• The Journal of the Acoustical Society of Korea
• /
• v.33 no.1
• /
• pp.81-86
• /
• 2014

This paper presents a study on the effectiveness of using the spectral and the temporal octave-band order statistics for musical genre classification. In order to represent the relative disposition of the harmonic and non-harmonic components, we utilize the octave-band order statistics of power spectral distribution. Experiments on the widely used two music datasets were performed; the results show that the octave-band order statistics improve genre classification accuracy by 2.61 % for one dataset and 8.9 % for another dataset compared with the mel-frequency cepstral coefficients and the octave-band spectral contrast. Experimental results show that the octave-band order statistics are promising for musical genre classification.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.12
• /
• pp.238-244
• /
• 2013

For musical genre classification, various types of feature vectors are utilized. Mel-frequency cepstral coefficient (MFCC), decorrelated filter bank (DFB), and octave-based spectral contrast (OSC) are widely used as short-term features, and their long-term variations are also utilized. In this paper, OSC features are extracted not only in the single-octave band domain, but also in the multiple-octave band one to capture the correlation between octave bands. As a baseline system, we select the genre classification system that won the fourth place in the 2012 music information retrieval evaluation exchange (MIREX) contest. By applying the OSC features based on multiple-octave bands, we obtain the better classification accuracy by 0.40% and 3.15% for the GTZAN and Ballroom databases, respectively.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.6 no.4
• /
• pp.23-29
• /
• 2002

Radar method can be effective in probing concrete structures damaged by earthquake. Data analysis is usually performed in time domain, by considering time delay of the wave due to the dielectric constant of concrete. In this study, improved data analysis has been performed using signal processing scheme of spectra analysis and filtering. Three antenna with 900MHz, 1㎓, and 1.5㎓ center frequency were used to detect a steel bar or delamination in specimens for obtaining data, Frequency spectrum was filtered in low pass, high pass, and band pass varying cutoff frequency with 1/3 octave in frequency domain. The most effective cutoff frequency for each frequency has been determined as the range for 2 octave lower to 1 octave higher and 2 octave lower to 1 octave lower. This result provided a basis in improving data analysis capability using frequency domain filtering.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04b
• /
• pp.439-441
• /
• 2001

현재 사용되고 있는 전자악기는 한 옥타브에 대해 여러 개의 샘플 모듈(sample module)을 메모리에 저장해두면서, 옥타브내의 다른 음들은 그 샘플 모듈을 피지 시프팅(pitch shifting)하여 생성한다[1]. 따라서, 하나의 악기에 대해 많은 샘플 모듈을 사용하게 되고 많은 메모리를 필요로 한다. 본 논문에서는 보다 적은 샘플 모듈을 사용하여 메모리를 절약할 수 있는 방법을 연구하였다. 피치 시프팅의 범위를 하나의 옥타브 이내로 제한하지 않고, 피치 시프팅에 의해 발생하는 원음과 오차 평균을 줄이면서 가장 적절한 샘플 모듈음을 찾아낸다. 본 논문에서는 전자악기의 악기음 중에서 피아노 음을 선택했으며, 피아노의 88개 음들 중에서 피치 시프팅을 했을 때 원음과 가장 가까운 음을 만들어내는 음들을 샘플 모듈로 사용한다. 이러한 방법으로 샘플 모듈음들을 선택하면 기존의 전자악기에서 사용하는 샘플 모듈 개수보다 훨씬 적으면서도 동일한 음질을 보장하고, 또한 메모리 공간도 절약할 수 있다.

• Proceedings of the KIEE Conference
• /
• 2008.07a
• /
• pp.1841-1842
• /
• 2008

본 논문은 Scale Invariant Feature Transform(SIFT)알고리즘으로부터 얻어진 로컬 특징점으로부터 물체를 인식하는 방법에 대하여 논하였다. SIFT알고리즘은 물체의 스케일, 회전에 강인하고, 또한 3차원 시점의 변화에도 부분적으로 강인한 특징점을 추출한다. SIFT 알고리즘은 입력영상에 크기가 다른 가우시안 함수를 적용하고, 블러링된 영상들의 차 영상에서 극값을 추출하여 특징점으로 사용한다. 하지만 SIFT알고리즘에서 가우시안 함수를 적용하는 것은 상당히 많은 연산을 필요로 하기 때문에 본 논문에서는 하나의 옥타브를 사용하여 연산시간을 단축하였다. 하나의 옥타브를 사용함으로써 물체의 스케일이 크게 변하였을 때는 문제가 발생한다. 이를 해결하기 위하여 대상 물체의 작은 스케일, 큰 스케일에서 추출된 특징점을 혼합하여 DB를 생성하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2014.01a
• /
• pp.425-426
• /
• 2014

충격소음(Impulse Noise) 측정 시 지그(Ear Simulator)가 자체적으로 음압측정에 미치는 영향을 확인하기 위해 지그를 이용하여 실내에서 충격소음을 측정하였다. 측정의 신뢰성과 정확성을 확보하기 위해 마이크로폰과 지그에서 동일조건에서 동시에 데이터를 수집하였다. 추가로 헤드&토르소에서 실험을 실시하였다. 최대음압레벨과 1/3옥타브 밴드 결과를 비교분석함으로써 측정 방법에 따른 차이점을 발견하였다.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.3
• /
• pp.98-103
• /
• 2001

The binaural auditory system of human has the capability of differentiating the direction and distance of sound sources. This feature is well characterised in terms of the inter-aural intensity difference (IID), the inter-aural time difference (ITD) and/or the spectral shape difference (SSD) arising from the acoustic transfer of a sound source to the outer ears. This paper proposes an effective way of extracting the three sound perception factors (IID, ITD, SSD) from the head-related transfer functions (HRTF's) that depends on the direction and distance of the acoustic source from the listener. It includes the estimation method of the equivalent ITD and 1/3-octave band-based IID factors and their usage to locate a sound source in space. Subjective and objective tests were carried out to examine the effectiveness of the proposed methodology and its applicability to real sound systems. Those experimental results are illustrated in this paper.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.49 no.4
• /
• pp.111-119
• /
• 2012

Time Stretched Pulse (TSP) is used for transmitting and analyzing the impulse signal over the designated spatial place. However, if transfer functions of transmitter and receiver are unknown, performance investigation of free field in temporal domain is barely possible due to the overlap between the direct and indirect signal from the space. Generally, the free field or hemi-free field is evaluated by the Annex A of ISO 3745 in which utilizing the inverse square law with one-third octave band signals. In this paper, the author performs analysis of free field via applying TSP with inverse square law and the results are compared with the one-third octave band signals. According to the analysis of deviation between the corresponding signal and inverse square law model, the proposed TSP method provides the comparable performance index to the one-third octave band signal with reduced measuring time. Provided that the pre-whitening can be implementable by employing the speaker and microphone transfer function, further analyses from TSP compression are able to be performed such as multipath separation from time domain data. The anechoic chamber used in this experiment is verified conformance with ISO 3745 for free field and hemi-free field condition for limited frequency of the signal.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2010.10a
• /
• pp.95-96
• /
• 2010

• Proceedings of the IEEK Conference
• /
• 2003.07a
• /
• pp.266-269
• /
• 2003

마이크로파 대역에서 넓은 공진 주파수 대역폭을 갖는 V 형 테이퍼드 슬롯 안테나와 1x4 선형 배열 안테나를 설계하였다. 설계주파수는 5.8 [GHz]이었고 비유 전율이 3.48 인 RO4350 기판을 사용하였으며 멀티옥타브 특성과 복사 패턴을 얻었다.