• Journal of KIISE
• /
• v.42 no.6
• /
• pp.769-780
• /
• 2015

Smoothing a digital curve by averaging its connected points is widely employed to minimize sharp changes of the curve that are generally introduced by noise. An appropriate degree of smoothing is critical since the area or features of the original shape can be distorted at a higher degree while the noise is insufficiently removed at a lower degree. In this paper, we provide a mathematical relationship between the parameters, such as the number of iterations, average distance between neighboring points, weighting factors for averaging and the moving distance of the point on the curve after smoothing. Based on these findings, we propose to control the smoothed curve such that its deviation is bounded particular error level as well as to significantly expedite smoothing for a pixel-based digital curve.

• Communications for Statistical Applications and Methods
• /
• v.2 no.1
• /
• pp.155-165
• /
• 1995

초모집단(superpopulation)으로 부터 반복적으로 유한모집단을 추출할 때, 이미 조사된 자료들을 이용하면 현재의 유한모집단 모수들을 ㄷ더 효율적으로 추정할 수 있다. 이러한 문제에 대하여 Ericson(1969)이 유한모집단 표본추출에서 베이지안 분석을 하였고, Ghosh와 Meeden(1986)은 정규 초모집단을 가정하여 유한모집단 평균의 경험적 베이즈 추정을 하였다. Nandram과 Sedransk (1993)는 Ghosh와 Meeden(1986)의 유한모집단들의 분산이 모두 같다는 가정들을 완화하여 유한집단 평균의 경험적 베이즈 추정을 하였다. 본 연구는 Nandram과 Sedransk의 결과를 층과표본추출의 경우로 일반화 하였다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.26 no.6
• /
• pp.445-451
• /
• 2016

This paper introduces a postprocessing method, an iteration method for melody, and an average neural network method for learning a large number of songs in order to improve musically insufficient parts in automatic composition using existing artificial neural network. The melody of songs composed by artificial neural networks is produced according to the melodies of trained songs, so it can not be a specific tonality and it is difficult to have a repetitive composition. In order to solve these problems, we propose a postprocessing method that converts the melody composed by artificial neural networks into a melody having a specific tonality according to music theory and an iteration method for melody by iteratively composing measure divisions of artificial neural networks. In addition, the existing training method of many songs has some disadvantages. To solve this problem, we adopt an average neural network that is made by averaging the weights of artificial neural networks trained each song. From some experiments, it was confirmed that the proposed method solves the existing problems.

• 한국기계연구소 소보
• /
• s.18
• /
• pp.169-175
• /
• 1988

The influence of mean strain on the behavior of early stage stress amplitude was investigated quantitatively in low cycle fatigue. It was obtained that the exponent of cyclic strain(n') and the coefficient of cyclic strain(C') decreased with increasing cycle numbers in compressive mean strain level. But it was the opposite in 0% mean strain and tensile mean strain level. And the cyclic yield strength ($\delta_(yc)$) was constant irrespective of mean strain or cyclic strain

• Magazine of the Korea Concrete Institute
• /
• v.10 no.3
• /
• pp.107-115
• /
• 1998

반복하중을 받는 무근 및 철근 콘크리트 부재는 하중의 반복작용에 의해 과도한 균열 및 처짐을 유발하고 결과적으로 전체 구조요소의 파괴를 일으킨다. 따라서 하중반복에 의한 누적손상의 진행과정을 정량적으로 평가하여 철근콘크리트 보의 사용성과 안전성을 평가하는 것이 중요하고 특히 취성적 거동 특성을 갖는 전단거동의 경우에 더욱 의미가 있다. 본 연구에서는 반복전단하중에 의한 누적손상의 정량적 분석을 위하여 평균변형도 및 평균응력개념에 기초한 수정압축장이론(modified compression field theory)에 의하여 누적손상 모델을 제안하였다. 본 연구에서 제안된 모델을 통하여 반복전단하중에 의한 처짐과 변형도의 변화를 평가하는데 유용하게 이용될 것으로 사료된다.

