Parallel Computer Vision Algorithms for Graphics Processing Units

La evolución de los teléfonos móviles inteligentes, dotados de cámaras digitales, está provocando una creciente demanda de aplicaciones cada vez más complejas que necesitan algoritmos de visión artificial en tiempo real; puesto que el tamaño de las señales de vídeo no hace sino aumentar y en cambio el rendimiento de los procesadores de un solo núcleo se ha estancado, los nuevos algoritmos que se diseñen para visión artificial han de ser paralelos para poder ejecutarse en múltiples procesadores y ser computacionalmente escalables. Una de las clases de procesadores más interesantes en la actualidad se encuentra en las tarjetas gráficas (GPU), que son dispositivos que ofrecen un alto grado de paralelismo, un excelente rendimiento numérico y una creciente versatilidad, lo que los hace interesantes para llevar a cabo computación científica. En esta tesis se exploran dos aplicaciones de visión artificial que revisten una gran complejidad computacional y no pueden ser ejecutadas en tiempo real empleando procesadores tradicionales. En cambio, como se demuestra en esta tesis, la paralelización de las distintas subtareas y su implementación sobre una GPU arrojan los resultados deseados de ejecución con tasas de refresco interactivas. Asimismo, se propone una técnica para la evaluación rápida de funciones de complejidad arbitraria especialmente indicada para su uso en una GPU. En primer lugar se estudia la aplicación de técnicas de síntesis de imágenes virtuales a partir de únicamente dos cámaras lejanas y no paralelas—en contraste con la configuración habitual en TV 3D de cámaras cercanas y paralelas—con información de color y profundidad. Empleando filtros de mediana modificados para la elaboración de un mapa de profundidad virtual y proyecciones inversas, se comprueba que estas técnicas son adecuadas para una libre elección del punto de vista. Además, se demuestra que la codificación de la información de profundidad con respecto a un sistema de referencia global es sumamente perjudicial y debería ser evitada. Por otro lado se propone un sistema de detección de objetos móviles basado en técnicas de estimación de densidad con funciones locales. Este tipo de técnicas es muy adecuada para el modelado de escenas complejas con fondos multimodales, pero ha recibido poco uso debido a su gran complejidad computacional. El sistema propuesto, implementado en tiempo real sobre una GPU, incluye propuestas para la estimación dinámica de los anchos de banda de las funciones locales, actualización selectiva del modelo de fondo, actualización de la posición de las muestras de referencia del modelo de primer plano empleando un filtro de partículas multirregión y selección automática de regiones de interés para reducir el coste computacional. Los resultados, evaluados sobre diversas bases de datos y comparados con otros algoritmos del estado del arte, demuestran la gran versatilidad y calidad de la propuesta. Finalmente se propone un método para la aproximación de funciones arbitrarias empleando funciones continuas lineales a tramos, especialmente indicada para su implementación en una GPU mediante el uso de las unidades de filtraje de texturas, normalmente no utilizadas para cómputo numérico. La propuesta incluye un riguroso análisis matemático del error cometido en la aproximación en función del número de muestras empleadas, así como un método para la obtención de una partición cuasióptima del dominio de la función para minimizar el error. ABSTRACT The evolution of smartphones, all equipped with digital cameras, is driving a growing demand for ever more complex applications that need to rely on real-time computer vision algorithms. However, video signals are only increasing in size, whereas the performance of single-core processors has somewhat stagnated in the past few years. Consequently, new computer vision algorithms will need to be parallel to run on multiple processors and be computationally scalable. One of the most promising classes of processors nowadays can be found in graphics processing units (GPU). These are devices offering a high parallelism degree, excellent numerical performance and increasing versatility, which makes them interesting to run scientific computations. In this thesis, we explore two computer vision applications with a high computational complexity that precludes them from running in real time on traditional uniprocessors. However, we show that by parallelizing subtasks and implementing them on a GPU, both applications attain their goals of running at interactive frame rates. In addition, we propose a technique for fast evaluation of arbitrarily complex functions, specially designed for GPU implementation. First, we explore the application of depth-image–based rendering techniques to the unusual configuration of two convergent, wide baseline cameras, in contrast to the usual configuration used in 3D TV, which are narrow baseline, parallel cameras. By using a backward mapping approach with a depth inpainting scheme based on median filters, we show that these techniques are adequate for free viewpoint video applications. In addition, we show that referring depth information to a global reference system is ill-advised and should be avoided. Then, we propose a background subtraction system based on kernel density estimation techniques. These techniques are very adequate for modelling complex scenes featuring multimodal backgrounds, but have not been so popular due to their huge computational and memory complexity. The proposed system, implemented in real time on a GPU, features novel proposals for dynamic kernel bandwidth estimation for the background model, selective update of the background model, update of the position of reference samples of the foreground model using a multi-region particle filter, and automatic selection of regions of interest to reduce computational cost. The results, evaluated on several databases and compared to other state-of-the-art algorithms, demonstrate the high quality and versatility of our proposal. Finally, we propose a general method for the approximation of arbitrarily complex functions using continuous piecewise linear functions, specially formulated for GPU implementation by leveraging their texture filtering units, normally unused for numerical computation. Our proposal features a rigorous mathematical analysis of the approximation error in function of the number of samples, as well as a method to obtain a suboptimal partition of the domain of the function to minimize approximation error.

