• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.3
• /
• pp.371-376
• /
• 2003

In this paper, we proposes a method for recognizing the facial expressions using ICA(Independent Component Analysis)-factorial representation method. Facial expression recognition consists of two stages. First, a method of Feature extraction transforms the high dimensional face space into a low dimensional feature space using PCA(Principal Component Analysis). And then, the feature vectors are extracted by using ICA-factorial representation method. The second recognition stage is performed by using the Euclidean distance measure based KNN(K-Nearest Neighbor) algorithm. We constructed the facial expression database for six basic expressions(happiness, sadness, angry, surprise, fear, dislike) and obtained a better performance than previous works.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.26 no.10B
• /
• pp.1444-1452
• /
• 2001

얼굴인식에서 정보 이론적 접근방법은 얼굴 영상을 기저 영상의 합으로 분해하는 것을 기초로 한다. 가장 많이 쓰이고 있는 방법은 Principal Component Analysis (PCA)를 기반으로 하는 eigenface 방법이다. PCA를 기반으로 하는 방법은 데이터의 2차 통계적 구조만을 고려하므로 화소 사이의 고차 통계적 의존성은 고려되지 않는다. Factorial code 표현법은 효과적인 정보 표현의 좋은 방법으로 알려져 있고 이것은 Independent Component Analysis (ICA)와 밀접한 관련이 있다. Factorial code 표현법은 eigenface 방법과 비교할 때 중요한 정보가 포함되어 있는 데이터의 고차 통계적 구조도 고려되어 더욱 효과적인 정보 표현을 기대할 수 있다. 이 논문에서는 PCA를 이용하여 찾아낸 저차원 특징 공간에서 Factorial code 표현법을 이용하여 얼굴인식을 위한 통계적 특징점을 찾아낸다. 얼굴 인식에 있어서 Factorial code 표현법이 eigenface 방법보다 성능이 우수함을 모의실험을 통하여 입증한다.

• Journal of Power System Engineering
• /
• v.22 no.1
• /
• pp.5-10
• /
• 2018

In order to employ factorial design on electronically-controlled diesel engine, effects of 5 factors on specific fuel consumption, nitrogen oxides and carbon monoxide were examined by fractional and full factorial design in this research. There were different results between fractional and full factorial design, then effect of variables as ambient condition and measurement of fuel consumption were confirmed. It was shown that ambient condition affected uniformly trend of nitrogen oxides and carbon monoxide. However, both ambient condition and measurement of fuel consumption had nothing to do with trend of specific fuel consumption and therefore it must be careful to employ factorial design on specific fuel consumption as response.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.12a
• /
• pp.329-332
• /
• 2002

본 논문에서는 효과적인 정보를 표현하는 ICA(Independent Component Analysis)-Factorial 표현 방법을 이용하여 얼굴감정인식을 수행한다. 얼굴감정인식은 두 단계인 특징추출과 인식단계에 의해 이루어진다. 먼저 특징추출방법은 PCA(Principal Component Analysis)을 이용하여 얼굴영상의 고차원 공간을 저차원 특징공간으로 변환한 후 ICA-factorial 표현방법을 통해 좀 더 효과적으로 특징벡터를 추출한다. 인식단계는 최소거리 분류방법인 유클리디안 거리를 이용하여 얼굴감정을 인식한다. 이 방법의 유용성을 설명하기 위해 6개의 기본감정(행복, 슬픔, 화남, 놀람, 공포, 혐오)에 대해 얼굴데이터베이스를 구축하고, 기존의 방법인 Eigenfaces, Fishefaces와 비교하여 좋은 인식성능을 보이고자 한다.

