• 천문학논총
• /
• 제32권1호
• /
• pp.209-211
• /
• 2017

We performed Principle Component Analysis (PCA) over 264 galaxies in the IRAS Revised Bright Galaxy Sample (Sanders et al., 2003) using 12, 25, 60 and $100{\mu}m$ flux data observed by IRAS and 9, 18, 65, 90 and $140{\mu}m$ flux data observed by AKARI. We found that (i)the first principle component was largely contributed by infrared to visible flux ratio, (ii)the second principal component was largely contributed by the flux ratio between IRAS and AKARI, (iii)the third principle component was largely contributed by infrared colors.

• 한국정보통신학회논문지
• /
• 제12권9호
• /
• pp.1547-1553
• /
• 2008

본 논문은 간소화된 주성분 벡터를 이용한 벡터 그래픽 캐릭터의 다양한 얼굴 표정 생성 방법을 제안한다. 먼저 Russell의 내적 정서 상태에 기반하여 재정의된 벡터 그래픽 캐릭터들의 9가지 표정에 대해 주성분 분석을 수행한다. 이를 통해 캐릭터의 얼굴 특성과 표정에 주된 영향을 미치는 주성분 벡터를 찾아내고, 간소화된 주성분 벡터로부터 얼굴 표정을 생성한다. 또한 캐릭터의 특성과 표정의 가중치 값을 보간함으로써 자연스러운 중간 캐릭터 및 표정을 생성한다. 이는 얼굴 애니메이션에서 종래의 키프레임 저장 공간을 상당히 줄일 수 있으며, 적은 계산량으로 중간 표정을 생성할 수 있다. 이에 실시간 제어를 요구하는 웹/모바일 서비스, 게임 등에서 캐릭터 생성 시스템의 성능을 상당히 개선할 수 있다.

• Structural Engineering and Mechanics
• /
• 제54권2호
• /
• pp.379-392
• /
• 2015

A number of acceleration-based damage detection methods have been developed but they have not been widely applied in engineering practices because the acceleration response is insensitive to minor damage of civil structures. In this article, a damage detection approach using the long-gauge strain sensing technology and the principle component analysis technology is proposed. The Long gauge FBG sensor has its special merit for damage detection by measuring the averaged strain over a long-gauge length, and it can be connected each other to make a distributed sensor network for monitoring the large-scale civil infrastructure. A new damage index is defined by performing the principle component analyses of the long-gauge strains measured from the intact and damaged structures respectively. Advantages of the long gauge sensing and the principle component analysis technologies guarantee the effectiveness for structural damage localization. Examples of a simple supported beam and a steel stringer bridge have been investigated to illustrate the successful applications of the proposed method for structural damage detection.

• 한국콘텐츠학회:학술대회논문집
• /
• 한국콘텐츠학회 2003년도 추계종합학술대회 논문집
• /
• pp.213-217
• /
• 2003

이 논문에서는 MLP(Multilayer Perceptron)가 지닌 잡음 강건성에 대한 통계학적 분석을 하였다. 또한, MLP의 잡음 강건성을 향상시키기 위한 선형적 전처리 단계로써, ICA(independent component analysis)와 PCA(principle component analysis)를 고려하여, 이들이 지닌 잡음처리 효과를 분석한후, MLP와 접목시 나타나는 잡음 강건성의 향상 여부를 필기체 숫자 인식의 시뮬레이션으로 확인하였다.

• 한국운동역학회지
• /
• 제25권2호
• /
• pp.157-165
• /
• 2015

Object : The goals of this research were to make Performance Enhanced Model(PE) taken the largest performance index (PI) through artificial variation of principle components calculated by principle component analysis for trial data, and to verify the effect through comparing kinematic factors between trial data (Raw) and PE. Method : Ten subjects (5 men, 5 women) were recruited and 80% of their maximal record was considered. The PI is a regression equation. In order to develop PE, we extracted Principle components from trial position data (by Principle Components Analysis (PCA)). Before PCA, we made 17 position data to 3 row matrix according to components. We calculated 3 eigen value (principle components) through PCA. And except Y (medial-lateral direction) component (because motion of Y component is small), principle components of X (anterior-posterior direction) and Z (vertical direction) components were changed as following. Changed principle components = principle components + principle components ${\times}$ k. After changing the each principle component, we reconstructed position data using the changed principle components and calculated performance index (PI). A Paired t-test was used to compare Raw data and Performance Enhanced Model data. The level of statistical significance was set at $p{\leq}0.05$. Result : The PI was significantly increased about 12.9kg at PE ($101.92{\pm}6.25$) when compared to the Raw data ($91.29{\pm}7.10$). It means that performance can be increased by optimizing 3D positions. The difference of kinematic factors as follows : the movement distance of the bar from start to lock out was significantly larger (about 1cm) for PE, the width of anterior-posterior bar position in full phase was significantly wider (about 1.3cm) for PE and the horizontal displacement toward the weightlifter after beginning of descent from maximal height was significantly greater (about 0.4cm) for PE. Additionally, the minimum knee angle in the 2-pull phase was significantly smaller (approximately 2.7cm) for the PE compared to that of the Raw. PE was decided at proximal position from the Raw (origin point (0,0)) of PC variation). Conclusion : PI was decided at proximal position from the Raw (origin point (0,0)) of PC variation). This means that Performance Enhanced Model was decided by similar motion to the Raw without a great change. Therefore, weightlifters could be accept Performance Enhanced Model easily, comfortably and without large stress. The Performance Enhance Model can provide training direction for athletes to improve their weightlifting records.

