• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2014.06a
• /
• pp.38-41
• /
• 2014

본 논문에서는 필기 문서 영상을 분석하여 단어 단위로 요소들을 분할하는 방법을 제안한다. 일반적으로 인쇄 문서에 비하여 필기 문서에서는 글자 간 간격이 일정하지 않을 뿐만 아니라 필기자 또는 작성된 언어에 따라 특성이 매우 다르게 나타나기 때문에 단어를 분리하는 것은 어려운 문제로 간주되었고 많은 연구가 진행되었다. 제안하는 방법은 이 문제를 해결하기 위하여 글자 획의 두께를 고려하여 정규화시킨 각 연결 요소간 간격과 간격 안에 존재하는 글자 픽셀의 수로 구성된 2 차원의 특징값을 추출하였다. 이 특징값을 바탕으로, 제안하는 방법은 k-평균 클러스터링을 이용하여 각 텍스트라인을 구성하는 연결 요소간 간격을 단어 사이의 간격과 단어 내부 글자간의 간격으로 분류하였다. ICDAR 2013 Handwriting Segmentation Contest 데이터베이스에 대한 실험 결과 제안하는 방법은 가장 우수한 성능을 나타내었다.

• Computational Structural Engineering
• /
• v.2 no.4
• /
• pp.15-20
• /
• 1989

이글에서는 2가지 예제를 통해 h-version과 p-version의 비교를 살펴보면서 p-version 해석이 h-version에 비해 상대적으로 많은 장점들을 가지고 있으며, 신뢰도, 정확도, 효율성, 경제성, 용장성 등 측면에서 우월함을 증명해 보였다. 특히 응력집중(stress concentration)이 일어나는 crack-tips, cut-outs, reentrant corners, presence of stiffners, mixed boundary conditions 등 많은 특이성(singularity) 문제에 더욱 적합함을 본 예제 외의 발표된 많은 논문들을 통해 알 수 있으며, 모델링의 단순성에 기인하여 사용이 매우 쉽다는 것도 무엇보다 큰 이점이라 하겠다. p-version은 h-version의 비효율성을 차수 p를 1, 2 또는 3으로 줄인 후 이 값을 고정시키고 다시 요소분할을 통해 진해(true solution)에 접근시키는 방식을 위하면 다시 종래의 h-version으로 환원되는 호환성을 갖고 있다는 것이다. 고로 구조해석에서 h-p version이 가장 이상적인 유한요소해석 방법이라 할 수 있겠는데, 다시 말하면 균열문제의 경우 균열선단(crack-tip)에서는 p-level을 높이고 (p=8, 9 or 10) 비교적 응력집중이 낮은 영역에서는 p-level을 낮춤으로써 (p=3, 4 or 5) 그 효율성을 극대화할 수 있겠다.

• Journal of the Korean Society for Precision Engineering
• /
• v.15 no.11
• /
• pp.166-171
• /
• 1998

본 논문에서는 고체구조물의 비선형변형해석에 대하여 일반이론으로 개발된 수치해법을 단순 고체 구조물인 일차원 봉 문제에 적용하여 그 변형해석을 수행 하였다. 정확한 해를 구하기 위하여 증분 뉴톤-랩슨방법이 수정 보완 사용되었다. 또한 개발된 비선형유한요소법의 검증을 위하여 수학적인 정해가 존재하는 균일한 체력이 작용하는 단순봉의 변형을 해석하여 그 결과를 수학적인 정해와 비교하였다. 비교 결과 본 논문을 통하여 개발된 비선형 유한요소법의 정확성이 입증되었다.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.17 no.1
• /
• pp.39-47
• /
• 2004

In this paper, the defects of two-dimensional anisotropic elastic bodies are identified by using the boundary element method. The use of numerical models that contain only boundary integral terns reduces the dimensionality of the problem by one. This advantage is particularly important in problems such as crack mechanics. Avoiding domain meshing is also particularly advantageous in the solution of inverse problems since it overcomes mesh perturbations and simplifies the procedure. In this paper, nondestructive approaches for the existing isotropic materials are extended to analyze the elastic bodies made of anisotropic materials such as composites. After verifying that the proposing boundary element model is in good agreement with numerical results reported by other investigators, the effect of noise in the measurements on the identifiability is studied with respect to different design parameters of layered composites. Sample studies are carried out for various layup configurations and loading conditions. The effects of the layup sequences in detecting flaw of composites is explored in this paper.