Reflectarray antennas for dual polarization and broadband telecom satellite applications

A reflectarray antenna with improved performance is proposed to operate in dual-polarization and transmit-receive frequencies in Ku-band for broadcast satellite applications. The reflectarray element contains two orthogonal sets of four coplanar parallel dipoles printed on two surfaces, each set combining lateral and broadside coupling. A 40-cm prototype has been designed, manufactured, and tested. The lengths of the coupled dipoles in the reflectarray cells have been optimized to produce a collimated beam in dual polarization in the transmit and receive bands. The measured radiation patterns confirm the high performance of the antenna in terms of bandwidth (27%), low losses, and low levels of cross polarization. Some preliminary simulations at 11.95 GHz for a 1.2-m antenna with South American coverage are presented to show the potential of the proposed antenna for spaceborne antennas in Ku-band.

Yeast microarrays for genome wide parallel genetic and gene expression analysis

We have developed high-density DNA microarrays of yeast ORFs. These microarrays can monitor hybridization to ORFs for applications such as quantitative differential gene expression analysis and screening for sequence polymorphisms. Automated scripts retrieved sequence information from public databases to locate predicted ORFs and select appropriate primers for amplification. The primers were used to amplify yeast ORFs in 96-well plates, and the resulting products were arrayed using an automated micro arraying device. Arrays containing up to 2,479 yeast ORFs were printed on a single slide. The hybridization of fluorescently labeled samples to the array were detected and quantitated with a laser confocal scanning microscope. Applications of the microarrays are shown for genetic and gene expression analysis at the whole genome level.

Military and government applications of human-machine communication by voice.

This paper describes a range of opportunities for military and government applications of human-machine communication by voice, based on visits and contacts with numerous user organizations in the United States. The applications include some that appear to be feasible by careful integration of current state-of-the-art technology and others that will require a varying mix of advances in speech technology and in integration of the technology into applications environments. Applications that are described include (1) speech recognition and synthesis for mobile command and control; (2) speech processing for a portable multifunction soldier's computer; (3) speech- and language-based technology for naval combat team tactical training; (4) speech technology for command and control on a carrier flight deck; (5) control of auxiliary systems, and alert and warning generation, in fighter aircraft and helicopters; and (6) voice check-in, report entry, and communication for law enforcement agents or special forces. A phased approach for transfer of the technology into applications is advocated, where integration of applications systems is pursued in parallel with advanced research to meet future needs.

Sequential and parallel synchronous alternating iterative methods

The so-called parallel multisplitting nonstationary iterative Model A was introduced by Bru, Elsner, and Neumann [Linear Algebra and its Applications 103:175-192 (1988)] for solving a nonsingular linear system Ax = b using a weak nonnegative multisplitting of the first type. In this paper new results are introduced when A is a monotone matrix using a weak nonnegative multisplitting of the second type and when A is a symmetric positive definite matrix using a P -regular multisplitting. Also, nonstationary alternating iterative methods are studied. Finally, combining Model A and alternating iterative methods, two new models of parallel multisplitting nonstationary iterations are introduced. When matrix A is monotone and the multisplittings are weak nonnegative of the first or of the second type, both models lead to convergent schemes. Also, when matrix A is symmetric positive definite and the multisplittings are P -regular, the schemes are also convergent.

Parallel processing research in the former Soviet Union /

Mode of access: Internet.

