• 한국산업정보학회논문지
• /
• 제29권2호
• /
• pp.15-25
• /
• 2024

효과적인 동물 생태계 분석을 위해서는 동물 서식 현황을 자동으로 파악할 수 있는 동물 관제 기술이 중요하다. 특히 울음소리로 종을 판별하는 동물 소리 분류 기술은 영상을 통한 판별이 어려운 환경에서 큰 주목을 받고 있다. 기존 연구들은 단일 딥러닝 모델을 사용하여 동물 소리를 분류하였으나, 야외 환경에서 수집된 동물 소리는 많은 배경 잡음을 포함하여 단일 모델의 판별력을 악화시키며, 종에 따른 데이터 불균형으로 인해 모델의 편향된 학습을 야기한다. 이에, 본 논문에서는 클래스의 데이터 수를 고려하여 페널티를 부여하는 Focal Loss를 사용한 여러 분류 모델의 예측결과를 앙상블을 통해 결합하여 잡음이 많은 동물 소리를 효과적으로 분류할 수 있는 기법을 제안한다. 공개 데이터 셋을 사용한 실험에서, 제안된 기법은 단일 모델의 평균 성능에 비해 Recall 기준으로 최대 22.6%의 성능 개선을 달성하였다.

• 한국음향학회지
• /
• 제25권3호
• /
• pp.144-150
• /
• 2006

심장병이 있는 환자들을 진료할 때 의사들은 청진기를 이용하여 심음 (heart sound)을 듣고 이를 기준으로 환자의 병의 유무나 질환의 종류에 대한 기초적인 판단을 하게 된다. 하지만, 심음은 환자의 상태나 외부 잡음의 영향에 따라서 신호의 특성이 변하고 또한 정상적인 심음과 질병을 나타내는 심음과의 차이가 비교적 구분하기 어려울 정도로 작기 때문에 숙달된 전문의가 아니면, 진단의 정확도가 떨어질 가능성이 있다. 따라서 신호처리 기법을 이용하여 심음을 분석해서 심음이 정상적인지의 유무를 자동으로 판단할 수 있다면, 진단을 하는 의사들에게 유용한 정보가 될 것이라 생각된다. 본 연구에서는 심음의 질병유무와 질병종류를 자동으로 판단하기 위해서 기존에 많이 사용되었던 artificial neural network (ANN) 대신에 hidden Markov model (HMM)을 사용하는 방법을 제안하였으며, 기초적인 실험결과 상당히 우수한 성능을 보임을 알 수 있었다.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2001년도 춘계학술대회논문집
• /
• pp.1003-1008
• /
• 2001

In beamforming method, source positions are predicted by MUSIC (Multiple Signal Classification) power method and composite sound fields can then be decomposed into each partial field by beamforming, detenninistically without restriction of the distance between reference microphones and sources. However, reference microphone array shape is important in both MUSIC and beamforming method. Thus the present paper describes the effect of the reference microphone array shape.

• 한국항행학회논문지
• /
• 제24권5호
• /
• pp.393-407
• /
• 2020

불법 무인기에 의한 위협을 줄이기 위해, 음향 기반 기법에 의한 추적시스템을 구현하였다. 드론 음향 추적 방식에는 3가지 주요 사항이 있다. 첫째, 가변 빔 형성을 통해 공간을 스캔하여 음원을 찾아 마이크 어레이를 사용하여 소리를 녹음한다. 둘째, 음원의 존재 유·무 여부를 알기 위해 은닉 마르코프 모델(HMM)로 분류한다. 마지막으로 음원이 드론인 경우, 적응형 빔 패턴을 기반의 추적기준 신호로 기록 및 저장된 음원을 사용한다. 시뮬레이션은 배경 노이즈 및 간섭 사운드가 없는 이상적인 상태와 배경 노이즈 및 간섭 사운드가 있는 비이상적인 조건 모두에서 수행되며 불법적인 드론의 추적 성능을 평가하였다. 드론 추적 시스템은 마이크 어레이 성능에 따른 탐색 거리 성능향상 및 음향 패턴 일치 정도에 따른 드론 유무 판정 기준을 설계하여 음성판독 회로설계에 반영하였다.

• 농촌계획
• /
• 제11권2호
• /
• pp.59-69
• /
• 2005

Because of increasing numbers of cars many highways are being constructed lively, and the noise of passing cars has influenced surrounding areas. In consideration of this, some alternatives and researches for soundproof facilities are proceeding, but aesthetic approach hasn't been considered. Therefore, this research is focused on soundproof effects for each types, effectual simulation methods, visual assessment and estimation between the landscape before simulation and the landscape after. Soundproof facilities are divided largely by the soundproof barrier, the soundproof mounding, the soundproof grove. The soundproof grove has three main function. First, leaves and branches absorbs sound vibrations. Second, leaves absorbs sound, and branches obstruct sounds. Third, by means of sounds of shaking leaves, forest can offset noises. This research was proceeded by means of classification of soundproof grove types and investigation of visual simulation methods. We made visual simulation for each types, and estimated the landscape for each types.

