• 한국정밀공학회지
• /
• 제16권4호통권97호
• /
• pp.129-137
• /
• 1999

A parametric curve interpolator has been proposed for machining curves instead of a linear interpolator in which curves are approximated by a set of line segment. The parametric curve interpolator is superior to linear interpolator in machining time and contour error and generate exact position commands directly from curve equations. In this paper, a new toolpath generation method is proposed based on the parametric curve interpolator. This method retains all the benefits of parametric curve interpolator and can bound the scallop height within a specified value. By interpolating curves and surfaces directly from the mathematical equations, the amount of data from CAD/CAM system to CNC controller can be significantly reduced. The proposed method was implemented on a CNC controller and was confirmed to give a better result than the other existing method.

• 대한음성학회:학술대회논문집
• /
• 대한음성학회 1996년도 2월 학술대회지
• /
• pp.137-155
• /
• 1996

The purpose of this paper is to generate a pitch contour which is affected by tile phonetic environment and the number of syllables in each Korean isolated word using a neural network. To do this, we analyzed a set of 513 Korean isolated words, consisting of 1-4 syllables and extracted the pitch contour and the duration of each phoneme in all the words. The total number of phonemes we analyzed is about 3800. After that we approximated the pitch contour with a 1st order polynominal by a regression analysis. We could get the slope, the initial pitch and the duration of each phoneme. We used these 3 parameters as the target pattern of the neural network and let the neural network learn the rule of the variation of the pitch and duration, which was affected by the phonetic environment of each phoneme. We used 7 consecutive phoneme strings as an input pattern for a neural network to make the network learn the effect of phonetic environment around the center phoneme. In the learning phase, we used 3545 items(463 words) as target patterns which contained the phonetic environment of front and rear 3 phonemes and the neural network showed the correctness rate of 98.43%, 98.59%, 97.7% in the estimation of the duration, the slope, the initial pitch. In the recall phase, we tested the performance of tile neural network with 251 items(50 words) which weren't need as learning data and we could get the good correctness rate of 97.34%, 95.45%, 96.3% in the generation of the duration, the slope, and the initial pitch of each phoneme.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제32권5호
• /
• pp.357-365
• /
• 2005

본 연구에서는 유사한 여러 물체들이 인접하여 나타나는 영상열로부터 물체들을 개별적으로 분리할 수 있는 B-spline 적합(fitting) 알고리즘을 제안한다. 기존의 스네이크(snake) 알고리즘들은 초기화의 어려움과 다수의 극점 존재로 인해서 이러한 영상자료에서 물체의 영역을 개별적으로 분리하는 데는 어려움이 있다. 본 연구에서는 이 문제를 극복하고 다양한 형태의 물체가 인접해 있는 유사한 물체들로부터 효과적으로 분할 할 수 있는 유전자(genetic) 알고리즘 기반 B-spline 적합방안을 제안한다. 실제 상황을 고려하여 생성된 영상자료와 실제 치아 CT 영상을 이용한 평가에서 제안된 방법은 서로 인접해 있는 유사한 형태와 자기의 물체들을 개별적으로 정확하게 분할할 수 있음을 보였다. 제안된 알고리즘의 결과는 이상적으로 추출된 영역과의 일치성과 false positive 오류 그리고 false negative오류가 계산되어 검증되었다.

• Archives of Plastic Surgery
• /
• 제37권4호
• /
• pp.351-360
• /
• 2010

Purpose: As the influence of mass media increases, the general standard of attractiveness or beauty of a face also changes. The primary purpose of the study is to find out the factors of the attractive and beautiful face recognized by public. Methods: We picked out standard model photography and operated with Adobe$^{(R)}$ Photoshop$^{(R)}$ and Monariza$^{(R)}$ virtual plastic surgery program. The contour of face, eye, nose, forehead, zygoma, chin and proportion of upper, middle, lower face were changed. The interview survey was conducted through structured standard photo for 310 respondents. That was utilized in the final analysis. Multiple regression analysis was executed by SPSS 12.0. It was used to deal with statistical data and all the other necessary analysis. Results: According to general characteristics of the respondents, many differences were found in preferred face and facial aesthetic subunits. The younger generation preferred the lozenge and inverted triangle shape contour. The respondents over 40 of age preferred the egg shape contour. In chin and zygoma contour, the respondents at the age of 20 preferred distinctly small chin and relatively small lower face. On the other hand, the respondents over 40 of age preferred the wide zygoma relatively. In the proportion of upper, middle, lower face, 51.0% of respondents answered 1 : 1 : 1. If they want to have an aesthetic operation, they preferred protruding forehead. Also they preferred the small chin and V-shaped chin in frontal view. Conclusion: Many respondents preferred to have a plastic surgery for the better facial subunit. The statistical evidence from this study suggests that the harmony and balance of facial aesthetic subunits make attractive and beautiful face.

