으로써 극대 및 극소 부분에서의 경향성을 보다 유연하게 감지하는 이유로 판단된다. QR 방법을 통한 경향성 평가는 평균 중심의 해석문제점을 개선할 수 있으며 자료가 정규분포를 따르지 않거나 왜곡된 분포형태를 갖는 자료의 수문학적 경향성 평가에 유용하게 사용될 수 있을 것으로 판단된다.