Model-based objects recognition in industrial environments for autonomous vehicles control

Behavior-based navigation of autonomous vehicles requires the recognition of the navigable areas and the potential obstacles. In this paper we describe a model-based objects recognition system which is part of an image interpretation system intended to assist the navigation of autonomous vehicles that operate in industrial environments. The recognition system integrates color, shape and texture information together with the location of the vanishing point. The recognition process starts from some prior scene knowledge, that is, a generic model of the expected scene and the potential objects. The recognition system constitutes an approach where different low-level vision techniques extract a multitude of image descriptors which are then analyzed using a rule-based reasoning system to interpret the image content. This system has been implemented using a rule-based cooperative expert system

Model-based objects recognition in man-made environments

We describe a model-based objects recognition system which is part of an image interpretation system intended to assist autonomous vehicles navigation. The system is intended to operate in man-made environments. Behavior-based navigation of autonomous vehicles involves the recognition of navigable areas and the potential obstacles. The recognition system integrates color, shape and texture information together with the location of the vanishing point. The recognition process starts from some prior scene knowledge, that is, a generic model of the expected scene and the potential objects. The recognition system constitutes an approach where different low-level vision techniques extract a multitude of image descriptors which are then analyzed using a rule-based reasoning system to interpret the image content. This system has been implemented using CEES, the C++ embedded expert system shell developed in the Systems Engineering and Automatic Control Laboratory (University of Girona) as a specific rule-based problem solving tool. It has been especially conceived for supporting cooperative expert systems, and uses the object oriented programming paradigm

Individual Differences in True and False Memory Retrieval Are Related to White Matter Brain Microstructure

We sometimes vividly remember things that did not happen, a phenomenon with general relevance, not only in the courtroom. It is unclear to what extent individual differences in false memories are driven by anatomical differences in memory-relevant brain regions. Here we show in humans that microstructural properties of different white matter tracts as quantified using diffusion tensor imaging are strongly correlated with true and false memory retrieval. To investigate these hypotheses, we tested a large group of participants in a version of the Deese-Roediger-McDermott paradigm (recall and recognition) and subsequently obtained diffusion tensor images. A voxel-based whole-brain level linear regression analysis was performedto relatefractional anisotropyto indices oftrue andfalse memory recall and recognition. True memory was correlated to diffusion anisotropy in the inferior longitudinal fascicle, the major connective pathway of the medial temporal lobe, whereas a greater proneness to retrieve false items was related to the superior longitudinal fascicle connecting frontoparietal structures. Our results show that individual differences in white matter microstructure underlie true and false memory performance.

Fair value versus historical cost-based valuation for biological assets: Predictability of financial information

There is an intense debate on the convenience of moving from historical cost (HC) toward the fair value (FV) principle. The debate and academic research is usually concerned with financial instruments, but the IAS 41 requirement of fair valuation for biological assets brings it into the agricultural domain. This paper performs an empirical study with a sample of Spanish farms valuing biological assets at HC and a sample applying FV, finding no significant differences between both valuation methods to assess future cash flows. However, most tests reveal more predictive power of future earnings under fair valuation of biological assets, which is not explained by differences in volatility of earnings and profitability. The study also evidences the existence of flawed HC accounting practices for biological assets in agriculture, which suggests scarce information content of this valuation method in the predominant small business units existing in the agricultural sector in advanced Western countries

Learning Preferences with Kernel-Based Methods

Learning of preference relations has recently received significant attention in machine learning community. It is closely related to the classification and regression analysis and can be reduced to these tasks. However, preference learning involves prediction of ordering of the data points rather than prediction of a single numerical value as in case of regression or a class label as in case of classification. Therefore, studying preference relations within a separate framework facilitates not only better theoretical understanding of the problem, but also motivates development of the efficient algorithms for the task. Preference learning has many applications in domains such as information retrieval, bioinformatics, natural language processing, etc. For example, algorithms that learn to rank are frequently used in search engines for ordering documents retrieved by the query. Preference learning methods have been also applied to collaborative filtering problems for predicting individual customer choices from the vast amount of user generated feedback. In this thesis we propose several algorithms for learning preference relations. These algorithms stem from well founded and robust class of regularized least-squares methods and have many attractive computational properties. In order to improve the performance of our methods, we introduce several non-linear kernel functions. Thus, contribution of this thesis is twofold: kernel functions for structured data that are used to take advantage of various non-vectorial data representations and the preference learning algorithms that are suitable for different tasks, namely efficient learning of preference relations, learning with large amount of training data, and semi-supervised preference learning. Proposed kernel-based algorithms and kernels are applied to the parse ranking task in natural language processing, document ranking in information retrieval, and remote homology detection in bioinformatics domain. Training of kernel-based ranking algorithms can be infeasible when the size of the training set is large. This problem is addressed by proposing a preference learning algorithm whose computation complexity scales linearly with the number of training data points. We also introduce sparse approximation of the algorithm that can be efficiently trained with large amount of data. For situations when small amount of labeled data but a large amount of unlabeled data is available, we propose a co-regularized preference learning algorithm. To conclude, the methods presented in this thesis address not only the problem of the efficient training of the algorithms but also fast regularization parameter selection, multiple output prediction, and cross-validation. Furthermore, proposed algorithms lead to notably better performance in many preference learning tasks considered.

