• 한국멀티미디어학회논문지
• /
• 제15권10호
• /
• pp.1257-1263
• /
• 2012

본 연구는 3D레이저 스캔 방식으로 계측된 인체 데이터를 대상으로 하여 인체의 여러 동작들에 대한 애니메이션 모듈 구현을 목표로 하였다. 이를 위하여 애니메이션 회전을 위한 기준점인 인체의 골격 기준점을 추출하고 추출된 기준점을 이용하여 골격을 잡고 각 골격에 따른 계층트리를 구성하였다. 구성된 계층트리의 골격에 해당되는 오브젝트 정점들을 골격과 연결하고 주어진 애니메이션 3차원 정점들에 행동 패턴을 적용하여 스캔데이터에 애니메이션을 구현하였다.

• 응용통계연구
• /
• 제25권3호
• /
• pp.415-422
• /
• 2012

For statistical microarray data analysis, clustering analysis is a useful exploratory technique and offers the promise of simultaneously studying the variation of many genes. However, most of the proposed clustering methods are not rigorously solved for a time-course microarray data cluster and for a fitting time covariate; therefore, a statistical method is needed to form a cluster and represent a linear trend of each cluster for each gene. In this research, we developed a modified hierarchical mixture of an experts model to suggest clustering data and characterize each cluster using a linear mixed effect model. The feasibility of the proposed method is illustrated by an application to the human fibroblast data suggested by Iyer et al. (1999).

• Molecules and Cells
• /
• 제25권2호
• /
• pp.279-288
• /
• 2008

The analysis of microarray data is essential for large amounts of gene expression data. In this review we focus on clustering techniques. The biological rationale for this approach is the fact that many co-expressed genes are co-regulated, and identifying co-expressed genes could aid in functional annotation of novel genes, de novo identification of transcription factor binding sites and elucidation of complex biological pathways. Co-expressed genes are usually identified in microarray experiments by clustering techniques. There are many such methods, and the results obtained even for the same datasets may vary considerably depending on the algorithms and metrics for dissimilarity measures used, as well as on user-selectable parameters such as desired number of clusters and initial values. Therefore, biologists who want to interpret microarray data should be aware of the weakness and strengths of the clustering methods used. In this review, we survey the basic principles of clustering of DNA microarray data from crisp clustering algorithms such as hierarchical clustering, K-means and self-organizing maps, to complex clustering algorithms like fuzzy clustering.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제4권3호
• /
• pp.224-242
• /
• 2010

Data routing in wireless sensor networks must be energy-efficient because tiny sensor nodes have limited power. A cluster-based hierarchical routing is known to be more efficient than a flat routing because only cluster-heads communicate with a sink node. Existing hierarchical routings, however, assume unrealistically large radio transmission ranges for sensor nodes so they cannot be employed in real environments. In this paper, by considering the practical transmission ranges of the sensor nodes, we propose a clustering and routing method for hierarchical sensor networks: First, we provide the optimal ratio of cluster-heads for the clustering. Second, we propose a d-hop clustering scheme. It expands the range of clusters to d-hops calculated by the ratio of cluster-heads. Third, we present an intra-cluster routing in which sensor nodes reach their cluster-heads within d-hops. Finally, an inter-clustering routing is presented to route data from cluster-heads to a sink node using multiple hops because cluster-heads cannot communicate with a sink node directly. The efficiency of the proposed clustering and routing method is validated through extensive simulations.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권5호
• /
• pp.1343-1356
• /
• 2013

Cloud computing has emerged as perhaps the hottest development in information technology at present. This new computing technology requires that the users ensure that their infrastructure is safety and that their data and applications are protected. In addition, the customer must ensure that the provider has taken the proper security measures to protect their information. In order to achieve fine-grained and flexible access control for cloud computing, a new construction of hierarchical attribute-based encryption(HABE) with Ciphertext-Policy is proposed in this paper. The proposed scheme inherits flexibility and delegation of hierarchical identity-based cryptography, and achieves scalability due to the hierarchical structure. The new scheme has constant size ciphertexts since it consists of two group elements. In addition, the security of the new construction is achieved in the standard model which avoids the potential defects in the existing works. Under the decision bilinear Diffie-Hellman exponent assumption, the proposed scheme is provable security against Chosen-plaintext Attack(CPA). Furthermore, we also show the proposed scheme can be transferred to a CCA(Chosen-ciphertext Attack) secure scheme.

• Journal of Electrical Engineering and Technology
• /
• 제10권2호
• /
• pp.699-707
• /
• 2015

Although digit character recognition has got a significant improvement in recent years, it is still challenging to achieve satisfied result if the data contains an amount of distracting factors. This paper proposes a novel digit character recognition approach using a multi-layer hierarchical model, Hybrid Restricted Boltzmann Machines (HRBMs), which allows the learning architecture to be robust to background distracting factors. The insight behind the proposed model is that useful high-level features appear more frequently than distracting factors during learning, thus the high-level features can be decompose into hybrid hierarchical structures by using only small label information. In order to extract robust and compact features, a stochastic 0-1 layer is employed, which enables the model's hidden nodes to independently capture the useful character features during training. Experiments on the variations of Mixed National Institute of Standards and Technology (MNIST) dataset show that improvements of the multi-layer hierarchical model can be achieved by the proposed method. Finally, the paper shows the proposed technique which is used in a real-world application, where it is able to identify digit characters under various complex background images.

