• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2001년도 춘계학술대회논문집
• /
• pp.380-384
• /
• 2001

The household reftigerator's noise which is closely related with resident environment can hardly be evaluated its sound quality using the Korean Standards. Radiation characteristics of compressor noise consist of tonal noise in low frequency range and, or narrow band noise in high frequency range. In this study, measuring method for detecting the abnormal and low-level noise in high frequency band is presented, and control method for its reduction is proposed. After installing wall which is similar to living condition, we determined a major concerning frequency band of noise. It can be found the directivity of high frequency noise radiated from compressor by using experiment and analysis. According to isolation of noise transfer path, remarkable noise reduction is achieved.

• 한국정밀공학회지
• /
• 제20권3호
• /
• pp.82-88
• /
• 2003

The nondestructive internal quality evaluation of agricultural products has been strongly required from the needs for individual inspection. Recently, the ultrasonic wave has been considered as a solution fur this problem, and an ultrasonic system was constructed for the ultrasonic NDE of fruits and vegetables in our previous work. In this paper, the practical applicability of our ultrasonic system is tested fur the inspection of internal defects (central cavity) in Atlantic potato. Sound speed and RMS of transmitted ultrasonic wave signal were measured and classification algorithm using 2 dimensional stochastic analysis. was presented. Experimental results showed greater value of sound speed and RMS (root mean square) of transmitted signal in normal samples than in abnormal samples with cavity. Also a stochastic method to distinguish normal and abnormal showed fault detection rate less than 5%.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 ITC-CSCC -2
• /
• pp.1045-1048
• /
• 2000

This paper presents the application of the wavelet transform analysis and the neural network method to the phonocardiogram (PCG) signal. Heart sound is a acoustic signal generated by cardiac valves, myocardium and blood flow and is a very complex and nonstationary signal composed of many source. Heart sound can be discriminated normal heart sound and heart murmur. Murmurs have broader frequency bandwidth than the normal ones and can occur at random position of cardiac cycle. In this paper, we classified the group of heart sound as normal heart sound(NO), pre-systolic murmur(PS), early systolic murmur(ES), late systolic murmur(LS), early diastolic murmur(ED). And we used the wavelet transform to shorten artifacts and strengthen the low level signal. The ANN system was trained and tested with the back- propagation algorithm from a large data set of examples-normal and abnormal signals classified by expert. The best ANN configuration occurred with 15 hidden layer neurons. We can get the accuracy of 85.6% by using the proposed algorithm.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2007년도 추계학술대회논문집
• /
• pp.318-321
• /
• 2007

Since human listening is very sensitive to sound, a subjective index of a sound quality is required. Therefore, in the analysis for each situation, the sound evaluation is composed with sound quality factor. Many researchers spends their effort to make a more reliable and more accurate of sound in term of sound quality index for various system noise. The previous methods to evaluation of the SQ about vehicle interior noise are linear regression analysis of subjective SQ metrics by statistics and the estimation of the subjective SQ values by neural network. But these are so depended on jury test very much that they result in many difficulties. So, to reduce jury test weight, we suggested a new method using Mahalanobis distance for SQ evaluation. Threrefore, in this study Mahalanobis distance for the vehicle interior noise was derived using the objective SQ except jury test. Finnaly, the results of the SQ evaluation was analyzed discrimination between reference and abnormal group.

• 한국소음진동공학회논문집
• /
• 제18권1호
• /
• pp.57-60
• /
• 2008

Since human listening is very sensitive to sound, for evaluating of a sound quality is required. Therefore, in the analysis for each situation, the sound evaluation is composed with sound quality factor. My researchers spends their effort to make a more reliable and more accurate of sound in term of sound quality index for various system noise. The previous methods to evaluation of the SQ about vehicle interior noise are linear regression analysis of subjective SQ metrics by statistics and the estimation of the subjective SQ values by neural network. But these are highly dependent on jury test and have many difficulties due to various environmental factors. So, to reduce jury test weight. we suggested a new method using Mahalanobis distance for SQ evaluation. Threrefore, in this study Mahalanobis distance for the vehicle interior noise was derived using the objective SQ except jury test. Finnaly, the results of the SQ evaluation was analyzed discrimination between reference and abnormal group.

• 전자공학회논문지
• /
• 제51권6호
• /
• pp.244-251
• /
• 2014

본 논문에서, 가스절연개폐장치(GIS)의 정상상태와 비정상상태에 대해 예방진단기술을 이용하여 자세하게 서술하였다. 이 기술은 지능형 GIS 감시시스템에 의해 저장된 GIS의 데이터의 분석과 진단에 근거한다. GIS음의 파형은 방전과 자체내의 코로나 방전음에 의해 발생되는 것으로 잡음과 비슷하다. 그러므로, 본 논문에서, GIS음의 정상 상태와 비정상 상태로 분류하는데, 정상과 비정상 상태를 레벨교차율(LCR)과 스펙트로그램 에너지비율로 이용하여 구분하였다.

