• 한국통신학회논문지
• /
• 제22권4호
• /
• pp.662-674
• /
• 1997

In order to post process the vector-quantized images employing the theory of projections onto convex sets or the constrained minimization technique, the the projector onto QCS(quantization constraint set) as well as the filter that smoothes the lock boundaries should be investigated theoretically. The basic idea behind the projection onto QCS is to prevent the processed data from diverging from the original quantization region in order to reduce the blurring artifacts caused by a filtering operation. However, since the Voronoi regions in order to reduce the blurring artifacts caused by a filtering operation. However, since the Voronoi regions in the vector quantization are arbitrarilly shaped unless the vector quantization has a structural code book, the implementation of the projection onto QCS is very complicate. This paper mathematically analyzes the projection onto QCS from the viewpoit of minimizing the mean square error. Through the analysis, it has been revealed that the projection onto a subset of the QCS yields lower distortion than the projection onto QCS does. Searching for an optimal constraint set is not easy and the operation of the projector is complicate, since the shape of optimal constraint set is dependent on the statistical characteristics between the filtered and original images. Therefore, we proposed a hyper-cube as a constraint set that enables a simple projection. It sill be also shown that a proper filtering technique followed by the projection onto the hyper-cube can reduce the quantization distortion by theory and experiment.

• 통합자연과학논문집
• /
• 제10권1호
• /
• pp.33-39
• /
• 2017

Consider the problem of estimating a $p{\times}1$ mean vector ${\theta}(p-q{\geq}3)$, $q=rank(P_V)$ with a projection matrix $P_v$ under the quadratic loss, based on a sample $X_1$, $X_2$, ${\cdots}$, $X_n$. We find a James-Stein type decision rule which shrinks towards projection vector when the underlying distribution is that of a variance mixture of normals and when the norm ${\parallel}{\theta}-P_V{\theta}{\parallel}$ is restricted to a known interval, where $P_V$ is an idempotent and projection matrix and rank $(P_V)=q$. In this case, we characterize a minimal complete class within the class of James-Stein type decision rules. We also characterize the subclass of James-Stein type decision rules that dominate the sample mean.

• Communications for Statistical Applications and Methods
• /
• 제27권5호
• /
• pp.523-533
• /
• 2020

This paper suggests three available methods for finding nonnegative estimates of variance components of the random effects in mixed models. The three proposed methods based on the concepts of projections are called projection method I, II, and III. Each method derives sums of squares uniquely based on its own method of projections. All the sums of squares in quadratic forms are calculated as the squared lengths of projections of an observation vector; therefore, there is discussion on the decomposition of the observation vector into the sum of orthogonal projections for establishing a projection model. The projection model in matrix form is constructed by ascertaining the orthogonal projections defined on vector subspaces. Nonnegative estimates are then obtained by the projection model where all the coefficient matrices of the effects in the model are orthogonal to each other. Each method provides its own system of linear equations in a different way for the estimation of variance components; however, the estimates are given as the same regardless of the methods, whichever is used. Hartley's synthesis is used as a method for finding the coefficients of variance components.

• 한국통신학회논문지
• /
• 제21권9호
• /
• pp.2341-2348
• /
• 1996

This paper proposes an image compression algorithm that adopts projection scheme on mean-residual metod. Sub-blocks of an image are encoded using mean-residual method where mean value is predicted according to that of neighboring blocks. Projection scheme with 8 directions is applied to the compression of residual signals of blocks. Projection vectors are finite-state vector quantized according to the projection angle of nighboring blocks in order to exploit the correlation among them. Side information to represent the repetition of projection is run-length coded while the information for projection direction is compressed using entropy encoding. The proposed scheme apears to be better in PSNR performance when compared with conventional projection scheme as well as in subjective quality preserving the edges of images better than most tranform methods which usually require heavy computation load.

• Journal of the Korean Data and Information Science Society
• /
• 제28권2호
• /
• pp.435-442
• /
• 2017

Consider the problem of estimating a $p{\times}1$ mean vector ${\theta}$ (p ${\geq}$ 3) under the quadratic loss from multi-variate normal population. We find a James-Stein type estimator which shrinks towards the projection vectors when the underlying distribution is that of a variance mixture of normals. In this case, the norm ${\parallel}{\theta}-K{\theta}{\parallel}$ is known where K is a projection vector with rank(K) = q. The class of this type estimator is quite general to include the class of the estimators proposed by Merchand and Giri (1993). We can derive the class and obtain the optimal type estimator. Also, this research can be applied to the simple and multiple regression model in the case of rank(K) ${\geq}2$.

