Moving Object Detection for Real-Time High-Quality Lightweight Applications on Smart Cameras

Here, a novel and efficient strategy for moving object detection by non-parametric modeling on smart cameras is presented. Whereas the background is modeled using only color information, the foreground model combines color and spatial information. The application of a particle filter allows the update of the spatial information and provides a priori information about the areas to analyze in the following images, enabling an important reduction in the computational requirements and improving the segmentation results

Mixing Snapshots and Fast Time Integration of PDEs

A local proper orthogonal decomposition (POD) plus Galerkin projection method was recently developed to accelerate time dependent numerical solvers of PDEs. This method is based on the combined use of a numerical code (NC) and a Galerkin sys- tem (GS) in a sequence of interspersed time intervals, INC and IGS, respectively. POD is performed on some sets of snapshots calculated by the numerical solver in the INC inter- vals. The governing equations are Galerkin projected onto the most energetic POD modes and the resulting GS is time integrated in the next IGS interval. The major computa- tional e®ort is associated with the snapshots calculation in the ¯rst INC interval, where the POD manifold needs to be completely constructed (it is only updated in subsequent INC intervals, which can thus be quite small). As the POD manifold depends only weakly on the particular values of the parameters of the problem, a suitable library can be con- structed adapting the snapshots calculated in other runs to drastically reduce the size of the ¯rst INC interval and thus the involved computational cost. The strategy is success- fully tested in (i) the one-dimensional complex Ginzburg-Landau equation, including the case in which it exhibits transient chaos, and (ii) the two-dimensional unsteady lid-driven cavity problem

Broadband Channel Long Delay Cluster Measurements and Analysis at 2.4GHz in Subway Tunnels

The delay caused by the reflected ray in broadband communication has a great influence on the communications in subway tunnel. This paper presents measurements taken in subway tunnels at 2.4 GHz, with 5 MHz bandwidth. According to propagation characteristics of tunnel, the measurements were carried out with a frequency domain channel sounding technique, in three typical scenarios: line of sight (LOS), Non-line-of-sight (NLOS) and far line of sight (FLOS), which lead to different delay distributions. Firstly IFFT was chosen to get channel impulse response (CIR) h(t) from measured three-dimensional transfer functions. Power delay profile (PDP) was investigated to give an overview of broadband channel model. Thereafter, a long delay caused by the obturation of tunnel is observed and investigated in all the scenarios. The measurements show that the reflection can be greatly remained by the tunnel, which leads to long delay cluster where the reflection, but direct ray, makes the main contribution for radio wave propagation. Four important parameters: distribution of whole PDP power, first peak arriving time, reflection cluster duration and PDP power distribution of reflection cluster were studied to give a detailed description of long delay characteristic in tunnel. This can be used to ensure high capacity communication in tunnels

Run-time Scalable Architecture for Deblocking Filtering in H.264/AVC-SVC Video Codecs

Systems relying on fixed hardware components with a static level of parallelism can suffer from an underuse of logical resources, since they have to be designed for the worst-case scenario. This problem is especially important in video applications due to the emergence of new flexible standards, like Scalable Video Coding (SVC), which offer several levels of scalability. In this paper, Dynamic and Partial Reconfiguration (DPR) of modern FPGAs is used to achieve run-time variable parallelism, by using scalable architectures where the size can be adapted at run-time. Based on this proposal, a scalable Deblocking Filter core (DF), compliant with the H.264/AVC and SVC standards has been designed. This scalable DF allows run-time addition or removal of computational units working in parallel. Scalability is offered together with a scalable parallelization strategy at the macroblock (MB) level, such that when the size of the architecture changes, MB filtering order is modified accordingly

Adaptive Multi-Camera System for Real Time Object Detection

In this paper we present an adaptive multi-camera system for real time object detection able to efficiently adjust the computational requirements of video processing blocks to the available processing power and the activity of the scene. The system is based on a two level adaptation strategy that works at local and at global level. Object detection is based on a Gaussian mixtures model background subtraction algorithm. Results show that the system can efficiently adapt the algorithm parameters without a significant loss in the detection accuracy.

Integrating software testing and run-time checking in an assertion verification framework

We have designed and implemented a framework that unifies unit testing and run-time verification (as well as static verification and static debugging). A key contribution of our approach is that a unified assertion language is used for all of these tasks. We first propose methods for compiling runtime checks for (parts of) assertions which cannot be verified at compile-time via program transformation. This transformation allows checking preconditions and postconditions, including conditional postconditions, properties at arbitrary program points, and certain computational properties. The implemented transformation includes several optimizations to reduce run-time overhead. We also propose a minimal addition to the assertion language which allows defining unit tests to be run in order to detect possible violations of the (partial) specifications expressed by the assertions. This language can express for example the input data for performing the unit tests or the number of times that the unit tests should be repeated. We have implemented the framework within the Ciao/CiaoPP system and effectively applied it to the verification of ISO-prolog compliance and to the detection of different types of bugs in the Ciao system source code. Several experimental results are presented that ¡Ilústrate different trade-offs among program size, running time, or levéis of verbosity of the messages shown to the user.

