• Communications for Statistical Applications and Methods
• /
• 제10권1호
• /
• pp.1-9
• /
• 2003

In this article, we consider a Bayesian estimation method for the geometric mean of $textsc{k}$ exponential parameters, Using the Tibshirani's orthogonal parameterization, we suggest an invariant prior distribution of the $textsc{k}$ parameters. It is seen that the prior, probability matching prior, is better than the uniform prior in the sense of correct frequentist coverage probability of the posterior quantile. Then a weighted Monte Carlo method is developed to approximate the posterior distribution of the mean. The method is easily implemented and provides posterior mean and HPD(Highest Posterior Density) interval for the geometric mean. A simulation study is given to illustrates the efficiency of the method.

• 인터넷정보학회논문지
• /
• 제4권2호
• /
• pp.31-37
• /
• 2003

최근 움직임 포착 시스템은 애니메이션 연구 분야에 가장 활발하게 적용되고 있다. 움직임 데이터의 포착은 움직임 포착 시스템은 실제 배우의 움직임과 연결하여 이루어진다. 본 연구는 기하학적 평균 필터를 이용하여 움직임 데이터를 편집하는 방법에 대해 다룬다. 포착한 움직임 데이터에 잡음이 끼면 부자연스러운 동작을 생성하기 때문에 동작을 자연스럽게 만들어 내는 중간 과정이 필요한데 기하학적 평균 필터를 이용하여 자연스러운 동작을 생성할 수 있도록 한 것이다. 실험 결과는 기하학적 평균필터가 다양한 공간필터 중에서 움직임 데이터의 잡음을 제거하는데 가장 우수함을 보여준다. 본 연구의 제안 방법은 실시간 애니메이션, 가상현신, 3D 응용분야등에 다양하게 응용 가능하다.

• 대한원격탐사학회지
• /
• 제30권6호
• /
• pp.769-776
• /
• 2014

동일 대상에 대한 두 영상의 등록을 위해서는 두 영상에 공통적으로 존재하는 특징점을 검출하고 검출된 특징점 간의 대응관계를 찾는 과정이 필수적이다. 본 논문에서는 화소의 밝기 변화를 측정할 수 있는 그레디언트 행렬의 고유치 기하평균에 기반한 새로운 특징점 검출기를 제안한다. 제안하는 특징점 검출기는 그레디언트 행렬의 두 고유치의 기하평균 크기에 비례하고 기하 평균 크기가 동일한 경유 두 고유치의 상대적인 차이에 비례하여 가변적으로 변하는 특성을 가진다. 제안한 특징점 검출기의 성능 평가를 위해 다양한 종류의 코너가 존재하는 합성 영상과 항공 영상을 기준 영상으로 사용하여 코너 검출의 위치 오차를 분석하였다. 제안한 검출기의 위치 오차는 Gaussian smoothing scale 조건하에서 대표적인 코너 검출기인 Harris detector의 위치 오차보다 작은 결과가 얻어졌다.

• 대한수학회논문집
• /
• 제31권1호
• /
• pp.95-100
• /
• 2016

In this paper, we establish a new refinement of the arithmetic-geometric mean inequality. Applying this result in information theory, we obtain a more precise upper bound for Shannon's entropy.

• 대한방사선기술학회지:방사선기술과학
• /
• 제20권1호
• /
• pp.35-38
• /
• 1997

Scintigraphic measurement of gastric emptying time has been reported to be influenced by the variation in depth of radionuclide within the stomach. This study was designed to clarify whether a part of the variability in gastric emptying could be ascribed to a relationship between anterior image, the total anteroposterior Image and the tissue attenuation correction(geometric mean). A dual-head scintillation camera(ADAC, USA) was used to investigate effect of such changes. We were performed 16 normal subject gastric emptying studies with $^{99m}TC$ labelled scramble egg, milk and solid meal(610 Kcal, 300 g) The results are as follows; On anterior Image, $T_{1/2}$ emptying time was delayed by 5 min, 6.5%(range $3{\sim}18\;min,\;5{\sim}31.4%$) compared with the geometric mean. But there was no different gastric emptying time between the total anteroposterior image and geometric mean. Therefore, if will be useful to use the method of geometric mean or the total anteroposterior image to evaluate the gastric emptying time accurately.

• 한국산업보건학회지
• /
• 제9권1호
• /
• pp.14-22
• /
• 1999

This study was conducted to evaluate workplace protection factors(WPF) for two half-facepiece respirators (HR-1, HR-2) in welding workplace and to provide data on the workplace performance of negative-pressure, half-facepiece respirators against airborne particulate contaminants. The outside iron oxide(Fe2O3) concentration of welding fume for the respirator HR-1 ranged from 0.177 to $12.508mg/m^3$ with a geometric mean of $1.118mg/m^3$ and the HR-2 respirator showed a iron oxide range of 0.500 to $3.494mg/m^3$ with a geometric mean of $1.082mg/m^3$. But the inside oxide concentration of welding fume for the respirator HR-1 ranged from 0.002 to $0.364mg/m^3$ with a geometric mean of $0.019mg/m^3$ and the HR-2 respirator showed a iron oxide range of 0.012 to $0.639mg/m^3$ with a geometric mean of $0.041mg/m^3$. The iron oxide inside concentrations were significantly less than $5mg/m^3$(TLV) for both of respirators. The WPF were ranged from 3 to 3744 with a geometric mean of 60 for HR-1 and range from 2 to 129 a geometric mean of 26 for HR-2. And, in this study, the 5th percentile of the workplace protection factors for half-facepiece aspirators (HR-1, HR-2) were 11.2 and 7.1, respectively. The correlation relationship between the Quantative Fit Factors(QNFT) and the WPF for half-mask negative pressure respirators were 0.099 and 0.460.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제54권1호
• /
• pp.40-47
• /
• 2005

This paper concerns on Filtered-x least mean square (FXLMS) algorithm for adaptive estimation of feedforward control parameters. The conditions for convergence in ensemble mean of the FXLMS algorithm are derived and the directional convergence properties are discussed from a new geometric vector analysis. The convergence and its directionality are verified along with some computer simulations.

• 충청수학회지
• /
• 제7권1호
• /
• pp.217-226
• /
• 1994

We introduce the notion of the mean Bocce dentable and provide the geometric characterization of the CCP. We show that X has the CCP if and only if every bounded subset of X is mean Bocce dentable.

• 한국지진공학회논문집
• /
• 제20권6호
• /
• pp.369-378
• /
• 2016

The design spectrum for Korea, which is known to belong to an intra-plate region, is developed from the ground motion records of the earthquakes occurred in Korea and overseas intra-plate regions. The horizontal spectrum is defined as geometric mean spectrum, GMRotI50. From the statistical analysis of the geometric mean spectra, a mean plus one standard deviation spectrum in lognormal distribution is obtained. Regression analysis is performed on this curve to determine the shape of spectrum including transition periods. The developed design spectrum is valid for the estimation both spectral acceleration and displacement.

• 대한수학회보
• /
• 제56권2호
• /
• pp.521-533
• /
• 2019

In this paper, we define a mean of two positive numbers called the k-golden mean and study some properties of it. Especially, we show that the 2-golden mean refines the harmonic and the geometric means. As an application, we define the k-golden ratio and give some properties of it as an generalization of the golden ratio. Furthermore, we define the matrix k-golden mean of two positive-definite matrices and give some properties of it. This is an improvement of Lim's results [2] for which the matrix golden mean.