In response to the drawbacks of low efficiency, cumbersome calculation, and easy-to-fall local optimal solutions in existing shallow water acoustic parameters inversion research, this paper proposes a method based on feedback (BP) neural network model. Firstly, theoretically predicted values sound pressure field are obtained through fast (FFM). Secondly, relationship model between ground parameter is established BP Finally, measured data brought into obtain results. The application results...

Underwater acoustic technology is essential for ocean observation, exploration and exploitation, its development based on an accurate predication of underwater wave propagation. In shallow sea environments, the geoacoustic parameters, such as seabed structure, sound speeds, densities, speed attenuations in layers, would significantly affect propagation characteristics. To obtain more inversion results these this study presents method using waveguide characteristic impedance Bayesian...

Side-scan sonar (SSS) is an important underwater imaging method that has high resolutions and convenient to use. However, due the restriction of conventional pulse compression technology, side-scan beam sidelobe in range direction relatively high, which affects definition contrast images. When working a shallow-water environment, image quality especially influenced by strong bottom reverberation or other targets on seabed. To solve this problem, for image-quality improvement based...

Geoacoustic parameter inversion is a crucial issue in underwater acoustic research for shallow sea environments and has increasingly become popular the recent past. This paper investigates geoacoustic parameters environment using single-receiver method based on Bayesian theory. In this context, seabed regarded as an elastic medium, pressure at different positions under low-frequency chosen study object, theoretical prediction value of described by Fast Field Method (FFM). The cost function...

Seabed geoacoustic parameters play an important role in underwater acoustic channel modeling. Traditional methods to determine these parameters, for example, drilling, are expensive and being replaced by inverse technology. An inversion method based on Bayesian theory is presented derive the structure of a layered seabed shallow sea. The was considered elastic medium. objective this research use sound pressure detected sensors at different positions nonlinear estimate their uncertainties...

This work is concerned with the characteristics of very low frequency sound propagation (VLF, ≤100 Hz) in shallow marine environment. Under these conditions, classical hypothesis considering sea bottom as a fluid environment no longer appropriate, and at should be also considered. Hence, based on finite element method (FEM), setting an elastic medium, proposed model which unifies water established, full waveguides can synchronously discussed. Using this model, effects topography various...

Abstract In this study, the improved Dynamic Time Warping (DTW) algorithm is used in MATLAB to perform simulation classification and recognition on sounds of Nycticorax nycticorax other 10 birds Wuzhishan Islands Reserve Zhoushan, Zhejiang. By setting slope path, path restricted a fixed parallelogram. The combines with global restrictions can effectively speed up search, reduce amount calculation, memory requirements, time, thus improve efficiency. Experimental results show that compared...

Most traditional sound field calculation methods regard the seabed as horizontal stratified liquid sea bottom and conduct simulation analysis based on frequency domain. Hence, generality of above research is limited to varying degrees. To accurately clarify propagation characteristics mechanism very low-frequency (VLF, ≤100 Hz) waves in shallow sea, a numerical model established using finite element time-domain method (FETD) three-dimensional cylindrical coordinate system. Using this model,...

The ocean ambient noise field experiences a stochastic process of many such sources and the respective interactions their wave fields with waveguide boundaries. At frequencies about 1 kHz higher, forward scattering from surface can strongly affect shallow water sound propagation. However, most available forecasting models do not consider effects multiple wave. Therefore, there is need for an accurate method predicting noises at middle high-frequency which account scatterings. Aiming...

A method of geo-acoustic parameter inversion based on the Bayesian theory is proposed for acquisition acoustic parameters in shallow sea with elastic seabed. Firstly, theoretical prediction value sound pressure field calculated by fast (FFM). According to theory, we establish misfit function between measured and field. It under assumption Gaussian data errors which are line likelihood function. Finally, posterior probability density (PPD) given as result inversion. Our research conducted...

Likelihood probability based phylogenetic analysis approaches have contributed to impressive advances in minimizing the variance of estimating evolutionary parameters. However, their actual applications are greatly limited due very time-consuming calculations Conditional Probabilities (CLPs). Accurately and quickly obtaining likelihoods massive tree samples can facilitate process. Inspired by recent advance machine learning techniques that improve performance many related prediction tasks,...

The manual interpretation of traditional side-scan sonar images has problems such as solid subjectivity, low efficiency, high labor cost, and large time consumption, the machine learning method requires feature selection, which lacks adaptability robustness. In this paper, we introduce a convolutional neural network method, can automatically learn features from submarine aircraft wreckage complete classification recognition. Using 48 in SeabedObjects (ship aircraft) dataset after...