Using combined static analysis and profiling for logic program execution time estimation

Predicting statically the running time of programs has many applications ranging from task scheduling in parallel execution to proving the ability of a program to meet strict time constraints. A starting point in order to attack this problem is to infer the computational complexity of such programs (or fragments thereof). This is one of the reasons why the development of static analysis techniques for inferring cost-related properties of programs (usually upper and/or lower bounds of actual costs) has received considerable attention.

The CDG, UDG, and MEL methods for automatic compile-time parallelization of logic programs for independent and-parallelism

There has been significant interest in parallel execution models for logic programs which exploit Independent And-Parallelism (IAP). In these models, it is necessary to determine which goals are independent and therefore eligible for parallel execution and which goals have to wait for which others during execution. Although this can be done at run-time, it can imply a very heavy overhead. In this paper, we present three algorithms for automatic compiletime parallelization of logic programs using IAP. This is done by converting a clause into a graph-based computational form and then transforming this graph into linear expressions based on &-Prolog, a language for IAP. We also present an algorithm which, given a clause, determines if there is any loss of parallelism due to linearization, for the case in which only unconditional parallelism is desired. Finally, the performance of these annotation algorithms is discussed for some benchmark programs.

Computational methods to create and analyze a digital gene expression atlas of embryo development from microscopy images

Abstract The creation of atlases, or digital models where information from different subjects can be combined, is a field of increasing interest in biomedical imaging. When a single image does not contain enough information to appropriately describe the organism under study, it is then necessary to acquire images of several individuals, each of them containing complementary data with respect to the rest of the components in the cohort. This approach allows creating digital prototypes, ranging from anatomical atlases of human patients and organs, obtained for instance from Magnetic Resonance Imaging, to gene expression cartographies of embryo development, typically achieved from Light Microscopy. Within such context, in this PhD Thesis we propose, develop and validate new dedicated image processing methodologies that, based on image registration techniques, bring information from multiple individuals into alignment within a single digital atlas model. We also elaborate a dedicated software visualization platform to explore the resulting wealth of multi-dimensional data and novel analysis algo-rithms to automatically mine the generated resource in search of bio¬logical insights. In particular, this work focuses on gene expression data from developing zebrafish embryos imaged at the cellular resolution level with Two-Photon Laser Scanning Microscopy. Disposing of quantitative measurements relating multiple gene expressions to cell position and their evolution in time is a fundamental prerequisite to understand embryogenesis multi-scale processes. However, the number of gene expressions that can be simultaneously stained in one acquisition is limited due to optical and labeling constraints. These limitations motivate the implementation of atlasing strategies that can recreate a virtual gene expression multiplex. The developed computational tools have been tested in two different scenarios. The first one is the early zebrafish embryogenesis where the resulting atlas constitutes a link between the phenotype and the genotype at the cellular level. The second one is the late zebrafish brain where the resulting atlas allows studies relating gene expression to brain regionalization and neurogenesis. The proposed computational frameworks have been adapted to the requirements of both scenarios, such as the integration of partial views of the embryo into a whole embryo model with cellular resolution or the registration of anatom¬ical traits with deformable transformation models non-dependent on any specific labeling. The software implementation of the atlas generation tool (Match-IT) and the visualization platform (Atlas-IT) together with the gene expression atlas resources developed in this Thesis are to be made freely available to the scientific community. Lastly, a novel proof-of-concept experiment integrates for the first time 3D gene expression atlas resources with cell lineages extracted from live embryos, opening up the door to correlate genetic and cellular spatio-temporal dynamics. La creación de atlas, o modelos digitales, donde la información de distintos sujetos puede ser combinada, es un campo de creciente interés en imagen biomédica. Cuando una sola imagen no contiene suficientes datos como para describir apropiadamente el organismo objeto de estudio, se hace necesario adquirir imágenes de varios individuos, cada una de las cuales contiene información complementaria respecto al resto de componentes del grupo. De este modo, es posible crear prototipos digitales, que pueden ir desde atlas anatómicos de órganos y pacientes humanos, adquiridos por ejemplo mediante Resonancia Magnética, hasta cartografías de la expresión genética del desarrollo de embrionario, típicamente adquiridas mediante Microscopía Optica. Dentro de este contexto, en esta Tesis Doctoral se introducen, desarrollan y validan nuevos métodos de procesado de imagen que, basándose en técnicas de registro de imagen, son capaces de alinear imágenes y datos provenientes de múltiples individuos en un solo atlas digital. Además, se ha elaborado una plataforma de visualization específicamente diseñada para explorar la gran cantidad de datos, caracterizados por su multi-dimensionalidad, que resulta de estos métodos. Asimismo, se han propuesto novedosos algoritmos de análisis y minería de datos que permiten inspeccionar automáticamente los atlas generados en busca de conclusiones biológicas significativas. En particular, este trabajo se centra en datos de expresión genética del desarrollo embrionario del pez cebra, adquiridos mediante Microscopía dos fotones con resolución celular. Disponer de medidas cuantitativas que relacionen estas expresiones genéticas con las posiciones celulares y su evolución en el tiempo es un prerrequisito fundamental para comprender los procesos multi-escala característicos de la morfogénesis. Sin embargo, el número de expresiones genéticos que pueden ser simultáneamente etiquetados en una sola adquisición es reducido debido a limitaciones tanto ópticas como del etiquetado. Estas limitaciones requieren la implementación de estrategias de creación de atlas que puedan recrear un multiplexado virtual de expresiones genéticas. Las herramientas computacionales desarrolladas han sido validadas en dos escenarios distintos. El primer escenario es el desarrollo embrionario temprano del pez cebra, donde el atlas resultante permite constituir un vínculo, a nivel celular, entre el fenotipo y el genotipo de este organismo modelo. El segundo escenario corresponde a estadios tardíos del desarrollo del cerebro del pez cebra, donde el atlas resultante permite relacionar expresiones genéticas con la regionalización del cerebro y la formación de neuronas. La plataforma computacional desarrollada ha sido adaptada a los requisitos y retos planteados en ambos escenarios, como la integración, a resolución celular, de vistas parciales dentro de un modelo consistente en un embrión completo, o el alineamiento entre estructuras de referencia anatómica equivalentes, logrado mediante el uso de modelos de transformación deformables que no requieren ningún marcador específico. Está previsto poner a disposición de la comunidad científica tanto la herramienta de generación de atlas (Match-IT), como su plataforma de visualización (Atlas-IT), así como las bases de datos de expresión genética creadas a partir de estas herramientas. Por último, dentro de la presente Tesis Doctoral, se ha incluido una prueba conceptual innovadora que permite integrar los mencionados atlas de expresión genética tridimensionales dentro del linaje celular extraído de una adquisición in vivo de un embrión. Esta prueba conceptual abre la puerta a la posibilidad de correlar, por primera vez, las dinámicas espacio-temporales de genes y células.

