• Smart Structures and Systems
• /
• 제29권6호
• /
• pp.777-783
• /
• 2022

Even though several deep learning-based methods have been applied in the field of acoustic source localization, the previous works have only been conducted using the two-dimensional representation of the beamforming maps, particularly with the planar array system. While the acoustic sources are more required to be localized in a spherical microphone array system considering that we live and hear in the 3D world, the conventional 2D equirectangular map of the spherical beamforming map is highly vulnerable to the distortion that occurs when the 3D map is projected to the 2D space. In this study, a 3D deep learning approach is proposed to fulfill accurate source localization via distortion-free 3D representation. A target function is first proposed to obtain 3D source distribution maps that can represent multiple sources' positional and strength information. While the proposed target map expands the source localization task into a point-wise prediction task, a PointNet-based deep neural network is developed to precisely estimate the multiple sources' positions and strength information. While the proposed model's localization performance is evaluated, it is shown that the proposed method can achieve improved localization results from both quantitative and qualitative perspectives.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2009년도 춘계학술대회 논문집
• /
• pp.490-495
• /
• 2009

An approach to 3D visualization of multiple sound sources has been developed with the application of a moving array technique. Frequency-domain beamforming algorithm is used to generate a beam power map and the sound source is modeled as a point source. When a conventional delay and sum beamformer is used, it is considered that 2D distribution of sensors leads to have deficiency in spatial resolution along a measurement distance. The goal of moving an array in this study is to form 3D array aperture surrounding multiple sound sources so that the improved spatial resolution in a virtual space can be expected. Numerical simulation was made to examine source localization capabilities of various shapes of array. The 3D beam power maps of hemispherical and spherical distribution are found to have very sharp resolution. For experiments, two sound sources were placed in the middle of defined virtual space and arc-shaped line array was rotated around the sources. It is observed that spherical array show the most accurate determination of multiple sources' positions.

• 한국소음진동공학회논문집
• /
• 제19권9호
• /
• pp.907-914
• /
• 2009

An approach to 3D visualization of multiple sound sources has been developed with the application of a moving array technique. Frequency domain beamforming algorithm is used to generate a beam power map and the sound source is modeled as a point source. When a conventional delay and sum beamformer is used, it is considered that 2D distribution of sensors leads to have deficiency in spatial resolution along a measurement distance. The goal of moving an array in this study is to form 3D array aperture surrounding multiple sound sources so that the improved spatial resolution in a virtual space can be expected. Numerical simulation was made to examine source localization capabilities of various shapes of array. The 3D beam power maps of hemispherical and spherical distribution are found to have very sharp resolution. For experiments, several sound sources were placed in the middle of defined virtual space and arc-shaped line array was rotated around the sources. It is observed that spherical array shows the most accurate determination of multiple sources' positions.

• 대한의용생체공학회:의공학회지
• /
• 제25권5호
• /
• pp.391-401
• /
• 2004

이차원 위상 어레이 변환자를 이용하는 실시간 3차원 영상 시스템은 많은 수의 채널 수를 가지기 때문에 고 비용의 매우 복잡한 빔집속부를 사용하여야 한다. 또한 각 주사선에 대해 초음파를 매번 송수신해야 하므로 볼륨 레이트 또한 낮게 된다. 이를 해결하기 위해 기존에 제안된 교차어레이를 이용한 3차원 영상화 기법은 실시간 3차원 영상을 위한 고속 주사가 가능하고 측방향으로는 동적집속이 가능하지만, 고도방향으로는 송신집속깊이를 제외하고는 고도방향의 해상도가 저하된다는 문제점이 있다. 이러한 문제점을 극복하기 위해 본 논문에서는 합성구경 기반의 교차 어레이를 사용한 3차원 영상화 기법을 제안한다. 제안한 방법에서는 고도방향으로 놓인 일차원 송신어레이의 각 변환소자를 한번에 하나씩 순차적으로 송신하고, 반사된 신호들을 측방향으로 놓인 일차원 수신어레이의 모든 변환소자를 이용하여 수신하게 된다. 수신시 측방향으로는 동적집속, 고도방향으로는 합성구경 기법을 이용하여 빔을 집속함으로써 모든 영상 점에 대해 측방향과 고도방향 모두 동적집속된 효과를 얻을 수 있다. 또한, 제안한 방법은 고도방향으로 합성구경 기법을 이용함으로써 송신어레이 변화소자 개수만큼의 초음파 송수신 과정을 통해 관심 영역의 입제영상을 임의의 요구되는 단면영상 수를 이용하여 구성할 수 있다. 제안한 방법을 통해 기존의 고정집속 기반의 교차어레이를 이용한 3차원 영상화 기법과 비교하여 측방향으로는 동일하고 고도방향으로는 훨씬 우수한 해상도의 영상을 획득할 수 있음을 컴퓨터 모사실험을 통해 점증하였다. 또한 동반논문에서는 제안한 방법에 비해 송신전력과 고도방향 해상도를 더욱 향상시킬 수 있는 선형파면 기반의 합성구경 기법을 제안한다.