• Journal of Advanced Marine Engineering and Technology
• /
• 제32권5호
• /
• pp.768-774
• /
• 2008

In many areas of speech signal processing such as automatic speech recognition and packet based voice communication technique, VAD (voice activity detection) plays an important role in the performance of the overall system. In this paper, we present a new feature parameter for VAD which is the product of energy of the signal and the difference of two types of entropies. For this end, we first define a Mel filter-bank based entropy and calculate its difference from the conventional entropy in frequency domain. The difference is then multiplied by the spectral energy of the signal to yield the final feature parameter which we call PEED (product of energy and entropy difference). Through experiments. we could verify that the proposed VAD parameter is more efficient than the conventional spectral entropy based parameter in various SNRs and noisy environments.

• 응용통계연구
• /
• 제24권2호
• /
• pp.383-391
• /
• 2011

이 논문에서는 역가우스분포의 적합을 위한 변형된 엔트로피 기반 검정을 제시한다. 이 검정은 자료생성분포와 역가우스분포의 엔트로피 차이에 기초를 두고 있으며 검정통계량은 엔트로피 차이의 추정량을 사용한다. 엔트로피 차이의 추정량은 자료생성분포에 대한 엔트로피 추정량으로 Vasicek의 표본엔트로피와 역가우스분포에 대한 엔트로피 추정량로 균일최소분산불편추정량을 사용하여 얻는다. 모의실험을 통해 얻은 표본크기와 윈도크기에 따른 검정통계량의 기각값들을 표의 형태로 제공한다. 제안한 검정의 검정력 알아보기 위해 여러 대립분포와 표본크기에 대해서 모의실험을 수행하고 기존의 엔트로피 기반 검정과 비교한다.

• 대한수학회논문집
• /
• 제14권3호
• /
• pp.581-595
• /
• 1999

In the present paper we provide a short of the uni-form CLT for the function-indexed martingale difference process under the uniformly integrable entropy by establishing a maximal inequality.

• 정보처리학회논문지B
• /
• 제16B권2호
• /
• pp.93-100
• /
• 2009

본 논문에서는 차영상 엔트로피 기반의 시선 인식 시스템을 제안한다. 차영상 엔트로피는 현재 입력된 영상과 참조 영상 또는 시선의 위치별 평균 영상들로부터 차영상을 획득하고, -255부터 +255까지의 히스토그램 빈도수를 이용하여 계산한다. 차영상 엔트로피 기반의 시선 인식방법은 2가지 방법이다. 1) 첫 번째 방법은 현재 입력된 영상과 시선 위치별 45개의 평균 영상들과의 차영상 엔트로피를 계산하여 현재 응시하고 있는 방향을 인식하고, 2) 두 번째 방법은 현재 입력된 영상과 45개의 참조 영상들과의 차영상 엔트로피를 계산하여 현재의 응시 방향을 인식한다. 참조 영상은 네 방향의 영상을 입력받아 시선 위치별 45개의 평균 영상을 이용하여 생성한다. 제안한 시스템의 성능을 평가하기 위해 PCA 기반의 시선 인식 시스템과 비교 실험을 하였고, 인식 방향은 좌상, 우상, 좌하, 우하 네 방향으로 하였으며, 45개의 참조 영상 또는 평균영상에 대하여 인식 영역을 변경하여 실험하였다. 실험 결과 차영상 엔트로피는 97.00%, PCA는 95.50%의 인식률을 보여 차영상 엔트로피 기반의 시스템이 PCA 기반의 시스템보다 1.50% 더 높은 인식률을 나타내었다.

• 대한전자공학회논문지
• /
• 제12권2호
• /
• pp.1-10
• /
• 1975

본 논문은 사진데이타에 쉽게 적용할수있는 한 adaptive data compression 방법을 노표시하였다. 이웃 sample data 사이의 높은 correlation때문에 발생할 부호화 복잡성을 간편한 sample difference data로 대처하였으며, 자단의 statistical nonstationarity에 적응키 위해서 여덟가지 부호(code)로 구성된 code set중에서 최적부호를 선택토록 하였다. code erst는 두가지 등장부호와 여섯가지 보완형 Shannon-Fanro 부호로 되었다. difference data의 확률분포는 Laplacian model로, entropy의 확률분포는 Gaussian model대 하였다. 부호선별 Paranleter로서 entropy와 Pr[차이값=0]=Po를 비교하였다. 콤퓨타 실험결과 이 adaptive coding 방법으로 2대 1의 데이타 감축비를 얻었다. 이 방법은 fixed coding에 비해서 데이타 감축비와 부호화효율에서 약 10%와 15%의 이득을 주었다. 또한 도는 entropy보다 휠신 편리한 부호선별 parameter인 중시에 entropy 경우와 1% 내외의 좋은 결과를 얻을수 있음이 확인되었다. This paper presents an adaptive data compression algorithm for video data. The coling complexity due to the high correlation in the given data sequence is alleviated by coding the difference data, sequence rather than the data sequence itself. The adaptation to the nonstationary statistics of the data is confined within a code set, which consists of two constant length cades and six modified Shannon·Fano codes. lt is assumed that the probability distributions of tile difference data sequence and of the data entropy are Laplacian and Gaussion, respectively. The adaptive coding performance is compared for two code selection criteria: entropy and pr[difference value=0]=po. It is shown that data compression ratio 2 : 1 is achievable with the adaptive coding. The gain by the adaptive coding over the fixed coding is shown to be about 10% in compression ratio and 15% in code efficiency. In addition, po is found to he not only a convenient criterion for code selection, but also such efficient a parameter as to perform almost like entropy.