Introduction Existing methods primarily focus on earth acoustic parameters inversion under specific layered structures. However, they face challenges with experimental data from unknown seabed stratification, hindering accurate parameter inversion. Methods To address this, a novel algorithm combines Back Propagation Neural Network (BPNN) for distinguishing stratification and inverting parameters. Simulated sound pressure disturb as input, enabling feature recognition training the neural...

A study concerning the influence of flow on Herschel–Quincke duct is presented here, which includes numerical model, acoustic source and absorption condition called Perfectly Matched Layer. For excitation a sound field, normal mode wave placed at inlet tube. The function PML to simulate infinite tube avoid reflection wave. To investigate field coupled calculation method combining finite element computational fluid dynamics used solve linearized Euler equation, named Galbrun equation....

Background: Deep learning object detection algorithm is widely used in the field of image classification and has become an indispensable part. With improvement accuracy, sonar target based on deep gradually focus more people's research. Objective: This article aims to provide a summary analysis learning-based algorithms, with hope offering insights for future research technology. Method: paper systematically summarizes algorithms learning. According method principle, existing are divided...

To clarify environment of shallow ocean front propagation mechanism and characteristic low frequency sound signal, using the Finite Element Method (FEM), set up a little bit source under action water waveguide all field numerical model, model simulation analysis different nature oceanic front, gradient slant range influence law energy flow characteristics. The results show that, compared with case without low-frequency will significantly leak to sea bottom in area transition from warm cold...

Fish resource assessment has always been an indispensable type of work in fishery surveys. In order to accurately assess the number fish fenced farms and guide farming more scientifically efficiently, a method using sonar technology count is proposed. The data collected by subjected image processing methods such as filtering denoising, edge extraction, target division, achieve purpose statistics on fish. Firstly, median filter used remove noise. Secondly, Guess-Laplace operator detect...

Abstract The sound field of a 3D shallow sea environment with conical submarine mountain is simulated and calculated using the finite element method. To verify accuracy calculation results, effects height source depth on sound-propagation characteristics are evaluated. simulation results indicate that owing to reflection concentration acoustic energy due mountain, before wave reaches subsea impact can reach 10 dB. In addition, hindered by seamount. Therefore, loss after higher than in...

The signal to noise ratio (SNR) of seismic waves is usually very low after long distance transmission. For this condition, improve the bearing estimation capability in SNR, a frequency domain polarization weighted ESPRIT method using single vector device proposed. parameters extracted from signals are used design function which applied received signals. angle and target estimated through ES-PRIT simulation experiment results show that presented can obtain accurate values under SNR with...

The vertical directionality of wind-generated noise in the direction is influenced by sea state, which turn dependent on acoustic source distribution associated with breaking surface waves and propagation environment. At present, many ocean ambient models are available, such as K/I model, CAREY CANARY model. And most treat a flat pressure-release boundary, effects bottom roughness not considered these models. In shallow water, both (sea state) can be factors influence noise. aim this paper...

This paper focuses on using the acoustic vector field physical properties for ocean bottom geoacoustic inversion. The characteristics of a shallow water waveguide with elastic are described according to Fast Field Models (FFM) at first. intensity and propagation direction power fluxes I in underwater is numerically investigated, correlation sound pressure particle velocity analyzed. For vertical contains more information from bottom, sensitivity test component coefficient parameters...

Abstract To discuss the effect of an ocean mesoscale vortex on underwater low-frequency sound propagation, a theoretical model field under action was established based acoustic wave theory and ray theory, using finite element method. The influences properties intensity characteristics propagation are discussed. simulation results indicate that, in cold vortex, convergence zone moves toward source, span decreases, gain increases. Alternatively, for warm away from increases, rays diffuse,...

Abstract In order to clarify the propagation mechanism and characteristics of low-frequency sound signals in shallow sea with elastic bottom, using Time-domain Finite Element Method (TDFEM), a time-domain numerical calculation model field full-waveguide under action point source is established. With this model, The influence seabed acoustic parameters’ topography’ changes on simulated analyzed. results show that when there shear wave effect seabed, waves will excite Scholte propagating...

In this paper, a geoacoustic parameters inversion method for shallow water based on Bayesian Information Criterion (BIC) is developed. study, the seabed n-layer horizontal elastic medium, and in each layer medium include density ρ <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">n</inf> , compression wave (P-wave) velocity c xmlns:xlink="http://www.w3.org/1999/xlink">pn</inf> shear (S-wave) xmlns:xlink="http://www.w3.org/1999/xlink">sn</inf>...