Integrating software testing and run-time checking in an assertion verification framework.

We have designed and implemented a framework that unifies unit testing and run-time verification (as well as static verification and static debugging). A key contribution of our approach is that a unified assertion language is used for all of these tasks. We first propose methods for compiling runtime checks for (parts of) assertions which cannot be verified at compile-time via program transformation. This transformation allows checking preconditions and postconditions, including conditional postconditions, properties at arbitrary program points, and certain computational properties. The implemented transformation includes several optimizations to reduce run-time overhead. We also propose a minimal addition to the assertion language which allows defining unit tests to be run in order to detect possible violations of the (partial) specifications expressed by the assertions. This language can express for example the input data for performing the unit tests or the number of times that the unit tests should be repeated. We have implemented the framework within the Ciao/CiaoPP system and effectively applied it to the verification of ISO-prolog compliance and to the detection of different types of bugs in the Ciao system source code. Several experimental results are presented that illustrate different trade-offs among program size, running time, or levels of verbosity of the messages shown to the user.

Computational Fluid Dynamics Expert System using Artificial Neural Networks

The design of a modern aircraft is based on three pillars: theoretical results, experimental test and computational simulations. As a results of this, Computational Fluid Dynamic (CFD) solvers are widely used in the aeronautical field. These solvers require the correct selection of many parameters in order to obtain successful results. Besides, the computational time spent in the simulation depends on the proper choice of these parameters. In this paper we create an expert system capable of making an accurate prediction of the number of iterations and time required for the convergence of a computational fluid dynamic (CFD) solver. Artificial neural network (ANN) has been used to design the expert system. It is shown that the developed expert system is capable of making an accurate prediction the number of iterations and time required for the convergence of a CFD solver.