• The Journal of the Acoustical Society of Korea
• /
• 제27권1E호
• /
• pp.25-29
• /
• 2008

The projection approximation subspace tracking (PAST) is one of the attractive subspace tracking algorithms, because it estimatesthe signal subspace adaptively and continuously. Furthermore, the computational complexity is relatively low. However, the algorithm still has room for improvement in the subspace estimation accuracy. In this paper, we propose a new algorithm to improve the subspace estimation accuracy using a normally ordered input vector and a reversely ordered input vector simultaneously.

• 한국정보과학회논문지:정보통신
• /
• 제28권4호
• /
• pp.651-656
• /
• 2001

히스토그램 인터섹션은 영상에서 컬러가 가지는 값의 빈도 수를 이용하여 간단하면서도 효율적으로 영상을 검색하는 방법으로, 영상의 글로블 특성을 잘 나타내는 반면 영상에서의 위치 정보가 누락되어 다른 영상을 동일 영상으로 인지하기 쉽고, 영상 내에 포함된 형상 정보를 표현하는 적절한 방법은 아니다. 영상에 대한 1차원 투영을 이용하면 영상의 개략적인 형상 정보와 함께 위치 정보를 나타낼 수 있어 히스토그램의 단점을 극복할 수 있지만, 영상 크기에 따라 투영 벡터의 길이가 달라져 색인 데이타로 사용하기에는 문제가 있다. 본 논문에서는 투영벡터에서 영상이 가지는 형상 정보의 첨두치를 이용하여 첨두치들 간의 상대거리 및 최대 첨두치에 관한 정보를 검색에 사용하였다. 검색 성능의 확인을 위하여 히스토그램 인터섹션 및 투영벡터만을 이용한 경우의 검색 결과와 비교하였고, 실험 결과를 이용하여 각 방법의 장단점을 분석하였다.

• 통합자연과학논문집
• /
• 제11권3호
• /
• pp.154-160
• /
• 2018

Consider the problem of estimating a $p{\times}1$ mean vector ${\theta}(p-r{\geq}3)$, r = rank(K) with a projection matrix K under the quadratic loss, based on a sample $Y_1$, $Y_2$, ${\cdots}$, $Y_n$. In this paper a James-Stein type estimator with shrinkage form is given when it's variance distribution is specified and when the norm ${\parallel}{\theta}-K{\theta}{\parallel}$ is constrain, where K is an idempotent and symmetric matrix and rank(K) = r. It is characterized a minimal complete class of James-Stein type estimators in this case. And the subclass of James-Stein type estimators that dominate the sample mean is derived.

• 한국통신학회논문지
• /
• 제25권7A호
• /
• pp.1044-1049
• /
• 2000

영상의 특징을 나타내는 방법의 하나인 투영은 영상의 근사화된 형상 및 위치 정보 등의 많은 유용한 정보를 포함하고 있다. 그러나 투영을 영상 검색을 위한 방법으로 사용할 경우, 사용되는 색인 데이터 량이 많고, 에디터 베이스의 영상 크기에 따라 토영된 벡터의 길이가 달라진다는 단점이 있다. 이에 본 논문에서는 투영기법이 안고 있는 이러한 문제점을 극복하는 방법으로 데이터베이스 영상을 투영한 후 투영 벡터의 국부화를 통하여 영상의 지역적 특성이 반영되도록 하였으며, 색인 데이터 량을 주리기 위하여 투영된 벡터의 분산 값을 색인 데이터로 활용하였다. 제안된 방법은 검색 시 투영 기법의 장점을 수용함과 동시에 영상의 통계적 특성을 활용할 수 있을 뿐 아니라 시스템 구현 시 질의 시간 내에 응답을 얻을 수 있다는 이점이 있다.

• 한국HCI학회논문지
• /
• 제1권2호
• /
• pp.27-33
• /
• 2006

본 논문은 회전 및 천이 불변 이미지 텍스처 검색의 새로운 방식을 소개한다. 주파수 영역의 극 좌표계에서 동일한 공간주파수에서 각도방향으로 투영을 함으로써 각도 투영법을 만들어 냈으며, 제안된 각도 투영법을 이용하여 주파수 영역에서 푸리에 계수의 합과 표준 편차를 특징벡터로 이용하였다. 각도 투영법을 쉽게 구현하기 위하여 극 좌표계에서 라돈변환이 수행된다. 실험 시 MPEG-7 데이터를 이용하였으며 그 결과는 여러 텍스처 이미지를 검 색하는데 있어서 특징을 잘 구별해 내는 결과를 보여준다. 또한 제안된 회전 및 천이불변 특징 추출 알고리듬은 등 방성 텍스처나 국부적인 방향성을 보이는 텍스처 영상 검색에서 효율적인 검색률을 보인다.