• 조명전기설비학회논문지
• /
• 제26권9호
• /
• pp.73-79
• /
• 2012

In this paper, we study on how to determine the number of hidden layer neurons in neural network for predicting defect size of steam generator tube. It was reported in the literature that the number of hidden layer neurons can be efficiently determined with the help of cross-validation. Although the cross-validation provides decent estimation performance in most cases, the performance depends on the selection of validation set and rather poor performance may be led to in some cases. In order to avoid such a problem, we propose to use multifold cross-validation. Through the simulation study, it is shown that the estimation performance of defect width (defect depth, respectively) attains 94% (99.4%, respectively) of the best performance achievable among the considered neuron numbers.

• 제어로봇시스템학회논문지
• /
• 제5권5호
• /
• pp.558-567
• /
• 1999

X-ray laminography and DT(digital tomosynthesis) that can form a cross-sectional image of 3-D objects promis to be good solutions for inspecting interior defects of industrial products. DT is a kind of laminography technique and the difference is in the fact that it synthesizes the several projected images by use of the digitized memory and computation. The quality of images acquired from the DT system varies according to image synthesizing methods, the number of images used in image synthesizing, and X-ray projection angles. In this paper, a new image synthesizing method named 'log-root method' is proposed to get clear and accurate cross-sectional images, which can reduce both artifact and blurring generated by materials out of focal plane. To evaluate the quality of cross-sectional images, two evaluating criteria : (1) shape accuracy and (2) clearness of the cross-sectional images are defined. Based on these criteria, a series of simulations are performed, and the results show the superiority of the new synthesizing method over the existing ones such as averaging and minimum methods.

• 대한전기학회논문지:전기물성ㆍ응용부문C
• /
• 제48권12호
• /
• pp.785-790
• /
• 1999

This research has been carried out to develop a low-energy large-aperture pulsed electron beam generator (LELA), 200keV 1A, for industrial applications. One of the most important feature of this electron beam generator is large electron beam cross section of $190cm^2$. Low energy electron beam generators have been used for water cleaning, flue gas cleaning, and pasteurization, etc. In these applications the cross sectionof the e-beam is related to reaction efficiency. Another important feature of this LELA EB generator is easy maintenance because of its simple structure and relatively low vacuum operation compared to the conventional EB generators. The conventional EB generators need to be scanned because the small cross section thermal electron emitters are used in the conventional EB generators which have small EB cross section. In this research, we use the secondary electrons generated by ion bombardment on the HV cathode surface as a electron source. Therefore we can make any shape of EB cross section without scanning.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2013년도 춘계학술대회
• /
• pp.807-810
• /
• 2013

본 논문에서는 영상 복원을 위해서 경계검출 알고리즘을 갖는 하중 격자형 메디안필터를 사용한 새로운 방법을 고려해 보았다. 이 방법은 경계를 검출하기 위한 간단한 가정의 시험들로 구성되어 있으며 움직이는 격자형 창을 사용하였다. 이 방법을 잡음에 의해서 손상된 영상에 적용하였으며 그 결과를 일반 메디안 필터와 격자형 메디안 필터의 결과와 비교 하였다. 실험 결과, 경계검출 알고리즘을 갖는 하중 격자형 메디안 필터를 사용하는 것이 다른 메디안 필터를 사용한 결과보다 매우 우수한 성능을 가짐을 확인하였다.

• 소성∙가공
• /
• 제22권7호
• /
• pp.401-406
• /
• 2013

The use of roll-formed products increases every year due to its advantages, such as high production rates, reduced tooling cost and improved quality. However, till now, it is limited to part profiles with constant cross section. In recent years, the flexible roll forming process, which allows variable cross sections of profiles by adaptive roll stands, was developed. In this study, an attempt to optimize profile design for the flexible roll forming process was performed. An equation that predicts the longitudinal strain for part geometries with variable cross-sections was proposed. The relationship between geometrical parameters and the longitudinal strain was analyzed and investigations on the optimal profile design were performed. Experiments were conducted with a lab-scale roll forming machine to validate the proposed equation. The results show that the profile design method proposed in this study is feasible and parts with variable cross sections can be successfully fabricated with the flexible roll forming process.

