Automatic compilation of MATLAB programs for synergistic execution on heterogeneous processors

MATLAB is an array language, initially popular for rapid prototyping, but is now being increasingly used to develop production code for numerical and scientific applications. Typical MATLAB programs have abundant data parallelism. These programs also have control flow dominated scalar regions that have an impact on the program's execution time. Today's computer systems have tremendous computing power in the form of traditional CPU cores and throughput oriented accelerators such as graphics processing units(GPUs). Thus, an approach that maps the control flow dominated regions to the CPU and the data parallel regions to the GPU can significantly improve program performance. In this paper, we present the design and implementation of MEGHA, a compiler that automatically compiles MATLAB programs to enable synergistic execution on heterogeneous processors. Our solution is fully automated and does not require programmer input for identifying data parallel regions. We propose a set of compiler optimizations tailored for MATLAB. Our compiler identifies data parallel regions of the program and composes them into kernels. The problem of combining statements into kernels is formulated as a constrained graph clustering problem. Heuristics are presented to map identified kernels to either the CPU or GPU so that kernel execution on the CPU and the GPU happens synergistically and the amount of data transfer needed is minimized. In order to ensure required data movement for dependencies across basic blocks, we propose a data flow analysis and edge splitting strategy. Thus our compiler automatically handles composition of kernels, mapping of kernels to CPU and GPU, scheduling and insertion of required data transfer. The proposed compiler was implemented and experimental evaluation using a set of MATLAB benchmarks shows that our approach achieves a geometric mean speedup of 19.8X for data parallel benchmarks over native execution of MATLAB.

Multi-objective optimization of satellite image registration using discrete Pparticle swarm optimisation

A new multi-sensor image registration technique is proposed based on detecting the feature corner points using modified Harris Corner Detector (HDC). These feature points are matched using multi-objective optimization (distance condition and angle criterion) based on Discrete Particle Swarm Optimization (DPSO). This optimization process is more efficient as it considers both the distance and angle criteria to incorporate multi-objective switching in the fitness function. This optimization process helps in picking up three corresponding corner points detected in the sensed and base image and thereby using the affine transformation, the sensed image is aligned with the base image. Further, the results show that the new approach can provide a new dimension in solving multi-sensor image registration problems. From the obtained results, the performance of image registration is evaluated and is concluded that the proposed approach is efficient.

Fracture processes of self compacting Concrete using image analysis

The mode I fracture toughness of concrete can be experimentally determined using three point bend beam in conjunction with digital image correlation (DIC). Three different geometrically similar sizes of beams are cast for this study. To study the influence of fly ash and silica fume on fracture toughness of SCC, three SCC mixes are prepared with and without mineral additions. The scanning electron microscope (SEM) images are taken on the fractured surface to add information on fracture process in SCC. From this study, it is concluded that the fracture toughness of SCC with mineral addition is higher when compared to those without mineral addition.

Evaluation of kissing bond in composite adhesive lap joints using digital image correlation: Preliminary studies

The assembly of aerospace and automotive structures in recent years is increasingly carried out using adhesives. Adhesive joints have advantages of uniform stress distribution and less stress concentration in the bonded region. Nevertheless, they may suffer due to the presence of defects in bond line and at the interface or due to improper curing process. While defects like voids, cracks and delaminations present in the adhesive bond line may be detected using different NDE methods, interfacial defects in the form of kissing bond may go undetected. Attempts using advanced ultrasonic methods like nonlinear ultrasound and guided wave inspection to detect kissing bond have met with limited success stressing the need for alternate methods. This paper concerns the preliminary studies carried out on detectability of dry contact kissing bonds in adhesive joints using the Digital Image Correlation (DIC) technique. In this attempt, adhesive joint samples containing varied area of kissing bond were prepared using the glass fiber reinforced composite (GFRP) as substrates and epoxy resin as the adhesive layer joining them. The samples were also subjected to conventional and high power ultrasonic inspection. Further, these samples were loaded till failure to determine the bond strength during which digital images were recorded and analyzed using the DIC method. This noncontact method could indicate the existence of kissing bonds at less than 50% failure load. Finite element studies carried out showed a similar trend. Results obtained from these preliminary studies are encouraging and further tests need to be done on a larger set of samples to study experimental uncertainties and scatter associated with the method. (C) 2013 Elsevier Ltd. All rights reserved.

Bilateral smoothing of gradient vector field and application to image segmentation

Medical image segmentation finds application in computer-aided diagnosis, computer-guided surgery, measuring tissue volumes, locating tumors, and pathologies. One approach to segmentation is to use active contours or snakes. Active contours start from an initialization (often manually specified) and are guided by image-dependent forces to the object boundary. Snakes may also be guided by gradient vector fields associated with an image. The first main result in this direction is that of Xu and Prince, who proposed the notion of gradient vector flow (GVF), which is computed iteratively. We propose a new formalism to compute the vector flow based on the notion of bilateral filtering of the gradient field associated with the edge map - we refer to it as the bilateral vector flow (BVF). The range kernel definition that we employ is different from the one employed in the standard Gaussian bilateral filter. The advantage of the BVF formalism is that smooth gradient vector flow fields with enhanced edge information can be computed noniteratively. The quality of image segmentation turned out to be on par with that obtained using the GVF and in some cases better than the GVF.

