Dissertations / Theses on the topic 'Mono camera'

Advanced driver assistance systems (ADAS) is a popular and evolving area of research and development. By providing assistance to the vehicle drivers, ADAS could significantly reduce the number of traffic accidents since 90 % of all accidentsare caused by the human factor. ADAS with cameras provides a wide field of view and thanks to today’s advanced image processing techniques, lots of informationcan be extracted from the camera image. This thesis proposes a method of estimating the heading of vehicles using a mono camera system. The method consists of an extended Kalman filter with a constant velocity motion model to predict the vehicle’s path, fed by classification measurements from machine learning algorithms together with angular rate measurements. Monte Carlo simulations performed in this thesis show promising results. The results on real-world data indicate that the method used to construct the angular rate measurements must be improved in order to reach the same results as obtained from the simulations. An additional measurement, the vehicle’s corners, is introduced in order to further provide the filter with information. The thesis shows that the mono camera system needs further improvements in order to reach the same level of performance as a stereo camera system.

Recently, sensors such as radars and cameras have been widely used in automotives, especially in Advanced Driver-Assistance Systems (ADAS), to collect information about the vehicle's surroundings. Stereo cameras are very popular as they could be used passively to construct a 3D representation of the scene in front of the car. This allowed the development of several ADAS algorithms that need 3D information to perform their tasks. One interesting application is Road Surface Preview (RSP) where the task is to estimate the road height along the future path of the vehicle. An active suspension control unit can then use this information to regulate the suspension, improving driving comfort, extending the durabilitiy of the vehicle and warning the driver about potential risks on the road surface. Stereo cameras have been successfully used in RSP and have demonstrated very good performance. However, the main disadvantages of stereo cameras are their high production cost and high power consumption. This limits installing several ADAS features in economy-class vehicles. A less expensive alternative are monocular cameras which have a significantly lower cost and power consumption. Therefore, this thesis investigates the possibility of solving the Road Surface Preview task using a monocular camera. We try two different approaches: structure-from-motion and Convolutional Neural Networks.The proposed methods are evaluated against the stereo-based system. Experiments show that both structure-from-motion and CNNs have a good potential for solving the problem, but they are not yet reliable enough to be a complete solution to the RSP task and be used in an active suspension control unit.

La diffusion Raman spontanée est un processus physique qui permet de caractériser la composition d’un système chimique au niveau moléculaire. Ne nécessitant pas de marqueur mais simplement une source lumineuse, elle est utilisée dans de nombreux domaines, aussi variés que la recherche biomédicale ou que le contrôle qualité industriel. Néanmoins, l’efficience de ce simple processus est faible, et les temps d’acquisition qui en résultent sont donc conséquents. De plus, les méthodes d’acquisition conventionnelles du signal Raman génèrent de larges volumes de données et utilisent des détecteurs parfois peu sensibles, ce qui amplifie les temps d’acquisition déjà longs. Ainsi, l’utilisation du Raman spontané – processus pourtant simple et efficace – est alors limitée, particulièrement en biologie. Il semble donc pertinent de chercher à rendre la technique plus efficiente, c’est-à-dire plus rapide tout en conservant sa simplicité. Pour ce faire, dans le cadre de cette thèse, nous nous plaçons dans le cas où les espèces chimiques contenues dans l’échantillon sont déjà connues, et nous cherchons à estimer leurs concentrations respectives. Nous concevons un montage optique en utilisant des outils de traitement du signal, afin de rendre l’acquisition plus efficiente. Nous montrons des résultats préliminaires qui suggèrent l’utilité de la technique pour certaines applications, dans les domaines de l’imagerie biomédicale, de l’industrie pharmaceutique et de l’environnement
Spontaneous Raman scattering is a physical process that provides a unique knowledge of materials at the molecular level. Since it does not require any label but simply a source of light, it is used in many different fields, ranging from biomedical research to industrial quality control. Nevertheless, this simple process has a limited efficiency and requires substantial acquisition times. In addition, the conventional methods to acquire the Raman signal generate overwhelmingly large data sets, and sometimes use detectors with limited sensitivity, which amplify the already long acquisition times. Thus, the use of spontaneous Raman – yet effective and simple – is limited, especially in biology. Therefore, it seems relevant to try to make the technique more efficient, that is to say faster while keeping its simplicity. To do so, in this thesis work, we consider that the chemical species contained in the sample are a priori known, and we wish to estimate their respective concentrations. We design an optical setup that uses signal processing tools to make the acquisition more efficient. We show some preliminary results that suggest the technique usefulness for fields related to biomedical imaging, pharmaceutical industry and the environment

