• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.11a
• /
• pp.305-309
• /
• 2000

In this paper, we presents a new methodology for face recognition after analysing conventional ICA(Independent Component Analysis) based approach. In the literature we found that ICA based methods have followed the same procedure without any exception, first PCA(Principal Component Analysis) has been used for feature extraction, next ICA learning method has been applied for feature enhancement in the reduced dimension. However, it is contradiction that features are extracted using higher order moments depend on variance, the second order statistics. It is not considered that a necessary component can be located in the discarded feature space. In the new methodology, features are extracted using the magnitude of kurtosis(4-th order central moment or cumulant). This corresponds to the PCA based feature extraction using eigenvalue(2nd order central moment or variance). The synergy effect of PCA and ICA can be achieved if PCA is used for noise reduction filter. ICA methodology is analysed using SVD(Singular Value Decomposition). PCA does whitening and noise reduction. ICA performs the feature extraction. Simulation results show the effectiveness of the methodology compared to the conventional ICA approach.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.20 no.1
• /
• pp.45-54
• /
• 2016

This study suggests a prediction model of ground motion spectral shape considering characteristics of earthquake records in Korea. Based on the Graizer and Kalkan's prediction procedure, a spectral shape model is defined as a continuous function of period in order to improve the complex problems of the conventional models. The approximate spectral shape function is then developed with parameters such as moment magnitude, fault distance, and average shear velocity of independent variables. This paper finally determines estimator coefficients of subfunctions which explain the corelation among the independent variables using the nonlinear optimization. As a result of generating the prediction model of ground motion spectral shape, the ground motion spectral shape well estimates the response spectrum of earthquake recordings in Korea.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.2
• /
• pp.459-465
• /
• 2009

For an inverse double Weibull distribution which is symmetric about zero, we obtain distribution and moment of ratio of independent inverse double Weibull variables, and also obtain the cumulative distribution function and moment of a skew-symmetric inverse double Weibull distribution. And we introduce a skew-symmetric inverse double Weibull generated by a double Weibull distribution.

• Communications for Statistical Applications and Methods
• /
• v.16 no.3
• /
• pp.541-547
• /
• 2009

We consider distribution, reliability and moment of ratio in two independent beta random variables X and Y, and reliability and $K^{th}$ moment of ratio are represented by a mathematical generalized hypergeometric function. We introduce an approximate maximum likelihood estimate(AML) of reliability and right-tail probability in the beta distribution.

• Journal of the Korean Data and Information Science Society
• /
• v.18 no.2
• /
• pp.543-552
• /
• 2007

We consider estimation of the right-tail probability in a half-triangle distribution, and also consider inference on reliability, and derive the k-th moment of ratio of two independent half-triangle distributions with different supports. As we define a skew-symmetric random variable from a symmetric triangle distribution about origin, we derive its k-th moment.

• Physical Therapy Korea
• /
• v.15 no.2
• /
• pp.64-72
• /
• 2008

The purpose of this study was to determine the differences in joint moment in the intact limb of uni-transfemoral amputees and to identify the implications of knee osteoarthritis. As an experimental method, three-dimensional gait analysis was performed on 10 uni-transfemoral amputees and 10 healthy males. Kinematics and kinetics at the hip, knee, and ankle joint were calculated. As a statistical method, independent t-tests were conducted to perform a comparison between the transfemoral amputee group and the control group. The results showed that the external knee adduction moment increased in the transfemoral amputee group (.22 Nm/kg) compared with that of the control group (.13 Nm/kg) at terminal stance (p=.008). External knee flexion moment also increased in the transfemoral amputee group (.24 Nm/kg) but this difference was not statistically significant. External hip flexion moment increased in the transfemoral amputee group (1.35 Nm/kg) compared with that of the control group (.45 Nm/kg) at initial stance, and external hip extension moment decreased in the transfemoral amputee group (-.26 Nm/kg) compared with that of the control group (-.76 Nm/kg) at terminal stance. Although external ankle plantarflexion moment of the transfemoral amputee group increased, it was not found to be statistically significant. The results suggest that the intact limb joint moment of the uni-transfemoral amputees during walking can be different from that of healthy subjects. In conclusion, it was found that there is a link between the increase of external knee adduction moment and the prevalence of knee osteoarthritis in uni-transfemoral amputees. This result is expected to provide some objective data for rehabilitation programs related to knee osteoarthritis in transfemoral amputees.

• The KIPS Transactions:PartB
• /
• v.13B no.1 s.104
• /
• pp.7-14
• /
• 2006

This paper presents an efficient face recognition method using both first moment of image and basis images. First moment which is a method for finding centroid of image, is applied to exclude the needless backgrounds in the face recognitions by shifting to the centroid of face image. Basis images which are the face features, are respectively extracted by principal component analysis(PCA) and fixed-point independent component analysis(FP-ICA). This is to improve the recognition performance by excluding the redundancy considering to second- and higher-order statistics of face image. The proposed methods has been applied to the problem for recognizing the 48 face images(12 persons*4 scenes) of 64*64 pixels. The 3 distances such as city-block, Euclidean, negative angle are used as measures when match the probe images to the nearest gallery images. The experimental results show that the proposed methods has a superior recognition performances(speed, rate) than conventional PCA and FP-ICA without preprocessing, the proposed FP-ICA has also better performance than the proposed PCA. The city-block has been relatively achieved more an accurate similarity than Euclidean or negative angle.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.1
• /
• pp.219-223
• /
• 2009

We shall derive a quotient distribution of two independent gamma variables and its moment and reliability are represented by hypergeometric function and Wittaker's function. And we shall consider an inference on the reliability in two independent gamma random variables.

• 한국데이터정보과학회:학술대회논문집
• /
• 2006.11a
• /
• pp.37-42
• /
• 2006

We shall consider an inference of the reliability P(Y

• Proceedings of the Earthquake Engineering Society of Korea Conference
• /
• 2002.03a
• /
• pp.3-11
• /
• 2002

Source parameters for forty nine recent earthquakes occurred in and around Korean Peninsula are determined and the relations among them are studied. The corner frequency and seismic moment are estimated from three different methods. The spectral fitting of the source displacement spectrum with the $\omega$-square source model of Brune(1970) and Snoke(1987)'s method are applied to all events and empirical Green's function method for two events are adopted. The source parameters determined in this study show different values depending on the adopted method and on the stations of which seismograms are recorded. It is interpreted that the disagreements principally originate from insufficient consideration of source radiation pattern and attenuation and amplification according to path direction. The corner frequencies and seismic moments are averaged to exclude the directional effects and other source parameters are estimated from the mean corner frequency and seismic moment. The static stress drops estimated in this study tend to be independent of seismic moment or magnitude for events above a certain size. For earthquakes with the size less than about 3.0$\times$10$^{21}$dyne-cm(nearly same as M$_{L}$=3.7), the stress drop tends to decrease with the decreasing moment. This fact suggests a breakdown of scaling law of source parameters below the threshold magnitude. The moment magnitudes calculated from source parameters appear to be slightly larger than the Richter's local magnitudes in the range above M$_{L}$=3.5.3.5.