Design and implementation of a windows-based parallel computing environment for large scale optimization

A parallel computing environment to support optimization of large-scale engineering systems is designed and implemented on Windows-based personal computer networks, using the master-worker model and the Parallel Virtual Machine (PVM). It is involved in decomposition of a large engineering system into a number of smaller subsystems optimized in parallel on worker nodes and coordination of subsystem optimization results on the master node. The environment consists of six functional modules, i.e. the master control, the optimization model generator, the optimizer, the data manager, the monitor, and the post processor. Object-oriented design of these modules is presented. The environment supports steps from the generation of optimization models to the solution and the visualization on networks of computers. User-friendly graphical interfaces make it easy to define the problem, and monitor and steer the optimization process. It has been verified by an example of a large space truss optimization. (C) 2004 Elsevier Ltd. All rights reserved.

Process systems modelling and applications in granulation: A review

Granulation is one of the fundamental operations in particulate processing and has a very ancient history and widespread use. Much fundamental particle science has occurred in the last two decades to help understand the underlying phenomena. Yet, until recently the development of granulation systems was mostly based on popular practice. The use of process systems approaches to the integrated understanding of these operations is providing improved insight into the complex nature of the processes. Improved mathematical representations, new solution techniques and the application of the models to industrial processes are yielding better designs, improved optimisation and tighter control of these systems. The parallel development of advanced instrumentation and the use of inferential approaches provide real-time access to system parameters necessary for improvements in operation. The use of advanced models to help develop real-time plant diagnostic systems provides further evidence of the utility of process system approaches to granulation processes. This paper highlights some of those aspects of granulation. (c) 2005 Elsevier Ltd. All rights reserved.

Focused, eight-element transceive phased array coil for parallel magnetic resonance imaging of the chest-theoretical considerations

A new transceive system for chest imaging for MRI applications is presented. A focused, eight-element transceive torso phased array coil is designed to investigate transmitting a focused radiofrequency field deep within the torso and to enhance signal homogeneity in the heart region. The system is used in conjunction with the SENSE reconstruction technique to enable focused parallel imaging. A hybrid finite-difference-time-domain/method-of-moments method is used to accurately predict the radiofrequency behavior inside the human torso. The simulation results reported herein demonstrate the feasibility of the design concept, which shows that radiofrequency field focusing with SENSE reconstruction is theoretically achievable. (c) 2005 Wiley-Liss, Inc.

Multiple-acquisition parallel imaging combined with a transceive array for the amelioration of high-field RF distortion: a modeling study

A new method for ameliorating high-field image distortion caused by radio frequency/tissue interaction is presented and modeled, The proposed method uses, but is not restricted to, a shielded four-element transceive phased array coil and involves performing two separate scans of the same slice with each scan using different excitations during transmission. By optimizing the amplitudes and phases for each scan, antipodal signal profiles can be obtained, and by combining both images together, the image distortion can be reduced several-fold. A hybrid finite-difference time-domain/method-of-moments method is used to theoretically demonstrate the method and also to predict the radio frequency behavior inside the human head. in addition, the proposed method is used in conjunction with the GRAPPA reconstruction technique to enable rapid imaging. Simulation results reported herein for IIT (470 MHz) brain imaging applications demonstrate the feasibility of the concept where multiple acquisitions using parallel imaging elements with GRAPPA reconstruction results in improved image quality. (c) 2006 Wiley Periodicals, Inc.

Parallel geostatistics for sparse and dense datasets

Very large spatially-referenced datasets, for example, those derived from satellite-based sensors which sample across the globe or large monitoring networks of individual sensors, are becoming increasingly common and more widely available for use in environmental decision making. In large or dense sensor networks, huge quantities of data can be collected over small time periods. In many applications the generation of maps, or predictions at specific locations, from the data in (near) real-time is crucial. Geostatistical operations such as interpolation are vital in this map-generation process and in emergency situations, the resulting predictions need to be available almost instantly, so that decision makers can make informed decisions and define risk and evacuation zones. It is also helpful when analysing data in less time critical applications, for example when interacting directly with the data for exploratory analysis, that the algorithms are responsive within a reasonable time frame. Performing geostatistical analysis on such large spatial datasets can present a number of problems, particularly in the case where maximum likelihood. Although the storage requirements only scale linearly with the number of observations in the dataset, the computational complexity in terms of memory and speed, scale quadratically and cubically respectively. Most modern commodity hardware has at least 2 processor cores if not more. Other mechanisms for allowing parallel computation such as Grid based systems are also becoming increasingly commonly available. However, currently there seems to be little interest in exploiting this extra processing power within the context of geostatistics. In this paper we review the existing parallel approaches for geostatistics. By recognising that diffeerent natural parallelisms exist and can be exploited depending on whether the dataset is sparsely or densely sampled with respect to the range of variation, we introduce two contrasting novel implementations of parallel algorithms based on approximating the data likelihood extending the methods of Vecchia [1988] and Tresp [2000]. Using parallel maximum likelihood variogram estimation and parallel prediction algorithms we show that computational time can be significantly reduced. We demonstrate this with both sparsely sampled data and densely sampled data on a variety of architectures ranging from the common dual core processor, found in many modern desktop computers, to large multi-node super computers. To highlight the strengths and weaknesses of the diffeerent methods we employ synthetic data sets and go on to show how the methods allow maximum likelihood based inference on the exhaustive Walker Lake data set.