Dans un travail réalisé au Centre de Robotique et à l'Institut VEDECOM, nous nous sommes intéressés aux systèmes robustes de localisation visuelle en milieu urbain pour la voiture autonome. Obtenir une pose exacte à partir d'une mono-caméra est difficile et insuffisant en terme de précision pour la voiture autonome actuelle. Nous nous sommes concentrés sur l'utilisation de Systèmes d'Information Géographiques (SIG) pour concevoir une approche fiable, précise et absolue de localisation en milieu urbain.Le développement de SIG publics nous a apporté un nouvel horizon pour résoudre le problème de la localisation, mais ses informations, telles que les cartes topologiques, sémantiques, métriques, les Street Views, les cartes de profondeur, les cartes cadastrales 3D et les cartes en haute définition, doivent être bien analysées et organisées pour extraire les informations pertinentes pour une voiture autonome. Notre première tâche consistait à concevoir une base de données hors ligne accessible par un robot à partir d'un SIG public dense, à savoir Google Maps, qui a l'avantage d'avoir une couverture mondiale. Nous générons une représentation topométrique compacte de l'environnement urbain en extrayant quatre données utiles du SIG, y compris : les topologies, les géo-coordonnées, les Street Views panoramiques et les cartes de profondeur associées. Dans le même temps, un ensemble de données en ligne a été acquis par une mono-caméra équipée sur les véhicules de VEDECOM. Afin de rendre les Street View sphériques compatibles avec l'imagerie en ligne, une transformation basée sur l'interpolation d'image est introduite pour obtenir des images rectilignes à partir de Street Views.Nous proposons deux méthodes de localisation : l'une est une approche de vision par ordinateur basée sur l'extraction de caractéristiques, l'autre est une méthode d'apprentissage basée sur les réseaux de neurones convolutionnels (convnet). En vision par ordinateur, l'extraction de caractéristiques est un moyen populaire de résoudre le positionnement à partir d'images. Nous tirons parti de Google Maps et utilisons ses données topo-métriques hors ligne pour construire un positionnement grossier à fin, à savoir un processus de reconnaissance de lieu topologique puis une estimation métrique de pose par optimisation de graphe. La méthode a été testée en environnement urbain et démontre à la fois une précision sous-métrique et une robustesse aux changements de point de vue, à l'illumination et à l'occlusion. Aussi, les résultats montrent que les emplacements éloignés de Street Views produisent une erreur significative dans la phase d'estimation métrique. Ainsi, nous proposons de synthétiser des Street Views artificielles pour compenser la densité des Street View originales et améliorer la précision.Cette méthode souffre malheureusement d'un temps de calcul important. Étant donné que le SIG nous offre une base de données géolocalisée à l'échelle mondiale, cela nous motive à régresser des localisations globales directement à partir d'un convnet de bout en bout. La base de données hors ligne précédemment construite est encore insuffisante pour l'apprentissage d'un convnet. Pour compenser cela nous densifions la base d'origine d'un facteur mille et utilisons la méthode d'apprentissage par transfert pour faire converger notre régresseur convnet et avoir une bonne performance. Le régresseur permet également d'obtenir une localisation globale à partir d'une seule image et en temps réel.Les résultats obtenus par ces deux approches nous fournissent des informations sur la comparaison et la relation entre les méthodes basées sur des caractéristiques et celles basées sur le convnet. Après avoir analysé et comparé les performances de localisation des deux méthodes, nous avons également abordé des perspectives pour améliorer la robustesse et la précision de la localisation face au problème de localisation urbaine assistée par SIG
In a work made at Centre de Robotique and Institut VEDECOM, we studied robust visual urban localization systems for self-driving cars. Obtaining an exact pose from a monocular camera is difficult and cannot be applied to the current autonomous cars. We mainly focused on fully leveraging Geographical Information Systems (GIS) to achieve a low-cost, robust, accurate and global urban localization.The development of public GIS's has brought us a new horizon to address the localization problem but their tremendous amount of information, such as topological, semantic, metric maps, Street Views, depth maps, 3D cadastral maps and High Definition maps, has to be well analyzed and organized to extract relevant information for self-driving cars. Our first task was to design a robotic accessible offline database from a dense public GIS, namely Google Maps, which has the advantage to propose a worldwide coverage. We make a compact topometric representation for the dynamic urban environment by extracting four useful data from the GIS, including topologies, geo-coordinates, panoramic Street Views, and associated depth maps. At the same time, an online dataset was acquired with a low-cost camera equipped on VEDECOM vehicles. In order to make spheric Street Views compatible with the online imagery, an image warping and interpolation based transformation is introduced to render rectilinear images from Street Views.We proposed two localization methods: one is a handcrafted-features-based computer vision approach, the other is a convolutional neural network (convnet) based learning technique. In computer vision, extracting handcrafted features is a popular way to solve the image based positioning. We take advantages of the abundant sources from Google Maps and benefit from the topometric offline data structure to build a coarse-to-fine positioning, namely a topological place recognition process and then a metric pose estimation by a graph optimization. The method is tested on an urban environment and demonstrates both sub-meter accuracy and robustness to viewpoint changes, illumination and occlusion. Moreover, we demonstrate that sparse Street View locations produce a significant error in the metric pose estimation phase. Thus our former framework is refined by synthesizing more artificial Street Views to compensate the sparsity of original Street Views and improve the precision.The handcrafted feature based framework requires the image retrieval and graph optimization. It is hard to achieve in a real-time application. Since the GIS offers us a global scale geotagged database, it motivates us to regress global localizations from convnet features in an end-to-end manner. The previously constructed offline database is still insufficient for a convnet training. We hereby augment the originally constructed database by a thousand factor and take advantage of the transfer learning method to make our convnet regressor converge and have a good performance. In our test, the regressor can also give a global localization of an input camera image in real time.The results obtained by the two approaches provide us insights on the comparison and connection between handcrafted feature-based and convnet based methods. After analyzing and comparing the localization performances of both methods, we also talked about some perspectives to improve the localization robustness and precision towards the GIS-aided urban localization problem

This master-thesis presents an approach to track and count the number of fruit incommercial mango orchards. The algorithm is intended to enable precision agri-culture and to facilitate labour and post-harvest storage planning. The primary objective is to develop an multi-view algorithm and investigate how it can beused to mitigate the effects of visual occlusion, to improve upon estimates frommethods that use a single central or two opposite viewpoints. Fruit are detectedin images by using two classification methods: dense pixel-wise cnn and regionbased r-cnn detection. Pair-wise fruit correspondences are established between images by using geometry provided by navigation data, and lidar data is used to generate image masks for each separate tree, to isolate fruit counts to individual trees. The tracked fruit are triangulated to locate them in 3D space, and spatial statistics are calculated over whole orchard blocks. The estimated tree counts are compared to single view estimates and validated against ground truth data of 16 mango trees from a Bundaberg mango orchard in Queensland, Australia. The results show a high R2-value of 0.99335 for four hand labelled trees and a highest R2-value of 0.9165 for the machine labelled images using the r-cnn classifier forthe 16 target trees.

The car has become bigger, faster and more advanced for each passing year since its first appearance, and the safety requirements have also become stricter. Computer vision based support is a growing area of safety features where the car is equipped with a mono- or stereo camera. It can be used for detecting pedestrians walking out in the street, give a warning for wild-life during a cold January night using night-vision cameras and much more. This master thesis investigates the problem of detecting and tracking overtaking vehicles. Vehicles that overtake are only partly visible in the beginning, rendering it hard for standard detection/classification algorithms to get a positive detection. The need to quickly detect an incoming vehicle is crucial to be able to take fast counter-measure, such as braking, if needed. A novel approach referred to as the \textit{Wall detector} is suggested, detecting incoming vehicles using one-dimensional optical flow. Under the assumption that an overtaking car is moving in parallel to the ego-vehicle, both cars are moving towards the vanishing point in the image. A detection wall, consisting of several detection lines moving towards the vanishing point, is created, making all objects that are moving parallel to the ego-vehicle move along these lines. The result is a light-weight and fast detector with good detection performance in real-time. Several approaches for the Wall detector are implemented and evaluated, revealing that a feature based approach is the best choice. The information from the system can be used as input to heavier algorithms, boosting the confidence or to initialize a track.

