2D Fast Fourier Transform Research Articles

Nanotechnology-assisted piezoelectrodes for cochlear stimulation

Event Abstract Back to Event Nanotechnology-assisted piezoelectrodes for cochlear stimulation Serena Danti1, Carlos Mota1, 2, Massimiliano Labardi3, Giada G. Genchi4, Dario Puppi2, Federica Chiellini2, Virgilio Mattoli4, Gianni Ciofani4, Luca Bruschini1 and Stefano Berrettini1 1 University of Pisa, Departmentof Surgical, Medical, Molecular Pathology and Emergency Medicine, Italy 2 University of Pisa, Department of Chemistry and Industrial Chemistry, Italy 3 National Council of Researches (CNR), Institute for Chemical-Physical Processes, Italy 4 Italian Institute of Technology (IIT), Center for Micro-BioRobotics @SSSA, Italy Introduction: Cochlear implants (CIs) are complex electronic devices which represent the only option to treat profound or severe sensorineural hearing loss. CIs show inherent drawbacks, like interference with magnetic fields and water-compatibility. It has been shown that piezoelectric materials alone can replace the function of cochlear sensory epithelium [1]. First studies used a silicon/polyvinylidene fluoride (PVDF) membrane, but the output voltage was still insufficient to excite the auditory neurons. This study is aimed at applying nanotechnology to achieve increased device sensitivity, via two strategies: (i) doping PVDF with piezoelectric nanoceramics to improve device performance, and (ii) tailoring architectural cues of materials to improve device efficiency at the neuron/material interface. Materials and Methods: Coaxial electrospinning (V = 20 kV, rotation velocity = 500-3.000 r.p.m) was used to obtain aligned ultrafine fibers from a PVDF solution (20% w/v dimethylformamide) mixed with barium titanate nanoparticles (BTNPs; composition range: 0%-20% w/w). Fiber diameter was measured via scanning electron microscopy (SEM) (n = 100). Fiber alignment was evaluated by 2D Fast Fourier Transform method. Presence and distribution of BTNPs in the fibers was observed with transmission electron microscopy (TEM). Piezoelectric crystallographic phases were assessed via X-ray diffraction. Piezoelectric properties were investigated with a set-up which measures the material deformation under an applied voltage. The substrates were cultured in vitro with PC12, as a model of neuronal cells, to assess cell viability (MTS assay) and differentiation (neurite sprout) under high frequency ultrasound (2 W, 5 s) for mechanic stimulation. Results and Discussion: Thin PVDF/BTNP fiber ribbons were produced. Fiber diameter varied depending on the rotation velocity and not on the BTNP concentration. Increasing rotation velocity gave rise to reduced fiber diameters and enhanced fiber alignment. Therefore, samples obtained at 3.000 r.p.m. were chosen for characterization (Figure 1). Piezoelectric crystal phases 2θ of PVDF and barium titanate were detected, and BTNPs resulted homogeneously dispersed in the fibers (Figure 1). In PVDF nanofiber ribbons, increasing BTNP concentration up to 20% (w/w) enhanced both direct and converse piezoelectric effects on a logarithmic scale (Figure 1), without exerting cytotoxic effects on PC12 cells (Figure 2). Neurite sprout was more pronounced on PVDF/BTNP 10% fibrous materials than in tissue culture plastics and averagely increased under ultrasound irradiation, with no statistical significance (Figure 2). Conclusion: Doping PVDF with nanoceramics can increase the piezoelectrode performance. Architectural features (e.g., aligned ultrafine fibers) can enhance the contact at the neuron/material interface. Obtaining highly sensitive piezoelectric CIs could substantially impact the quality of life of deaf people and reduce the healthcare costs. Italian Ministry of Health – Finalized Research Program 2011

Assessment of Damage Detection in Composite Structures Using 3D Vibrometry

Carbon fibre reinforced polymers (CFRP) have been used significantly more in recent years due to their increased specific strength over aluminium structures. One major area in which their use has grown is the aerospace industry where many now use CFRP in their construction. One major problem with CFRP's is their low resistance to impacts. Structural health monitoring (SHM) aims to continually monitor a structure throughout its entire life and can allow aircraft owners to identify impact damage as it occurs. This means that it can be repaired prior to growth, saving weight with the repair and the time that aircraft is grounded. Two areas of SHM being researched are Acoustic Emission (AE) monitoring and AcoustoUltrasonics (AU) both based on an understanding of the propagation of ultrasonic waves. 3D Scanning laser vibrometry was used to monitor the propagation of AU waves with the aim of gaining a better understanding their interaction with delamination in carbon fibre reinforced polymers. Three frequencies were exited with a PZT transducer and the received signal analysed by a cross correlation method. The results from this and the vibrometer scans revealed 100 kHz as the most effective propagating frequency of the three. A high resolution scan was then conducted at this frequency where it could be seen that only the out of plane component of the wave interacted with the damage, in particular the A0 mode. A 3D Fast Fourier Transform was then plotted, which identified the most effective frequency as 160 kHz.