• Proceedings of the IEEK Conference
• /
• 2001.09a
• /
• pp.765-768
• /
• 2001

본 논문에서는 터보 TCM(Turbo Trellis Coded Modulation)에서 복호기 입력 전에 수신데이터를 이용하여 SNR을 추정하고, SNR에 따라 요구하는 성능을 만족하는 반복 복호수를 적응적으로 복호기 앞단에서 미리 설정하는 반복복호 제어기법을 제안한다. 성능분석결과 최대 반복 복호를 수행했을 때와 비교하여 성능 감소 없이 평균 반복 복호수를 줄일 수 있었다. 따라서 제안구조는 터보 TCM의 복호화 과정에서 문제점 중 하나인 복호 계산량과 지연을 성능 저하 없이 효율적으로 감소시킬 수 있다.

• The Journal of the Acoustical Society of Korea
• /
• v.17 no.6
• /
• pp.37-42
• /
• 1998

본 논문에서는 SCHMM에 적용된 관찰 확률 최대화에 의한 화자 적응 알고리즘을 제안한다. 제안된 알고리즘은 SCHMM의 관찰 확률 밀도들이 새로운 화자의 음성 특징을 잘 표현하지 못하는 경우 인식 성능이 저하되는 것을 막기 위하여, 적응 데이터의 각 특징 벡터들이 최대의 관찰 확률을 가질 수 있도록 관찰 확률 밀도를 결정하는 평균 벡터 μ와 분산 행렬 Σ를 기울기 탐색(gradient search) 알고리즘에 의해 반복적으로 적응시켜 주는 방법이다. SCHMM의 상태 천이 확률 A와 혼합 밀도 계수 C는 관찰 확률 밀도 적응 과정 을 거친 후, 적응 데이터로부터 구한 확률과 기존 확률의 가중 평균을 취하는 과정을 반복 하여 적응시켜 주었다. 제안된 화자 적응 알고리즘을 사용하여 단독음 인식 실험을 수행한 결과, 화자 적응을 수행하지 않았을 때와 비교하여 화자 독립 시스템에서는 평균 9.8%, 남 성 화자 종속 시스템에서는 평균 46.0%, 여성 화자 종속 시스템에서는 평균 52.7%의 인식 률 향상을 보였다.

• Journal of Korea Multimedia Society
• /
• v.10 no.11
• /
• pp.1516-1522
• /
• 2007

We had proposed the FPA(Fixed Partition Averaging) method in order to improve the storage requirement and classification rate of the Memory Based Reasoning. The algorithm worked not bad in many area, but it lead to some overhead for memory usage and lengthy computation in the multi classes area. We propose an Repetitive FPA algorithm which repetitively partitioning pattern space in the multi classes area. Our proposed methods have been successfully shown to exhibit comparable performance to k-NN with a lot less number of patterns and better result than EACH system which implements the NGE theory.

• Journal of the Korean Geotechnical Society
• /
• v.31 no.11
• /
• pp.15-23
• /
• 2015

An experimental study has been conducted to evaluate the effects of average and cyclic shear stresses on the undrained failure behaviors of dense marine silty sand by using the Cyclic Direct Simple Shear apparatus. The results show that when the average shear stress ratio is zero, symmetric cyclic shear deformation is the major component of deformation, and permanent shear deformation is relatively small. On the other hand, when the average shear stress ratio is larger than zero, asymmetric permanent shear deformation is the major component, and cyclic shear deformation does not change much as the number of cyclic loads increases. The average shear stress ratio has less effects on the number of cyclic loads needed to fail, as compared with the cyclic shear stress ratio. The proposed stress-dependent failure contour can effectively be used to assess the cyclic shear strength of soil beneath the foundation for the design of offshore structures.

• Journal of Korean Society of Transportation
• /
• v.28 no.2
• /
• pp.45-52
• /
• 2010

The objective of this paper is to develop and apply a method that estimates the amount of traffic congestion (vehicle hours of delay) caused by traffic accidents that occur on freeways in Korea. A key feature of this research is the development of a method to separate the non- recurrent delay from any recurrent delay that is present on the road at the time and place of a reported accident. The main idea to separate these two delays is to use the speed difference between speed under accident condition and speed under normal flow condition. For the case study application, two datasets were combined to accomplish the objective of the study: (1) accident data and (2) traffic flow data. Eventually, the results can be useful for the performance evaluation of accident reduction program, for strategic plans to cope with congestion caused by traffic accidents, and for rectification of the estimation method for traffic congestion costs.