Development of an in-air circular synthetic aperture sonar system as an educational tool

Blanford, Thomas E.; McKay, John; Brown, Daniel C.; Park, J. Daniel; Johnson, Shawn F.

doi:10.1121/2.0001025

Cited by 8 publications

(8 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…AirSAS measurements: In addition to simulated measurements, we capture in-air SAS measurements using AirSAS [9]. AirSAS consists of a tweeter and microphone directed towards a turntable.…”

Section: Methodsmentioning

confidence: 99%

“…To deconvolve SAS images, we define an analysis-by-synthesis optimization (i.e., our method does not require training data) that optimizes neural network weights to model a function that maps point scattering positions (x, y) to real-valued amplitudes. We validate our method on simulated SAS data generated with a ray-based scattering model and on real data captured from an in-air circular SAS system, called AirSAS [9]. We provide a comparison of our reconstructions to delayand-sum beamformed images, and show that our method reconstructs more detailed image features by deconvolving side lobes from the transmitted waveform.…”

Section: Introductionmentioning

confidence: 90%

See 1 more Smart Citation

Implicit Neural Representations for Deconvolving SAS Images

Reed¹,

Blanford²,

Brown³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Synthetic aperture sonar (SAS) image resolution is constrained by waveform bandwidth and array geometry. Specifically, the waveform bandwidth determines a point spread function (PSF) that blurs the locations of point scatterers in the scene. In theory, deconvolving the reconstructed SAS image with the scene PSF restores the original distribution of scatterers and yields sharper reconstructions. However, deconvolution is an ill-posed operation that is highly sensitive to noise. In this work, we leverage implicit neural representations (INRs), shown to be strong priors for the natural image space, to deconvolve SAS images. Importantly, our method does not require training data, as we perform our deconvolution through an analysis-bysynthesis optimization in a self-supervised fashion. We validate our method on simulated SAS data created with a point scattering model and real data captured with an in-air circular SAS. This work is an important first step towards applying neural networks for SAS image deconvolution.

show abstract

“…AirSAS measurements: In addition to simulated measurements, we capture in-air SAS measurements using AirSAS [9]. AirSAS consists of a tweeter and microphone directed towards a turntable.…”

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 90%

Implicit Neural Representations for Deconvolving SAS Images

Reed¹,

Blanford²,

Brown³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…While, to the best of our knowledge, neural networks have not been applied to SAS deconvolution, many works leverage Fig. 2: Our Pipeline: We evaluate our deconvolution method on images captured with an in-air CSAS platform called AirSAS [14]. We reconstruct CSAS measurements, compute the scene PSF, and enhance the image quality using our SINR deconvolution approach.…”

Section: B Deconvolutionmentioning

confidence: 99%

“…1: Our Method: Our proposed deconvolution pipeline is capable of reconstructing scenes from extremely lowbandwidth measurements (5 kHz) by using a neural network to deconvolve the effects of a non-ideal PSF. From left to right: an object on the AirSAS turn-table [14], the image reconstructed using a conventional SAS reconstruction algorithm (delay-and-sum) containing significant side-lobe energy distorting the object geometry, and our proposed reconstruction using our deconvolution pipeline (SINR) that leverages implicit neural representations to deconvolve the scene.…”

Section: Introductionmentioning

confidence: 99%

“…To address this, we design a neural network capable of predicting a complex scattering field to perform a coherent deconvolution that accounts for a scene's spatially-varying phase. We evaluate our pipeline on simulated measurements created with a point scattering model and real measurements captured with an in-air SAS system called AirSAS [14] (see Figure 1). Our specific contributions are as follows:…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

SINR: Deconvolving Circular SAS Images Using Implicit Neural Representations

Reed¹,

Blanford²,

Brown³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Circular Synthetic aperture sonars (CSAS) capture multiple observations of a scene to reconstruct high-resolution images. We can characterize resolution by modeling CSAS imaging as the convolution between a scene's underlying point scattering distribution and a system-dependent point spread function (PSF). The PSF is a function of the transmitted waveform's bandwidth and determines a fixed degree of blurring on reconstructed imagery. In theory, deconvolution overcomes bandwidth limitations by reversing the PSF-induced blur and recovering the scene's scattering distribution. However, deconvolution is an illposed inverse problem and sensitive to noise. We propose a selfsupervised pipeline (does not require training data) that leverages an implicit neural representation (INR) for deconvolving CSAS images. We highlight the performance of our SAS INR pipeline, which we call SINR, by implementing and comparing to existing deconvolution methods. Additionally, prior SAS deconvolution methods assume a spatially-invariant PSF, which we demonstrate yields subpar performance in practice. We provide theory and methods to account for a spatially-varying CSAS PSF, and demonstrate that doing so enables SINR to achieve superior deconvolution performance on simulated and real acoustic SAS data. We provide code 1 to encourage reproducibility of research.

show abstract