Discrete visual features modeling via leave-one-out likelihood estimation and applications

Discrete data are an important component in many image processing and computer vision applications. In this work we propose an unsupervised statistical approach to learn structures of this kind of data. The central ingredient in our model is the introduction of the generalized Dirichlet distribution as a prior to the multinomial. An estimation algorithm, based on leave-one-out likelihood and empirical Bayesian inference, for the parameters is developed. This estimation algorithm can be viewed as a hybrid expectation–maximization (EM) which alternates EM iterations with Newton–Raphson iterations using the Hessian matrix. We propose then the use of our model as a parametric basis for support vector machines within a hybrid generative/discriminative framework. In a series of experiments involving scene modeling and classification using visual words, and color texture modeling we show the efficiency of the proposed approach.     

Symbolic-to-statistical hybridization: extending generation-heavy machine translation

The last few years have witnessed an increasing interest in hybridizing surface-based statistical approaches and rule-based symbolic approaches to machine translation (MT). Much of that work is focused on extending statistical MT systems with symbolic knowledge and components. In the brand of hybridization discussed here, we go in the opposite direction: adding statistical bilingual components to a symbolic system. Our base system is Generation-heavy machine translation (GHMT), a primarily symbolic asymmetrical approach that addresses the issue of Interlingual MT resource poverty in source-poor/target-rich language pairs by exploiting symbolic and statistical target-language resources. GHMT’s statistical components are limited to target-language models, which arguably makes it a simple form of a hybrid system. We extend the hybrid nature of GHMT by adding statistical bilingual components. We also describe the details of retargeting it to Arabic–English MT. The morphological richness of Arabic brings several challenges to the hybridization task. We conduct an extensive evaluation of multiple system variants. Our evaluation shows that this new variant of GHMT—a primarily symbolic system extended with monolingual and bilingual statistical components—has a higher degree of grammaticality than a phrase-based statistical MT system, where grammaticality is measured in terms of correct verb-argument realization and long-distance dependency translation.     

Optical resources provisioning: multipoint‐to‐point lightpaths mapping in all‐optical networks

In wavelength‐division multiplexing (WDM) optical networks, the bandwidth request of a traffic stream can be much lower than the capacity of a lightpath. Efficiently grooming low‐speed connections onto high‐capacity lightpaths will improve the network throughput and reduce the network cost. In this paper, we propose and evaluate a new concept of traffic aggregation in WDM mesh networks that aims to eliminate both the bandwidth under‐utilization and scalability concerns that are typical in all‐optical wavelength routed networks. This approach relies on the multipoint‐to‐point lightpath concept. In order to assess the efficiency of our proposal, all underlying network costs are compared. To achieve this aim, we devise a new provisioning algorithm to map the multipoint‐to‐point lightpaths in the network. Our results show that the proposed aggregation technique can significantly improve the network throughput while reducing its cost. Copyright © 2005 John Wiley & Sons, Ltd.     

Experimental design methodology applied to electro-Fenton treatment for degradation of herbicide chlortoluron

The degradation of herbicide chlortoluron in aqueous medium by electro-Fenton process using a carbon felt cathode and a platinum anode was studied. The great oxidation ability of this process is due to the large production of hydroxyl radical (OH) by electrochemically induced Fenton's reagent. Hydroxyl radicals are very powerful oxidizing agents which react on organics up to complete mineralization. The influence of some experimental parameters such as initial concentration, current intensity and processing time on the degradation and mineralization rate of chlortoluron by hydroxyl radicals has been investigated. The evolution of chlortoluron concentration with processing time shows a pseudo first order kinetics (k_abs = (4.8 ± 0.2) × 10⁹ mol⁻¹ L s⁻¹). A Doehlert matrix was applied for determination of the optimal working conditions. Optimal parameters for maximum mineralization efficiency (TOC removal ratio of 98%) was achieved after 8 h of treatment using a chlortoluron initial concentration of 0.125 mM and an applied current of 300 mA. The mineralization of aqueous chlortoluron solutions was confirmed by identification of the end-products such as carboxylic acids and inorganic ions. Their evolution during electro-Fenton treatment was studied.     

Power Utility Nontechnical Loss Analysis With Extreme Learning Machine Method

This paper presents a new approach to nontechnical loss (NTL) analysis for utilities using the modern computational technique extreme learning machine (ELM). Nontechnical losses represent a significant proportion of electricity losses in both developing and developed countries. The ELM-based approach presented here uses customer load-profile information to expose abnormal behavior that is known to be highly correlated with NTL activities. This approach provides a method of data mining for this purpose, and it involves extracting patterns of customer behavior from historical kWh consumption data. The results yield classification classes that are used to reveal whether any significant behavior that emerges is due to irregularities in consumption. In this paper, ELM and online sequential-ELM (OS-ELM) algorithms are both used to achieve an improved classification performance and to increase accuracy of results. A comparison of this approach with other classification techniques, such as the support vector machine (SVM) algorithm, is also undertaken and the ELM performance and accuracy in NTL analysis is shown to be superior.     