La caméra Compton est un dispositif permettant d’imager les sources de rayonnement gamma. Ses avantages sont sa sensibilité (absence de collimateur mécanique) et la possibilité de reconstruire des images 3D avec un dispositif immobile. Elle également adaptée pour des sources à large spectre énergétique. Ce dispositif est un candidat prometteur en médecine nucléaire et en hadronthérapie. Ces travaux, financés par le projet européen ENVISION (European NoVel Imaging Systems for ION therapy) Coopération-FP7, portent sur le développement de méthodes de reconstruction d’images pour la caméra Compton pour la surveillance de la thérapie par ions. Celle-ci nécessite idéalement une reconstruction temps réel avec une précision millimétrique, même si le nombre de données acquises est relativement faible. Nous avons développé des méthodes analytiques et itératives. Leurs performances sont analysées dans le contexte d’acquisitions réalistes (géométrie de la caméra, nombre d’événements). Nous avons développé une méthode analytique de rétroprojection filtrée. Cette méthode est rapide mais nécessite beaucoup de données. Nous avons également développé des méthodes itératives utilisant un algorithme de maximisation de la fonction de vraisemblance. Nous avons proposé un modèle probabiliste pour l’estimation des éléments de la matrice système nécessaire à la reconstruction et nous avons développé différentes approches pour le calcul de ses éléments : l’une néglige les incertitudes de mesure sur l’énergie, l’autre les prend en compte en utilisant une distribution gaussienne. Nous avons étudié une méthode simplifiée utilisant notre modèle probabiliste. Plusieurs reconstructions sont menées à partir de données simulées, obtenues avec Geant4, mais provenant aussi de plusieurs prototypes simulés de caméra Compton proposés par l’Institut de Physique Nucléaire de Lyon (IPNL) et par le Centre de recherche de Dresde-Rossendorf en Allemagne. Les résultats sont prometteurs et des études plus poussées, à partir de données encore plus réalistes, viseront à les confirmer
The Compton camera is a device for imaging gamma radiation sources. The advantages of the system lie in its sensitivity, due to the absence of mechanical collimator, and the possibility of imaging wide energy spectrum sources. These advantages make it a promising candidate for application in hadrontherapy. Funded by the european project ENVISION, FP7-Cooperation Work Program, this work deals with the development of image reconstruction methods for the Compton camera. We developed both analytical and iterative methods in order to reconstruct the source from cone-surface projections. Their performances are analyzed with regards to the context (geometry of the camera, number of events). We developped an analytical method using a Filtered BackProjection (FBP) formulation. This method is fast but really sensitive to the noise. We have also developped iterative methods using a List Mode-Maximum Likelihood Expectation Maximization (LM-MLEM) algorithm. We proposed a new probabilistic model for the computation of the elements of the system matrix and different approaches for the calculation of these elements neglecting or not the measurement uncertainties. We also implemented a simplified method using the probabilistic model we proposed. The novelty of the method also lies on the specific discretization of the cone-surface projections. Several studies are carried out upon the reconstructions of simulated data worked out with Geant4, but also simulated data obtained from several prototypes of Compton cameras under study at the Institut de Physique Nucléaire de Lyon (IPNL) and at the Research Center of Dresden-Rossendorf. Results are promising, and further investigations on more realistic data are to be done

The purpose of this master thesis was to study computer vision algorithms for vehicle detection in monochrome images captured by mono camera. The work has mainly been focused on detecting rear-view cars in daylight conditions. Previous work in the literature have been revised and algorithms based on edges, shadows and motion as vehicle cues have been modified, implemented and evaluated. This work presents a combination of a multiscale edge based detection and a shadow based detection as the most promising algorithm, with a positive detection rate of 96.4% on vehicles at a distance of between 5 m to 30 m. For the algorithm to work in a complete system for vehicle detection, future work should be focused on developing a vehicle classifier to reject false detections.

Vision based active safety systems have become more frequently occurring in modern vehicles to estimate depth of the objects ahead and for autonomous driving (AD) and advanced driver-assistance systems (ADAS). In this thesis a lightweight deep convolutional neural network performing real-time depth estimation on single monocular images is implemented and evaluated. Many of the vision based automatic brake systems in modern vehicles only detect pre-trained object types such as pedestrians and vehicles. These systems fail to detect general objects such as road debris and roadside obstacles. In stereo vision systems the problem is resolved by calculating a disparity image from the stereo image pair to extract depth information. The distance to an object can also be determined using radar and LiDAR systems. By using this depth information the system performs necessary actions to avoid collisions with objects that are determined to be too close. However, these systems are also more expensive than a regular mono camera system and are therefore not very common in the average consumer car. By implementing robust depth estimation in mono vision systems the benefits from active safety systems could be utilized by a larger segment of the vehicle fleet. This could drastically reduce human error related traffic accidents and possibly save many lives. The network architecture evaluated in this thesis is more lightweight than other CNN architectures previously used for monocular depth estimation. The proposed architecture is therefore preferable to use on computationally lightweight systems. The network solves a supervised regression problem during the training procedure in order to produce a pixel-wise depth estimation map. The network was trained using a sparse ground truth image with spatially incoherent and discontinuous data and output a dense spatially coherent and continuous depth map prediction. The spatially incoherent ground truth posed a problem of discontinuity that was addressed by a masked loss function with regularization. The network was able to predict a dense depth estimation on the KITTI dataset with close to state-of-the-art performance.

