• Korean Journal of Remote Sensing
• /
• v.31 no.4
• /
• pp.303-319
• /
• 2015

The objectives of this paper are to measure surface imperviousness using three different classification methods: per-pixel, sub-pixel, and object-oriented classification. They are tested on high-spatial resolution QuickBird data at 2.4 meters (four spectral bands and three principal component bands) as well as a medium-spatial resolution Landsat TM image at 30 meters. To measure impervious surfaces, we selected 30 sample sites with different land uses and residential densities across image representing the city of Phoenix, Arizona, USA. For per-pixel an unsupervised classification is first conducted to provide prior knowledge on the possible candidate spectral classes, and then a supervised classification is performed using the maximum-likelihood rule. For sub-pixel classification, a Linear Spectral Mixture Analysis (LSMA) is used to disentangle land cover information from mixed pixels. For object-oriented classification several different sets of scale parameters and expert decision rules are implemented, including a nearest neighbor classifier. The results from these three methods show that the object-oriented approach (accuracy of 91%) provides more accurate results than those achieved by per-pixel algorithm (accuracy of 67% and 83% using Landsat TM and QuickBird, respectively). It is also clear that sub-pixel algorithm gives more accurate results (accuracy of 87%) in case of intensive and dense urban areas using medium-resolution imagery.

• Journal of Korea Multimedia Society
• /
• v.25 no.6
• /
• pp.886-893
• /
• 2022

Outlier detection refers to the task of detecting data that deviate significantly from the normal data distribution. Most outlier detection methods compute an outlier score which indicates the degree to which a data sample deviates from normal. However, setting a threshold for an outlier score to determine if a data sample is outlier or normal is not trivial. In this paper, we propose a binary prediction method for outlier detection based on spectral clustering and one-class SVM ensemble. Given training data consisting of normal data samples, a clustering method is performed to find clusters in the training data, and the ensemble of one-class SVM models trained on each cluster finds the boundaries of the normal data. We show how to obtain a threshold for transforming outlier scores computed from the ensemble of one-class SVM models into binary predictive values. Experimental results with high dimensional text data show that the proposed method can be effectively applied to high dimensional data, especially when the normal training data consists of different shapes and densities of clusters.

• Journal of the Society of Naval Architects of Korea
• /
• v.49 no.4
• /
• pp.350-358
• /
• 2012

This paper deals with accuracy of accumulated fatigue damage estimation using stochastic fatigue analysis method based on Rayleigh PDF. From full scale measurement data on an 8100TEU container vessel, zero-order spectral moments for wave- and vibration-induced energy spectral densities are determined on the probability level of 99%. 80 simulation cases in total are prepared according to the variation of ratio of zero-order spectral moments and center frequency of vibration ESD. By using inverse Fourier transformation and rainflow cycle counting for the combined ESD of wave and vibration, exact fatigue damages are derived. Fatigue damages in frequency domain based on Rayleigh PDF show large conservativeness compared to exact fatigue damages in times domain. The main cause of the excessive conservativeness is analyzed by two aspects: ratio of zero crossing and peak frequencies and ratio of initial zero order spectral moments and zero order spectral moments from rainflow stress range distributions. Finally, a guideline of applicability of Rayleigh PDF is proposed for wide band processes.

• Wind and Structures
• /
• v.23 no.5
• /
• pp.385-403
• /
• 2016

The across-wind dynamic loads on L-shaped tall buildings with various geometric dimensions were investigated through a series of wind tunnel testing. The lift coefficients, power spectral densities and vertical correlation coefficients of the across-wind loads were analyzed and discussed in details. Taking the side ratio and terrain category as key variables, empirical formulas for estimating the across-wind dynamic loads on L-shaped tall buildings were proposed on the basis of the wind tunnel testing results. Comparisons between the predictions by the empirical formulas and the wind tunnel test results were made to verify the accuracy and applicability of the proposed formulas. Moreover, a simplified procedure to evaluate the across-wind dynamic loads on L-shaped tall buildings was derived from the proposed formulas. This study aims to provide a simple and reliable way for the estimation of across-wind dynamic loads on L-shaped tall buildings.

