• 한국정보통신학회논문지
• /
• 제23권12호
• /
• pp.1506-1513
• /
• 2019

최근 드론과 같은 멀티로터 UAV(Unmanned Aerial Vehicle, 무인항공기)의 산업 범위가 크게 확대됨에 따라, UAV를 활용한 데이터의 수집 및 처리, 분석에 대한 요구도 함께 증가하고 있다. 그러나 UAV를 이용해서 수집된 음향 데이터는 UAV의 모터 소음과 바람 소리 등으로 크게 손상되어, 음향 데이터의 처리 및 분석이 어렵다는 단점이 있다. 따라서 본 논문에서는 UAV에 연결된 마이크를 통해 수신된 음향 신호로부터 목표 음향 신호의 품질을 향상시킬 수 있는 방법에 대해 연구하였다. 본 논문에서는 기존의 단일 채널 음향 향상 기술 중 하나인 densely connected dilated convolutional network를 음향 신호의 채널 간 특성을 반영할 수 있도록 확장하였으며, 그 결과 SDR, PESQ, STOI과 같은 평가 지표에서 기존 연구 대비 좋은 성능을 보였다.

• 대한임베디드공학회논문지
• /
• 제10권4호
• /
• pp.189-198
• /
• 2015

Most IoT-related approaches have tried to establish the relation by connecting the network between things. The proposed research will present how the pervasive interaction of eco-system formed by touching the objects between humans and things can be recognized on purpose. By collecting and sharing the detected patterns among all kinds of things, we can construct the environment which enables individualized interactions of different objects. To perform the aforementioned, we are going to utilize technical procedures such as event-driven signal processing, pattern matching for signal recognition, and hardware in the loop simulation. We will also aim to implement the prototype of sensor processor based on Arduino MCU, which can be integrated with system using Arduino-Matlab/Simulink hybrid-interoperation environment. In the experiment, we use piezo transducer to detect the vibration or vibrates the surface using acoustic wave, which has specific frequency spectrum and individualized signal shape in terms of time axis. The signal distortion in time and frequency domain is recorded into memory tracer within sensor processor to extract the meaningful pattern by comparing the stored with lookup table(LUT). In this paper, we will contribute the initial prototypes for the acoustic touch processor by using off-the-shelf MCU and the integrated framework based on Matlab/Simulink model to provide the individualization of the touch-sensing for the user on purpose.

• 한국음향학회지
• /
• 제41권2호
• /
• pp.235-241
• /
• 2022

수동 음향 관측을 통해 수집된 방대한 양의 데이터에서 해양포유류의 소리를 탐지하고 식별하기 위해 합성곱 신경망(Convolutional Neural Network, CNN)을 활용한 연구가 많이 수행되고 있다. 본 연구는 2017년 8월부터 2018년 8월까지 동시베리아 해에서 수집된 수중음향 스펙트럼 이미지를 기반으로 CNN을 활용하여 턱수염물범 소리의 분류 자동화 가능성을 확인해 보았다. 학습 데이터로서 다른 소음이 거의 포함되지 않은 뚜렷한 턱수염물범 소리를 사용하였을 때, 암기로 인한 과적합이 발생하였다. 일부 데이터를 소음이 포함된 데이터로 교체하여 학습시켜 수집된 전체 데이터로 평가한 결과 정확도(0.9743), 정밀도(0.9783), 재현율(0.9520)으로 모델이 이전보다 일반화되어 과적합이 방지되는 것을 확인하였다. 본 연구를 통해 물범신호 분류는 학습 데이터에 소음이 포함되었을 때 성능이 증가하는 것으로 나타났다.

• ETRI Journal
• /
• 제33권3호
• /
• pp.423-433
• /
• 2011

This paper introduces an adaptive and integrated utterance verification (UV) framework using minimum verification error (MVE) training as a new set of solutions suitable for real applications. UV is traditionally considered an add-on procedure to automatic speech recognition (ASR) and thus treated separately from the ASR system model design. This traditional two-stage approach often fails to cope with a wide range of variations, such as a new speaker or a new environment which is not matched with the original speaker population or the original acoustic environment that the ASR system is trained on. In this paper, we propose an integrated solution to enhance the overall UV system performance in such real applications. The integration is accomplished by adapting and merging the target model for UV with the acoustic model for ASR based on the common MVE principle at each iteration in the recognition stage. The proposed iterative procedure for UV model adaptation also involves revision of the data segmentation and the decoded hypotheses. Under this new framework, remarkable enhancement in not only recognition performance, but also verification performance has been obtained.

• 한국지진공학회:학술대회논문집
• /
• 한국지진공학회 2000년도 춘계 학술발표회 논문집 Proceedings of EESK Conference-Spring
• /
• pp.63-69
• /
• 2000

Korea Institute of Geology Mining and Materials(KIGAM) cooperating with Southern Methodist University(SMU) has been operating seismo-acoustic array in Chul-Won area to discriminate man-made explosions from natural earthquakes since at the end of July 1999. In order to characterize propagation parameters of detected seismo-acoustic signal and to associate these signals as a blast event accompanying seismic and acoustic signals simultaneously it is necessary to understand infrasound wave propagation in the atmosphere. Two comparable Effective Sound Velocity Structures(ESVS) in atmosphere were constructed by using empirical model (MSISE90 and HWM93) and by aerological observation data of Korea Meteorological Administration (KMA) at O-San area. Infrasound propagation path computed by empirical model resulted in rare arival of refracted waves on ground less than 200km from source region. On the other hand Propagation paths by KMA more realistic data had various arrivals at near source region and well agreement with analyzed seismo-acoustic signals from Chul-Won data. And infrasound propagation in specific direction was very influenced by horizontal wind component in that direction. Linear travel time curve drawn up by 9 days data of the KMA in autumn season showed 335.6m/s apparent sound velocity in near source region. The propagation characteristics will be used to associate seismo-acoustic signals and to calculate propagation parameters of infrasound wave front.