• Structural Engineering and Mechanics
• /
• 제58권1호
• /
• pp.21-38
• /
• 2016

Many studies have been carried out to investigate the important factors in calculating the realistic entropy amount of water distribution networks, but none of them have considered both mechanical and hydraulic characteristics of the networks. Also, the entropy difference in various networks has not been calculated exactly. Therefore, this study suggested a modified entropy function to calculate the informational entropy of water distribution networks so that the order of demand nodes and entropy difference among various networks could be calculated by taking into account both mechanical and hydraulic characteristics of the network. This modification was performed through defining a coefficient in the entropy function as the amount of outflow at each node to all dissipated power in the network. Hence, a more realistic method for calculating entropy was presented by considering both mechanical and hydraulic characteristics of network while keeping simplicity. The efficiency of the suggested method was evaluated by calculating the entropy of some sample water networks using the modified function.

• 대한수학회보
• /
• 제47권1호
• /
• pp.39-51
• /
• 2010

In this paper we consider the uniform central limit theorem for a martingale-difference array of a function-indexed stochastic process under the uniformly integrable entropy condition. We prove a maximal inequality for martingale-difference arrays of process indexed by a class of measurable functions by a method as Ziegler [19] did for triangular arrays of row wise independent process. The main tools are the Freedman inequality for the martingale-difference and a sub-Gaussian inequality based on the restricted chaining. The results of present paper generalizes those of Ziegler [19] and other results of independent problems. The results also generalizes those of Bae and Choi [3] to martingale-difference array of a function-indexed stochastic process. Finally, an application to classes of functions changing with n is given.

• Journal of the Korean Physical Society
• /
• 제73권9호
• /
• pp.1303-1309
• /
• 2018

In the present paper, we study the entropy analysis of boundary layer flow over a slender stretching sheet under the action of a non uniform magnetic field that is acting perpendicular to the flow direction. The effects of viscous dissipation and Joule heating are included in the energy equation. Using similarity transformation technique the momentum and thermal boundary layer equations to a system of nonlinear differential equations. Numerical solutions are obtained using the shooting and fourth-order Runge-Kutta method. The expressions for the entropy generation number and Bejan number are also obtained using a suggested similarity transformation. The main objective of this article is to investigate the effects of different governing parameters such as the magnetic parameter ($M^2$), Prandtl number (Pr), Eckert number (Ec), velocity index parameter (m), wall thickness parameter (${\alpha}$), temperature difference parameter (${\Omega}$), entropy generation number (Ns) and Bejan number (Be). All these effects are portrayed graphically and discussed in detail. The analysis reveals that entropy generation reduces with decreasing wall thickness parameter and increasing temperature difference between the stretching sheet and the fluid outside the boundary layer. The viscous and magnetic irreversibilities are dominant in the vicinity of the stretching surface.

• International Journal of Internet, Broadcasting and Communication
• /
• 제5권2호
• /
• pp.5-10
• /
• 2013

We proposed a novel similarity measure using weighted difference entropy (WDE) to improve the performance of the CF system. The proposed similarity metric evaluates the entropy with a preference score difference between the common rated items of two users, and normalizes it based on the Gaussian, tanh and sigmoid function. We showed significant improvement of experimental results and environments. These experiments involved changing the number of nearest neighborhoods, and we presented experimental results for two data sets with different characteristics, and results for the quality of recommendation.

• 대한수학회보
• /
• 제51권2호
• /
• pp.317-328
• /
• 2014

In Bae et al. [2], we have considered the uniform CLT for the martingale difference arrays under the uniformly integrable entropy. In this paper, we prove the same problem under the bracketing entropy condition. The proofs are based on Freedman inequality combined with a chaining argument that utilizes majorizing measures. The results of present paper generalize those for a sequence of stationary martingale differences. The results also generalize independent problems.