• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.05a
• /
• pp.204-207
• /
• 2013

Music genre classification is an essential component for music information retrieval system. There are two important components to be considered for better genre classification, which are audio feature extraction and classifier. This paper incorporates two different kinds of features for genre classification, timbral texture and rhythmic content features. Timbral texture contains several spectral and Mel-frequency Cepstral Coefficient (MFCC) features. Before choosing a timbral feature we explore which feature contributes less significant role on genre discrimination. This facilitates the reduction of feature dimension. For the timbral features up to the 4-th order central moments and the covariance components of mutual features are considered to improve the overall classification result. For the rhythmic content the features extracted from beat histogram are selected. In the paper Extreme Learning Machine (ELM) with bagging is used as classifier for classifying the genres. Based on the proposed feature sets and classifier, experiment is performed with well-known datasets: GTZAN databases with ten different music genres, respectively. The proposed method acquires the better classification accuracy than the existing approaches.

• The Journal of the Acoustical Society of Korea
• /
• v.22 no.3E
• /
• pp.124-132
• /
• 2003

We propose a new method for speech/non-speech classifiers based on concepts of the adaptive boosting (AdaBoost) algorithm in order to detect speech for robust speech recognition. The method uses a combination of simple base classifiers through the AdaBoost algorithm and a set of optimized speech features combined with spectral subtraction. The key benefits of this method are the simple implementation, low computational complexity and the avoidance of the over-fitting problem. We checked the validity of the method by comparing its performance with the speech/non-speech classifier used in a standard voice activity detector. For speech recognition purpose, additional performance improvements were achieved by the adoption of new features including speech band energies and MFCC-based spectral distortion. For the same false alarm rate, the method reduced 20-50% of miss errors.

• Journal of Astronomy and Space Sciences
• /
• v.18 no.2
• /
• pp.95-100
• /
• 2001

We have presented an ab initio model of the $2{mu}m$ spectral features of $(H_2)_2$ based on the far-infrared models of McKellar & Schaefer (1991). We have shown that the intensity variations of the $2{mu}m;(H_2)_2$ features depend on the ortho/para ratios of $H_2$ We have discussed the applicability of the variations to the atmospheres of the giant planets for the derivations of the ortho/para ratios. The signal to noise ratios of currently available spectra of the giant planets are not sufficient enough to derive accurate ortho/para ratios of these planets. Observations with longer exposure times and larger telescope apertures are required to obtain better spectra for the derivations of the ortho/para ratios of $H_2$ in the atmospheres of the giant planets.

• Proceedings of the Korea Committee for Ocean Resources and Engineering Conference
• /
• 2006.11a
• /
• pp.113-120
• /
• 2006

In this paper, generation mechanisms of ocean freak waves are briefly introduced in the context of wave-current interaction phenomena. As an accurate and efficient numerical tool, the spectral element method is presented with general features and specific treatment for the wave-current interaction problem. The present model of the fluid motion is based on the Navier-Stokes equations incorporating a velocity-pressure formulation. In order to deal with the free surface motion, an Arbitrary Lagrangian-Eulerian (ALE) description is adopted. As an intermediate stage of development, solution procedure and characteristic aspects of the present modeling and numerical method features are addressed in detail, and numerical results for wave-current interaction is left as further study.

• The Bulletin of The Korean Astronomical Society
• /
• v.41 no.1
• /
• pp.59.4-60
• /
• 2016

One important aspect of the late stage stellar evolution is the mass loss processes, where a significant amount of stellar material will be returned to the interstellar space to be used for stars of the next generation. Raman scattered O VI and He II by atomic hydrogen in symbiotic stars and young planetary nebulae are found to be excellent tools to investigate the mass loss processes and estimate the mass loss rate. These features appear near hydrogen Balmer emission lines due to the huge cross section in the vicinity of Lyman resonance transitions. With the capability of high spectral resolution and broad spectral coverage, BOES is an ideal instrument to perform Raman spectroscopy of these objects. In this talk, a cursory overview of our research activities on Raman spectroscopy of symbiotics and PNe using the BOES is presented.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2009.10a
• /
• pp.1022-1024
• /
• 2009

Stereoscopic Images provide depth information with the relative distances between the objects in the images. There are many different ways to extract disparity maps from the visible spectral images. For the infrared spectral range, the same approach cannot be utilized for the innate low resolution and colorless features because typical methods require corresponding features between the images. The authors suggest a new approach that makes use of image segmentation to obtain depth information for stereoscopic millimeter-wave images. For image segmentation a selective visual attention model based on the theory of a feature-integration of attention is used. Experimental results show the proposed method provides reasonable depth information for object shape recognition and display.

• Journal of the Optical Society of Korea
• /
• v.14 no.3
• /
• pp.190-198
• /
• 2010

A set of radiative transfer software named CART (Combined Atmospheric Radiative Transfer) has been developed to rapidly calculate atmospheric transmittance and background radiance. The spectral resolution of CART is $1cm^{-1}$, and the spectral region covers from 1 to $25000cm^{-1}$. CART has five characteristic features, and it can be applied to many fields. The features and applications of CART are summarized in detail.

• The Bulletin of The Korean Astronomical Society
• /
• v.38 no.2
• /
• pp.86.2-86.2
• /
• 2013

Ellerman bombs(EBs) are emission features at the wings of the H alpha spectral line. They are believed to be a kind of a magnetic reconnection feature in the low chromosphere or near photosphere. It was previously reported that surges often occur in association with EBs. However, previous observations were restricted to imaging observation. Using Fast Imaging Solar Spectrograph installed in New Solar Telescope at Big Bear Solar Observatory, California, we observed 5 EBs and associated surges with high-spatial and high-spectral resolutions. In this presentation, we will show the results and discuss the physical properties.

• Proceedings of the Optical Society of Korea Conference
• /
• 1989.02a
• /
• pp.147-151
• /
• 1989

A spectral inversion scheme, based on the moment-method procudure with series-expanded field within each cell by dividing the dielectric object into a small number os rectangular cells with large area, is developed to regularize the ill-posedness inherent in inverse scattering problems. One of the interesting features on the presented scheme is that the relative dielectric constant may be obtained by averaging over each cell. This averaging is expected to play an important role in regularizing the high-frequency effect due to noise.

• The Bulletin of The Korean Astronomical Society
• /
• v.33 no.1
• /
• pp.27.1-27.1
• /
• 2008