• 한국컴퓨터정보학회논문지
• /
• 제26권11호
• /
• pp.21-31
• /
• 2021

최근 코로나(Covid-19)의 영향으로 스마트 헬스케어 관련 산업과 비대면 방식의 원격 진단을 통한 질환 분류 예측 연구의 필요성이 증가하고 있다. 일반적으로 호흡기 질환의 진단은 비용이 많이 들고 숙련된 의료 전문가를 필요로 하여 현실적으로 조기 진단 및 모니터링에 한계가 있다. 따라서, 간단하고 편리한 청진기로부터 수집된 호흡음을 딥러닝 기반 모델을 활용하여 높은 정확도로 분류하고 조기 진단이 필요하다. 본 연구에서는 청진을 통해 수집된 폐음 데이터를 이용하여 이상 호흡음 분류모델을 제안한다. 데이터 전처리로는 대역통과필터(BandPassFilter)방법론을 적용하고 로그 멜 스펙트로그램(Log-Mel Spectrogram)과 Mel Frequency Cepstral Coefficient(MFCC)을 이용하여 폐음의 특징적인 정보를 추출하였다. 추출된 폐음의 특징에 대해서 효과적으로 분류할 수 있는 병렬 합성곱 신경망 네트워크(Parallel CNN network)모델을 제안하고 다양한 머신러닝 분류기(Classifiers)와 결합한 스태킹 앙상블(Stacking Ensemble) 방법론을 이용하여 이상 호흡음을 높은 정확도로 분류하였다. 본 논문에서 제안한 방법은 96.9%의 정확도로 이상 호흡음을 분류하였으며, 기본모델의 결과 대비 정확도가 약 6.1% 향상되었다.

• 한국전자통신학회논문지
• /
• 제11권8호
• /
• pp.759-766
• /
• 2016

본 논문은 비정상 상황 시 발생하는 음원에 대해 주변 환경 음에 강인한 음원 구간을 검출하여, 구간내의 신호를 이용한 음원 인식 과 위치 추적 시스템 설계에 관한 연구이다. 강인한 음원 구간 검출은 수신되는 오디오 신호로부터 단 구간 가중 평균 델타 에너지를 계산하여, 저역 통과 필터에 입력 후, 출력되는 결과 값들의 비교를 통해 배경음에 강인한 구간을 정의 하며, 음원 인식은 검출된 구간 내 데이터로부터 종래의 인식 방법인 HMM(: Hidden Markov Model)을 이용해, 음원 인식 정보를 생성하여 학습 및 인식을 한다. 이는 주변 배경음이 포함된 음원 신호에 대해 기존 신호의 에너지를 이용해 구간을 검출 후, HMM을 통한 인식에 비해 3.94% 상향된 인식률을 보인다. 또한 인식 결과를 바탕으로 구간내의 신호간의 TDOA(: Time Delay of Arrival)를 이용한 위치 파악은 실제 발생 위치와의 각도와 97.44%일치함을 보인다.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2008년도 춘계학술대회논문집
• /
• pp.863-868
• /
• 2008

This paper investigates the abnormal noise of polygon mirror scanner motor due to the mechanical contact. In the high speed polygon mirror scanner motor the vibration of polygon mirror scanner motor is one of the main sources of abnormal noise, because structure-borne noise due to the vibration is bigger than aerodynamic noise, especially when the rotating part contacts the stationary parts. This research determines the main harmonics of structure-borne noise by using sound quality evaluation. It also develops an experimental set-up to measure the mechanical contact and vibration of polygon mirror scanner motor simultaneously. This paper also show that mechanical contact between rotating shaft and stationary sleeve is one of the dominant vibration sources of structure-borne noise which cause the abnormal noise of the high speed polygon mirror scanner motor by using the developed experimental set-up.

• 제어로봇시스템학회논문지
• /
• 제20권1호
• /
• pp.48-57
• /
• 2014

It is suggested that the curvature trajectory be used to estimate the real-time moving sound sources and efficiently the robot estimating the sound sources. Since the target points of the real-time moving sound sources change, the mobile robot continuously estimates the changed target points. In such a case, the robot experiences a slip phenomenon due to the abnormal velocity and the changes of the navigating state. By selecting an appropriate curvature and navigating the robot gradually by using it, it is possible to enable the robot to reach the target points without having much trouble. In order to recognize the sound sources in real time, three microphones need to be organized in a straight form. Also, by applying the cross-correlation algorithm to the TDOA base, the signals can be analyzed. By using the analyzed data, the locations of the sound sources can be recognized. Based on such findings, the sound sources can be estimated. Even if the mobile robot is navigated by selecting the gradual curvature based on the changed target points, there could be errors caused by the inertia and the centrifugal force related to the velocity. As a result, it is possible to control the velocity of both wheels of the robot through the velocity PID controller in order to compensate for the slip phenomenon and minimize the estimated errors. In order to examine whether the suggested curvature trajectory is appropriate for estimating the sound sources, two mobile robots are arranged to carry out an actual experiment. The first robot is moved by discharging the sound sources, while the second robot recognizes and estimates the locations of the discharged sound sources in real time.