• 대한음성학회지:말소리
• /
• 제53호
• /
• pp.75-92
• /
• 2005

Prosody Generation Based on C-ToBI Representation for Text-to-SpeechSeungwon Kim, Yu Zheng, Gary Geunbae Lee, Byeongchang KimProsody modeling is critical in developing text-to-speech (TTS) systems where speech synthesis is used to automatically generate natural speech. In this paper, we present a prosody generation architecture based on Chinese Tone and Break Index (C-ToBI) representation. ToBI is a multi-tier representation system based on linguistic knowledge to transcribe events in an utterance. The TTS system which adopts ToBI as an intermediate representation is known to exhibit higher flexibility, modularity and domain/task portability compared with the direct prosody generation TTS systems. However, the cost of corpus preparation is very expensive for practical-level performance because the ToBI labeled corpus has been manually constructed by many prosody experts and normally requires a large amount of data for accurate statistical prosody modeling. This paper proposes a new method which transcribes the C-ToBI labels automatically in Chinese speech. We model Chinese prosody generation as a classification problem and apply conditional Maximum Entropy (ME) classification to this problem. We empirically verify the usefulness of various natural language and phonology features to make well-integrated features for ME framework.

• Elastomers and Composites
• /
• 제36권3호
• /
• pp.153-161
• /
• 2001

궤도차량의 고무패드는 상당한 크기의 진폭과 주파수를 가지고 반복되는 동적 변형을 받고 있으므로, 이로 인하여 열발생과 더불어 상당한 내부온도의 상승을 초래한다. 이러한 열발생은 고무 재료의 점탄성 특성에 기인한 것으로, 점성효과에 의하여 변형시의 기계적 에너지의 일부가 히스테리시스 루프의 면적에 해당되는 열에너지로 변환되기 때문이다. 발생한 열은 발산조건이 충분하지 못할 경우 내부온도의 상승을 초래하며, 온도가 과다할 경우에는 고무제품의 성능이나 수명에 중대한 장애요인으로 작용하게 된다. 본 연구에서는 전차 궤도고무에 작용하는 동적 하중에 근거하여 각 부품별 열발생률을 실험을 통하여 측정하고, 이를 입력자료로 하여 궤도패드에 발생하는 온도분포를 유한요소법으로 해석하였다.

• 산업융합연구
• /
• 제16권1호
• /
• pp.35-40
• /
• 2018

절차적 생성(Procedural Generation) 알고리즘은 일련의 규칙을 반복적으로 수행하여 게임에 활용되는 콘텐츠를 자동으로 생성하는 알고리즘이다. 게임의 규모와 함께 게임 내에 활용되는 콘텐츠의 양이 늘어나기 때문에 콘텐츠 제작에 많은 시간이 소요된다. 이에 따라 절차적 생성과 같은 게임 인공지능을 이용해 게임의 콘텐츠를 자동으로 생성하는 인공지능 연구가 활발하게 진행되고 있다. 본 논문에서는 절차적 생성을 이용해 3차원 게임 맵을 제작하는 알고리즘을 제안한다. 제안하는 알고리즘은 노이즈가 점진적으로 변화되는 펄린 노이즈(Perlin-Noise)를 활용하여 경로가 자연스럽게 연결되는 2차원 등고선을 생성한다. 생성된 2차원 등고선을 바탕으로 3차원의 높이 맵(Height-Map)을 제작한다. 생성된 높이 맵은 맵의 형태가 이질적이지 않고, 수작업으로 제작한 게임월드와 마찬가지로 플레이어가 모든 구간을 돌아다닐 수 있게 제작된 것을 확인하였다. 향후에는 알고리즘의 성능을 향상시켜 실제 게임에 적용할 것이다.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 1991년도 추계학술발표회 논문집 지반공학에서의 컴퓨터 활용 COMPUTER UTILIZATION IN GEOTECHNICAL ENGINEERING
• /
• pp.179-194
• /
• 1991

This paper explains outline of a behavior analysis program for underground structures, and its application to a tunnel problem. The program can deal with elasto-plastic behavior of medium and supporting structures, discontinuous behavior due to existing joint, creation and propagation of cracks. in-situ loading condition, and incremental behavior due to stepwise excavation, etc. The program also has additional capabilities such as graphic output of mesh, displacement pattern, stress condition, and safety factor contour, and automatic mesh generation during the excavation steps.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1995년도 춘계학술대회 논문집
• /
• pp.157-163
• /
• 1995

This paper presents an algorithm to deternine the tool path height for rough machining of sculptured surface. To minimize rough machining of sculptured surface, it is necessary to determine the tool path heights of contour planes. the proposed algorithm searches for the height at which maximum metal removal rate is obtained. This bisection method is accomplished until all shoulder heights are within roughing tolerance. The machining experiment demonstrates the superiority of the algorithm presented in this thesis.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 하계학술대회 논문집 E
• /
• pp.2022-2024
• /
• 1999

PDP(plasma display panel)is expected to be the standard of the next generation display devices. However, there is a serious problem. That is disturbances are occasionally observed in gray scales and colors when moving images are displayed on PDP. Various methods have been presented to reduce these disturbances. But the picture quality of PDP is not yet sufficient. We developed optimized-modified binary coded light emission scheme. Simulated result show the improving image quality. Also we have simulated and estimated various proposed methods.