• 방송공학회논문지
• /
• 제23권6호
• /
• pp.855-865
• /
• 2018

음향 이벤트 인식은 다수의 음향 이벤트가 발생하는 환경에서 이를 인식하고 각각의 발생과 소멸 시점을 판단하는 기술로써 인간의 청각적 인지 특성을 모델화하는 연구다. 음향 장면 및 이벤트 인식 연구 그룹인 DCASE는 연구자들의 참여 유도와 더불어 음향 인식 연구의 활성화를 위해 챌린지를 진행하고 있다. 그러나 DCASE 챌린지에서 제공하는 데이터 세트는 이미지 인식 분야의 대표적인 데이터 세트인 이미지넷에 비해 상대적으로 작은 규모이며, 이 외에 공개된 음향 데이터 세트는 많지 않아 알고리즘 개발에 어려움이 있다. 본 연구에서는 음향 이벤트 인식 기술 개발을 위해 실내외에서 발생할 수 있는 이벤트를 정의하고 수집을 진행하였으며, 보다 큰 규모의 데이터 세트를 확보하였다. 또한, 인식 성능 개선을 위해 음향 이벤트 존재 여부를 판단하는 보조 신경망을 추가한 이중 CNN 구조의 알고리즘을 개발하였고, 2016년과 2017년의 DCASE 챌린지 기준 시스템과 성능 비교 실험을 진행하였다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2004년도 하계학술대회 논문집 C
• /
• pp.1748-1750
• /
• 2004

This paper deals with the detection method of the Quench phenomenon for superconducting magnet using the Acoustic Emission (AE) sensor. AE sensor is the elements, which is used to change the Acoustic signal to the voltage value. This signal may be used to detect whether the superconducting magnet has been at the Quench state or not. Recently, the development of the Quench detection technique, which is the using voltage and current signals, fiber-optic sensor, and so on, for the superconducting applications is widely studying. This method for the Quench detection of the superconducting magnet is also studying at some kinds of institute in Japan and the united state. Because of the large-scale superconducting magnet like International Thermonuclear Experimental Reactor(ITER) is charged a lot of energy, when the Quench phenomenon is being at the superconducting magnet it is happen to the problem of the protection for the applications. In this paper, we concluded that the Quench detection was possible when the mechanical stress by means of the local heat is generated at the part of inside superconducting magnets.

• 대한전자공학회논문지SP
• /
• 제44권3호
• /
• pp.97-104
• /
• 2007

적응 반향제거기는 휴대전화나 음성 인식 시스템과 같은 음성 통신 시스템에서 중요한 부분의 하나로 자리잡았다. 이러한 응용에서 반향경로는 긴 임펄스 응답을 가지게 된다. 본 논문에서는 음향반향제거를 위해 Moving-Average Least Mean Square(MVLMS) 알고리즘을 제안하였다. 유색 입력 모델을 이용한 실험 결과는 MVLMS 검출 알고리즘이 Least Mean Square 검출 알고리즘에 비해 수렴 성능이 우위에 있음을 입증하였다. MVLMS 알고리즘은 약간의 계산 복잡도 향상이 있지만, 표준 LMS 검출 알고리즘에 비해 월등한 안정성 향상을 가져온다.

• 한국군사과학기술학회지
• /
• 제18권5호
• /
• pp.538-547
• /
• 2015

The goal of this study is to develop an algorithm to propose optimal deployment of detection sensor nodes in the target area, based on a performance surface, which represents detection performance of active and passive acoustic sonar systems. The performance surface of the active detection system is calculated from the azimuthal average of maximum detection ranges, which is estimated with a transmission loss and a reverberation level predicted using ray-based theories. The performance surface of the passive system is calculated using the transmission loss model based on a parabolic equation. The optimization of deployment configurations is then performed by a hybrid method of a virtual force algorithm and a particle swarm optimization. Finally, the effectiveness of deployment configurations is analyzed and discussed with the simulation results obtained using the algorithm proposed in this paper.

• 한국멀티미디어학회논문지
• /
• 제19권3호
• /
• pp.557-566
• /
• 2016

Heart sounds generated by the beating heart and blood flow reflect the turbulence created when the heart valves snap shut. Cardiac diagnosis is typically started by an auscultation using a stethoscope, from which a medical doctor, depending on his hearing capabilities and training, listens and interprets the acoustic signal. This method of diagnostic is uncertain, mostly due to the fact that human ear loses the acoustic frequency sensitivity through the years. Even though an auscultation has some weaknesses like uncertainty, it is considered as a primary tool due to its simplicity. In this paper, heart murmur detection algorithm is proposed using time and frequency characteristics of heart sound. The propose heart murmur detection method adapted conventional primary heart sound detection method in time domain and modified spectral flatness method in frequency domain for detecting heart murmurs. From experimental results, it is confirmed that the proposed algorithm detect the heart murmurs efficiently.