• Computational Structural Engineering
• /
• v.10 no.1
• /
• pp.193-200
• /
• 1997

Even though there are many analysis methods of circular tanks on elastic foundation, the finite element method is widely used for that purpose. But the finite element method requires a number of memory spaces, computation time to solve large stiffness equations. In this study many the simplified methods(Analogy of Beam on Elastic Foundation, Foundation Stiffness Matrix, Finite Element Method and Transfer Matrix Method) are applied to analyze a circular tank on elastic foundation. By the given analysis methods, BEF analogy and foundation matrix method, the circular tank was transformed into the skeletonized frame structure. The frame structure was divided into several finite elements. The stiffness matrix of a finite element is related with the transfer matrix of the element. Thus, the transfer matrix of each finite element utilized the transfer matrix method to simplify the analysis of the tank. There were no significant difference in the results of two methods, the finite element method and the transfer matrix method. The transfer method applied to a circular tank on elastic foundation resulted in four simultaneous equations to solve completely.

• Journal of the Korean Geophysical Society
• /
• v.5 no.1
• /
• pp.63-71
• /
• 2002

The integral equation method is a powerful tool for electromagnetic numerical modeling. But the difficulty of this technique is the size of their linear equations, which demands excessive memory and calculation time to invert. This limitation of the integral equation method becomes critical in inverse problem. To overcome this limitation, a lot of approximation and series methods, such as conventional Born, modifed Born and extended Born, were developed. But all the methods need volume integration of Green tensor, which is very time consuming. In electromagnetic theory, Green tensor rapidly decreases as the distance between source and field cell increases. Therefore, the source cell which are far away from the field cell does not make an effect on the electric field of the field cell. Consequently, by ignoring the effect of Green tensor due to far away source cells, computing time for electromagnetic numerical modeling can be reduced dramatically. Comparisons of this new method against a full integral equation, extended Born approximation and series code show that the method is accurate enough much less time consuming.

• Journal of the Korean Magnetics Society
• /
• v.24 no.2
• /
• pp.66-69
• /
• 2014

The spreading tube of X-ray cathode tube displayed with an electromagnetic finite element method was designed. To analyze a feature design and the concrete coordinate performance of soft X-ray tube modeling, the orbit of electron beam was simulated by OPERA-3D SW program. The fixed conditions were the applied voltage, the temperature, the work function of thermal electron between cathode and anode of tungsten. Through the analysis of distribution of electron beam and the variation of dividing region, the design of soft X-ray spreading tube equipped with two cross filaments was optimized.

• Geotechnical Engineering
• /
• v.14 no.4
• /
• pp.129-140
• /
• 1998

Elasto-plastic finite element analysis of geotechnical boundary value problems necessitate the stress integration for the known strain increments. For the elasto-plastic constitutive model, the stress integration is generally achieved by numerical schemes, because analytical integration is impossible for general strain path. In this case, the accuracy of numerical stress integration has an important role on the overall accuracy of nonlinear finite element solution. In this study, the Sloan's substepping method which is one of explicit integration methods has been adopted and iris applicability has been checked. The unstability and inaccuracy of ifs results initiated from initial stress level were revealed. So. a new modified numerical integration method which employs the basic concept of modified Euler scheme for error control is proposed and accuracy and stability of the solutions are confirmed by triaxial test simulation.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.5 no.6
• /
• pp.11-19
• /
• 2001

A series of test on concrete-filled composite columns was preformed to evaluate structural performance under axial compression and cyclic lateral loading. It was presented that concrete-filled composite columns had high strength, high stiffness and large energy-absorption capacity on account of mutual confinement between the steel plate and filled-in concrete. A cross section analysis procedure developed to predict the moment-curvature relation of composite columns was proven to be on accurate and effective method. The ductility factor and the response modification factor were evaluated for the seismic design of concrete-filled composite columns. It was shown that concrete-filled composite columns could be used as a very efficient earthquake-resistant structural member.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.29 no.1 s.232
• /
• pp.145-155
• /
• 2005

In this study, a finite element method program code which can be accomodate boundary conditions on the complex surfaces has been developed to simulate polymer extruder processes. The analysis method includes the fractional 4-step method for efficient computation time and compact usage of memory storage to solve the velocities and the pressure values from the Navier-Stokes equation. By using the developed program which was verified with simple Poiseuille flow mixture phenomena in single-and twin-screw extruder are analyzed. It is concluded that the proposed method resulte Poiseuille Poiseuille d in fair agreement with the exact solution of simple flow and the back flow near the entrance happens in single-screw model. It is identified that the location and values of maximum pressure in the twin screw extruder model. It is expected that the Velocity field found can be used to predict the degree of mixture in the extruder barrel.