Evolution and Stellar Content of AGN Host Galaxies

In this dissertation, Active Galactic Nuclei (AGN) and their host galaxies are discussed. Together with transitional events, such as supernovae and gamma-ray bursts, AGN are the most energetic phenomena in the Universe. The dominant fraction of their luminosity originates from the center of a galaxy, where accreting gas falls into a supermassive black hole, converting gravitational energy to radiation. AGN have a wide range of observed properties: e.g. in their emission lines, radio emission, and variability. Most likely, these properties depend significantly on their orientation to our line-of-sight, and to unify AGN into physical classes it is crucial to observe their orientation-independent properties, such as the host galaxies. Furthermore, host galaxy studies are essential to understand the formation and co-evolution of galactic bulges and supermassive black holes. In this thesis, the main focus is on observationally characterizing AGN host galaxies using optical and near-infrared imaging and spectroscopy. BL Lac objects are a class of AGN characterized by rapidly variable and polarized continuum emission across the electromagnetic spectrum, and coredominated radio emission. The near-infrared properties of intermediate redshift BL Lac host galaxies are studied in Paper I. They are found to be large elliptical galaxies that are more luminous than their low redshift counterparts suggesting a strong luminosity evolution, and a contribution from a recent star formation episode. To analyze the stellar content of galaxies in more detail multicolor data, especially observations at blue wavelengths, are essential. In Paper III, optical - near-infrared colors and color gradients are derived for low redshift BL Lac host galaxies. They show bluer colors and steeper color gradients than inactive ellipticals which, most likely, are caused by a relatively young stellar population indicating a different evolutionary stage between AGN hosts and inactive ellipticals. In Paper II, near-infrared imaging of intermediate redshift radio-quiet quasar hosts is used to study their luminosity evolution. The hosts are large elliptical galaxies, but they are systematically fainter than the hosts of radio-loud quasars at similar redshifts, suggesting a link between the luminosity of the host galaxies and the radio properties of AGN. In Paper IV, the characteristics of near-infrared stellar absorption features of low redshift radio galaxies are compared with those of inactive early-type galaxies. The comparison suggests that early-type galaxies with AGN are in a different evolutionary stage than their inactive counterparts. Moreover, radio galaxies are found to contain stellar populations containing both old and intermediate age components.

High-Performance-Tensile-Strength Alpha-Grass Reinforced Starch-Based Fully Biodegradable Composites

Though there has been a great deal of work concerning the development of natural fibers in reinforced starch-based composites, there is still more to be done. In general, cellulose fibers have lower strength than glass fibers; however, their specific strength is not far from that of fiberglass. In this work, alpha-fibers were obtained from alpha-grass through a mild cooking process. The fibers were used to reinforce a starch-based biopolymer. Composites including 5 to 35% (w/w) alpha-grass fibers in their formulation were prepared, tested, and subsequently compared with those of wood- and fiberglass-reinforced polypropylene (PP). The term “high-performance” refers to the tensile strength of the studied composites and is mainly due to a good interphase, a good dispersion of the fibers inside the matrix, and a good aspect ratio. The tensile strength of the composites showed a linear evolution for fiber contents up to 35% (w/w). The strain at break of the composites decreased with the fiber content and showed the stiffening effects of the reinforcement. The prepared composites showed high mechanical properties, even approaching those of glass fiber reinforced composites

Web-based corporate social responsibility reporting

Due to concerns regarding globalisation and sustainable development, corporate social responsibility (CSR) is topical in the business context and in the field of accounting. The main objective of this study was to review previous academic literature in the field of CSR reporting and develop an insight into CSR reporting in the Web-based environment. The main purpose was to find out what Web-based CSR reporting is like and how companies are utilising the Internet to communicate on responsibility issues. I did not, however, collect empirical research data but limited my study into theoretical and descriptive examination. In order to create an insight into Web-based reporting, I examined the development, motives and current practices of CSR reporting. I concluded that the Internet is a unique, interactive communication channel that is used differently compared with annual reports. The amount of companies engaging in Web-based CSR reporting is increasing and the reporting practices in terms of e.g. content and accessibility of information vary. I also concluded that many companies have not yet discovered the true potential of the Web as an interactive communication medium.