• International Journal of Contents
• /
• 제11권2호
• /
• pp.57-62
• /
• 2015

Relative entropy is a divergence measure between two probability density functions of a random variable. Assuming that the random variable has only two alphabets, the relative entropy becomes a cross-entropy error function that can accelerate training convergence of multi-layer perceptron neural networks. Also, the n-th order extension of cross-entropy (nCE) error function exhibits an improved performance in viewpoints of learning convergence and generalization capability. In this paper, we derive a new divergence measure between two probability density functions from the nCE error function. And the new divergence measure is compared with the relative entropy through the use of three-dimensional plots.

• 식물조직배양학회지
• /
• 제24권5호
• /
• pp.257-262
• /
• 1997

Brassica와 Raphanus 간에 속간 교잡종을 육성하기 위해서 34개의 교배 조합을 설정한 다음 수분 12일이 경과된 배주를 BA와 GA 1.0 ㎎/L가 첨가된 MS배지에 접종하고 배의 발아율을 조사하였다. BA나 GA 첨가는 배의 발아에 큰 영향을 미치지 않았다. 무 속의 '내병총태'와 '중국청피'를 부계로 교잡한 조합에서 배의 발아율이 낮았으나 갓(Brassica juncea)을 부계로 교잡한 조합에서는 비교적 배의 발아율이 높았다. 이러한 결과는 배주배양시 배지에 첨가한 casein hydrolysate, malt extract, BA, kinetin 및 glutamine이 배의 발아에 미치는 영향보다는 부계나 모계의 교잡조합에 따라 배의 발아율에 상당한 차이가 있다는 것을 의미한다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제8권4호
• /
• pp.239-247
• /
• 2008

In a resource-limited wireless communication environment, various approaches to meet the ever growing application requirements in an efficient and transparent manner, are being researched and developed. Amongst many approaches, cross layer technique is by far one of the significant contributions that has undoubtedly revolutionized the way conventional layered architecture is perceived. In this paper, we propose a Universal Cross Layer Framework based on vertical layer architecture. The primary contribution of this paper is the functional architecture of the vertical layer which is primarily responsible for cross layer interaction management and optimization. The second contribution is the use of optimization cycle that comprises awareness parameters collection, mapping, classification and the analysis phases. The third contribution of the paper is the decomposition of the parameters into local and global network perspective for opportunistic optimization. Finally, we have shown through simulations how parameters' variations can represent local and global views of the network and how we can set local and global thresholds to perform opportunistic optimization.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2015년도 추계학술대회
• /
• pp.711-714
• /
• 2015

영상 복원을 위해서 에지(edge) 검출 알고리즘을 갖는 가중 격자형 메디안필터(median filter)를 사용한 새로운 방법을 제안하였다. 에지를 검출하기 위한 시험들로 구성되어 있으며 격자형 창을 사용하였다. 잡음에 의해서 손상된 영상에 제안한 방법을 적용 하였으며 그 결과를 일반 메디안 필터와 격자형 메디안 필터의 결과들과 비교 하였다. 에지 검출 알고리즘을 갖는 가중격자형 메디안 필터를 사용하는 것이 다른 메디안 필터를 사용한 결과보다 매우 우수한 성능을 가짐을 확인하였다.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2013년도 추계학술대회
• /
• pp.1006-1009
• /
• 2013

본 논문에서는 영상 향상을 위해서 에지 검출 알고리즘을 갖는 격자형 메디안필터를 사용한 새로운 방법을 제안하였다. 이 방법은 에지를 검출하기 위한 간단한 시험들로 구성되어 있으며 격자형 창을 사용하였다. 잡음에 의해서 손상된 영상에 제안한 방법을 적용 하였으며 그 결과를 일반 메디안 필터와 격자형 메디안 필터의 결과와 비교 하였다. 실험 결과, 에지 검출 알고리즘을 갖는 격자형 메디안 필터를 사용하는 것이 다른 메디안 필터를 사용한 결과보다 매우 우수한 성능을 가짐을 확인하였다.