• 대한전자공학회논문지SP
• /
• 제46권5호
• /
• pp.162-166
• /
• 2009

본 논문은 적응 반향 제거기인 parametric wiener filter (PWF)의 반향신호 제거 파라미터를 구간 조건에 따라 차등적으로 적용시키는 새로운 음향학적 반향제거 (Acoustic Echo Suppression, AES) 알고리즘을 제안한다. PWF는 반향 신호 추정의 부정확성을 보상하고자 반향신호 제거 파라미터를 사용한다. 기존의 방법은 단일통화 구간과 동시통화 구간에 관계없이 동일한 반향신호 제거파라미터 값을 적용하여 동시통화 환경에서 근단 신호의 왜곡을 발생 시킨다. 본 논문에서는 상관계수 (Correlation Coefficient)를 이용한 동시통화(Double-Talk) 검출 알고리즘과 음성구간검출기 (Voice Activity Detector, VAD)의 결과로부터, 원단 화자만 있는 구간과 동시통화 구간을 구분하고 각각의 경우에 반향제거 파라미터를 달리 적용하여 동일한 반향신호 제거 파라미터를 사용함에 따른 동시통화 구간내의 근단 화자신호의 음질 저하 없이 반향신호를 보다 효과적으로 제거하는 방법을 제안한다. 제안된 방법의 결과는 객관적인 실험을 통해 비교 평가한 결과 우수한 성능을 보였다.

• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 2009년 춘계학술대회논문집
• /
• pp.70-74
• /
• 2009

The helicopter development technology is improved by the increasing of computing power and advanced test facilities. The increasing efficiency of fuel consuming by the developing of improved lift-to-drag rotor system is the major issue, the noise reduction for ecology(civil area) and increase of survivability to reduce noise detection(battlefield) also are important. This investigation shows the helicopter external noise flight test methods, conventional military helicopter's(UH-60) experimental results and the numerical modeling method for aero-acoustic of rotor blade and the result of CAA(Computational Aero-Acoustic) for main rotor blade.

• Journal of Biosystems Engineering
• /
• 제23권6호
• /
• pp.621-628
• /
• 1998

In order to develop an inspection algorithm for an automatic eggshell inspection system, acoustic impulse response with neural network method was studied. An improved error backpropagation algorithm was selected as a loaming rule of neural network, and three layer network was chosen for the neural network architecture. Acoustic signals in time domain and theirs power spectrum were studied as the input to the neural network. The classification feasibility and success rate were investigated in terms of statistical analysis and neural network approach. As a result, the success rate was 95% with the statistical model having five independent variables. Among the neural network models studied, the power spectrum of acoustic signal as the input with 64 input neurons and the two impact data showed the success rate of 95.5% which was slightly higher than of statistical analysis.

• Structural Engineering and Mechanics
• /
• 제54권6호
• /
• pp.1075-1095
• /
• 2015

The use of acoustic emission (AE) technique for detecting and monitoring damages and the progress on damages in different structures is widely used and has earned a reputation as one of the most reliable and well-established technique in non-destructive testing (NDT). Acoustic Emission is a very efficient and effective technology used for fracture behavior and fatigue detection in metals, fiberglass, wood, composites, ceramics, concrete and plastics. It can also be used for detecting faults and pressure leaks in vessels, tanks, pipes, as well as for monitoring the progression of corrosion in welding. This paper reviews major research developments over the past few years in application of acoustic emission in numerous engineering fields, including manufacturing, civil, aerospace and material engineering.

• 한국안전학회지
• /
• 제18권2호
• /
• pp.6-15
• /
• 2003

A method is presented for diagnosing the converter bearings by using acoustic emission. The flaking mechanism causing the large-scale bearing for furnace to flaw is investigated and a possibility of defect is verified by Finite Element method. he diagnosis logic is proposed fir detecting the flaw of a non-continuous rotating machine. It is proved that the acoustic emission energy can be used as a representative parameter for an acoustic event. Applying the method to the tilting bearings for steel mill in operation, the effectiveness of this logic is evaluated. It is shown that AE signal is generated only when the bearing is tilting, and the trend analysis can be focused upon this process.