Improved Privacy-Preserving P2P Multimedia Distribution Based on Recombined Fingerprints

Peer-reviewed

Secure and Anonymous Multimedia Content Distribution in Peer-to-Peer Networks

Peer-reviewed

Selective and non-extractive spectrophotometric determination of cefdinir in formulations based on donor-acceptor complex formation

Cefdinir has broad spectrum of activity and high prescription rates, hence its counterfeiting seems imminent. We have proposed a simple, fast, selective and non-extractive spectrophotometric method for the content assay of cefdinir in formulations. The method is based on complexation of cefdinir and Fe under reducing condition in a buffered medium (pH 11) to form a magenta colored donor-acceptor complex (λ max = 550 nm; apparent molar absorptivity = 3720 L mol-1 cm-1). No other cephalosporins, penicillins and common excipients interfere under the test conditions. The Beer's law is followed in the concentration range 8-160 µg mL-1.

Application of ftir in the determination of acrylate content in poly(sodium acrylate-co-acrylamide) superabsorbent hydrogels

Hydrogels have been prepared by free-radical solution copolymerization of acrylamide and sodium acrylate (NaAc), with molar ratio ranging from 25/75 to 80/20, respectively, using methylene bisacrylamide as the crosslinking agent. A FTIR spectroscopy procedure to determine the acrylate/acrylamide ratio in these hydrogels was proposed based on absorbance at 1410 cm-1 (nCOO-) and 2940 cm-1 (nCH and nCH2). A straight line with a good linear correlation coefficient (0.998) was obtained by plotting the acrylate content (Ac%) versus relative absorbance (Arel = A1410/A2940). Results were confirmed by the amount of sodium cation released in acid medium determined by atomic absorption spectrometry.

Validation of an hplc method for the determination of 1,7-dihydroxy-2,3-methylenedioxyxanthone in extracts obtained from Polygala cyparissias A. St. Hil. & Moq. (Polygalaceae) and evaluation of the influence of its content on the gastroprotective activity

Polygala cyparissias is a plant widespread in Southern Latin America. Recently, we demonstrated the gastroprotective activity of the extract, as well as for one of the isolated metabolites-1,7-dihydroxy-2,3-methylenedioxyxanthone (MDX). In this study, a HPLC method for the quantification of MDX was validated. The HPLC method was linear (0.5-24 µg mL-1 of MDX) with good accuracy, precision and robustness. The content of MDX in the extracts from whole and different parts of the plant ranged from 0 to 5.4 mg g-1 and the gastroprotective index ranged from 72.1 to 99.1%. Thus, the method might be used for the standardization of the extracts based on the MDX marker.

A NOVEL METHOD FOR THE DETERMINATION OF TRACE THORIUM BY DISPERSIVE LIQUID-LIQUID MICROEXTRACTION BASED ON SOLIDIFICATION OF FLOATING ORGANIC DROP

In this study, dispersive liquid-liquid microextraction based on the solidification of floating organic droplets was used for the preconcentration and determination of thorium in the water samples. In this method, acetone and 1-undecanol were used as disperser and extraction solvents, respectively, and the ligand 1-(2-thenoyl)-3,3,3-trifluoracetone reagent (TTA) and Aliquat 336 was used as a chelating agent and an ion-paring reagent, for the extraction of thorium, respectively. Inductively coupled plasma-optical emission spectrometry was applied for the quantitation of the analyte after preconcentration. The effect of various factors, such as the extraction and disperser solvent, sample pH, concentration of TTA and concentration of aliquat336 were investigated. Under the optimum conditions, the calibration graph was linear within the thorium content range of 1.0-250 µg L-1 with a detection limit of 0.2 µg L-1. The method was also successfully applied for the determination of thorium in the different water samples.

Pathogenicity, molecular characterization, and cercosporin content of Brazilian isolates of Cercospora kikuchii

Cercospora kikuchii, involved with the defoliation of soybean (Glycine max) plants, is normally associated with Septoria glycines in late season. Seventy-two isolates from different regions in Brazil, obtained mainly from purple stained seeds, showed phenotypic variation. Cercosporin content and rate of colony growth was higly variable among isolates. A strong correlation between cercosporin content and virulence was identified. Genetic variation among and within populations was evaluated based on 86 RAPD loci. The RAPD analysis clustered all isolates into seven groups. No relationship was observed between RAPD groups and geographic origin or cercosporin content. The sequence of the internal spacer regions (ITS1-5.8S-ITS2) from 13 isolates chosen according to the previous RAPD and clustering analysis showed high similarity (97%-100%) to the GenBank sequences of C. kikuchii (AY266160, AY266161, AY152577 and AF291708). It is clear from this work that Brazilian isolates of C. kikuchii from different geographic regions, are variable in relation to virulence, RAPD profiles and cercosporin content. Cercosporin content could be a good parameter for choosing an adequate isolate for screening resistant or tolerant cultivars. Considering that this pathogen is easily seed-borne, findings are expected to show the same haplotypes in different regions. Migration could be favoured by infected seeds as demonstrated by RAPD analysis.