• 한국컴퓨터정보학회논문지
• /
• 제19권7호
• /
• pp.9-16
• /
• 2014

Dark channel prior를 이용한 안개제거 알고리즘으로 만족할만한 연구결과가 발표된 이후로 이 알고리즘의 처리 속도를 높이기 위한 많은 연구들이 진행되었다. 이들 중에서 median dark channel prior를 이용한 알고리즘이 주목을 받고 있지만 여전히 낮은 처리속도의 한계를 갖고 있다. 그래서 본 논문에서는 고화질 영상에서 고속 안개 제거를 위한 SIMD 구조에 적합한 병렬메모리 모델을 제안한다. 제안하는 병렬메모리 모델은 n개의 화소들에 동시에 접근할 수 있으며, 3, 5, 7 또는 11의 크기를 갖는 4가지 종류의 median filter를 위한 간격들을 허용한다. 그래서 충분한 데이터 대역폭을 지원하기에 median dark channel prior를 이용한 알고리즘을 고속으로 처리할 수 있다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권5호
• /
• pp.2529-2543
• /
• 2019

Scanning electron microscopy (SEM) image can link with the microscopic world through reflecting interaction between electrons and materials. The SEM images are easily subject to blurring distortions during the imaging process. Inspired by the fact that dark channel prior captures the changes to blurred SEM images caused by the blur process, we propose a method to evaluate the SEM images sharpness based on the dark channel prior. A SEM image database is first established with mean opinion score collected as ground truth. For the quality assessment of the SEM image, the dark channel map is generated. Since blurring is typically characterized by the spread of edge, edge of dark channel map is extracted. Then noise is removed by an edge-preserving filter. Finally, the maximum gradient and the average gradient of image are combined to generate the final sharpness score. The experimental results on the SEM blurred image database show that the proposed algorithm outperforms both the existing state-of-the-art image sharpness metrics and the general-purpose no-reference quality metrics.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제36권2호
• /
• pp.109-121
• /
• 2009

사전정보는 클러스터링 성능을 유도할 수 있는 요인이지만, 활용 방법에 따라 차이는 발생한다. 특히, 사전정보를 초기 중심으로 활용할 때, 사전정보 간 유사도에 대해 고려하는 것이 필요하다. 레이블이 같더라도 낮은 유사도를 갖는 사전정보로 인해 초기 중심 설정 시 문제가 발생할 수 있기 때문에, 이들을 구분하여 활용하는 방법이 필요하다. 따라서 본 논문은 낮은 유사도를 갖는 사전정보를 구분하여 문제를 해결하는 방법을 제시한다. 또한 유사도에 의해 구분된 사전정보는 다양하게 활용함으로써 생성되는 다양한 클러스터링 결과를 연관규칙에 기반하여 앙상블 함으로써 통합된 하나의 분석 결과를 도출하여 클러스터링 분석 성능을 더욱 개선시킬 수 있다.

• 한국의류학회지
• /
• 제33권4호
• /
• pp.519-530
• /
• 2009

The purpose of this study is to investigate the effects of the prior knowledge, the market position and negative information of the alliance apparel product on consumer attitude including preference, purchase intent and utility. Smart wear with MP3 was selected for the alliance apparel product. Negative information was manipulated into two types-product related and brand related negative information. 251 subjects participated for the study. For the data analysis, reliability test and three way analysis of variance were conducted. The results showed that when the partner brand has the higher market position, subjects with high prior knowledge revealed preference for the alliance apparel product more. When the partner brand has the higher market position, the subjects who were given the negative information on the alliance apparel product reported preference and utility more than the ones who were given the negative information on the company. The findings of the study imply that apparel industries should make an effort to establish the positive corporate image as well as to produce high quality apparel product. Also marketers should provide consumers with the knowledge about brand new alliance apparel product.

