Field modelling acceleration on ultrasonic systems using graphic hardware

Romero-Laorden, David; Martínez-Graullera, Óscar; Martín, Carlos Javier Blanco; Pérez, Manuel Tomás Pastor; Ullate, L.G.

doi:10.1016/j.cpc.2010.10.032

Cited by 6 publications

(3 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nowadays, new massively parallel architectures can empower computational softwares, at the expense of adapting algorithms to the specificities of those architectures to highlight and improve parallel computation steps. The exploitation of both multicore-GPP and GPU capabilities results in a new intensively parallel algorithm of simulation of UT inspection [2] [3]. This new model relies on analytical solution to the beam propagation.…”

Section: Introductionmentioning

confidence: 99%

A fast ultrasonic simulation tool based on massively parallel implementations

Lambert¹,

Rougeron²,

Lacassagne³

et al. 2014

AIP Conference Proceedings

View full text Add to dashboard Cite

This paper presents a CIVA optimized ultrasonic inspection simulation tool, which takes benefit of the power of massively parallel architectures : graphical processing units (GPU) and multi-core general purpose processors (GPP). This tool is based on the classical approach used in CIVA : the interaction model is based on Kirchoff, and the ultrasonic field around the defect is computed by the pencil method. The model has been adapted and parallelized for both architectures. At this stage, the configurations addressed by the tool are : multi and mono-element probes, planar specimens made of simple isotropic materials, planar rectangular defects or side drilled holes of small diameter. Validations on the model accuracy and performances measurements are presented.

show abstract

Section: Introductionmentioning

confidence: 99%

A fast ultrasonic simulation tool based on massively parallel implementations

Lambert¹,

Rougeron²,

Lacassagne³

et al. 2014

AIP Conference Proceedings

View full text Add to dashboard Cite

show abstract

“…The exploitation of both multicore-GPP and GPU capabilities results in a new intensively parallel algorithm of simulation of UT beam. Those architectures have already been used to provide consistent speedup over wave propagation [1], or field modeling for a probe alone [2] but they have not been applied to UT field simulation yet.…”

Section: Introductionmentioning

confidence: 99%

High performances simulation of ultrasonic fields for Non Destructive Testing

Lambert

Lacassagne

Rougeron

et al. 2014

SNA + MC 2013 - Joint International Conference on Supercomputing in Nuclear Applications + Monte Carlo

View full text Add to dashboard Cite

Ultrasonic imaging is a commonly used method to detect and identify defects in a mechanical part in nuclear applications. Nowadays massively parallel architectures enable the simulation of ultrasonic field emitted by a phased array transducer inspecting a part across a coupling medium. In this paper, regular field computation model will be discussed along its implementations on General Purpose Processors (GPP) and Graphic Processing Units (GPU).

show abstract

“…Some previous works as [2] show toolboxes for beamforming computation over CPUs and multi-core CPUs obtaining very good timing results. Our research group has been working on GPUs applied for field modeling acceleration [3] [4] as well as for fast beamforming [5]- [7] achieving speed ups of 150× over conventional CPU beamforming. Nowadays, GPU beamformers are a reality and there are lot of research groups working on solutions for NDT and medical applications [8].…”

Section: Introductionmentioning

confidence: 99%

Cheetah: A Library for Parallel Ultrasound Beamforming in Multi-Core Systems

Romero-Laorden¹,

Martín-Arguedas²,

Villazón-Terrazas³

et al. 2015

JAMP

View full text Add to dashboard Cite

Developing new imaging methods needs to establish some proofs of concept before implementing them on real-time scenarios. Nowadays, the high computational power reached by multi-core CPUs and GPUs have driven the development of software-based beamformers. Taking this into account, a library for the fast generation of ultrasound images is presented. It is based on Synthetic Aperture Imaging Techniques (SAFT) and it is fast because of the use of parallel computing techniques. Any kind of transducers as well as SAFT techniques can be defined although it includes some pre-built SAFT methods like 2R-SAFT and TFM. Furthermore, 2D and 3D imaging (slicebased or full volume computation) is supported along with the ability to generate both rectangular and angular images. For interpolation, linear and polynomial schemes can be chosen. The versatility of the library is ensured by interfacing it to Matlab, Python and any programming language over different operating systems. On a standard PC equipped with a single NVIDIA Quadro 4000 (256 cores), the library is able to calculate 262,144 pixels in ≈105 ms using a linear transducer with 64 elements, and 2,097,152 voxels in ≈ 5 seconds using a matrix transducer with 121 elements when TFM is applied.

show abstract

Field modelling acceleration on ultrasonic systems using graphic hardware

Cited by 6 publications

References 22 publications

A fast ultrasonic simulation tool based on massively parallel implementations

A fast ultrasonic simulation tool based on massively parallel implementations

High performances simulation of ultrasonic fields for Non Destructive Testing

Cheetah: A Library for Parallel Ultrasound Beamforming in Multi-Core Systems

Contact Info

Product

Resources

About