Les systèmes d’imagerie infrarouge suivent depuis plusieurs années la même tendance de miniaturisation que les systèmes d’imagerie visible. Aujourd’hui cette miniaturisation se rapproche d’une limite physique qui amène la communauté à se tourner vers une nouvelle approche : la fonctionnalisation, c’est-à-dire l’apport de fonctions d’imagerie avancées aux systèmes telles que l’imagerie 3D.En infrarouge, la fonction d’imagerie 3D est très recherchée car elle pourrait apporter à un fantassin un outil de télémétrie passive fonctionnant de nuit comme de jour, ou encore permettre l’autonomie en environnements complexes à des systèmes tels que les drones. Cependant, le cout d’une caméra infrarouge hautes-performances est élevé. Multiplier le nombre de cameras n’est donc pas une solution acceptable pour répondre à ce besoin.C’est dans ce contexte que se situe ce travail qui consiste à apporter une fonction de vision 3D à des caméras infrarouges possédant un unique plan focal.Au cours de cette thèse, j’ai identifié la technologie d’imagerie 3D la plus adaptée à ce besoin : la camera plénoptique. J’ai montré que cette dernière permet de proposer, en intégrant une matrice de microlentilles dans le cryostat, un bloc de détection infrarouge avec une fonction d’imagerie 3D. L’environnement scellé du bloc de détection m’a amené à développer un modèle de dimensionnement rigoureux que j’ai appliqué pour concevoir et réaliser une camera plénoptique infrarouge refroidie. J’ai ensuite mis au point une méthode de caractérisation originale et intégré les mesures dans une série d’algorithmes de traitement d’image afin de remonter à la distance des objets observés
For a few years now, infrared cameras have been following the same miniaturization trend introduced with visible cameras. Today, this miniaturization is nearing a physical limit, leading the community to take a different approach called functionalization: that is bringing an advanced imaging capability to the system.For infrared cameras, one of the most desired functions is 3D vision. This could be used to bring soldiers a passive telemetry tool or to help UAVs navigate a complex environment, even at night. However, high performance infrared cameras are expensive. Multiplying the number of cameras would thus not be an acceptable solution to bring 3D vision to these systems.That is why this work focuses on bringing 3D vision to cooled infrared cameras using only a single focal plane array.During this PhD, I have first identified the plenoptic technology as the most suitable for our need of 3D vision with a single cooled infrared sensor. I have shown that integrating a microlens array inside the dewar could bring this function to the infrared region. I have then developed a complete design model for such a camera and used it to design and build a cooled infrared plenoptic camera. I have created a method to characterize our camera and integrated this method into the image processing algorithms necessary to generate refocused images and derive the distance of objects in the scene

The purpose of this thesis is the study and characterization of 12 image of the volcano Arsia Mons captured by the Cassis camera on board of the ESA’s TGO satellite. Arsia is the second biggest volcano on Mars in terms of volume of materials erupted and is composed by a large summit caldera and 2 huge aprons on the side, one extending toward N/N-E and the other toward S/S-W. Cassis camera provides color image with a great resolution of 4.5 m/pixel covering about 3% of the martian surface. The fact that the image are colored helps understanding the different processes that have occurred during the Martian history. This study analyzed every image using different tools and programs such as QGIS for the morphological analysis, Google Mars for the general overview and HiView for small details study of the HiRISE image where it was possible. The final result present a description of every image pointing out every structure and morphology as well as the overview of the area because some of the structures cannot be analyzed on a detailed scale. At the end we described the potential genesis of some pit craters due to volcanic (such as lava tube or rill) or tectonic processes. Moreover, there are some pits that may result from different not yet described processes such as piping which is a particular type of subsurface water erosion of loose pyroclastic materials following a collapse on the surface, or magmatic dyke intrusion of mantle materials causing surface sinkholes. There are a lot of terrestrial analogues easy to study on site that will help understanding better the correlations between Earth and Mars. The conclusions needs further investigations that could be useful for future robotic and human missions to Mars and other rocky planetary bodies that experienced volcanism.

Le saumon atlantique (Salmo salar) est une espèce tant emblématique que menacée à l’échelle internationale, au même titre que la plupart des espèces de poissons diadromes. La présence de barrages, en entraînant une rupture de la continuité écologique des cours d’eau, participe à ce déclin. Le remplacement du barrage en porte à flot sur l’estuaire du Couesnon (Normandie, France) par un barrage procédant à des lâchers d’eau à l’étale de basse mer, est susceptible d’impacter l’abondance et la dynamique migratoire des saumons. Pour étudier les remontées de saumons adultes sur ce fleuve, un échosondeur à faisceau partagé a été installé en amont du barrage et a enregistré en continu tous les passages de saumons lors des périodes estivales de 2010 à 2014. L’hydroacoustique, méthode non intrusive et indépendante des conditions de turbidité de l’eau, permet l’acquisition d’informations difficilement accessibles par d’autres techniques. Pour évaluer la réelle influence du barrage sur les migraticours d’eau voisin dont l’estuaire est libre, à l’aide d’un outil hydroacoustique bénéficiant de larges innovations technologiques, une caméra acoustique DIDSON. Les méthodologies développées soulignent les avantages et les limites des deux outils, mettant en lumière le gain d’information apporté par les caméras acoustiques. L’interprétation des résultats met en évidence que si le fonctionnement du barrage n’impacte pas le maintien de la population sur ce fleuve, la dynamique migratoire des saumons y est modifiée. Cette étude souligne les grandes capacités d’adaptation des saumons atlantique
The Atlantic salmon (Salmo salar) is as emblematic as in a global decline, as most of the diadromous fish species. Dams have direct consequences on the river ecological continuity and indirect effects, such as the modification of the hydrological and hydrosedimentary dynamics which affects these populations. In Normandy (France), the Couesnon River’s estuary flood-gate dam has recently been replaced by a new dam proceeding to water releases at low tide stall. This functioning may decrease the salmon abundance or disturb their migration dynamic on this river. To study the dam influence, a split-beam echosounder has been settled upstream the dam and has recorded Philippe BOëTpermanently the salmon migrations during the summer period from 2010 to 2014. The hydroacoustic methods, non-intrusive and independent from the turbidity conditions, provide useful information, neither available nor quantifiable with other existing monitoring technics. To evaluate the disturbance of the estuarine damthe migration dynamic of the same species has also been monitored on a nearby river, the Sélune River, whose estuary is free-flowing. The hydroacoustic tool settled on this river, a DIDSON, belongs to the acoustic camera generation and benefits from recent technological improvements. The defined methodologies highlight the advantages and limits of both tools, revealing the information gain brought by the acoustic cameras. A dam functioning influence on the Atlantic salmon migration has been revealed, but no negative impact has been described on the salmon annual recruitment level. This study