• Journal of Electrical Engineering and Technology
• /
• 제10권5호
• /
• pp.2179-2188
• /
• 2015

Recently, many researches have been done to solve the challenging problem of Blind Source Separation (BSS) problems in the underdetermined cases, and the “Two-step” method is widely used, which estimates the mixing matrix first and then extracts the sources. To estimate the mixing matrix, conventional algorithms such as Single-Source-Points (SSPs) detection only exploits the sparsity of original signals. This paper proposes a new underdetermined mixing matrix estimation method for time-delayed mixtures based on the receiver prior exploitation. The prior information is extracted from the specific structure of the complex-valued mixing matrix, which is used to derive a special criterion to determine the SSPs. Moreover, after selecting the SSPs, Agglomerative Hierarchical Clustering (AHC) is used to automaticly cluster, suppress, and estimate all the elements of mixing matrix. Finally, a convex-model based subspace method is applied for signal separation. Simulation results show that the proposed algorithm can estimate the mixing matrix and extract the original source signals with higher accuracy especially in low SNR environments, and does not need the number of sources before hand, which is more reliable in the real non-cooperative environment.

• Communications for Statistical Applications and Methods
• /
• 제14권2호
• /
• pp.431-442
• /
• 2007

In this paper, we develop the noninformative priors when the parameter of interest is the difference between quantiles of two exponential distributions. We want to develop the first and second order probability matching priors. But we prove that the second order probability matching prior does not exist. It turns out that Jeffreys' prior does not satisfy the first order matching criterion. The Bayesian credible intervals based on the first order probability matching prior meet the frequentist target coverage probabilities much better than the frequentist intervals of Jeffreys' prior. Some simulation and real example will be given.

• 대한의용생체공학회:학술대회논문집
• /
• 대한의용생체공학회 1997년도 춘계학술대회
• /
• pp.279-282
• /
• 1997

Bayesian reconstruction methods for emission computed tomography have been a topic of interest in recent years, partly because they allow for the introduction of prior information into the reconstruction problem. Early formulations incorporated priors that imposed simple spatial smoothness constraints on the underlying object using Gibbs priors in the form of four-nearest or eight-nearest neighbors. While these types of priors, known as "membrane" priors, are useful as stabilizers in otherwise unstable ML-EM reconstructions, more sophisticated prior models are needed to model underlying source distributions more accurately. In this work, we investigate whether the "thin plate" model has advantages over the simple Gibbs smoothing priors mentioned above. To test and compare quantitative performance of the reconstruction algorithms, we use Monte Carlo noise trials and calculate bias and variance images of reconstruction estimates. The conclusion is that the thin plate prior outperforms the membrane prior in terms of bias and variance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권10호
• /
• pp.4835-4855
• /
• 2018

In this paper, we study the sparse signal recovery that uses information of both support and amplitude of the sparse signal. A convergent iterative algorithm for sparse signal recovery is developed using Majorization-Minimization-based Non-convex Optimization (MM-NcO). Furthermore, it is shown that, typically, the sparse signals that are recovered using the proposed iterative algorithm are not globally optimal and the performance of the iterative algorithm depends on the initial point. Therefore, a modified MM-NcO-based iterative algorithm is developed that uses prior information of both support and amplitude of the sparse signal to enhance recovery performance. Finally, the modified MM-NcO-based iterative algorithm is used to estimate the time-varying sparse wireless channels with temporal correlation. The numerical results show that the new algorithm performs better than related algorithms.

• 대한산업공학회지
• /
• 제5권1호
• /
• pp.53-57
• /
• 1979

The concept of AOQL in designing Dodge's continuous sampling plans is modified to include probabilistic consideration reflecting the prior knowledge about the process average fraction defectives, and a new design criterion called AOQL, which eliminates some of the drawbacks of the AOQL criterion is proposed. AOQL, approach provides more economical sampling plans in many cases, and can be used even when only limited amount of prior information is available.

• 소음진동
• /
• 제9권6호
• /
• pp.1240-1246
• /
• 1999

A prior error estimator which is solving structural dynamic problems and which is based on the generalized-method, is developed. Since the proposed error estimator is computed with only previous information, the time step size can be adaptively selected without the feedback mechanism. This paper shows that the automatic time stepping algorithm using the error estimator performs an efficient time integration. To verify its efficiency, several examples are numerically investigated.