Novo‐G#: a multidimensional torus‐based reconfigurable cluster for molecular dynamics

SummaryMolecular dynamics (MD) is a large‐scale, communication‐intensive problem that has been the subject of high‐performance computing research and acceleration for years. Not surprisingly, the most success in accelerating MD comes from specialized systems such as the Anton machine. In this paper, we describe Novo‐G# (novo‐jee‐sharp), a multi‐node reconfigurable system designed for the acceleration of communication‐intensive scientific problems in general, and MD in particular. This system provides a high‐bandwidth, low‐latency 3D torus network to allow direct communication between kernels running on multiple field‐programmable gate arrays. We also present a performance model for Novo‐G# running the 3D Fast Fourier Transform (FFT) kernel that forms the core of MD simulations. We validate the model against published Anton performance data and through initial hardware experiments on Novo‐G#. Finally, through simulation studies, we show that this system at scale performs better than specialized systems like Anton and outperforms established CPU‐based clusters like Blue Gene/Q by an order of magnitude for the 3D FFT kernel, with greater flexibility and lower costs. Copyright © 2015 John Wiley & Sons, Ltd.

Femtosecond laser-induced periodic surface nanostructuring of sputtered platinum thin films

In this work, submicro and nanostructures self-formed on the surface of Platinum thin films under femtosecond laser-pulse irradiation are investigated. A Ti:Sapphire laser system was used to linearly scan 15mm lines with 100fs pulses at a central wavelength of 800nm with a 1kHz repetition rate. The resulting structures were characterized by scanning electron microscopy (SEM) and 2D-Fast Fourier Transform (2D-FFT) analysis. This analysis of images revealed different types of structures depending on the laser irradiation parameters: random nanostructures, low spatial frequency LIPSS (LSFL) with a periodicity from about 450 to 600nm, and high spatial frequency LIPSS (HSFL) with a periodicity from about 80 to 200nm. Two different modifications regimes have been established for the formation of nanostructures: (a) a high-fluence regime in which random nanostructures and LSFL are obtained and (b) a low-fluence regime in which HSFL and LSFL are obtained.

Spectral turning bands for efficient Gaussian random fields generation on GPUs and accelerators

SummaryA random field (RF) is a set of correlated random variables associated with different spatial locations. RF generation algorithms are of crucial importance for many scientific areas, such as astrophysics, geostatistics, computer graphics, and many others. Current approaches commonly make use of 3D fast Fourier transform (FFT), which does not scale well for RF bigger than the available memory; they are also limited to regular rectilinear meshes.We introduce random field generation with the turning band method (RAFT), an RF generation algorithm based on the turning band method that is optimized for massively parallel hardware such as GPUs and accelerators. Our algorithm replaces the 3D FFT with a lower‐order, one‐dimensional FFT followed by a projection step and is further optimized with loop unrolling and blocking. RAFT can easily generate RF on non‐regular (non‐uniform) meshes and efficiently produce fields with mesh sizes bigger than the available device memory by using a streaming, out‐of‐core approach. Our algorithm generates RF with the correct statistical behavior and is tested on a variety of modern hardware, such as NVIDIA Tesla, AMD FirePro and Intel Phi. RAFT is faster than the traditional methods on regular meshes and has been successfully applied to two real case scenarios: planetary nebulae and cosmological simulations. Copyright © 2015 John Wiley & Sons, Ltd.

Methods for Evaluating Pictures and Extracting Music by 2D DFA and 2D FFT

This study proposes the following two methods applying two-dimensional DFA (detrended fluctuation analysis) and two-dimensional FFT (fast Fourier transform) algorithm: (1) a method for finding pleasant photographs of local tourist spots, and (2) a method for generating music from these photographs. We define “pleasant image” as the photograph containing 1/ f noise components since it has been suggested that the 1/ f -noise structure in visual art as well as in music can stimulate the perception of pleasant. The method for extracting music from the picture is developed as follows: (a) the image is divided into some squares and frequency components are extracted from each square by two-dimensional FFT, and then (b) the frequency components of each square are transformed into musical pitches. A method to arrange the music by reflecting features of the image is also provided.

