• 대한전자공학회논문지SP
• /
• 제41권2호
• /
• pp.41-50
• /
• 2004

MPEG-4 표준에서는 객체 단위의 부호화를 수행하기 위해 자연영상으로부터 비디오 객체를 분리하는 영상분할(segmentation) 기술이 필요하다. 영상분할 방법은 크게 자동 영상분할(automatic segmentation)과 반자동 영상분할(semi-automatic segmentation)의 두 부류로 나눌 수 있다. 지금까지 개발된 대부분의 자동 영상분할 방법은 비디오 객체의 명확한 수학적인 모델을 제시하기 곤란하며 한 화면에서 개별 객체를 추출하기 어렵기 때문에 그 성능에 한계가 있다. 본 논문에서는 이러한 문제점을 극복하기 위해 active contour 알고리즘을 이용한 반자동 영상분할 알고리즘을 제안한다. 초기 곡선으로부터 변화 가능한 모든 곡선의 집합을 모양공간으로 정의하고 그 공간을 선형공간이라고 가정하면, 모양공간(shape space)은 모양 행렬에 의해 행(column) 공간과 남은 빈(left null) 공간으로 나뉘어진다. 본 논문에서 제안하는 알고리즘은 행공간의 모양공간 벡터를 이용하여 초기 곡선으로부터 영상의 특징점까지의 변화를 기술하고 동적 그래프 검색 알고리즘을 이용하여 객체의 세밀한 부분을 묘사한다. 모양 행렬과 객체의 윤곽을 추정하기 위한 SUSAN 연산자의 사용으로 제안한 알고리즘은 저수준 영상처리로부터 생성되는 불필요한 특징점을 무시할 수 있다. 또한, 모양 행렬의 사용으로 생긴 제약은 동적 그래프 검색 알고리즘으로 보상한다.

• 한국산학기술학회논문지
• /
• 제15권10호
• /
• pp.6276-6281
• /
• 2014

유전자 알고리즘은 탐색과 최적화 문제에 대한 효과적인 방법으로 이용되고 있으나 다수의 정점이 있는 다중정점 함수에 대한 응용에 있어서는 지역해에 조기 수렴하여 고착되는 등 전역 최적해를 찾는데 어려움이 있다. 이러한 문제는 탐색공간을 충분히 탐색할 수 있는 모집단의 다양성이 부족한 데 기인하는 것이며 해결방법으로 니칭 방법과 크라우딩 방법 등이 소개되고 있다. 개체군의 다양성을 증가시키는 방법으로 지역해에 고착되지 않고 전역 최적해로 수렴되도록 하는 데 기본을 두고 있다. 본 논문에서는 다중정점 함수의 전역 최적해에 수렴하고 수렴속도를 높이는 방법으로 진화과정의 매 세대마다 탐색영역에 충분히 분포되도록 임의로 생성된 보조 모집단을 공급함으로서 안정적으로 전역 최적해로 수렴하는 방법을 제안하였다. 컴퓨터 모의실험을 통하여 본 논문에서 제안한 방법을 입증하였다.

• International Journal of Control, Automation, and Systems
• /
• 제4권6호
• /
• pp.775-781
• /
• 2006

An intelligent optimization method for designing Fractional Order PID(FOPID) controllers based on Particle Swarm Optimization(PSO) is presented in this paper. Fractional calculus can provide novel and higher performance extension for FOPID controllers. However, the difficulties of designing FOPID controllers increase, because FOPID controllers append derivative order and integral order in comparison with traditional PID controllers. To design the parameters of FOPID controllers, the enhanced PSO algorithms is adopted, which guarantee the particle position inside the defined search spaces with momentum factor. The optimization performance target is the weighted combination of ITAE and control input. The numerical realization of FOPID controllers uses the methods of Tustin operator and continued fraction expansion. Experimental results show the proposed design method can design effectively the parameters of FOPID controllers.

• 한국과학교육학회지
• /
• 제15권2호
• /
• pp.185-193
• /
• 1995

The purpose of this article is to find out psychological mechanism on the core of misconceptions. Theoretical frame to search for the core of misconceptions is based on the Pascual-Leone's neo Piagetian theory. Although Pascual-Leone's theory is a cognitive developmental theory, its psychological mechanism gives us new insights on misconception. According to the comparison between Pascual-Leone's psychological mechanism and the common specifics of misconceptions and their items, conclusions could be summarized as follows: 1) Children's misconceptions and LC learning structures have the same nature. 2) Structures in items of misconceptions and misleading factor structures in cognitive tasks affect mental process with the same mechanism. 3) LC learning structures was activated preferentially in knowledge structures by F operator, with the same activation mechanism, the process children's misconceptions was activated firstly among other conceptions could be explained.

