• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.10a
• /
• pp.317-320
• /
• 2004

디스플레이 장치간의 색 재현 차이를 극복하기 위하여 다양한 색역폭 사상 기법이 사용되고 있다. 기존 색역폭 사상 방법은 일반적으로 색 공간 변환과 같은 복잡한 비선형 변환을 여러 단계 거치므로 실시간 처리 구현이 어렵다. 본 논문에서는 신경 회로망을 이용하여 기존의 색역폭 사상 방법을 학습하고 근사화한 방법을 이용한다. 이를 위해 주어진 디스플레이 장치의 표현 가능한 모든 색상에 대해 미리 색역폭 사상을 수행하고 그 결과를 학습 데이터로 이용하게 되며, 학습된 신경망은 이종 디스플레이 장치간의 색역폭 사상에 사용된다. 제안된 색역사상을 실시간 처리하기 위해서 학습 과정은 오프라인을 통해서 이루어지게 되고, 구해진 신경망은 프로세서의 메모리를 이용, 1차원의 Look-Up Table로 구성한다. 제안한 방법을 색역폭 사상에 적절하도록 최적화시키면 고속의 색역폭 사상이 가능하다.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.20 no.2
• /
• pp.264-271
• /
• 2017

Link-K message is the Tactical Data Link message standard developed by the Agency for Defense Development which is optimized for the Joint Operation of ROK Armed Force. KMTF message is the standard message format of Battle Management Information System. Interconversion and propagation between these messages are definitely needed to have efficient warfare such as situation data propagation using network, the convergence of collected situation data, common situational awareness, cooperative engagement. Therefore, this study suggests a way of rule and process for format conversion between Link-K and KMTF messages.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.19 no.3
• /
• pp.713-723
• /
• 1995

In intelligent machine tools, a computer based control system, which can adapt the machining parameters in an optimal fashion based on sensor measurements of the machining process, should be incorporated. In this paper, the technology for adaptively optimizing the cutting conditions to maximize the material removal rate in face milling operations is proposed using the exterior penalty function method combined with multilayered neural networks. Two neural networks are introduced ; one for estimating tool were length, the other for mapping input and output relations from experimental data. Then, the optimization of cutting conditions is adaptively implemented using tool were information and predicted process output. The results are demonstrated with respect to each level of machining such as rough, fine and finish cutting.

• Journal of Ocean Engineering and Technology
• /
• v.10 no.1
• /
• pp.65-74
• /
• 1996

Up to now, the existing crack growth modelling has used a mathematical approximation but an assumed function have a great influence on this method. Especially, crack growth behavior that shows very strong nonlinearity needed complicated function which has difficulty in setting parameter of it. The main characteristics of neural network modelling to engineering field are simple calculations and absence of assumed function. In this paper, after discussing learning and generalization of neural networks, we performed crack growth modelling on the basis of above learning algorithms. J'-da/dt relation predicted by neural networks shows that test condition with unlearned data is simulated well within estimated mean error(5%).

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10b
• /
• pp.233-237
• /
• 2006

This study proposes an innovative measure for evaluating the performance of text clustering. In using K-means algorithm and Kohonen Networks for text clustering, the number clusters is fixed initially by configuring it as their parameter, while in using single pass algorithm for text clustering, the number of clusters is not predictable. Using labeled documents, the result of text clustering using K-means algorithm or Kohonen Network is able to be evaluated by setting the number of clusters as the number of the given target categories, mapping each cluster to a target category, and using the evaluation measures of text. But in using single pass algorithm, if the number of clusters is different from the number of target categories, such measures are useless for evaluating the result of text clustering. This study proposes an evaluation measure of text clustering based on intra-cluster similarity and inter-cluster similarity, what is called CI (Clustering Index) in this article.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.48 no.2
• /
• pp.82-88
• /
• 1999

In this paper, we propose a new structure optimization method of multilayer neural networks which begin and carry out learning from a bigger network. This method redundant links and neurons according to the rough set theory. In order to find redundant links, we analyze the variations of all weights and output errors in every step of the learning process, and then make the decision table from their variation of weights and output errors. We can find the redundant links from the initial structure by analyzing the decision table using the rough set theory. This enables us to build a structure as compact as possible, and also enables mapping between input and output. We show the validity and effectiveness of the proposed algorithm by applying it to the XOR problem.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.49 no.6
• /
• pp.253-258
• /
• 2000

This paper presents an application of Kohonen neural networks to assess the dynamic security of power systems. The dynamic security assessment(DSA) is an important factor in power system operation, but conventional techniques have not achieved the desired speed and accuracy. The critical clearing time(CCT) is an attribute which provides significant information about the quality of the post-fault system behaviour. The function of Kohonen networks is a mapping of the pre-fault system conditions into the neurons based on the CCTs. The power flow on each line is used as the input data, and an activated output neuron has information of the CCT of each contingency. The trajectory of the activated neurons during load changes can be used in on-line DSA efficiently. The applicability of the proposed method is demonstrated using a 9-bus example.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.4
• /
• pp.1988-1992
• /
• 2013

In this paper, we propose a method of object recognition to real image object which enter into an area. We needs this method for an application module to detect and trace the moving pattern of some objects entered into an specific area. A scheme to the object recognition is adopted to some applied modules that it is moved from only real image information recognition to real coordination recognition, the mapping between the GPS coordination and real image information provides object coordination.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2002.10a
• /
• pp.383-386
• /
• 2002

It is very important to inspect color of printed texture in the textile process. To distinguish the color of the printed texture, RGB color values obtained from a scanner must be transformed to the standard colorimetric system used in the textile industry. It is XYZ color system that is defined by CIE(Commission Internationale do 1Eclairage). The mapping from RGB to XYZ color values is not simple and the scanner has even a positional deviation of RGB colors. In this paper an automatic color inspection method using a general scanning machine is presented. We used a U(neural network) model to map RGB to XYZ and compensate the positional error. In the real experiments, this inspection system shows to get very exact XYZ values from the traditional scanner regardless of the measuring position.

• Structural Engineering and Mechanics
• /
• v.44 no.3
• /
• pp.325-337
• /
• 2012

In this article, the multi-objective optimization of cylindrical aluminum tubes under axial impact load is presented. The specific absorbed energy and the maximum crushing force are considered as objective functions. The geometric dimensions of tubes including diameter, length and thickness are chosen as design variables. D/t and L/D ratios are constricted in the range of which collapsing of tubes occurs in concertina or diamond mode. The Non-dominated Sorting Genetic Algorithm-II is applied to obtain the Pareto optimal solutions. A back-propagation neural network is constructed as the surrogate model to formulate the mapping between the design variables and the objective functions. The finite element software ABAQUS/Explicit is used to generate the training and test sets for the artificial neural networks. To validate the results of finite element model, several impact tests are carried out using drop hammer testing machine.