Memory Efficient Two-Pass 3D FFT Algorithm for Intel® Xeon PhiTM Coprocessor

Equipped with 512-bit wide SIMD instructions and large numbers of computing cores, the emerging x86-based Intel® Many Integrated Core (MIC) Architecture provides not only high floating-point performance, but also substantial off-chip memory bandwidth. The 3D FFT (three-dimensional fast Fourier transform) is a widely-studied algorithm; however, the conventional algorithm needs to traverse the data array three times. In each pass, it computes multiple 1D FFTs along one of three dimensions, giving rise to plenty of non-unit strided memory accesses. In this paper, we propose a two-pass 3D FFT algorithm, which mainly aims to reduce the amount of explicit data transfer between the memory and the on-chip cache. The main idea is to split one dimension into two sub-dimensions, and then combine the transform along each sub-dimension with one of the rest dimensions respectively. The difference in amount of TLB misses resulting from decomposition along different dimensions is analyzed in detail. Multi-level parallelism is leveraged on the many-core system for a high degree of parallelism and better data reuse of local cache. On top of this, a number of optimization techniques, such as memory padding, loop transformation and vectorization, are employed in our implementation to further enhance the performance. We evaluate the algorithm on the Intel® Xeon PhiTM coprocessor 7110P, and achieve a maximum performance of 136 Gflops with 240 threads in offload mode, which beats the vendor-specific Intel® MKL library by a factor of up to 2.22X.

Canopy pruning grade classification Based on fast Fourier transform and artificial neural network

<abstract><title><italic>Abstract.</italic></title> Canopy architecture optimization and pruning management are agricultural operations crucial for plant growth and fruit production. Classifying pruning grade and optimizing the tree canopy accordingly are essential for these operations. If the work is done properly, it can result in higher yields of quality fruit. In this article, we present a method utilizing fast Fourier transform (FFT) and a back-propagation artificial neural network (BP-ANN) to classify different pruning grades in cherry orchards with an upright fruiting offshoots (UFO) training system to illustrate our approach. The approach was implemented automatically by first using a discrete FFT to extract frequency information from images of a cherry tree canopy and then applying a band filter to digitize the 2D FFT spectrum to a 1D array. A BP-ANN model was then used to classify the pruning grade of the trees. By combining image processing and ANN-based classification techniques, our approach was resilient to details, such as specific leaf shapes and leaf vein structure, and achieved an accuracy rate of over 80% in classifying pruning grades for UFO cherry trees with respect to human expert grading. Principal component analysis (PCA) was also applied to simplify the prediction model complexity while maintaining a similar prediction accuracy rate with a much less complicated input data set. Experimental results showed that our method could provide real-time classification of pruning grades for UFO cherry trees with reasonable prediction accuracy.

Formation of laser-induced periodic surface structures on niobium by femtosecond laser irradiation

The surface morphology of a Niobium sample, irradiated in air by a femtosecond laser with a wavelength of 800 nm and pulse duration of 100 fs, was examined. The period of the micro/nanostructures, parallel and perpendicularly oriented to the linearly polarized fs-laser beam, was studied by means of 2D Fast Fourier Transform analysis. The observed Laser-Induced Periodic Surface Structures (LIPSS) were classified as Low Spatial Frequency LIPSS (periods about 600 nm) and High Spatial Frequency LIPSS, showing a periodicity around 300 nm, both of them perpendicularly oriented to the polarization of the incident laser wave. Moreover, parallel high spatial frequency LIPSS were observed with periods around 100 nm located at the peripheral areas of the laser fingerprint and overwritten on the perpendicular periodic gratings. The results indicate that this method of micro/nanostructuring allows controlling the Niobium grating period by the number of pulses applied, so the scan speed and not the fluence is the key parameter of control. A discussion on the mechanism of the surface topology evolution was also introduced.

Study of the Relation between the Spiral Arm Pitch Angle and the Kinetic Energy of Random Motions of the Host Spiral Galaxies, A

In this work, we report a relation between the kinetic energy of random motions of the corresponding host galaxies and spiral arm pitch angles (Mdynσ2- P), (M*σ2- P) where Mdyn is the bulge dynamical mass, M* is bulge stellar mass, and σ is the velocity dispersion of the host galaxy bulge. We measured the spiral arm pitch angle (P) for a sample of Spitzer/IRAC 3.6-μm images of 54 spiral galaxies, estimated by using a 2D Fast Fourier Transform decomposition technique (2DFFT). We selected a sample of nearly face-on spiral galaxies and used IRAF ellipse to determine the ellipticity and major-axis position angle in order to deproject the images to face-on, and using a 2D Fast Fourier Transform decomposition technique, we determined the spiral arm pitch angles. We estimated the kinetic energy of random motions of the corresponding host galaxies (Mdynσ, M*σ2) by using Mdyn, M*, and σ, where the stellar velocity dispersion (σ) of the bulge was taken from the literature. We determined the bulge dynamical mass (Mdyn) using the virial theorem, and the bulge stellar mass (M*) was estimated by using the bulge 3.6-μm luminosity with the appropriate stellar mass-to-light ratio (M/L).