• Korean Journal of Human Ecology
• /
• v.18 no.2
• /
• pp.381-395
• /
• 2009

This study was conducted to investigate effects of self concept of children consumers on consumption propensity of online items according to gender by considering that self, which means a thought on ownself, is expressed as online items in cyber space. Totally, 760 copies of the questionnaire were distributed and 716 ones were finally used as data for analysis in this study. The data were analyzed by using SPSS 14.0 and factorial analysis, Cronbach' ${\alpha}$, t-test, cross tabulation analysis and multiple regression were conducted. The results of this study could be summarized as follows. First, the factorial analysis performed to examine types of self concept found four factorial structures - justness, social, academic and family self. In addition, the factorial analysis done to determine compositional factors of irrational consumption propensity of children consumers showed three factorial structures - impulse buying, conspicuous consumption and imitative consumption. When the effects of personal variables, item-transaction variables and self concept on irrational consumption propensity of children consumers were investigated according to gender, significant factors affecting consumption propensity of both of the males and females were a monthly pocket money, time using the Internet and major transaction items. In addition, parents' permission for purchasing items, justness self and family self were found to be variables affecting consumption propensity in the male children. On the contrary, self concept of female children did not affect irrational consumption propensity.

• Journal of Korean Society for Quality Management
• /
• v.28 no.3
• /
• pp.59-67
• /
• 2000

In analysis of resolution IV fractional factorial experiments, the main effects only are analyzed, even though we can get some useful information on the confounded 2-factor interactions. In this paper, we introduce an exploiting method of the confounded structure of interactions, especially for the near minimal resolution IV 3$^{t}$ fractional factorial designs developed by Anderson and Thomas (1979). Moreover, in this paper the application way of the proposed method is also discussed by analyzing some simulated data.

• Communications for Statistical Applications and Methods
• /
• v.12 no.1
• /
• pp.47-59
• /
• 2005

Expected mean square(EMS) is an important part of conducting the analysis of variance in the experimental design problem, especially in mixed or random models. We present a set of EMS rules for balanced factorial designs under no restriction on interaction. Also we point out how to use the variance component of SPSS or SAS procedure to obtain EMS.

• Journal of Korean Society of Water and Wastewater
• /
• v.19 no.2
• /
• pp.181-192
• /
• 2005

This study evaluated the effect of factors influencing the initial biofilm formation in drinking water distribution pipe by running experiments using a $2^{4-1}$ fractional factorial experimental design with a replicate. Important variables used for assessing biofilm formation included BDOC(biodegradable dissolved organic carbon), viable heterotrophic bacteria present in drinking water, water temperature, and shear stress at two levels each. Based on the statistical analysis of biofilm levels measured as attached HPC(heterotrophic plate count) and community-level assay, the main factors that have significant effects on biofilm formation were found to be viable heterotrophic bacteria and BDOC. Water temperature only exhibited significant effect on the levels of attached HPC, while shear stress was not a significant factor under given conditions. Moreover, the statistical analysis revealed that interactions between the important variables were not statistically significant at a 0.05 significance level.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.1 no.1
• /
• pp.133-140
• /
• 2002

The object of this paper is to predict the surface roughness using the experiment equation of surface roughness, which is developed with a full factorial design in turning. $3^3$ full factorial design has been used to study main and interaction effects of main cutting parameters such as cutting speed, feed rate, and depth of cut, on surface roughness. For prediction of surface roughness, the arithmetic average (Ra) is used, and stepwise regression has been used to check the significance of all effects of cutting parameters. Using the result of these, the experimental equation of surface roughness, which consists of significant effects of cutting parameters, has been developed. The coefficient of determination of this equation is 0.9908. And the prediction ability of this equation was verified by additional experiments. The result of that, the coefficient of determination is 0.9718.

• Journal of Korean Society for Quality Management
• /
• v.38 no.1
• /
• pp.42-51
• /
• 2010

For the fractional factorial designs, the resolution-IV designs can be used when we want to estimate the main effects and to investigate the structure of the non-negligible two-factor interaction effects, when the three-factor and higher order interaction effects are all negligible. However we need to add the additional treatment combinations in order to identify the influential interactions for the resolution-IV fractional factorial designs. In this paper we investigate the statistical structure for 3-level resolution-IV designs constructed by fold-over scheme and introduce a method for analyzing the influential two-factor interactions.