Made available in DSpace on 2017-07-24T19:22:10Z (GMT). No. of bitstreams: 1 Yasmine Bitencourt Emilio Mendes.pdf: 6295050 bytes, checksum: ec73165b3475fb2d502d87ebf6e6717e (MD5) Previous issue date: 2010-02-11
Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
The aim of this study was to assess the effect of different modes of application of a conventional adhesive system AdperTM Single Bond [SB] and of two self-etch adhesives AdperTM SE Plus [SE] and AdheSE® [AD] and influence of dentin depth on mechanical properties of hardness and elastic modulus of adhesive layer using the nanoindentation technique. The surfaces occlusal of 48 caries-free human third molars were removed and the smear layer standardized for the obtaining of a flat superficial dentin surface, in two levels: superficial dentine (DS) or deep dentine (DP). The adhesives were applied on the dentin surface in two layers, under active (AA) or passive (PA) application and photoactivated (10 s / 500 mW/cm2). After composite build-ups were constructed incrementally and were stored in water by 24 h. The teeth were sectioned in perpendicular direction to obtain specimens in form of bonded sticks with a cross-sectional area of 0.8 ± 0.1mm2, separating the central bonded sticks of each tooth for they be fastened in sample holder and metallographic polishing. The test of nanoindentation was made on composite resin, adhesive system and dentin regions. The results of hardness and elastic modulus (GPa) of adhesive layer were analyzed by a three-way ANOVA and Tukey’s multiple comparison tests (α=0.05) and also took place the two-way ANOVA for each adhesive separately. The mean and standard deviation of hardness and elastic modulus values of composite resin (GPa) were respectively (0.94 ± 0.05) and (13.41 ± 0.52). In dentin, the hardness was (0.89 ± 0.06) and the elastic modulus (17.51 ± 0.71). In the adhesive layer, the highest hardness (0.32 ± 0.02) and elastic modulus (5.54 ± 0.47) of SB were obtained in DS group. The elastic modulus in the adhesive layer of SE (4.11 ± 0.10) was lowest in AA group. For the adhesive layer of AD, just the mode of application was statistically significant (p=0.0041), being AA (0.29 ± 0.01) larger than PA (0.26 ± 0.01) (p=0.0042). It was concluded that: 1) passive application in superficial dentine got greaters values of hardness and modulus of elasticity of the SB and of SE, 2) the active application in deep dentine resulted in bigger values of the properties in study of SB, 3) active application significantly influenced the values of hardness in the area of the adhesive layer of the AD ones in such a way in superficial dentine how much in deep, but didn’t influence the modulus.
O objetivo deste trabalho foi avaliar o efeito da forma de aplicação do sistema adesivo convencional AdperTM Single Bond 2 [SB] e dos sistemas adesivos autocondicionantes AdperTM SE Plus [SE] e AdheSE® [AD] e da profundidade dentinária nas propriedades mecânicas de dureza e módulo de elasticidade da camada adesiva por meio do teste de nanoindentação. As superfícies oclusais de 48 terceiros molares humanos hígidos foram seccionadas e a smear layer padronizada para a obtenção de uma superfície dentinária plana, em dois níveis: dentina superficial (DS) ou dentina profunda (DP). Os adesivos foram aplicados sobre a superfície dentinária em duas camadas de forma ativa (AA) ou forma passiva (AP) e fotoativados (10 s / 500 mW/cm2). Em seguida, foram restaurados incrementalmente e armazenados em água destilada por 24h. Os dentes foram seccionados em eixos “x” e “y” para obtenção de espécimes em forma de palitos de aproximadamente 0,8 + 0,1 mm2 de secção transversal, separando o palito central de cada dente para serem fixados em porta-amostra e polidos metalograficamente. O teste de nanoindentação foi realizado nas regiões da resina composta, camada adesiva e dentina. Os valores de dureza e módulo de elasticidade (GPa) da camada adesiva foram submetidos a uma análise de variância de três fatores e ao teste de Tukey para contraste das médias (α=0,05), e também realizou-se a análise de variância de 2 fatores para cada adesivo isoladamente. As médias e os desvios padrões da dureza e do módulo de elasticidade (GPa) da resina composta foram respectivamente (0,94 ± 0,05) e (13,41 ± 0,52). Em dentina, a dureza foi (0,82 ± 0,06) e o módulo de elasticidade (17,51 ± 0,71). Para o grupo do SB na camada adesiva, os maiores valores de dureza (0,32 ± 0,02) e módulo de elasticidade (5,54 ± 0,47) foram obtidos no grupo DS. O módulo de elasticidade da camada adesiva do SE (4,11 ± 0,10) foi menor para o grupo AA. Para a região da camada adesiva do AD, apenas a forma de aplicação foi estatisticamente significante (p=0,0041), sendo a AA (0,29 ± 0,01) maior que a AP (0,26 ± 0,01) (p=0,0042). Concluiu-se que: 1) aplicação passiva em dentina superficial obteve maiores valores de dureza e módulo de elasticidade do SB e do SE, 2) a aplicação ativa em dentina profunda resultou em maiores valores das propriedades em estudo do SB, 3) aplicação ativa influenciou significativamente os valores de dureza na região da camada adesiva do AD tanto em dentina superficial quanto em profunda, porém não influenciou o módulo de elasticidade.