Study of the modification of the properties of (PP/EPR) blends with a view to preserving natural resources when elaborating new formulation and recycling polymers

The aim of the present work is to study how CaCO₃, very abundant on earth, can be blended with polypropylene/ethylene propylene rubber (PP/EPR) either to preserve natural resources when elaborating virgin formulations or to simulate mixtures of waste made of CaCO₃ filled PP with PP/EPR which could result from end-of-life-vehicles (ELV). The article focuses on the studies of PP/EPR blends, used in the manufacture of automobile fenders, in the presence of nano-CaCO₃ and compatibilizers. Blends of various compositions (with and without compatibilizer and nanoparticles) were prepared using a corotating twin-screw extruder. The results were compared with the ones presented by a commercial (PP/EPR) blend. The experiments included mechanical tests, differential scanning calorimetry, scanning electron microscopy, and dynamic mechanical analysis experiments. The presence of the compatibilizers in the (PP/EPR) blends decreases the size of elastomer particles, improves the adherence to the interface and improves the mechanical properties. The nano-CaCO₃ would also seem to act as a nucleating agent in the PP/EPR matrix; it increases the cristallinity and the Young modulus of the blends. POLYM. COMPOS., 2009. © 2008 Society of Plastics Engineers     

New terpenoid dyes, 3-arylazocamphor derivatives: Dyeing performance and fastness properties

The 3-arylazocamphor (I–VI) and 3-arylazocamphor-2-oxime (VII–XI) dyes were prepared, for use as azodisperse dyes by coupling camphor or camphoroxime respectively with the appropriate diazonium salt in cold ethanol containing sodium acetate solution (10%). The structure of these compounds (I–XI) was established by studying their elemental analysis, i.r. and u.v. spectra. The effect of the nature and position of substituent in the arylazo moiety on the colour, dyeing behaviour, rate of exhaustion and fastness properties when applied to protein and man-made fibres were discussed. The effect of after-treatment with different metal salts on the dyed woollen fibre was also investigated.     

Monte Carlo Estimation of a Joint Density Using Malliavin Calculus, and Application to American Options

We use the Malliavin integration by parts formula in order to provide a family of representations of the joint density (which does not involve Dirac measures) of (X_θ, X _{θ + δ}), where X is a d-dimensional Markov diffusion (d ≥ 1), θ > 0 and δ > 0. Following Bouchard et al. (2004), the different representations are determined by a pair of localizing functions. We discuss the problem of variance reduction within the family of separable localizing functions: We characterize a pair of exponential functions as the unique integrated-variance minimizer among this class of separable localizing functions. We test our method on the d-dimensional Brownian motion and provide an application to the problem of American options valuation by the quantization tree method introduced by Bally et al. (2002). MSC 1991 Subject Classifications: Primary 60H07 65C05, secondary 49-00     

System Design of Fast Pet Scanners Utilizing Time-of-Flight

Recent advances in PET designs have shown that a gain in signal-to-noise ratio can be expected by incorporating time-of-flight data in positron emission tomography over the conventional PET mode. It has also been shown that cesium fluoride (CsF) offers the potential of faster timing and high detection efficiency which would be required for a clinical scanner utilizing time-of-flight information. Our research with CsF and the results of a feasibility study of time-off-light positron emission tomography reconstruction have shown that, indeed, a significant improvement in image quality results from such an approach and that coincidence resolving times of less than 500 psec FWHM are easily achievable with CsF detectors. However, the design of fast tomographic systems with multiple detectors which maintain this fast coincidence timing poses a challenging technical problem. The solution to this problem requires a departure from the conventional mode of PET designs to a fast on-line microprocessor based system which is capable of compressing and correcting the data for timing differences, normalization and image function. Such a system is described in this paper and its advantages and disadvantages are discussed.     

Multi-objective optimization of a flexible slider-crank mechanism synthesis,based on dynamic responses

This work deals with a multi-body system synthesis. A flexible slider crank mechanism has been investigated as an illustrative application. The main interest is focused on the mechanism design variables’ identification based on its dynamic responses. Three responses have been involved such as the slider velocity, the slider acceleration and the mid-point transversal deflection of the flexible connecting rod. Each of these responses has been embroiled separately in a mono-objective optimization. Subsequently, the multi-objective optimization subsuming these responses has been established. Two different optimization methods have been studied namely the genetic algorithm (GA) and the particle swarm optimization (PSO) technique. It has been proved that the multi-objective optimization presents more accurate results beside the mono-objective optimization. Compared to the GA, the PSO is more powerful and is able to identify the mechanism design variable with better accuracy, in spite of the affordable computational time allowed with the GA optimization.