• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 1998.03a
• /
• pp.172-177
• /
• 1998

An experimental study on the prediction of grinding wheel wear by dual detection methods was conducted by the laser displacement and acoustic emission(AE) system. The laser displacement sensor was located above the head of the grinding wheel and the AE sensor was set under the workpiece, where the wheel were condition can be detected. It was found that the dual detection methods by laser displacement system and AE system made it possible to predict the wheel wear. From the experiments, the root mean square(RMS) values both methods was found to be proportional to the grinding wheel wear.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.1
• /
• pp.24-31
• /
• 2020

In this paper, the deep convolutional neural network for sound event detection is studied. Especially, the end-to-end neural network, which generates the detection results from the input audio waveform, is studied for weakly supervised problem that includes weakly-labeled and unlabeled dataset. The proposed system is based on the network structure that consists of deeply-stacked 1-dimensional convolutional neural networks, and enhanced by the skip connection and gating mechanism. Additionally, the proposed system is enhanced by the sound event detection and post processings, and the training step using the mean-teacher model is added to deal with the weakly supervised data. The proposed system was evaluated by the Detection and Classification of Acoustic Scenes and Events (DCASE) 2019 Task 4 dataset, and the result shows that the proposed system has F₁-scores of 54 % (segment-based) and 32 % (event-based).

• Journal of Broadcast Engineering
• /
• v.22 no.5
• /
• pp.632-642
• /
• 2017

In this paper, we propose a DNN based speech detection system using acoustic characteristics and context information of media audio. The speech detection for discriminating between speech and non-speech included in the media audio is a necessary preprocessing technique for effective speech processing. However, since the media audio signal includes various types of sound sources, it has been difficult to achieve high performance with the conventional signal processing techniques. The proposed method improves the speech detection performance by separating the harmonic and percussive components of the media audio and constructing the DNN input vector reflecting the acoustic characteristics and context information of the media audio. In order to verify the performance of the proposed system, a data set for speech detection was made using more than 20 hours of drama, and an 8-hour Hollywood movie data set, which was publicly available, was further acquired and used for experiments. In the experiment, it is shown that the proposed system provides better performance than the conventional method through the cross validation for two data sets.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.41 no.1
• /
• pp.54-63
• /
• 2005

Detection ranges of acoustic instruments mainly used for fisheries and their research are derived as the range bordered by a certain signal-to-noise ration (SNR) thershold. The SNR is depicted by several factors on transmitting and receiving, sound propagation, scattering by objects, and mainly self-ship noise. The detection ranges are shown for several fisheries instrument, such as echo sounder, quantiative echo sounder, and bio-telemetry system. The results can be used for designing the instruments, examining the capability of user's own instruments, and interpreting obtained data or echograms. Increasing transmitting power is not as effective for high frequencies as for low frequencies to increase the detection range. Comparison of volume backscattering strengths obtained by the quantitative echo sounder at several frequencies should be done within the same detection range. By applying the concept of the detection range for the bio-telemetry receiver beams, the number of the beams and the beamwidths can be determined.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.35 no.5
• /
• pp.307-313
• /
• 2015

Leak detection in a pipeline usually involves acoustic emission sensors such as contact type sensors. These contact type sensors pose difficulties for installation and cannot operate in areas having high temperature and radiation. Therefore, recently, many researchers have studied the leak detection phenomenon by using a camera. Leak detection by using a camera has the advantages of long distance monitoring and wide area surveillance. However, the conventional leak detection method by using difference images often mistakes the vibration of a structure for a leak. In this paper, we propose a method for steam leakage detection by using the moving average of difference images and histogram analysis. The proposed method can separate the leakage and the vibration of a structure. The working performance of the proposed method is verified by comparing with experimental results.

• Journal of the korean Society of Automotive Engineers
• /
• v.8 no.3
• /
• pp.28-36
• /
• 1986

This study was investigated the feasibility of AE application on in-process detection of tool fracture and chipping. Carbon steel SM45C workpiece with longitudinal slots was turned interruptedly on a lathe. AE RMS signal at tool fracture was observed and also the tangential force and the feed observed at the time of tool fracture, the levels of tangential force and the feed force at the time of fracture decrease considerably. In chipping, high level AE signal was observed but there were no changes of cutting force. Peak AE RMS squared is proportional to the area of tool fracture and resultant force. Fracture model of tool fracture is proposed as $V_{p}$ = $C_{1}$ $E_{1}$F(.DELTA. A)$_{0.5}$ and peak AE RMS shows strong correlation with the fracture parameter F(.DELTA.A)$^{0.5}$.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.3E
• /
• pp.79-87
• /
• 2009

This paper reviews several multi-source localization methods which estimate ITDs based on zero crossings (ZCs). Employing signal-to-noise ratio (SNR) estimation from ITD variances, these ZC-based source localization methods are more robust to diffuse noise than the cross-correlation (CC)-based one with less computational complexity. In order to take reverberant environments into account, two approaches detect intervals which dominantly contain direct-path components from sources to sensors because they may effectively provide reliable ITDDs corresponding to source directions. One accomplishes the detection by comparing the original and cepstral-prefiltering-processed envelopes, and the other searches sudden increase of acoustic energy by considering typical characteristics of acoustic reverberation. Experiments for comparison of these methods demonstrate that the approach with energy-based detection efficiently achieves multi-source localization in reverberant environments.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.27 no.3
• /
• pp.262-267
• /
• 2007

Detection of gas leak from a tube is a very important issue in the quality control of machines such as the heat exchanger of an air-conditioner, because leakage of operating gas directly reduces the performance of machines. The acoustic emission (AE) method is a common way to detect leak of gas, however its application under the environment of mechanical vibration is restricted since most AE detectors are very sensitive to external vibration noise. In order to overcome this problem, we propose a method based on the mode analysis of the Lamb wave. In this method, the dominant Lamb mode and its frequency are found first, and then a proper band-pass filter is used to retain only this frequency component. In this way, we could improve the SNR (signal-to-noise ratio) of AE signal generated by gas leak from the tube even under vibratory environment.

• Journal of KSNVE
• /
• v.7 no.6
• /
• pp.931-936
• /
• 1997

Optical fiber sensor is a subject which has been attracted considerable attention in recent years. Detection of sound pressure with optical fibers positioned in the arms of a Mach-Zehnder interferometer is presented in this paper. A fiber length of the order of 150m is wounded is made by hollow cylinder type. To increase the sound signal 3${\times}3$ directional coupler is used. Fiber optic hydrophone is the underwater tank with 2kHz sound source. Finally, it is shown that the fiber optic hydeophone can stably detec 2kHz sound.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 1998.04a
• /
• pp.617-621
• /
• 1998

Optical sensing techniques have been associated with high sensitivity and precise measurements and attracted considerable attention in recent years. In this paper, two channels TDM(Time Division Multiplexing) fiber-optic hydrophone array for the underwater applications was fabricated and their acoustic characteristics were investigated by using the acoustic water tank. A fiber length of the order of 100m is wounded at the hollow cylinder type aluminum mandrel. An unbalanced interferometer (discrete Mach-Zehnder type) was used. Sound detection performance is tested in the underwater tank with 2kHz continuous sound source. Finally, it is shown that two channels TDM(Time Division Multiplexing) fiber-optic hydrophone array can detect 2kHz sound stably.