Neste trabalho foi analisado o comportamento de um transistor UTBB FD SOI MOSFET (Ultra-Thin-Bodyand-Buried-Oxide Fully-Depleted Silicon-on-Insulator Metal-Oxide-Semiconductor Field-Effect-Transistor) planar, operando em modo convencional, de tensão de limiar dinâmica (DT2-UTBB, onde a tensão de substrato é igual à de porta, VB=VG) e modo DT2 melhorado (kDT, onde a tensão de substrato é um múltiplo da tensão de porta, VB=kVG). O princípio de funcionamento desses modos foi estudado e a influência de diferentes tendências e tecnologias atuais foram analisadas nessas condições de polarização como a presença do plano de terra (Ground Plane - GP), o escalamento da camada de silício e a ausência de uma região de extensão de fonte e dreno. Também foi proposto neste trabalho o modo kDT inverso, onde a tensão de porta é um múltiplo da tensão de substrato (VG=kVB). O efeito do superacoplamento foi identificado e analisado a partir de diferentes técnicas, como nas curvas de capacitância, sua influência no efeito de corpo e no transistor UTBB operando em modo DT2 e DT2 melhorado. Finalmente o efeito da alta temperatura também foi estudado em transistores UTBB nos modos DT2 e DT2 melhorado, bem como nas curvas de capacitâncias. A operação em DT2 apresentou melhores resultados que o método convencional, principalmente para canais mais curtos (redução da inclinação de sublimiar em 36%, elevação da transcondutância máxima em 23% e diminuição do DIBL, Drain Induced Barrier Lowering, em 57%). A presença do GP acentuou esta melhora (redução da inclinação de sublimiar em 51%, elevação da transcondutância máxima em 32% e diminuição do DIBL em 100%), uma vez que há um maior acoplamento entre o substrato e o canal, fortalecendo sua atuação na redução da tensão de limiar. O modo melhorado da tensão de limiar dinâmica apresentou melhores parâmetros elétricos que no modo DT2, devido à redução mais expressiva da tensão de limiar para uma mesma varredura da tensão de porta. No modo kDT inverso, os parâmetros também foram ainda melhores (60% menor SS e 147% maior gm,max para os dispositivos sem GP e 68% menor SS e 189% maior gm,max nos dispositivos com GP) devido ao óxido de porta ser mais fino que o óxido enterrado. Com relação ao escalamento do filme de silício, para maiores valores de sobretensão de porta, a redução da camada de silício apresenta uma maior resistência série e uma maior degradação da mobilidade, reduzindo a corrente de dreno. Já para tensões de porta negativas, o GIDL (Gate Induced Drain Leakage) é mais elevado para menores espessuras do filme de silício. Entretanto, a menor espessura da camada de silício (tSi) mostrou ser vantajoso no modo kDT, devido ao acoplamento mais forte. O filme de silício mais fino melhorou principalmente o DIBL (a espessura de 6nm apresentou um DIBL 3 vezes menor que o dispositivo de 14nm para k=5), diminuindo o campo elétrico do dreno, e o SS (a espessura de 6nm apresentou um SS 7% menor que o dispositivo de 14nm para k=5), onde o campo elétrico vertical não é suficiente para degradar o parâmetro do dispositivo. O superacoplamento mostrou-se benéfico em transistores UTBB operando em modo DT2 e kDT, amplificando o efeito da inversão de volume e elevando consideravelmente a transcondutância e a mobilidade (melhora de até 131% para k=5, NMOS e tSi=7nm, tomando o caso VB=0V como referência). O superacoplamento também apresentou resultados positivos no estudo do escalamento dos dispositivos, apresentando um excelente acoplamento ainda para o menor comprimento de canal medido (0,076 para comprimento de 20nm contra 0,09 para L=1µm). Com relação à engenharia de fonte e dreno, os melhores resultados foram obtidos para os dispositivos sem a implantação da região de extensão (extensionless) e com comprimento dos espaçadores de 20nm. Os mesmos transistores extensionless também demonstraram serem mais suscetíveis com o aumento do fator k, apresentando o melhor comportamento na região de sublimiar (inclinação de sublimiar, SS, até 59% menor), desempenho analógico (elevação de mais de 300% no ganho intrínseco de tensão, AV, e de mais de 600% na tensão Early, VEA) e aplicação em baixas tensões (menor inclinação de sublimiar e tensão de limiar). A única desvantagem observada para a operação em DT2 e kDT foi a elevada corrente de GIDL (elevação de uma ordem de grandeza entre os transistores auto-alinhados com k=5 em relação ao auto-alinhado com k=0), entretanto, os dispositivos sem a implantação da região de extensão de fonte e dreno apresentaram um menor GIDL (redução de 1 ordem de grandeza para os dispositivos sem a implantação de 20nm com k=5 em relação ao dispositivo auto-alinhado com k=5) devido ao menor campo elétrico da porta para o dreno, o que pode ser uma solução para essa desvantagem. A região de extensão mais longa (sem a implantação) e, principalmente a operação em modo kDT, melhoram os parâmetros (elevação de 82% na transcondutância máxima, gm,max, redução de 45% no SS, de 41% no DIBL, elevação de 303% no AV e de 97% no VEA), superando a degradação observada pelo aumento da temperatura (porcentagens apresentadas já estão considerando a degradação da temperatura). Além disso, os modos kDT reduziram a tensão de porta do ponto ZTC (Zero-Temperature-Coefficient) em até 57%, sendo interessante em aplicações de baixa tensão. O modo kDT também permitiu o ajuste da tensão de limiar e da tensão de polarização, ainda com o nível de corrente independente com a temperatura e com o fator k.
In this work, it was analyzed the behavior of a planar UTBOX FD SOI NMOSFET (Ultra-Thin-Buried-Oxide Fully-Depleted Silicon-on-Insulator Metal-Oxide- Semiconductor Field-Effect-Transistor), operating in conventional (VB=0V), dynamic threshold (DT2-UTBB, where the back-gate bias is equal to the front-gate one, VB=VG) and enhanced DT (kDT-UTBB, where the back-gate bias is a multiple value of the front-gate one, VB=kVG) modes. The working principle of these modes has been studied and the effect of different technologies and current trends were analyzed under such biasing conditions as the presence of the ground plane (ground plane - GP), the scaling of the silicon layer and the absence of a doped extended source and drain region. It was also proposed in this paper the inverse kDT-UTBB mode, where the gate voltage is a multiple of the back-gate one (VB=kVG). The supercoupling effect was identified and analyzed through different techniques, such as the capacitance curves, its influence on the body effect and in UTBB SOI transistors operating in DT2 and kDT modes. Finally, the high temperature influence was also studied in UTBB SOI transistors operating in DT2 and kDT modes, as well as on capacitance characteristics. The operation DT2 showed better results than the conventional method, mainly for shorter channels (reduced subthreshold slope, SS, in 36%, increased maximum transconductance, gm,max, in 23% and reduced Drain Induced Barrier Lowering, DIBL, 57%). The presence of GP intensified this improvement (reducing SS by 51%, raising gm,max by 32% and reduced DIBL by 100%), due to the greater coupling of the substrate on the channel, strengthening its influence on reducing the threshold voltage. The kDT mode showed better electrical parameters than the DT2 due to a remarkable reduction of the threshold voltage for the same VG sweep. In the inverse kDT mode, the parameters were also better (60% lower SS and 147% higher gm,max for devices without GP and 68% lower SS and 189% higher gm,max on devices with GP) due to the thinner gate oxide than the buried oxide. With regard to the silicon film scaling, for higher values of gate voltage, the thinner silicon layer presented a larger series resistance and a greater mobility degradation, reducing the drain current. For negative gate biases, the GIDL (Gate Induced Drain Leakage) is higher for smaller thicknesses of the silicon film. However, the lower silicon film thickness showed to be advantageous in kDT due to the stronger coupling. The thinner silicon thickness has improved the DIBL (thickness of 6nm presented a DIBL 3 times smaller than the device of 14nm for k = 5), reducing the drain electric field, and the SS (thickness of 6nm presented an SS 7% smaller than 14nm device for k = 5), where the vertical electric field is not enough to degrade the device parameter. The supercoupling demonstrated beneficial results in UTBB transistors in DT2 and kDT operations, amplifying the volume inversion effect and rising significantly the transconductance and the mobility (improvement of up to 131% for k=5, 7nm-NMOS, taking VB=0V as the reference). Measurements and simulations have also shown positive results in the scalability study, presenting an excellent coupling for the shortest channel considered (0.076 for L=20nm against 0.09 for L=1µm). With respect to source and drain engineering, the best results were obtained for devices without the extension implantation and spacer length of 20nm. They also demonstrated to be more susceptible to the increase of k factor, showing the best behavior in the subthreshold region (59% lower), analog performance (300% higher intrinsic voltage gain, AV and 600% higher Early voltage, VEA) and for low voltages applications (reduced SS and VT). The only drawback observed for operation in kDT was the higher GIDL current (increase of 1 order of magnitude between self-aligned transistors with k=5 and self-aligned ones with k=0). However, the devices without the extension region implantation had a lower GIDL (1 order of magnitude lower for 20nm-extensionless devices with k=5, taking the self-aligned ones with k=5 as the reference) due to the lower gate-to-drain electric field, which can be a solution to this disadvantage. The longer extension region (without implantation) and, mainly, the kDT operation improved the parameters (increase of 82% in gm,max, reduction of 45% in SS, 41% reduced DIBL, rising of 303% in AV and 97% increased VEA), surpassing the degradation caused by rising the temperature (the last percentages is already considering the temperature degradation). Moreover, the DT2 and kDT operations reduced the gate bias of the ZTC point (Zero-Temperature-Coefficient) in 57%, being interesting for low voltage applications. The kDT mode also allowed the threshold voltage and the biases tunning, still with the current level independent of the temperature and the k-factor.