• 대한산업공학회지
• /
• 제31권1호
• /
• pp.10-19
• /
• 2005

The minimum spanning tree (MST) problem is one of the traditional optimization problems. Unlike the MST, the degree constrained minimum spanning tree (DCMST) of a graph cannot, in general, be found using a polynomial time algorithm. So, finding the DCMST of a graph is a well-known NP-hard problem of importance in communications network design, road network design and other network-related problems. So, it seems to be natural to use evolutionary algorithms for solving DCMST. Especially, when applying an evolutionary algorithm to spanning tree problems, a representation and search operators should be considered simultaneously. This paper introduces a new tree representation scheme and a genetic operator for solving combinatorial tree problem using evolutionary algorithms. We performed empirical comparisons with other tree representations on several test instances and could confirm that the proposed method is superior to other tree representations. Even it is superior to edge set representation which is known as the best algorithm.

• 한국군사과학기술학회지
• /
• 제18권2호
• /
• pp.189-200
• /
• 2015

This paper gives an overview on the some potential applications and key technologies of biomimetic underwater robot for naval operations. Unlike most manned underwater naval systems, biomimetic underwater robots can be especially useful in near-land or harbour areas due to their ability to operate in shallow water effectively. Biomimetic underwater robot provide advantages in reaching locations that would be difficult or too dangerous for a manned vehicle to reach, as well as providing a level of autonomy that can remove the requirement for dedicated human operator support. Using multiple or schools of underwater robots would provide increased flexibility for navigation, communication and surveillance ability. And it alleviate some of the restrictions associated with speed and endurance design constraints.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회 2001년도 추계학술대회 논문집
• /
• pp.358-361
• /
• 2001

Fixed-mobile convergence services can be defined as the combination of previously separate fixed and mobile services, and networks and commercial practices. Examples of fixed-mobile convergence services include single voicemail box, single number and unified messaging across fixed and mobile networks. Recently as more voice is transferred to mobile networks, convergence services between fixed and mobile become more important. In Korea convergence services are only starting to become established, and are likely to become an important part of any operator's offering. In this paper, I search the different levels of fixed-mobile convergence services and the trends and regulations for fixed-mobile convergence services in major countries. And at the same time I also try to find a new direction of future regulatory principles related to fixed-mobile convergence services.

• 동력기계공학회지
• /
• 제2권3호
• /
• pp.60-67
• /
• 1998

Edge detection is the first step and very important step in image analysis. We cast edge detection as a problem in cost minimization. This is achieved by the formulation of a cost function that evaluates the quality of edge configurations. The cost function can be used as a basis for comparing the performances of different detectors. This cost function is made of desirable characteristics of edges such as thickness, continuity, length, region dissimilarity. And we use a simulated annealing algorithm for minimum of cost function. Simulated annealing are a class of adaptive search techniques that have been intensively studied in recent years. We present five strategies for generating candidate states. Experimental results(building image and test image) which verify the usefulness of our simulated annealing approach to edge detection are better than other operator.

• Journal of Electrical Engineering and Technology
• /
• 제3권1호
• /
• pp.125-129
• /
• 2008

A new genetic algorithm for the solution of a multi-tasking problem is presented in this paper. The approach introduces innovative genetic operation that guides the genetic algorithm more directly towards better quality of the population. A wide variety of standard genetic parameters are explored, and results allow the comparison of performance for cases both with and without the new operator. The proposed algorithm improves the convergence speed by reducing the number of generations required to identify a near-optimal solution, significantly reducing the convergence time in each instance.

• 대한산업공학회지
• /
• 제30권2호
• /
• pp.130-145
• /
• 2004

This paper addresses the integrated problem of process planning and scheduling in FMS (Flexible Manufacturing System). The integration of process planning and scheduling is important for an efficient utilization of manufacturing resources. In this paper, a new method using an artificial intelligent search technique, called asymmetric multileveled symbiotic evolutionary algorithm, is presented to handle the two functions at the same time. Efficient genetic representations and operator schemes are considered. While designing the schemes, we take into account the features specific to each of process planning and scheduling problems. The performance of the proposed algorithm is compared with those of a traditional hierarchical approach and existing evolutionary algorithms. The experimental results show that the proposed algorithm outperforms the compared algorithms.