효과적인 다중 차량 추적을 위한 객체 특징 추출 및 매칭

A vehicle tracking system makes it possible to induce the vehicle movement path for avoiding traffic congestion and to prevent traffic accidents in advance by recognizing traffic flow, monitoring vehicles, and detecting road accidents. To track the vehicles effectively, those which appear in a sequence of video frames need to identified by extracting the features of each object in the frames. Next, the identical vehicles over the continuous frames need to be recognized through the matching among the objects` feature values. In this paper, we identify objects by binarizing the difference image between a target and a referential image, and the labelling technique. As feature values, we use the center coordinate of the minimum bounding rectangle(MBR) of the identified object and the averages of 1D FFT(fast Fourier transform) coefficients with respect to the horizontal and vertical direction of the MBR. A vehicle is tracked in such a way that the pair of objects that have the highest similarity among objects in two continuous images are regarded as an identical object. The experimental result shows that the proposed method outperforms the existing methods that use geometrical features in tracking accuracy.

Spectral 6DOF Registration of Noisy 3D Range Data with Partial Overlap

We present Spectral Registration with Multilayer Resampling (SRMR) as a 6 Degrees Of Freedom (DOF) registration method for noisy 3D data with partial overlap. The algorithm is based on decoupling 3D rotation from 3D translation by a corresponding resampling process of the spectral magnitude of a 3D Fast Fourier Transform (FFT) calculation on discretized 3D range data. The registration of all 6DOF is then subsequently carried out with spectral registrations using Phase Only Matched Filtering (POMF). There are two main aspects for the fast and robust registration of Euler angles from spherical information in SRMR. First of all, there is the permanent use of phase matching. Second, based on the FFT on a discrete Cartesian grid, not only one spherical layer but also a complete stack of layers are processed in one step. Experiments are presented with challenging datasets with respect to interference and overlap. The results include the fast and robust registration of artificially transformed data for ground-truth comparison, scans from the Stanford Bunny dataset, high end 3D laser range finder (LRF) scans of a city center, and range data from a low-cost actuated LRF in a disaster response scenario.

Rapid interactive modeling of 3D magnetic anomalies

I implement a MATLAB® function (for_3DFFT_mag) for calculating magnetic anomalies from a 3D distribution of magnetization, which can be loaded interactively through an user-friendly graphic interface. The forward calculation engine is based on a 3D Fast Fourier Transform computation, that gives accurate results in a very short computing time, making the use of this program particularly suitable for 3D interactive modeling of observed magnetic anomalies.

P3DFFT: A Framework for Parallel Computations of Fourier Transforms in Three Dimensions

Fourier and related transforms is a family of algorithms widely employed in diverse areas of computational science, notoriously difficult to scale on high-performance parallel computers with large number of processing elements (cores). This paper introduces a popular software package called P3DFFT implementing Fast Fourier Transforms (FFT) in three dimensions (3D) in a highly efficient and scalable way. It overcomes a well-known scalability bottleneck of 3D FFT implementations by using two-dimensional domain decomposition. Designed for portable performance, P3DFFT achieves excellent timings for a number of systems and problem sizes. On Cray XT5 system P3DFFT attains 45% efficiency in weak scaling from 128 to 65,536 computational cores. Library features include Fourier and Chebyshev transforms, Fortran and C interfaces, in- and out-of-place transforms, uneven data grids, single and double precision. P3DFFT is available as open source at http://code.google.com/p/p3dfft/. This paper discusses P3DFFT implementation and performance in a way that helps guide the user in making optimal choices for parameters of their runs.

FPGA based Reconfigurable 2D FFT System

This paper develops a system level architecture for implementing a cost-efficient, FPGA-based reconfigurable two dimensional (2D) FFT system. The adopted approach considers both the hardware cost (in terms of FPGA resource requirements), and performance (in terms of throughput). These two extremes are optimized based on using run time reconfiguration, double buffering technique, shared Dual Ported RAM (DPRAM) modules and the “hardware virtualization” to reuse the available processing components. The system employs two one Dimensional (1D) FFT processor each with sixteen reconfigurable parallel FFT cores. Each core represents a 16 complex point parallel FFT engine. Thus the architecture supports transform length of 256X256 complex points, as a demonstrator to the design idea, using fixed-point arithmetic and has been developed using radix-4 butterfly architecture. The simulation results that have been performed using VHDL modeling language and ModelSim software shows that the full design can be implemented using single FPGA platform requiring about 50,000 Slices. Keywords: 2D Fast Fourier Transform Radix-4. Run Time Reconfiguration