Submitted by Cleide Dantas (cleidedantas@ufpa.br) on 2014-03-19T14:51:45Z No. of bitstreams: 2 license_rdf: 23898 bytes, checksum: e363e809996cf46ada20da1accfcd9c7 (MD5) Dissertacao_TerritorioRecursoTerritorio.pdf: 976707 bytes, checksum: e2cd80625370be03c7b9dc66b3c113e1 (MD5)
Approved for entry into archive by Ana Rosa Silva (arosa@ufpa.br) on 2014-06-04T16:16:27Z (GMT) No. of bitstreams: 2 license_rdf: 23898 bytes, checksum: e363e809996cf46ada20da1accfcd9c7 (MD5) Dissertacao_TerritorioRecursoTerritorio.pdf: 976707 bytes, checksum: e2cd80625370be03c7b9dc66b3c113e1 (MD5)
Made available in DSpace on 2014-06-04T16:16:27Z (GMT). No. of bitstreams: 2 license_rdf: 23898 bytes, checksum: e363e809996cf46ada20da1accfcd9c7 (MD5) Dissertacao_TerritorioRecursoTerritorio.pdf: 976707 bytes, checksum: e2cd80625370be03c7b9dc66b3c113e1 (MD5) Previous issue date: 2010
Aborda-se no presente trabalho o processo de valorização do açaí no município de Cametá. Para isso foi utilizado como recorte territorial duas ilhas Cacoal e Cuxipiari que suas populações vivem em função da extração do açaí, nelas vivem quatro comunidades ribeirinhas. O principal objetivo desta pesquisa foi investigar como tem se dado o processo de valorização e expansão do açaí no mercado externo no modo de vida das comunidades ribeirinhas do município de Cametá; o que mudou em termos sociais, econômicos, culturais na vida das comunidades ribeirinhas que extraem o açaí e fazem disso a sua principal ocupação. Busca-se identificar o uso do território como recurso e como abrigo, usado pelas comunidades ribeirinhas, e por atores sociais como cooperativas e empresas beneficiadoras do açaí. A metodologia utilizada foi a aplicação de questionários semi-estruturados e entrevistas com mp4, assim como a análise de diversos documentos oficiais e relatórios institucionais. O recorte temporal utilizado foi a partir da década de 1990, por ser esse o momento que o açaí ganha maior visibilidade no mercado externo e intensifica o seu processo de mercadorização.
In the present work, we approach the valuation process of the açaí in the municipality of Cametá. Thus, two islands – Cacoal and Cuxipiarí – which live on the açaí extraction were used as a territorial delimitation. And on those islands, there are four riverine communities. The main aim of this research is to investigate how the expansion and valuation process of the açaí happens in the foreign market in the way of life of the riverine communities of the municipality of Cametá; and what has changed as regards the society, economy, and culture in the life of those communities which extract the açaí – their main working activity. We search for identifying the use of the territory as resource and shelter, used by the riverine communities, and by social actors such as cooperative and processing company of the açaí. The methodology used was that of the application of semi-estructured questionnaires and interviews with MP4, as well as the analysis of several official documents and institutional reports. The period considered was from the 1990’s because at that moment the açaí gains a broader acceptance in the foreign market and intensifies its commercialization process.

