L'intérêt croissant pour les technologies Web génère de nouveaux défis. Le format XML s'est imposé comme une référence pour le stockage et l'échange de données. Certains documents XML ont acquis une taille telle, qu'il est inefficace voire impossible de les stocker en mémoire centrale. Cela amène à repenser les algorithmes prévus pour traiter ces documents. Une solution consiste à considérer un document XML comme un flux, qui correspond à une lecture unidirectionnelle de ce document. Ce flux est alors traité à la volée. Ainsi le document n'est jamais stocké en mémoire centrale, et uniquement les parties utiles y sont mémorisées. L'un des traitements effectués sur les fichiers XML est la sélection d'information par des requêtes. Ceci constitue une étape de base pour la transformation de documents XML, permettant ainsi à des applications utilisant différents schémas XML d'échanger des informations. Cette thèse étudie l'évaluation de requêtes sur des flux XML. Deux formalismes de requêtes sont considérés· le standard XPath, et les automates d'arbres Pour cela, une mesure de la faculté d'une requête à être évaluée sur des flux XML est introduite. A l'aune de cette mesure, les requêtes XPath et par automates ne sont pas adaptées à une évaluation de flux XML. Pour chacun des deux formalismes de requêtes, de larges fragments adaptés à ce type d'évaluation sont définis et étudiés. Pour les requêtes par automates d'arbres, deux autres critères liés à l'évaluation de flux XML sont montrés décidables en temps polynomial / The growing interest for Web technologies leads to new challenges. XML is now a reference for storing and exchanging data. Some XML documents are now so large, that il is inefficient or even impossible to store them in main memory. This calls for new paradigms to treat these data. One of them consists in considering an XML document as a stream, corresponding to a one-way reading of this document. This stream is then processed on-the-f1y. Hence the document is never stored in main memory, and only the useful parts are memorized. One task of XML processing is to retrieve information, using queries. This is the base step for XML document transformation, that allows applications using distinct XML schemas to exchange data. This thesis studies the query answering problem on XML streams. Two query classes are considered: the XPath standard, and tree automata. For this purpose, a measure of streamability of a query is introduced. This one shows that queries defined by XPath expressions or tree automata are not streamable. For both query formalisms, large streamable fragments are introduced and studied. For queries defined by tree automata, Iwo other streamability criteria are proved to be decidable in polynomial time.

No description available.

La reconnaissance du comportement et la prédiction des activités des personnes depuis la vidéo sont des préoccupations majeures dans le domaine de la vision par ordinateur. L'objectif principal de mon travail de thèse est de proposer des algorithmes qui permettent d'analyser des objets en mouvement à partir de la vidéo pour extraire des comportements humains. Cette analyse est effectuée dans des environnements intérieurs ou extérieurs filmés par des simples webcams ou par des caméras plus sophistiquée. La scène analysée peut être de deux types en fonction du nombre de personnes présentes. On distingue les scènes de foule où le nombre de personnes est important. Dans ce type de scène, nous nous intéressons aux problèmes de la détection d'évènements de foule, à l'analyse des flux et à l'extraction des motifs de mouvement. Le deuxième type de scène se caractérise par la présence d'une seule personne à la fois dans le champ de la caméra. Elle est appelée scène individuelle. Nous y traitons le problème de reconnaissance d'actions humaines. Pour atteindre ces objectifs, nous proposons une approche basée sur trois niveaux d'analyse. Le premier est l'extraction des caractéristiques de bas niveau récupérés les images constituant un flux vidéo (ex. les zones en mouvement). Le deuxième construit des descripteurs pour l’analyse du comportement humain (ex. la direction et la vitesse de mouvement moyennes). Le niveau le plus haut se sert des descripteurs de l'étape intermédiaire afin de fournir aux utilisateurs des résultats concrets sur l'analyse du comportement humain (ex. telle personne marche, une autre court, etc.). Des expérimentations sur des benchmarks connus ont validé nos approches, avec un positionnement très intéressant par rapport à l'état de l'art. / The recognition and prediction of people activities from videos are major concerns in the field of computer vision. The main objective of my thesis is to propose algorithms that analyze human behavior from video. This problem is also called video content analysis or VCA. This analysis is performed in outdoor or indoor environments using simple webcams or more sophisticated surveillance cameras. The video scene can be of two types depending on the number of people present. The first type is characterized by the presence of only one person at a time in the video. We call this an individual scene where we will tackle the problem of human action recognition. The second type of scene contains a large number of persons. This is called a crowd scene where we will address the problems of motion pattern extraction, crowd event detection and people counting. To achieve our goals, we propose an approach based on three levels of analysis. The first level is the detection of low-level descriptors retrieved from the images of the video (e.g. areas in motion). The second level retrieves descriptors for modeling human behavior (e.g. average speed and direction of movement). The top level uses the descriptors of the intermediate step to provide users with concrete results on the analysis of behavior (e.g. this person is running, that one is walking, etc.). Experimentation on well-known benchmarks have validated our approaches, with very satisfying results compared to the state of the art.