Determination of Age and Gender Based on Features of Human Motion Using AdaBoost Algorithms

Automated human identification by their walking behavior is a challenge attracting much interest among machine vision researchers. However, practical systems for such identification remain to be developed. In this study, a machine learning approach to understand human behavior based on motion imagery was proposed as the basis for developing pedestrian safety information systems. At the front end, image and video processing was performed to separate foreground from background images. Shape-width was then analyzed using 2D discrete wavelet transformation and 2D fast Fourier transformation to extract human motion features. Finally, an adaptive boosting (AdaBoost) algorithm was performed to classify human gender and age into its class based on spatiotemporal information. The results demonstrated the capability of the proposed systems to classify gender and age highly accurately.

Ultra-fast FFT protein docking on graphics processors

Modelling protein-protein interactions (PPIs) is an increasingly important aspect of structural bioinformatics. However, predicting PPIs using in silico docking techniques is computationally very expensive. Developing very fast protein docking tools will be useful for studying large-scale PPI networks, and could contribute to the rational design of new drugs. The Hex spherical polar Fourier protein docking algorithm has been implemented on Nvidia graphics processor units (GPUs). On a GTX 285 GPU, an exhaustive and densely sampled 6D docking search can be calculated in just 15 s using multiple 1D fast Fourier transforms (FFTs). This represents a 45-fold speed-up over the corresponding calculation on a single CPU, being at least two orders of magnitude times faster than a similar CPU calculation using ZDOCK 3.0.1, and estimated to be at least three orders of magnitude faster than the GPU-accelerated version of PIPER on comparable hardware. Hence, for the first time, exhaustive FFT-based protein docking calculations may now be performed in a matter of seconds on a contemporary GPU. Three-dimensional Hex FFT correlations are also accelerated by the GPU, but the speed-up factor of only 2.5 is much less than that obtained with 1D FFTs. Thus, the Hex algorithm appears to be especially well suited to exploit GPUs compared to conventional 3D FFT docking approaches. http://hex.loria.fr/ and http://hexserver.loria.fr/.

An Open Source, Fast Ultrasound B-Mode Implementation for Commodity Hardware

This document describes an open source, high performance ultrasound B-Mode implementation based on the Insight Toolkit (ITK). ITK extensions are presented to calculate the radio-frequency (RF) signal envelope. A variety of 1D Fast Fourier Transform options are introduced including VNL, FFTW, and an OpenCL solution. Scan conversion is implemented for phased array or curvilinear transducers. The entire image processing pipeline is streamable to limit memory consumption during multi-frame or 3D acquisitions with the introduction of an itk::StreamingResampleImageFilter.

Using 3D FFT fractal dimension estimator to analyze the complexity of fetal cortical surface from MR images

This study aims to investigate the complexity of developing fetal cortical surface using a 3D fast Fourier transform (FFT) fractal dimension (FD) estimator. Thirty-five fetal MR images were chosen at 22–36 weeks of gestational age (GA) including 30 normal brains and 5 testing brains. Observations of testing brains indicate that: cortical dysplasia (developmental delay) and twins have lower FD than that of normal fetal brains. Results of FD from normal brains show that are positively correlated to GA and evidently indicate cortical complexity increasing with developing weeks of GA, which are in good agreement with fetal brain development. It could be said that the 3D FFT FD estimator is a promising means for the quantification of complexity of the fetal cortical surface.

A Raw Signal Simulator for Bistatic SAR

This article proposes a new efficient raw signal simulator for the bistatic synthetic aperture radar (SAR) based on 2D fast Fourier transform (FFT) to deal with cases of both ideal trajectory and trajectory deviation. It begins with analyzing the geometric configuration and the range history of the bistatic SAR in side-looking and squint modes of ideal trajectory as well as trajectory deviation. Then a detailed and mathematical study is conducted on the equivalence relation of bistatic-to-monostatic applications (BTMA) in the case of ideal trajectory and trajectory deviation. Also a set of formulas are derived for the equivalence relation between bistatic SAR and monostatic SAR on some reasonable assumptions. Therefore, the application of the simulation method based on the 2D FFT for the monostatic SAR can be extended to the case of bistatic SAR. Finally, the simulation results prove the validity of this method. By comparing the efficiency of the proposed method with that of the time domain method, it is shown that the former is a few orders of magnitude higher.