SVis: A computational steering visualization environment for surface structure determination

The arrangement of atoms at the surface of a solid accounts for many of its properties: Hardness, chemical activity, corrosion, etc. are dictated by the precise surface structure. Hence, finding it, has a broad range of technical and industrial applications. The ability to solve this problem opens the possibility of designing by computer materials with properties tailored to specific applications. Since the search space grows exponentially with the number of atoms, its solution cannot be achieved for arbitrarily large structures. Presently, a trial and error procedure is used: an expert proposes an structure as a candidate solution and tries a local optimization procedure on it. The solution relaxes to the local minimum in the attractor basin corresponding to the initial point, that might be the one corresponding to the global minimum or not. This procedure is very time consuming and, for reasonably sized surfaces, can take many iterations and much effort from the expert. Here we report on a visualization environment designed to steer this process in an attempt to solve bigger structures and reduce the time needed. The idea is to use an immersive environment to interact with the computation. It has immediate feedback to assess the quality of the proposed structure in order to let the expert explore the space of candidate solutions. The visualization environment is also able to communicate with the de facto local solver used for this problem. The user is then able to send trial structures to the local minimizer and track its progress as they approach the minimum. This allows for simultaneous testing of candidate structures. The system has also proved very useful as an educational tool for the field.

Computational Methods for Identification of Vibrating Structures

System identification deals with the problem of building mathematical models of dynamical systems based on observed data from the system" [1]. In the context of civil engineering, the system refers to a large scale structure such as a building, bridge, or an offshore structure, and identification mostly involves the determination of modal parameters (the natural frequencies, damping ratios, and mode shapes). This paper presents some modal identification results obtained using a state-of-the-art time domain system identification method (data-driven stochastic subspace algorithms [2]) applied to the output-only data measured in a steel arch bridge. First, a three dimensional finite element model was developed for the numerical analysis of the structure using ANSYS. Modal analysis was carried out and modal parameters were extracted in the frequency range of interest, 0-10 Hz. The results obtained from the finite element modal analysis were used to determine the location of the sensors. After that, ambient vibration tests were conducted during April 23-24, 2009. The response of the structure was measured using eight accelerometers. Two stations of three sensors were formed (triaxial stations). These sensors were held stationary for reference during the test. The two remaining sensors were placed at the different measurement points along the bridge deck, in which only vertical and transversal measurements were conducted (biaxial stations). Point estimate and interval estimate have been carried out in the state space model using these ambient vibration measurements. In the case of parametric models (like state space), the dynamic behaviour of a system is described using mathematical models. Then, mathematical relationships can be established between modal parameters and estimated point parameters (thus, it is common to use experimental modal analysis as a synonym for system identification). Stable modal parameters are found using a stabilization diagram. Furthermore, this paper proposes a method for assessing the precision of estimates of the parameters of state-space models (confidence interval). This approach employs the nonparametric bootstrap procedure [3] and is applied to subspace parameter estimation algorithm. Using bootstrap results, a plot similar to a stabilization diagram is developed. These graphics differentiate system modes from spurious noise modes for a given order system. Additionally, using the modal assurance criterion, the experimental modes obtained have been compared with those evaluated from a finite element analysis. A quite good agreement between numerical and experimental results is observed.

Selection of TDOA Parameters for MDM Speaker Diarization

Several methods to improve multiple distant microphone (MDM) speaker diarization based on Time Delay of Arrival (TDOA) features are evaluated in this paper. All of them avoid the use of a single reference channel to calculate the TDOA values and, based on different criteria, select among all possible pairs of microphones a set of pairs that will be used to estimate the TDOA's. The evaluated methods have been named the "Dynamic Margin" (DM), the "Extreme Regions" (ER), the "Most Common" (MC), the "Cross Correlation" (XCorr) and the "Principle Component Analysis" (PCA). It is shown that all methods improve the baseline results for the development set and four of them improve also the results for the evaluation set. Improvements of 3.49% and 10.77% DER relative are obtained for DM and ER respectively for the test set. The XCorr and PCA methods achieve an improvement of 36.72% and 30.82% DER relative for the test set. Moreover, the computational cost for the XCorr method is 20% less than the baseline.

Time and space partition platform for safe and secure flight software.

There are a number of research and development activities that are exploring Time and Space Partition (TSP) to implement safe and secure flight software. This approach allows to execute different real-time applications with different levels of criticality in the same computer board. In order to do that, flight applications must be isolated from each other in the temporal and spatial domains. This paper presents the first results of a partitioning platform based on the Open Ravenscar Kernel (ORK+) and the XtratuM hypervisor. ORK+ is a small, reliable real-time kernel supporting the Ada Ravenscar Computational model that is central to the ASSERT development process. XtratuM supports multiple virtual machines, i.e. partitions, on a single computer and is being used in the Integrated Modular Avionics for Space study. ORK+ executes in an XtratuM partition enabling Ada applications to share the computer board with other applications.