A new method of manufacturing heat flux sensors was developed using a combination of copper etching and stencil printing nickel/silver conductive ink thermocouple materials onto a thin-film polyimide Kapton® substrate. The semi-automated production capabilities of this manufacturing process significantly decrease the cost of producing thin-film heat flux sensors while still maintaining acceptable performance characteristics. Material testing was performed to first determine the most appropriate materials as well as the theoretical sensitivity and time response of the final sensor. Seebeck coefficient of a thermocouple formed using the combination of EMS CI-1001 silver and EMS CI-5001 nickel ink was measured to be 18.3 ± 0.9 uV/ deg C. Calibrations were then performed on a sample of sensors produced using the novel manufacturing process to verify theoretical values for both sensitivity and time response. The printed heat flux sensor (PHFS) made using this process has a nominal voltage output sensitivity of 4.10 ± 0.23 mV/(W/cm2) and first order time constant response time of 0.592 ± 0.026 seconds. Lastly, a cost analysis was performed to estimate that the final cost to produce the PHFS is approximately $7.73 per sensor. This cost is significantly lower than commercially available sensors which range from $210 upwards to $3000. / Master of Science

A new thin film heat flux array (HFA) was designed and constructed using a series of nickel/copper thermocouples deposited onto a thin Kapton® polyimide film. The HFA is capable of withstanding temperatures up to 300 °C and produces signals of 42 μV/(W/cm²). As a result of its thin film construction, the HFA has a first order time constant of 32 ms. Calibrations were completed to determine the gage's output as well as its time response. In order to measure the signal from the HFA amplifiers were designed to increase the magnitude of the voltage output. An example case is given where the HFA is used in an experiment to correlate time-resolved heat flux and velocities. / Master of Science

A novel method for measuring heat flux of boiling water is designed and built to study critical heat flux (CHF) and observe the response of a heat flux sensor when CHF occurs. A high temperature heat flux sensor is embedded in the wall of a pipe to get a direct measurement of the surface heat flux and sensor temperature. By submerging the pipe in water and applying a controlled heat flux to the inside diameter over the area where the sensor is located, boiling is created on the outer surface while measuring the heat flux. The heat flux is gradually increased up to CHF and the heat flux response is observed to determine if the heat flux sensor could sense CHF when it occurred. The heat flux sensor is able to consistently measure the value for CHF, which is approximately 510 kW/m" for this system. It is also observed during the experiments that the heat flux response undergoes an inflection of the heat transfer coefficient at a consistent temperature just before reaching CHF. This observed inflection caused the heat flux response to deviate from its cubic relationship with the temperature and drastically increase for a very small change in temperature. This inflection response can be used as an indication for approaching CHF and can also be used to approximate its value without prior knowledge of when it occurs. / Master of Science

No description available.

No description available.

No description available.

No description available.