• Proceedings of the PSK Conference
• /
• 2002.10a
• /
• pp.402.2-402.2
• /
• 2002

An algorithm for finding the best regression models has been developed using NIR spectral data. In cases of regression analysis for quantitation with NIR spectral data, it is very critical to find essential features from the spectral data. This task was accessed in two ways. The first one was to use all-possible combinations of varibles (wavelengths). Correlation coefficients at each spectral points were calculated to get initial set of variables and all of the possible combinations of variable sets were tested with SEC. SEP and/or $R^2$. (omitted)

• Wind and Structures
• /
• v.2 no.2
• /
• pp.105-117
• /
• 1999

The various spectral density functions of wind are applied in the wind process simulation by the spectral representation method. In view of the spectral density functions, the characteristics of the simulated processes are compared. The ensemble spectral density functions constructed from the simulated sample processes are revealed to have the similarity not only in global shape but also in the maximum values with the target spectral density functions with a high accuracy. For the correlation structure to be satisfied in the circumferential direction on the cooling tower shell, a new formula is suggested based on the mathematical expression representing the circumferential distribution of the wind pressure on the cooling tower shell. The simulated wind processes are applied in the dynamic analysis of cooling tower shell in the time domain and the fluctuating stochastic behavior of the cooling tower shell is investigated.

• The Journal of the Acoustical Society of Korea
• /
• v.31 no.2
• /
• pp.93-99
• /
• 2012

For speech coding, a vocal tract is modeled using Linear Predictive Coding (LPC) coefficients. The LPC coefficients are typically transformed to Line Spectral Frequency (LSF) parameters which are advantageous for linear interpolation and quantization. If multidimensional LSF data are quantized directly using Vector-Quantization (VQ), high rate-distortion performance can be obtained by fully utilizing intra-frame correlation. In practice, since this direct VQ system cannot be used due to high computational complexity and memory requirement, Split VQ (SVQ) is used where a multidimensional vector is split into multilple sub-vectors for quantization. The LSF parameters also have high inter-frame correlation, and thus Predictive SVQ (PSVQ) is utilized. PSVQ provides better rate-distortion performance than SVQ. In this paper, to implement the optimal predictors in PSVQ for voice storage devices, we propose Multi-Frame AR-model based SVQ (MF-AR-SVQ) that considers the inter-frame correlations with multiple previous frames. Compared with conventional PSVQ, the proposed MF-AR-SVQ provides 1 bit gain in terms of spectral distortion without significant increase in complexity and memory requirement.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.10
• /
• pp.142-149
• /
• 2013

In general, a color of an image is represented by using red, green, and blue channels in a RGB camera system. However, only information of three channels are limited to estimate a spectral reflectance of a real scene. Because of this, the RGB camera system can not accurately represent the color. To overcome this limitation and represent an accurate color, researches to estimate the spectral reflectance by using a multi-channel camera system are being actively proceeded. Recently, a reflectance estimation method adaptively constructing a similar training set from a traditional training set according to a camera response by using a spectral similarity was introduced. However, in this method, an accuracy of the similar training set is reduced because the spectral similarity based on an average and a maximum distances was applied. In this paper, a reflectance estimation method applied a spectral similarity based on a correlation coefficient is proposed to improve the accuracy of the similar training set. Firstly, the correlation coefficient between the similar training set and the spectral reflectance obtained by Wiener estimation method is calculated. Secondly, the similar training set is constructed from the traditional training set according to the correlation coefficient. Finally, Wiener estimation method applied the similar training set is performed to estimate the spectral reflectance. To evaluate a performance of the proposed method with previous methods, experimental results are compared. As a result, the proposed method showed the best performance.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.15 no.3
• /
• pp.52-65
• /
• 2012

The net biomass accumulation (or net primary production, NPP) and gross primary production (GPP) have closely related with carbon accumulations(or carbon exchange) in vegetation. There are many approaches to estimate biomass using remote sensing techniques. The vegetation indices (VIs) can be a methodology to estimate biomass which assumes total chlorophyll contents. Various VIs were characterized with difference development conditions as vegetation species, input datasets. The hyperspectral data have also different spatial/spectral resolutions for aerial surveying. Additionally they need particular spectral bands selection difficulty to calculate the VIs. The objective of this study is to evaluate the correlations with airborne hyperspectral data (compact airborne spectrographic imager, CASI) and spectral unmixing model (or spectral mixture analysis, SMA) to characterize vegetation indices in forest area. The spectral mixture analysis was used to model the spectral purity of each pixel as an endmember. The endmembers are the fraction components derived from hyperspectral data through the SMA. In this study, we choose three endmembers represented vegetation pixels in the hyperspectral data. These endmembers were compared with 9 VIs by the Pearson's correlation coefficient. The results show MTVI1 and TVI have same correlation coefficient with 0.877. The MCARI, especially has very high relationship with vegetation endmembers as 0.9061 at less vegetation and soil distributed site. The MTVI1 and TVI have high correlations with the vegetation endmembers as 0.757 in whole test sites.