Parallel implementation of image segmentation algorithms on a network of transputers

Image segmentation is one of the most computationally intensive operations in image processing and computer vision. This is because a large volume of data is involved and many different features have to be extracted from the image data. This thesis is concerned with the investigation of practical issues related to the implementation of several classes of image segmentation algorithms on parallel architectures. The Transputer is used as the basic building block of hardware architectures and Occam is used as the programming language. The segmentation methods chosen for implementation are convolution, for edge-based segmentation; the Split and Merge algorithm for segmenting non-textured regions; and the Granlund method for segmentation of textured images. Three different convolution methods have been implemented. The direct method of convolution, carried out in the spatial domain, uses the array architecture. The other two methods, based on convolution in the frequency domain, require the use of the two-dimensional Fourier transform. Parallel implementations of two different Fast Fourier Transform algorithms have been developed, incorporating original solutions. For the Row-Column method the array architecture has been adopted, and for the Vector-Radix method, the pyramid architecture. The texture segmentation algorithm, for which a system-level design is given, demonstrates a further application of the Vector-Radix Fourier transform. A novel concurrent version of the quad-tree based Split and Merge algorithm has been implemented on the pyramid architecture. The performance of the developed parallel implementations is analysed. Many of the obtained speed-up and efficiency measures show values close to their respective theoretical maxima. Where appropriate comparisons are drawn between different implementations. The thesis concludes with comments on general issues related to the use of the Transputer system as a development tool for image processing applications; and on the issues related to the engineering of concurrent image processing applications.

An intelligent approach for deploying applications in WSN's

Wireless sensor networks have been identified as one of the key technologies for the 21st century. They consist of tiny devices with limited processing and power capabilities, called motes that can be deployed in large numbers of useful sensing capabilities. Even though, they are flexible and easy to deploy, there are a number of considerations when it comes to their fault tolerance, conserving energy and re-programmability that need to be addressed before we draw any substantial conclusions about the effectiveness of this technology. In order to overcome their limitations, we propose a middleware solution. The proposed scheme is composed based on two main methods. The first method involves the creation of a flexible communication protocol based on technologies such as Mobile Code/Agents and Linda-like tuple spaces. In this way, every node of the wireless sensor network will produce and process data based on what is the best for it but also for the group that it belongs too. The second method incorporates the above protocol in a middleware that will aim to bridge the gap between the application layer and low level constructs such as the physical layer of the wireless sensor network. A fault tolerant platform for deploying and monitoring applications in real time offers a number of possibilities for the end user giving him in parallel the freedom to experiment with various parameters, in an effort towards the deployed applications running in an energy efficient manner inside the network. The proposed scheme is evaluated through a number of trials aiming to test its merits under real time conditions and to identify its effectiveness against other similar approaches. Finally, parameters which determine the characteristics of the proposed scheme are also examined.

Education Complex "Multi-agent Technologies for Parallel and Distributed Information Processing in Telecommunication Networks"

The paper describes education complex "Multi-agent Technologies for Parallel and Distributed Information Processing in Telecommunication Networks".

Parallel between Definition of Chess Playing Program and Definition of AI

In this report we will explain some earlier papers [1, 2] which are about definition of Artificial Intelligence and about perfect AI. The definition of AI is intuitive in [1] and formal in [2]. The perfect AI is a program that satisfies the definition for AI but which is absolutely useless because of the combinatory explosion. Most people do not understand these papers because they never saw AI and that is why for them the notion of AI is too abstract. In this report we will make parallel between definition of chess playing program and definition of AI. Of course, the definition of chess playing program is useless because people already know what this is. Anyway, we will give you this definition because its construction follows closely the construction of the definition of AI. Also the results are almost the same with the only difference that we can optimise the perfect chess playing program in order to obtain a real chess playing program, but for the moment we cannot optimise the perfect AI in order to obtain a real AI. In this report we will not speak about AI. The only matter which we will observe will be about chess playing programs. If you understand the construction and the results about chess playing programs then you can read the papers [1, 2] and to see similar results about AI.