• 한국철도학회논문집
• /
• 제9권4호
• /
• pp.432-440
• /
• 2006

Prior to the service evaluation, many kinds of its attributes must be identified on the basis of rational decision owing to complexity and ambiguity inherent in logistics service. there are so many evaluation methods but they are not applicable to logistics service the have property of non-additivity and overlapped attributes. Therefore, probability measure can not used to evaluate logistics service but Fuzzy Measure is required. And the method should be easy to calculate it Recently Fuzzy theory has been applied in Various evaluation problem. Fuzzy evaluation based on Fuzzy theory can accommodate fuzziness in judgement with people through introducing Fuzzy Measure. In this paper, Hierarchical Fuzzy Process is applied to evaluate level of logistics service in Korail and the biggest six logistics companies in the korea which is called 3PL Company. Also Fuzzy Relation Equation which is problem identifying evaluation value at these fuzzy evaluation is applied to verify relation between Input and Output data through @-operation. Therefore, we apply this Fuzzy Relation Equation to Hierarchical Fuzzy Process and verify evaluation value which objects of evaluation are able to possess.

• 한국지능시스템학회논문지
• /
• 제24권4호
• /
• pp.355-359
• /
• 2014

본 논문은 단순한 후반부 구조를 가진 퍼지 모델을 계층적 구조로 결합한 퍼지 패턴 분류기를 제안한다. 계층적 구조를 가진 퍼지 패턴 분류기의 기본 구조는 단순한 후반부 구조를 가진 퍼지 모델을 사용하여 전체 패턴 분류기의 구조적 복잡성을 높이지 않도록 설계 하였다. 입력공간을 계층적으로 분할하기 위하여 대표적인 퍼지 클러스터링 알고리즘인 Fuzzy C-Means clustering 기법을 이용하였다. 분할된 퍼지 입력 공간의 하위 구조를 분석하기 위하여 conditional Fuzzy C-Means 클러스터링 기법을 이용하였다. 계층적으로 분할된 퍼지 입력공간에 간단한 구조를 가진 퍼지 패턴 분류기를 적용하여 계층적 구조를 가진 패턴 분류기를 설계한다. 계층적으로 퍼지 모델들을 결합함으로써 입력 공간의 정보 분석을 거시적인 관점에서 시작하여 세부적으로 분석이 가능하게 되었다. 제안된 퍼지 패턴 분류기의 성능을 평가하기 위하여 다양한 기계 학습 데이터를 사용하였다.

• 한국산업정보학회논문지
• /
• 제11권1호
• /
• pp.77-85
• /
• 2006

본 논문에서는 내용기반 영상 분류를 위한 방법론으로써 신경망을 이용한 계층적 분류 방법을 제안한다. 분류 대상 영상은 인터넷상의 다양한 영상들 중에서 전경과 배경의 구분이 있는 객체 영상이다. 전처리 과정에서 영역 분할을 이용하여 영상 내에서 배경을 제거하고 객체 영역을 추출한다. 분류를 위한 특징으로는 웨이블릿 변환 후 추출된 형태 특징과 질감 특징을 이용한다. 추출된 특징 값들을 Principal Component Analysis(PCA)와 K-means를 이용해서 군집화 시키고 유사한 군집들을 묶으면서, 5단계의 계층적 분류기를 구성한다. 계층적 분류기는 BP를 학습 알고리즘으로 사용하는 59개의 신경망분류기로 구성된다. 배경을 제거하고 질감특징 중 가장 높은 분류율을 보이는 대각 모멘트를 사용하여 실험하였을 때, 100종류에서 각 10개씩, 총 1000개의 학습 데이터와 1000개의 테스트 데이터에 대하여 각각 81.5%와 75.1%의 정분류율을 보였다.

• 대한전자공학회논문지TC
• /
• 제44권1호
• /
• pp.10-17
• /
• 2007

본 논문에서는 기존의 지상파 DMB 방식과 역 호환성을 유지하는 동시에 고전송효율, 고품질 서비스를 제공하기 위한 계층 변조 기법을 소개한다. 계층 변조를 통해 기존의 데이터 스트림에 부가적인 데이터 스트림을 삽입함에 따라 개선된 지상파 DMB 시스템은 보다 더 많은 유효 전송률을 확보하고 고품질 비디오 서비스를 제공할 수 있으며, 동시에 보다 많은 방송 서비스 채널을 제공할 수 있다. 그러나 기본 서비스의 품질을 유지하면서 부가적인 서비스의 품질을 보장하기 위해서 강력한 오류정정 방식이 요구된다. 따라서 본 논문에서는 다양한 오류정정 방식에 따른 계층변조가 적용된 개선된 지상파 DMB 시스템의 오류 성능을 조사하고 분석한다.