Abstract:In this paper, we propose a modification of the standard Equivalent Source Method (ESM) for Near-Field Acoustic Holography (NAH). As in EMS, we aim at modeling the acoustic pressure radiated from a vibrating object, and its surface velocity, as the joint effect of a set of equivalent sources located within or dose to the object itself. The estimation of the equivalent source strengths (weigths) comes from the solution of a highly ill-conditioned problem. Rather than solving this problem in the least-squares se… Show more
“…To the authors knowledge, only rules of thumb are proposed, which are not, however, applicable in some contexts. In order to deal with this problem, ESM techniques based on CS [ 20 , 21 , 22 ] have been proposed with the aim of finding small and sparse subsets of equivalent sources.…”
Section: Introductionmentioning
confidence: 99%
“…For example, Canclini et al [ 22 ] proposed building a dictionary of equivalent sources in order to solve NAH. This technique, called dictionary-based esm (DESM), exploits synthetic data for finding the equivalent sources.…”
Section: Introductionmentioning
confidence: 99%
“…Recently, a new approach to NAH based on deep learning [ 23 ] has been proposed in [ 24 ]. The authors, inspired by the effectiveness of learned features for NAH [ 22 ] and the well-known feature learning capabilities of deep neural networks (DNN) in the context of acoustics [ 25 , 26 , 27 , 28 , 29 ], proposed a convolutional neural network (CNN) [ 30 ] for performing NAH. The promising approach of [ 24 ] provides accurate results, but the evaluation is limited to rectangular plates of isotropic material only.…”
In this manuscript, we describe a novel methodology for nearfield acoustic holography (NAH). The proposed technique is based on convolutional neural networks, with autoencoder architecture, to reconstruct the pressure and velocity fields on the surface of the vibrating structure using the sampled pressure soundfield on the holographic plane as input. The loss function used for training the network is based on a combination of two components. The first component is the error in the reconstructed velocity. The second component is the error between the sound pressure on the holographic plane and its estimate obtained from forward propagating the pressure and velocity fields on the structure through the Kirchhoff–Helmholtz integral; thus, bringing some knowledge about the physics of the process under study into the estimation algorithm. Due to the explicit presence of the Kirchhoff–Helmholtz integral in the loss function, we name the proposed technique the Kirchhoff–Helmholtz-based convolutional neural network, KHCNN. KHCNN has been tested on two large datasets of rectangular plates and violin shells. Results show that it attains very good accuracy, with a gain in the NMSE of the estimated velocity field that can top 10 dB, with respect to state-of-the-art techniques. The same trend is observed if the normalized cross correlation is used as a metric.
“…To the authors knowledge, only rules of thumb are proposed, which are not, however, applicable in some contexts. In order to deal with this problem, ESM techniques based on CS [ 20 , 21 , 22 ] have been proposed with the aim of finding small and sparse subsets of equivalent sources.…”
Section: Introductionmentioning
confidence: 99%
“…For example, Canclini et al [ 22 ] proposed building a dictionary of equivalent sources in order to solve NAH. This technique, called dictionary-based esm (DESM), exploits synthetic data for finding the equivalent sources.…”
Section: Introductionmentioning
confidence: 99%
“…Recently, a new approach to NAH based on deep learning [ 23 ] has been proposed in [ 24 ]. The authors, inspired by the effectiveness of learned features for NAH [ 22 ] and the well-known feature learning capabilities of deep neural networks (DNN) in the context of acoustics [ 25 , 26 , 27 , 28 , 29 ], proposed a convolutional neural network (CNN) [ 30 ] for performing NAH. The promising approach of [ 24 ] provides accurate results, but the evaluation is limited to rectangular plates of isotropic material only.…”
In this manuscript, we describe a novel methodology for nearfield acoustic holography (NAH). The proposed technique is based on convolutional neural networks, with autoencoder architecture, to reconstruct the pressure and velocity fields on the surface of the vibrating structure using the sampled pressure soundfield on the holographic plane as input. The loss function used for training the network is based on a combination of two components. The first component is the error in the reconstructed velocity. The second component is the error between the sound pressure on the holographic plane and its estimate obtained from forward propagating the pressure and velocity fields on the structure through the Kirchhoff–Helmholtz integral; thus, bringing some knowledge about the physics of the process under study into the estimation algorithm. Due to the explicit presence of the Kirchhoff–Helmholtz integral in the loss function, we name the proposed technique the Kirchhoff–Helmholtz-based convolutional neural network, KHCNN. KHCNN has been tested on two large datasets of rectangular plates and violin shells. Results show that it attains very good accuracy, with a gain in the NMSE of the estimated velocity field that can top 10 dB, with respect to state-of-the-art techniques. The same trend is observed if the normalized cross correlation is used as a metric.
“…The main problem of ESM is the computation of the optimal set (in terms of number and location) of equivalent sources. In order to deal with this problem, ESM techniques based on CS [15]- [17] have been proposed with the aim of finding small and sparse subsets of equivalent sources.…”
Section: Introductionmentioning
confidence: 99%
“…In [17] a dictionary-based ESM (DESM) is proposed in order to consider a sparse domain for solving ESM limitations. The ESM solution space is restricted to a suitable compressed dictionary whose components are retrieved from several sets of equivalent sources.…”
Near-field Acoustic Holography (NAH) is a wellknown problem aimed at estimating the vibrational velocity field of a structure by means of acoustic measurements. In this paper, we propose a NAH technique based on Convolutional Neural Network (CNN). The devised CNN predicts the vibrational field on the surface of arbitrary shaped plates (violin plates) with orthotropic material properties from a limited number of measurements. In particular, the architecture, named super resolution CNN (SRCNN), is able to estimate the vibrational field with a higher spatial resolution compared to the input pressure. The pressure and velocity datasets have been generated through Finite Element Method simulations. We validate the proposed method by comparing the estimates with the synthesized ground truth and with a state-of-the-art technique. Moreover, we evaluate the robustness of the devised network against noisy input data.
In practical acoustic measurement for large cylindrical surfaces, it is difficult to keep conformal and coaxial between the holographic surface and the reconstruction surface. To overcome this problem, a combined sound field reconstruction method using non-conformal plane measurement is proposed in this paper. Based on the sound pressure measured on the holographic plane, the combined method first reconstructs the sound field on the cylindrical conformal surface using statistically optimal planar near-field acoustic holography, and then reconstructs the sound field on the cylindrical reconstruction surface using statistically optimal cylindrical near-field acoustic holography. And a least square optimization method is proposed to determine the optimal position of the conformal surface. In addition, to overcome ill-posed problems, an error reduction method combining truncated singular value decomposition and Tikhonov regularization is proposed. Finally, the proposed method is applied to a test bed, and the accuracy and robustness of the sound field reconstruction for large cylindrical surfaces are obviously improved, which can provide reliable evidences for noise monitoring and control of mechanical systems.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.