• Journal of Astronomy and Space Sciences
• /
• v.38 no.4
• /
• pp.193-202
• /
• 2021

We present analysis results of Korean VLBI Network (KVN) four-band data for the highly variable blazar 3C 279. We measured the 22, 43, 86, and 129 GHz flux densities and spectral indices of the source using contemporaneous data taken over 5.6 years. We used the discrete correlation function to investigate correlations between the radio emission properties and those measured in the optical (2 × 10¹⁴ - 1.5 × 10¹⁵ Hz), X-ray (0.3-10 keV), and gamma-ray (0.1-300 GeV) bands. We found a significant correlation between the radio spectral index and gamma-ray flux without a time delay and interpreted the correlation using an extended jet scenario for blazar emission.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 1992.04a
• /
• pp.26-29
• /
• 1992

Modern tall buildings are subject to wind induced oscillations. Those oscillations can cause discomfort to the occupants. To control these motions, tuned mass dampers have been used. In this paper, component node synthesis, based on Lagrange multipliers formulation. is applied to the along-wind motion of tall buildings with multiple tuned mass dampers. Spectral densities of accelerations of top floor are compared by changing the numbers and locations of tuned mass dampers. It is found that multiple tuned mass dampers can be more effective than single tuned mass damper in reducing the acceleration response.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.26 no.8
• /
• pp.1128-1136
• /
• 1989

The power spectral densities of run-length-limited(RLL) minimum-bandwidth(MB) line codes are obtained for some values of the zero probability of input data. They can be obtianed by implementing a computer program for the G.L. Cariolaro's algorithm. Using the program we calculate the power spectra of the known RLL MB line codes, such as VDB5, TIB5, VMDB5, RMD35, DRMD35, RMD45, and RMD33, and compare the characteristics of them. Specially, the relationship between eye widths and power spectra is analyzed.

• Journal of The Korean Astronomical Society
• /
• v.35 no.4
• /
• pp.187-196
• /
• 2002

A spectral line survey is performed from 159.7 to 164.7 GHz toward Orion-KL, as an extension of our previous line survey from 138.3 to 150.7 GHz with the same 14 m radio telescope of Taeduk Radio Astronomy Observatory. Typical system temperatures were 260 - 1000 K to achieve a sensitivity of about 0.02 - 0.04 K in TA unit. A total of 63 line spectra are detected in this survey. Among them, 54 lines lines are found to be the first detections towards an astronomical source and only 9 spectral lines have been previously identified from other observations. Forty-eight of 54 lines are believed to be from the known transitions of the known molecules, while 6 lines are 'unidentified'. All detected lines are found to be from a total of 10 molecular species and their isotopic variants. The molecular species with most numerous detected transitions are $HCOOCH_3$ (22), followed by $CH_3OCH_3$ (7), $C_2H_5CN$ (7), and $SO_2$ (6). The LTE rotation diagram analysis using all homogeneous data with those from previous survey gives more reliable determination of physical quantities. The derived values of the rotation temperatures and column densities for $HCOOCH_3$, $CH_3OCH_3$, and $SO_2$ are are 75 $\~$ 197 K and $1.5 {\~}18 {\times} 10^{15}\;cm^{-2}$, respectively.

• Journal of Advanced Navigation Technology
• /
• v.10 no.4
• /
• pp.333-341
• /
• 2006

This paper analyzes the spectral power properties of various time hopping UWB signals and shows that the power spectral densities of the various signals could have to be determined by the PSD of the pulse used in the signal. The pulse design method by which the FCC emission mask can be utilized fully is proposed. The method combines the arbitrary derivative Gaussian pulse linearly. The coefficients of the linear combination are calculated by the LSE(Least Square Error) method. Various parameters such as the number of coefficients and the types of the basic pulses are considered when calculating the PSD and pulse shapes of the new pulses.

• Journal of The Korean Astronomical Society
• /
• v.34 no.3
• /
• pp.167-179
• /
• 2001

A molecular line survey towards the UC H II region G34.3+0.15 from 155.3 to 165.3GHz has been conducted with the TRAO 14-m radio telescope. Combined with our previous observations from 84.7 to 115.6GHz and 123.5 to 155.3GHz (Paper I), the spectral coverage of this survey in G34.3+0.15 now runs from 85 to 165 GHz. From these latest observations, a total of 18 lines from 6 species were detected. These include four new lines corresponding to ${\Delta}$J = 0, ${\Delta}$K = 1 transitions of the $CH_3OH$ E-type species, and two new lines corresponding to transitions from $SO_2$ and $HC_3N$. These 6 new lines are $CH_3OH$[1(1) - 1(0)E], $CH_3OH$[2(1) - 2(0)E], $CH_3OH$[3(1) - 3(0)E], $CH_3OH$[4(1) - 4(0)E], $SO_2$[14(1, 13) -14(0, 14)] and $HC_3N$[18 -17]. We applied a rotation diagram analysis to derive rotation temperatures and column densities from the methanol transitions detected, and combined with NRAO 12-m data from Slysh et al. 1999. Applying a two-component fit, we find a cold component with temperature 13-16K and column density $3.3-3.4 {\times} 10^{14} cm^{-2}$, and a hot component with temperature 64 - 83K and column density $9.3{\times}10^{14} - 9.7 {\times} 10^{14} cm^{-2}$. On the other hand, applying just a one-component fit yields temperatures in the 47 -62 K range and column densities from $7.5-1.1 {\times} 10^{15} cm^{-2}$.