• 한국통신학회논문지
• /
• 제34권8C호
• /
• pp.772-781
• /
• 2009

This paper proposes a novel algorithm for the efficient classification and retrieval of medical images. After color and edge features are extracted from medical images, these two feature vectors are then applied to a multi-class Support Vector Machine, to give membership vectors. Thereafter, the two membership vectors are combined into an ensemble feature vector. Also, to reduce the search time, Correlated Categories Vector is proposed for similarity matching. The experimental results show that the proposed system improves the retrieval performance when compared to other methods.

• 대한전자공학회논문지TC
• /
• 제40권10호
• /
• pp.59-66
• /
• 2003

본 논문에서는 다수 트래픽 클래스들에게 최소 대역폭을 보장해 줄 수 있는 광버스트스위칭 (Optical Burst Switching, OBS) 기법을 제안한다. 현재까지 제안되었던 QoS (Quality of Service) 관련 OBS 기법들은 클래스간 차별화는 가능하였으나 여러 클래스들이 존재하는 경우 하위 우선순위의 트래픽 클래스는 상급 클래스와 충돌로 최소 대역폭을 보장해 주진 못 하였다. 본 논문에서는 클래스별로 데이터 채널의 일정 시간구간을 할당하여, 최소한 그 영역에서는 최상급 클래스보다도 높은 우선순위를 부여하여 최소 대역폭을 보장받도록 하였다. 이런 동작을 효율적으로 가능하게 하기 위해서 새로운 버스트 어셈블리 알고리즘과, 데이터 채널의 시간영역을 관리방법을 제안한다. 또한 제안된 기법의 단대단 지연시간과 성능을 시뮬레이션을 통해 검증하였다.

• 정보처리학회논문지B
• /
• 제11B권5호
• /
• pp.611-618
• /
• 2004

본 논문에서는 인터넷상의 대용량 자료에서 원하는 정보를 검색하기 위한 지원 기능으로서 분산 객체지향 데이타베이스에서 클래스를 분할하여 여러 사이트에 분산시키는 기법에 관하여 설계 및 구현하였다 제안된 분산 기법은 클래스의 분할 과정과 할당 과정으로 구성된다. 클래스의 분할 과정에서는 메소드, 계승 및 복합 객체와 같은 객체지항 데이터베이스의 특성을 반영하여 클래스를 분할하였으며 할당 과정에서는 저장, 질의 처리 및 전송비용을 고려하여 할당수식을 정의하였으며 이를 유전자 알고리즘을 이용하여 구현하였다.

• 경영과학
• /
• 제17권2호
• /
• pp.75-86
• /
• 2000

In evolutionary algorithm, there exist various models for the evolution of the population with respect to schemes and strategies for reproduction. In the application of the algorithm to a specific problem, one model suitable to the problem is to be properly chosen and a program expert or a software is needed to help implement and test a designed algorithm. In this study, abject oriented modeling and the class library for simple evolutionary algorithms(SEA) with one population is developed. The library proposed here can be used as a generalized tool for solving problems in a wide range of domains.

• 한국정보전자통신기술학회논문지
• /
• 제9권6호
• /
• pp.546-552
• /
• 2016

We present a quantitative evaluation of an algorithm for model-based face recognition. The algorithm actively learns how individual faces vary through video sequences, providing on-line suppression of confounding factors such as expression, lighting and pose. By actively decoupling sources of image variation, the algorithm provides a framework in which identity evidence can be integrated over a sequence. We demonstrate that face recognition can be considerably improved by the analysis of video sequences. The method presented is widely applicable in many multi-class interpretation problems.

• ETRI Journal
• /
• 제29권6호
• /
• pp.769-778
• /
• 2007

This paper proposes a multiplication algorithm for $F_{p^m}$, which can be efficiently applied to many pairs of characteristic p and extension degree m except for the case that 8p divides m(p-1). It uses a special class of type- Gauss period normal bases. This algorithm has several advantages: it is easily parallelized; Frobenius mapping is easily carried out since its basis is a normal basis; its calculation cost is clearly given; and it is sufficiently practical and useful when parameters k and m are small.

• 한국지능시스템학회논문지
• /
• 제14권3호
• /
• pp.363-368
• /
• 2004

In this paper, the robustness property of 2nd-order iterative learning control(ILC) method for a class of linear and nonlinear discrete-time dynamic systems is studied. 2nd-order ILC method has the PD-type learning algorithm based on both time-domain performance and iteration-domain performance. It is proved that the 2nd-order ILC method has robustness in the presence of state disturbances, measurement noise and initial state error. In the absence of state disturbances, measurement noise and initialization error, the convergence of the 2nd-order ILC algorithm is guaranteed. A numerical example is given to show the robustness and convergence property according to the learning parameters.

• 대한수학회지
• /
• 제45권5호
• /
• pp.1323-1339
• /
• 2008

In this paper, a new class of $(h,{\eta})$-proximal for proper functionals in Hilbert spaces is introduced. The existence and Lip-schitz continuity of the $(h,{\eta})$-proximal mappings for proper functionals are proved. A class of generalized nonlinear quasi-variational-like inclusions in Hilbert spaces is introduced. A perturbed three-step iterative algorithm with errors for the generalized nonlinear quasi-variational-like inclusion is suggested. The existence and uniqueness theorems of solution for the generalized nonlinear quasi-variational-like inclusion are established. The convergence and stability results of iterative sequence generated by the perturbed three-step iterative algorithm with errors are discussed.

• 한국수학교육학회지시리즈B:순수및응용수학
• /
• 제14권1호
• /
• pp.37-48
• /
• 2007

In this paper, we introduce and study a class of generalized multivalued quasivariational inclusions for fuzzy mappings, and establish its equivalence with a class of fuzzy fixed-point problems by using the resolvent operator technique. We suggest a new iterative algorithm for the generalized multivalued quasivariational inclusions. Further, we establish a few existence results of solutions for the generalized multivalued quasivariational inclusions involving $F_r$-relaxed Lipschitz and $F_r$-strongly monotone mappings, and discuss the convergence criteria for the algorithm.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2004년도 ICEIC The International Conference on Electronics Informations and Communications
• /
• pp.447-450
• /
• 2004

This paper proposes remotely sensed image classification method by fuzzy c-means clustering algorithm using average intra-cluster distance. The average intra-cluster distance acquires an average of the vector set belong to each cluster and proportionates to its size and density. We perform classification according to pixel's membership grade by cluster center of fuzzy c-means clustering using the mean-values of training data about each class. Fuzzy c-means algorithm considered membership degree for inter-cluster of each class. And then, we validate degree of overlap between clusters. A pixel which has a high degree of overlap applies to the maximum likelihood classification method. Finally, we decide category by comparing with fuzzy membership degree and likelihood rate. The proposed method is applied to IKONOS remote sensing satellite image for the verifying test.