• 대한의용생체공학회:학술대회논문집
• /
• 대한의용생체공학회 1997년도 추계학술대회
• /
• pp.75-78
• /
• 1997

In this paper we present new sound spectral analysis methods or prosthetic heart valve sounds. Phonocardiograms(PCG) of prosthetic heart valve were analyzed in order to derive frequency domain feature suitable or the classification of the valve state. The fast orthogonal search method and MUSIC (MUltiple SIgnal Classification) method are described or finding the significant frequencies in PCG. The fast orthogonal search method is effective with short data records and cope with noisy, missing and unequally-spaced data. MUSIC method's key to the performance is the division of the information in the autocorrelation matrix or the data matrix into two vector subspaces, one a signal subspace and the other a noise subspace.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제7권3호
• /
• pp.91-98
• /
• 2018

집단으로 사육되는 돼지 농장에서 돼지 소모성 질환의 자동 탐지는 매우 중요한 문제이다. 특히, 밀집된 돈사에서 사육되는 돼지들의 호흡기 질환은 축산 농가의 막대한 경제적 손실을 야기하는 대표적 질병들 중 하나이다. 본 논문에서는 소리 신호 해석에 기반하여 돼지의 호흡기 질환을 조기 탐지 및 식별하는 잡음에도 강인한 시스템을 제안한다. 제안하는 시스템은, 먼저 1차원의 소리 신호를 2차원의 회색조 영상으로 변환한 후, DNS기법으로 질감 특징 정보를 갖는 이미지를 생성한다. 마지막으로, 이를 CNN에 입력함으로써 잡음에도 강인한 돼지 호흡기 질병 탐지 및 식별 시스템을 구현하고자 한다. 실제 국내 돈사에서 취득한 돼지의 발성음을 이용하여 제안하는 시스템의 성능을 실험적으로 검증한바, 제안된 시스템은 경제적인 비용(저가의 소리 센서)과 시스템 정확도(96.0% 정확도)로 다양한 잡음 환경에서도 돼지의 호흡기 질병들을 탐지할 수 있음을 실험적으로 확인하였다. 제안된 시스템은 독자적인 혹은 기존 방법들의 보완책으로 사용될 수 있다.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2004년도 추계학술대회논문집
• /
• pp.802-805
• /
• 2004

It is requures to understand the meaning of sound and cultural background of target spaces for soundscape design. The purpose of this study is to derive the general soundscape preserved in the mind of Korean people. Four kinds of traditional Korean songs, that is to say, Sijo (Korean verse), folk-song, Pansori (song for drama by Chang reciter), and popular songs, were used as tools of research subjects. That is because the sounds expressed in literature are very useful subjects for soundscape survey. Paragraphs containing description of sounds were selected from the lyrics of the songs. Then 11 sound categories were named, and 13 meaning categories were classified according to referential and emotional aspects, respectively.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2007년도 춘계학술대회논문집
• /
• pp.515-518
• /
• 2007

SAFRS(Spontaneous Acoustic Field Reproduction System) is a system to sense changes of surroundings and produce sounds which can go well with environment elements sensed by the system in to the space. The sounds were judged by individual evaluation and, the classification of the preferred sounds according to the mood of the space was suggested in the former study. Effectiveness of SAFRS with field application was validated by prior studies which dealt with researching acoustic environment, evaluating images of sounds, and rating environment with existence and nonexistence of sound resources such as fountains and the system after applied in D university. In this study, for more effective field application of SAFRS, research for the acoustic environment around sound resources and subjective evaluation of the preference of the sounds from the resources were made and it was considered that the results of the experiments should be primary information to propose proper sound level to be offered by the system. The results of the study are as follows; 1) It was considered that the ambience of the center road was dependent upon produced sounds by the system and water sounds of the fountain and that of walk way was mostly dependent upon produced sounds. 2) The results of the subjective evaluation showed that the distance from sound resources was suggestive; the more distant from produced sounds the less full and clear the sounds, the less distant from the sounds of water the more delight and idyllic ambience, and the less distant from the forest the more idyllic ambient and diversity. 3) The results upwards were telling that an average value of six elements for the evaluation was even at the place set back 10.2m from center road and walk way. And harmony of all sounds of the place should be considered to propose suitable sound level of SAFRS.

• 한국소음진동공학회논문집
• /
• 제17권6호
• /
• pp.547-552
• /
• 2007

SAFRS(spontaneous acoustic field reproduction system) is a system to sense changes of surroundings and produce sounds which can go well with environment elements sensed by the system in to the space. The sounds were judged by individual evaluation and, the classification of the preferred sounds according to the mood of the space was suggested in the former study. Effectiveness of SAFRS with field application was validated by prior studies which dealt with researching acoustic environment, evaluating images of sounds, and rating environment with existence and nonexistence of sound resources such as fountains and the system after applied in D university. In this study, for more effective field application of SAFRS, research for the acoustic environment around sound resources and subjective evaluation of the preference of the sounds from the resources were made and it was considered that the results of the experiments should be primary information to propose proper sound level to be offered by the system. The results of the study are as follows; 1) It was considered that the ambience of the center road was dependent upon produced sounds by the system and water sounds of the fountain and that of walk way was mostly dependent upon produced sounds. 2) The results of the subjective evaluation showed that the distance from sound resources was suggestive; the more distant from produced sounds the less full and clear the sounds, the less distant from the sounds of water the more delight and idyllic ambience, and the less distant from the forest the more idyllic ambient and diversity. 3) The results upwards were telling that an average value of six elements for the evaluation was even at the place set back 10.2m from center road and walk way. And harmony of all sounds of the place should be considered to propose suitable sound level of SAFRS.