Made available in DSpace on 2016-06-02T20:23:11Z (GMT). No. of bitstreams: 1 3236.pdf: 10667295 bytes, checksum: 4577e702fe0c4a20dd47793beb779bec (MD5) Previous issue date: 2010-08-02
The motion picture cameras have changed through the years. However, they have kept their main design which has not changed during this period. One of the goals of this text is to understand how that design was created, and also the changes it has suffered without losing its essence. Besides that, this text aims to understand how different cameras connected with the Brazilian cinema production in the fifties and sixties. A change on camera s use is noticeable in that period in Brazil. There is a change in appreciation from the more traditional cameras linked to the studio system to the European cameras developed in the thirties and forties, which were lighter and more portable. This issue coincides with a specific characteristic in Brazilian cinema at that time, when people started to believe in the independent cinema production as an answer to the studio system, which was the main thought until then. In order to show this entire context, it has been chosen the Vera Cruz and Cinema Novo, their movies and shootings, to confront and connect them to camera models and their relation to the mode of production
As câmeras cinematográficas passaram por mudanças ao longo dos anos. Porém, manteve um design que se perpetuou durante este período. Um dos objetivos do presente trabalho é entender como este design foi criado, e as variações pelas quais ele passou, sem perder a sua essência. Em paralelo a isso, entender como estas diferentes câmeras travaram relação com a produção do cinema brasileiro dos anos 1950 e 1960. É nesta época que se observa no país uma troca de postura em relação às câmeras. Passa-se de uma valorização das câmeras mais tradicionais, ligadas ao sistema de estúdio, à valorização das câmeras européias criadas no anos 1930 e 1940, que eram mais leves e portáteis. Isso coincide com um momento específico do cinema brasileiro, aquele em que se passa a crer numa solução de cinema independente como resposta ao cinema de estúdio, que era o pensamento majoritário até então. Para ilustrar todo este contexto, escolhemos a Vera Cruz e o Cinema Novo, seus filmes e filmagens, para relacioná-los e confrontá-los em relação aos tipos de câmeras e sua relação ao modo de produção.

Today, the main research field for the automotive industry is to find solutions for active safety. In order to perceive the surrounding environment, tracking nearby traffic objects plays an important role. Validation of the tracking performance is often done in staged traffic scenarios, where additional sensors, mounted on the vehicles, are used to obtain their true positions and velocities. The difficulty of evaluating the tracking performance complicates its development. An alternative approach studied in this thesis, is to record sequences and use non-causal algorithms, such as smoothing, instead of filtering to estimate the true target states. With this method, validation data for online, causal, target tracking algorithms can be obtained for all traffic scenarios without the need of extra sensors. We investigate how non-causal algorithms affects the target tracking performance using multiple sensors and dynamic models of different complexity. This is done to evaluate real-time methods against estimates obtained from non-causal filtering. Two different measurement units, a monocular camera and a LIDAR sensor, and two dynamic models are evaluated and compared using both causal and non-causal methods. The system is tested in two single object scenarios where ground truth is available and in three multi object scenarios without ground truth. Results from the two single object scenarios shows that tracking using only a monocular camera performs poorly since it is unable to measure the distance to objects. Here, a complementary LIDAR sensor improves the tracking performance significantly. The dynamic models are shown to have a small impact on the tracking performance, while the non-causal application gives a distinct improvement when tracking objects at large distances. Since the sequence can be reversed, the non-causal estimates are propagated from more certain states when the target is closer to the ego vehicle. For multiple object tracking, we find that correct associations between measurements and tracks are crucial for improving the tracking performance with non-causal algorithms.

Tese de mestrado, Ciências Geofísicas (Geofísica Interna), Universidade de Lisboa, Faculdade de Ciências, 2009
O método de sísmica de reflexão multicanal, desenvolvido pela indústria de exploração de hidrocarbonetos, possibilita a criação de imagens de estruturas termohalinas com uma resolução lateral (~10m) não conseguida pelas ferramentas oceanográficas convencionais. Esta técnica está-se a tornar uma nova ferramenta no estuda da circulação oceânica e da distribuição de calor e massa no nosso planeta que influenciam o nosso sistema climático. Para que seja possível aos oceanógrafos e geofísicos a interpretação dos dados sísmicos obtido se a sua relação com as estruturas termohalinas, é necessário obter a melhor imagem possível da coluna de água. O objectivo deste trabalho foi aplicar um conjunto de técnicas de processamento que permitissem acentuar as reflexões na camada de água. A sequência de processamento incluiu a eliminação da onda directa, normal moveout, stack, migração pós-stacke conversão tempo-profundidade. A supressão da onda directa é um dos passos mais importantes pois esta vai obscurecer as reflexões mais superficiais. Os dados analisados foram adquiridos na zona do Golfo de Cádiz em 2000. Esta região foi escolhida para testar a sísmica de reflexão em oceanografia devido à existência de uma grande variedade de características topográficas e oceanográficas, pela forte assinatura da Água Mediterrânica e pela extensa base de dados sísmicos e oceanográficos já existentes.
Multichannel seismic reflection techniques, developed by the hydrocarbon industry, can provide images of thermohaline structures with exceptional lateral resolution (~10 m) beyond that available from conventional oceanographic tool. This technique is becoming a new tool to better understand the ocean circulation and the distribution of heat and mass on our planet tha tinfluences our climate system. For the oceanographers and geophysicists to be able to interpret the seismic data and relate the seismic reflections with thermohaline structures there is the need to obtain a good image of the water column. The objective of this assignment was to obtain the best image possible, with enhanced reflected energy, using seismic processing techniques. The processing sequence included the elimination of the direct wave, normal move out, stacking, post-stack migration and time-depth conversion. The direct wave suppression is an important step because this strong arrival obscures the shallow reflections. The analyzed data was acquired in the Gulf of Cadiz in the year 2000. This region was chosen to test seismic reflection in oceanography because of the variety of topographic and oceanographic features, the strong signature of Mediterranean Outflow Water and the extensive archived oceanographic and seismic database.