• 통상정보연구
• /
• 제12권4호
• /
• pp.279-302
• /
• 2010

A better selection model on management innovative enterprises is needed since the Korean government provides multi benefits to those selected enterprises. However, the selection model's propriety is suspicious because of the shortage of consideration of assessment items. In particular, the most important two assessment items, strategy and performance are suspected of multicollinearity because of high correlation scores. No consideration on multicollinearity among those items leads to erroneous selection which doubly counts the same components with different item names. The principle component analysis is applied to factor out the uncorrelated items. Using the resulted principle components, the new estimations are carried out. The comparison between estimated results from using principle components and non principle components shows that the present selection model overly considers the performance items compared to the real effect of items, which is a result of multicollinearity between performance and strategy.

• International Journal of Advanced Culture Technology
• /
• 제6권3호
• /
• pp.142-150
• /
• 2018

This paper describes the algorithm that lowers the dimension, maintains the object recognition and significantly reduces the eigenspace configuration time by combining the edge orientation histogram and principle component analysis. By using the detected object region as a recognition input image, in this paper the object recognition method combined with principle component analysis and the multi-layer network which is one of the intelligent classification was suggested and its performance was evaluated. As a pre-processing algorithm of input object image, this method computes the eigenspace through principle component analysis and expresses the training images with it as a fundamental vector. Each image takes the set of weights for the fundamental vector as a feature vector and it reduces the dimension of image at the same time, and then the object recognition is performed by inputting the multi-layer neural network.

• 대한약침학회지
• /
• 제15권4호
• /
• pp.42-51
• /
• 2012

Objectives: This study used the basic principle of Oriental medicine, the sovereign, minister, assistant and courier principle (君臣佐使論) to investigate the effects of the component of ONGABO, which is composed of Ginseng Radix (Red Ginseng), Angelica Gigantis Radix, Schisandrae Fructus, Cuscuta Semen and Curcumae tuber on the viability of HepG2 cells. Methods: Single and mixed extracts of the component of ONGABO were prepared by lypohilizing powder of Red Ginseng (6-year root from Kanghwa), Angelica Gigantis Radix, Schisandrae Fructus, Cuscuta Semen, Curcumae Tuber (from Omniherb Co., Ltd., Korea) at the laboratory of herbal medicine in Woosuk University and were eluted after being macerated with 100% ethanol for three days. The cell viability of HepG2 was determined by using an absorptiometric analysis with PrestoBlue (Invitrogen) reagent after the plate had been incubated for 48 hours. All of the experiments were repeated three times to obtain the average value and standard deviation. The statistical analysis was done and the correlation factor was obtained by using Microsoft Office Excel 2007 and Origin 6.0 software. Results: Although Ginseng Radix (Red Ginseng) and Schisandrae Fructus did not enhance the viability of HepG2 cells, they were shown to provide protection of those cells. On the other hand, Angelica Gigantis Radix decreased the viability of HepG2 cells significantly, Cuscuta Semen and Curcumae Tuber had a small or no effect on the viability of HepG2 cells. Conclusions: In the sovereign, minister, assistant and courier principle (君臣佐使論), Ginseng Radix (Red Ginseng) corresponds to the sovereign component because it provides cell protection effects, Angelica Gigantis Radix corresponds to minister medicinal because it kills cells, Schisandrae Fructus corresponds to the assistant medicinal to help red ginseng having cell protect effects. Cuscuta Semen and Curcumae Tuber correspond to the courier medicinal having no effect in cell viability in HepG2. We hope this study provides motivation for advanced research on the sovereign, minister, assistant and courier principle.

• 한국통신학회논문지
• /
• 제29권1C호
• /
• pp.65-71
• /
• 2004

본 논문은 주성분분석(PCA, Principle Component Analysis) 혹은 독립성분분석(ICA, Independent Principle Component Analysis)를 이용하여 HMM(Hidden Markov Model) 파라메타의 차수를 감소시킴으로써 MLLR(Maximum Likelihood Linear Regression) 화자 적응 알고리즘을 개선하였다. 데이터의 특징을 잘 나타내는 PCA와 ICA를 통해 모델 mixture component의 상관관계를 줄이고 상대적으로 데이터의 분포가 적은 축을 삭제함으로써 추정해야 하는 적응 파라메타의 수를 줄였다. 기존의 MLLR 알고리즘은 SI(Speaker Independent)모델 보다 좋은 인식성능을 나타내기 위해 30초 이상의 적응 데이터가 요구되었고, 반면 제안한 알고리즘은 적응 파라메타의 수를 감소시킴으로써 10초 이상의 적응데이터가 요구되었다. 또한, 36차의 HMM 파라메타는 기존의 MLLR 알고리즘과 비슷한 인식성능을 나다내는 10차의 주성분이나 독릭성분을 사용함으로써 MLLR 알고리즘에서 적응파라메타를 추정할 때 요구되는 연산량을 1/167로 감소시켰다.

• 대한전자공학회논문지SP
• /
• 제46권2호
• /
• pp.99-107
• /
• 2009

본 논문에서는 자동차 번호판 숫자의 특징을 추출하기 위해 강화된 독립성분분석(independent component analysis)의 혼합모델을 제안한다 독립성분분석은 고차 통계적 특성만을 이용하기 때문에 고차 통계적 특성과 숫자 종류별 상관관계에 대한 특성을 고려하지 못한다. 이러한 독립성분분석의 한계를 극복하기 위해, 본 논문에서는 주성분분석(principle component analysis)과 선형판별분석(linear discriminant analysis)을 조합한 혼합 모델 형태의 독립성분분석을 제안한다. 실험 결과, 제안된 혼합 모델은 독립성분분석이나 다른 혼합 모델들보다 특징 추출과 인식에서 우수한 성능을 보임을 확인하였다.