Benchmarking recognition results on camera captured word image data sets

We have benchmarked the maximum obtainable recognition accuracy on five publicly available standard word image data sets using semi-automated segmentation and a commercial OCR. These images have been cropped from camera captured scene images, born digital images (BDI) and street view images. Using the Matlab based tool developed by us, we have annotated at the pixel level more than 3600 word images from the five data sets. The word images binarized by the tool, as well as by our own midline analysis and propagation of segmentation (MAPS) algorithm are recognized using the trial version of Nuance Omnipage OCR and these two results are compared with the best reported in the literature. The benchmark word recognition rates obtained on ICDAR 2003, Sign evaluation, Street view, Born-digital and ICDAR 2011 data sets are 83.9%, 89.3%, 79.6%, 88.5% and 86.7%, respectively. The results obtained from MAPS binarized word images without the use of any lexicon are 64.5% and 71.7% for ICDAR 2003 and 2011 respectively, and these values are higher than the best reported values in the literature of 61.1% and 41.2%, respectively. MAPS results of 82.8% for BDI 2011 dataset matches the performance of the state of the art method based on power law transform.

Multiobjective discrete particle swarm optimization for multisensor image alignment

A new technique is proposed for multisensor image registration by matching the features using discrete particle swarm optimization (DPSO). The feature points are first extracted from the reference and sensed image using improved Harris corner detector available in the literature. From the extracted corner points, DPSO finds the three corresponding points in the sensed and reference images using multiobjective optimization of distance and angle conditions through objective switching technique. By this, the global best matched points are obtained which are used to evaluate the affine transformation for the sensed image. The performance of the image registration is evaluated and concluded that the proposed approach is efficient.

Integration of speckle de-noising and image segmentation using synthetic aperture radar image for flood extent extraction

Flood is one of the detrimental hydro-meteorological threats to mankind. This compels very efficient flood assessment models. In this paper, we propose remote sensing based flood assessment using Synthetic Aperture Radar (SAR) image because of its imperviousness to unfavourable weather conditions. However, they suffer from the speckle noise. Hence, the processing of SAR image is applied in two stages: speckle removal filters and image segmentation methods for flood mapping. The speckle noise has been reduced with the help of Lee, Frost and Gamma MAP filters. A performance comparison of these speckle removal filters is presented. From the results obtained, we deduce that the Gamma MAP is reliable. The selected Gamma MAP filtered image is segmented using Gray Level Co-occurrence Matrix (GLCM) and Mean Shift Segmentation (MSS). The GLCM is a texture analysis method that separates the image pixels into water and non-water groups based on their spectral feature whereas MSS is a gradient ascent method, here segmentation is carried out using spectral and spatial information. As test case, Kosi river flood is considered in our study. From the segmentation result of both these methods are comprehensively analysed and concluded that the MSS is efficient for flood mapping.

Automatic speech segmentation using probabilistic latent component modeling

Latent variable methods, such as PLCA (Probabilistic Latent Component Analysis) have been successfully used for analysis of non-negative signal representations. In this paper, we formulate PLCS (Probabilistic Latent Component Segmentation), which models each time frame of a spectrogram as a spectral distribution. Given the signal spectrogram, the segmentation boundaries are estimated using a maximum-likelihood approach. For an efficient solution, the algorithm imposes a hard constraint that each segment is modelled by a single latent component. The hard constraint facilitates the solution of ML boundary estimation using dynamic programming. The PLCS framework does not impose a parametric assumption unlike earlier ML segmentation techniques. PLCS can be naturally extended to model coarticulation between successive phones. Experiments on the TIMIT corpus show that the proposed technique is promising compared to most state of the art speech segmentation algorithms.

Efficient gradient-free simplex method for estimation of optical properties in image-guided diffuse optical tomography

Typical image-guided diffuse optical tomographic image reconstruction procedures involve reduction of the number of optical parameters to be reconstructed equal to the number of distinct regions identified in the structural information provided by the traditional imaging modality. This makes the image reconstruction problem less ill-posed compared to traditional underdetermined cases. Still, the methods that are deployed in this case are same as those used for traditional diffuse optical image reconstruction, which involves a regularization term as well as computation of the Jacobian. A gradient-free Nelder-Mead simplex method is proposed here to perform the image reconstruction procedure and is shown to provide solutions that closely match ones obtained using established methods, even in highly noisy data. The proposed method also has the distinct advantage of being more efficient owing to being regularization free, involving only repeated forward calculations. (C) 2013 Society of Photo-Optical Instrumentation Engineers (SPIE)