• Speech Sciences
• /
• v.9 no.3
• /
• pp.3-16
• /
• 2002

We examined some problems in speaker identification procedures: transformation of acoustic parameters into auditory scales, invalid measurement values, and comparability of spectral energy values across the frequency range. To resolve those problems, we analyzed the acoustic spectral energy of three Korean numbers produced by ten female students from narrow-band spectrograms at 19 proportional time points of each voiced segment. Then, cells of the first five spectral matrices were averaged to form a matrix model for each speaker. The correlation coefficients and sum of the absolute amplitude difference in each pair of the spectral models of the ten subjects were obtained. Also, some individual matrix models were compared to those of the same subject or the other subject with a similar spectral model. Results showed that in numbers '2' and '9' subjects could not be clearly distinguished from the others but in number '4' it shed some possibility of setting threshold values for speaker identification if we employed the coefficients and the sum of absolute difference. Further studies would be desirable on various combinations of the range of long-term average spectra and the degree of signal pre-emphasis.

• Phonetics and Speech Sciences
• /
• v.3 no.4
• /
• pp.109-115
• /
• 2011

The purposes of this study were to identify acoustic parameters of connected speech and to contribute to acoustic analysis of dysphonic voice about patient's natural speech voice as well as sustained phonation of vowels. Acoustic parameters of sentences included LTAS (long-term average spectrum) mean and spectral slope over frequence ranges such as 0-4kHz, 0-6kHz, 0-8kHz, 0-12.5kHz as well as HNR. Acoustic parameters of the vowel 'a' included jitter, RAP, shimmer, NHR, and HNR. Based on 'G' of GRBAS for the severity of dysphonia, two experienced raters judged and classified as four groups including controls, mild, moderate and severe dysphonic group. Connected speech was two sentences extracted from 'stroll' passage. Parameters of the vowel and LTAS mean of the sentences were measured by CSL. The spectral slope of the sentences and HNR of the vowel and the sentences were measured by Praat. Data were statistically analyzed by Spearman correlation and Kruskal-Wallis test using SPSS 12.0. The results of this study are as follows: First, jitter, RAP, shimmer and NHR were significantly different between the groups. Second, for several frequencies, LTAS mean and spectral slope of the sentences were significantly different between the groups. Third, the HNR of the sentences were significantly different between the groups. Forth, there was a presence of correlation between HNR and NHR of the vowel and HNR of the sentences. Accordingly, this study concluded that LTAS, spectral slope, and HNR were predictive parameters of connected speech voice for dysphonic voice.

• Wind and Structures
• /
• v.16 no.5
• /
• pp.517-540
• /
• 2013

This paper presents applications of proper orthogonal decomposition in both the time and frequency domains based on both cross spectral matrix and covariance matrix branches to analyze multi-variate unsteady pressure fields on prisms and to study spanwise and chordwise pressure distribution. Furthermore, modification of proper orthogonal decomposition is applied to a rectangular spanwise coherence matrix in order to investigate the spanwise correlation and coherence of the unsteady pressure fields. The unsteady pressure fields have been directly measured in wind tunnel tests on some typical prisms with slenderness ratios B/D=1, B/D=1 with a splitter plate in the wake, and B/D=5. Significance and contribution of the first covariance mode associated with the first principal coordinates as well as those of the first spectral eigenvalue and associated spectral mode are clarified by synthesis of the unsteady pressure fields and identification of intrinsic events inside the unsteady pressure fields. Spanwise coherence of the unsteady pressure fields has been mapped the first time ever for better understanding of their intrinsic characteristics.

• Journal of Pharmaceutical Investigation
• /
• v.37 no.6
• /
• pp.377-395
• /
• 2007

Near infrared(NIR) spectral data have been used for the noninvasive analysis of various biological samples. Nonetheless, absorption bands of NIR region are overlapped extensively. It is very difficult to select the proper wavelengths of spectral data, which give the best PCR(principal component regression) models for the analysis of constituents of biological samples. The NIR data were used after polynomial smoothing and differentiation of 1st order, using Savitzky-Golay filters. To find the best PCR models, all-possible combinations of available principal components from the given NIR spectral data were derived by in-house programs written in MATLAB codes. All of the extensively generated PCR models were compared in terms of SEC(standard error of calibration), $R^2$, SEP(standard error of prediction) and SECP(standard error of calibration and prediction) to find the best combination of principal components of the initial PCR models. The initial PCR models were found by SEC or Malinowski's indicator function and a priori selection of spectral points were examined in terms of correlation coefficients between NIR data at each wavelength and corresponding concentrations. For the test of the developed program, aqueous solutions of BSA(bovine serum albumin) and glucose were prepared and analyzed. As a result, the best PCR models were found using a priori selection of spectral points and the final model selection by SEP or SECP.

• Journal of Astronomy and Space Sciences
• /
• v.33 no.4
• /
• pp.247-256
• /
• 2016

An analysis of light curves and spectra of observed gamma-ray bursts in gamma-ray ranges is frequently demanded because the prompt emission contains immediate details regarding the central engine of gamma-ray bursts (GRBs). We have revisited the relationship between the collimation-corrected peak luminosity and the spectral lag, investigating the lag-luminosity relationships in great detail by focusing on spectral lags resulting from all possible combinations of channels. Firstly, we compiled the opening angle data and demonstrated that the distribution of opening angles of 205 long GRBs is represented by a double Gaussian function having maxima at ~ 0.1 and ~ 0.3 radians. We confirmed that the peak luminosity and the spectral lag are anti-correlated, both in the observer frame and in the source frame. We found that, in agreement with our previous conclusion, the correlation coefficient improves significantly in the source frame. It should be noted that spectral lags involving channel 2 (25-50 keV) yield high correlation coefficients, where Swift/Burst Alert Telescope (BAT) has four energy channels (channel 1: 15-25 keV, channel 2: 25-50 keV, channel 3: 50-100 keV, channel 4: 100-200 keV). We also found that peak luminosity is positively correlated with peak energy.