• ETRI Journal
• /
• 제44권3호
• /
• pp.462-475
• /
• 2022

Although researchers have proposed numerous techniques for speech emotion recognition, its performance remains unsatisfactory in many application scenarios. In this study, we propose a speech emotion recognition model based on a genetic algorithm (GA)-decision tree (DT) fusion of deep and acoustic features. To more comprehensively express speech emotional information, first, frame-level deep and acoustic features are extracted from a speech signal. Next, five kinds of statistic variables of these features are calculated to obtain utterance-level features. The Fisher feature selection criterion is employed to select high-performance features, removing redundant information. In the feature fusion stage, the GA is is used to adaptively search for the best feature fusion weight. Finally, using the fused feature, the proposed speech emotion recognition model based on a DT support vector machine model is realized. Experimental results on the Berlin speech emotion database and the Chinese emotion speech database indicate that the proposed model outperforms an average weight fusion method.

• Nuclear Engineering and Technology
• /
• 제53권1호
• /
• pp.148-163
• /
• 2021

Timely fault identification is important for safe and reliable operation of the electric valve system. Many research works have utilized different data-driven approach for fault diagnosis in complex systems. However, they do not consider specific characteristics of critical control components such as electric valves. This work presents an integrated shallow-deep fault diagnostic model, developed based on signals extracted from DN50 electric valve. First, the local optimal issue of particle swarm optimization algorithm is solved by optimizing the weight search capability, the particle speed, and position update strategy. Then, to develop a shallow diagnostic model, the modified particle swarm algorithm is combined with support vector machine to form a hybrid improved particle swarm-support vector machine (IPs-SVM). To decouple the influence of the background noise, the wavelet packet transform method is used to reconstruct the vibration signal. Thereafter, the IPs-SVM is used to classify phase imbalance and damaged valve faults, and the performance was evaluated against other models developed using the conventional SVM and particle swarm optimized SVM. Secondly, three different deep belief network (DBN) models are developed, using different acoustic signal structures: raw signal, wavelet transformed signal and time-series (sequential) signal. The models are developed to estimate internal leakage sizes in the electric valve. The predictive performance of the DBN and the evaluation results of the proposed IPs-SVM are also presented in this paper.

• Geomechanics and Engineering
• /
• 제12권4호
• /
• pp.627-637
• /
• 2017

The high positive correlation between plastic strain of loaded coal-rock and AE (acoustic emission) characteristic parameter was studied and proved through AE experiment during coal-rock uniaxial compression process. The results show that plastic strain in the whole process of uniaxial compression can be gained through the experiment. Moreover, coal-rock loaded process can be divided into four phases through analyzing the change of the plastic strain curve : pressure consolidation phase, apparent linear elastic phase, accelerated deformation phase, rupture and development phase, which corresponds to conventional elastic-plastic change law of loaded coal-rock. The theoretical curve of damage constitutive model is in high agreement with the experimental curve. So the damage evolution law of coal rock damage can be indicated by both acoustic emission and plastic strain. The results have great academic and realistic significance for further study of both AE signal characteristics during loaded coal-rock damaged process and the forecasting of coal-rock dynamic disasters.

• 유체기계공업학회:학술대회논문집
• /
• 유체기계공업학회 1999년도 유체기계 연구개발 발표회 논문집
• /
• pp.208-217
• /
• 1999

The present work describes the prediction method for the unsteady flow field and the acoustic pressure field of a ducted axial fan. The prediction method is comprised of time-marching free-wake method, acoustic analogy, and the Helmholtz-Kirchhoff BEM. The predicted sound signal of a rotor is similar to the experiment one. We assume that the rotor rotates with a constant angular velocity and the flow field around the rotor is incompressible and inviscid. Then, a time-marching free-wake method is used to model the fan and to calculate the flow field. The force of each element on the blade is calculated by the unsteady Bernoulli equation. Lowson's method is used to predict the acoustic source. The newly developed Helmholtz-Kirchhoff BEM for thin body is used to calculate the sound field of the ducted fan. The ducted fan with 6 blades is analysed and the sound field around the duct is calculated.

• 한국컴퓨터정보학회논문지
• /
• 제24권9호
• /
• pp.21-27
• /
• 2019

In this paper, we propose a deep auto-encoder-based pipe leak detection (PLD) technique from time-series acoustic data collected by microphone sensor nodes. The key idea of the proposed technique is to learn representative features of the leak-free state using leak-free time-series acoustic data and the deep auto-encoder. The proposed technique can be used to create a PLD model that detects leaks in the pipeline in an unsupervised learning manner. This means that we only use leak-free data without labeling while training the deep auto-encoder. In addition, when compared to the previous supervised learning-based PLD method that uses image features, this technique does not require complex preprocessing of time-series acoustic data owing to the unsupervised feature extraction scheme. The experimental results show that the proposed PLD method using the deep auto-encoder can provide reliable PLD accuracy even considering unsupervised learning-based feature extraction.