DSA image registration using non-uniform MRF model and pivotal control points

In order to reduce the motion artifacts in DSA, non-rigid image registration is commonly used before subtracting the mask from the contrast image. Since DSA registration requires a set of spatially non-uniform control points, a conventional MRF model is not very efficient. In this paper, we introduce the concept of pivotal and non-pivotal control points to address this, and propose a non-uniform MRF for DSA registration. We use quad-trees in a novel way to generate the non-uniform grid of control points. Our MRF formulation produces a smooth displacement field and therefore results in better artifact reduction than that of registering the control points independently. We achieve improved computational performance using pivotal control points without compromising on the artifact reduction. We have tested our approach using several clinical data sets, and have presented the results of quantitative analysis, clinical assessment and performance improvement on a GPU. (C) 2013 Elsevier Ltd. All rights reserved.

Taylor series expansion based multidimensional image reconstruction for confocal and 4pi microscopy

We propose and experimentally demonstrate a three-dimensional (3D) image reconstruction methodology based on Taylor series approximation (TSA) in a Bayesian image reconstruction formulation. TSA incorporates the requirement of analyticity in the image domain, and acts as a finite impulse response filter. This technique is validated on images obtained from widefield, confocal laser scanning fluorescence microscopy and two-photon excited 4pi (2PE-4pi) fluorescence microscopy. Studies on simulated 3D objects, mitochondria-tagged yeast cells (labeled with Mitotracker Orange) and mitochondrial networks (tagged with Green fluorescent protein) show a signal-to-background improvement of 40% and resolution enhancement from 360 to 240 nm. This technique can easily be extended to other imaging modalities (single plane illumination microscopy (SPIM), individual molecule localization SPIM, stimulated emission depletion microscopy and its variants).

Classification, representation, and automatic extraction of deformation features in sheet metal parts

This paper presents classification, representation and extraction of deformation features in sheet-metal parts. The thickness is constant for these shape features and hence these are also referred to as constant thickness features. The deformation feature is represented as a set of faces with a characteristic arrangement among the faces. Deformation of the base-sheet or forming of material creates Bends and Walls with respect to a base-sheet or a reference plane. These are referred to as Basic Deformation Features (BDFs). Compound deformation features having two or more BDFs are defined as characteristic combinations of Bends and Walls and represented as a graph called Basic Deformation Features Graph (BDFG). The graph, therefore, represents a compound deformation feature uniquely. The characteristic arrangement of the faces and type of bends belonging to the feature decide the type and nature of the deformation feature. Algorithms have been developed to extract and identify deformation features from a CAD model of sheet-metal parts. The proposed algorithm does not require folding and unfolding of the part as intermediate steps to recognize deformation features. Representations of typical features are illustrated and results of extracting these deformation features from typical sheet metal parts are presented and discussed. (C) 2013 Elsevier Ltd. All rights reserved.

Fast and efficient automatic memory management for GPUs using compiler-assisted runtime coherence scheme

Exploiting the performance potential of GPUs requires managing the data transfers to and from them efficiently which is an error-prone and tedious task. In this paper, we develop a software coherence mechanism to fully automate all data transfers between the CPU and GPU without any assistance from the programmer. Our mechanism uses compiler analysis to identify potential stale accesses and uses a runtime to initiate transfers as necessary. This allows us to avoid redundant transfers that are exhibited by all other existing automatic memory management proposals. We integrate our automatic memory manager into the X10 compiler and runtime, and find that it not only results in smaller and simpler programs, but also eliminates redundant memory transfers. Tested on eight programs ported from the Rodinia benchmark suite it achieves (i) a 1.06x speedup over hand-tuned manual memory management, and (ii) a 1.29x speedup over another recently proposed compiler--runtime automatic memory management system. Compared to other existing runtime-only and compiler-only proposals, it also transfers 2.2x to 13.3x less data on average.

Image reconstruction enables high resolution imaging at large penetration depths in fluorescence microscopy

Imaging thick specimen at a large penetration depth is a challenge in biophysics and material science. Refractive index mismatch results in spherical aberration that is responsible for streaking artifacts, while Poissonian nature of photon emission and scattering introduces noise in the acquired three-dimensional image. To overcome these unwanted artifacts, we introduced a two-fold approach: first, point-spread function modeling with correction for spherical aberration and second, employing maximum-likelihood reconstruction technique to eliminate noise. Experimental results on fluorescent nano-beads and fluorescently coated yeast cells (encaged in Agarose gel) shows substantial minimization of artifacts. The noise is substantially suppressed, whereas the side-lobes (generated by streaking effect) drops by 48.6% as compared to raw data at a depth of 150 mu m. Proposed imaging technique can be integrated to sophisticated fluorescence imaging techniques for rendering high resolution beyond 150 mu m mark. (C) 2013 AIP Publishing LLC.