Cette thèse consiste en l'utilisation d'outils d'informatiques pour recueillir des informations concernant l'écologie d'espèces animales. L'objectif de départ était d'assembler des algorithmes capables de traiter des enregistrements acoustiques et de détecter, lister et dénombrer les sons éventuellement présents d'insectes, amphibiens et oiseaux. Pour ce faire nous avons testé de manière non exhaustive différents classifieurs et descripteurs de signal audio9 pour (première partie) organiser et participer à trois concours internationaux de reconnaissance automatique de sons d'animaux et (seconde partie) construire un outil de suivi d'abondance de deux espèces d'oiseaux marins pélagiques sur l'île de la Réunion. La première moitié de la thèse (chapitre 7) a été dédiée à la construction et au test de modèles de reconnaissance multi-classes (92 espèces animales : 82 espèces d'oiseaux (dont 66 passériformes), 9 espèces d'insectes, et 1 espèce d'amphibien, Pelophylax kl. grafi). La seconde moitié de la thèse (chapitre 8) s'est concentrée sur la construction de détecteurs de cris de deux espèces d'oiseaux protégées dont les colonies sont particulièrement difficiles d'accès et menacées par le développement et les éclairages urbains : Le Pétrel de Barau (Pterodroma baraui, endémique de la Réunion et en danger d'extinction depuis 2008 d'après l'UICN) et le Puffin tropical (Puffinus bailloni). / The present manuscript deals with computer science applied to ecology. The main objective was to assembly algorithms able to analyse acoustic recordings and automatically detect, list and count sounds of insects, amphibiansand birds. We tested a non exhaustive list of audio features and classifiers to (first part) organize and participate to three international challenges of automatic regnotion of animal's sounds and (second part) build a automatic and passive acoustic monitoring of two species of pelagic seabirds on the Reunion island.

nÃo hà / Esta dissertaÃÃo trata da localizaÃÃo de placas de licenciamento brasileiras de veÃculos particulares. Imagens de carros em movimento, advindas de diversos sistemas analÃgicos de fiscalizaÃÃo de trÃnsito, foram usadas para testar os algoritmos. Estas imagens foram digitalizadas por um scanner e convertidas para o formato JPG ou JPEG (Joint Photographic Experts Group). O sistema proposto à classificado como sistema de reconhecimento de placa de licenciamento veicular (SRPLV), cujo propÃsito à a fiscalizaÃÃo de trÃnsito. O sistema à capaz de identificar a placa de veÃculos que avanÃaram sinais vermelhos ou excederam o limite de velocidade. VÃrios algoritmos, que constituem o sistema, foram analisados e comparados com o desempenho de algoritmos similares presentes na literatura. As principais contribuiÃÃes deste trabalho sÃo: o algoritmo de localizaÃÃo da placa, baseado em uma rede neural; a utilizaÃÃo do algoritmo Min/Max no pÃs-processamento das imagens binÃrias, e o uso do mÃtodo das linhas e colunas na etapa de separaÃÃo dos caracteres. Para avaliar o desempenho do sistema, testes foram realizados usando os dois mÃtodos propostos para a localizaÃÃo, baseados na correlaÃÃo mÃxima e em uma rede neural sem realimentaÃÃo. ConcluÃmos que o Ãltimo apresentou resultados superiores ao primeiro, de acordo com os resultados obtidos para o conjunto de imagens utilizadas. / This dissertation deals with the localization of Brazilian license plates of private vehicles, and recognition of their characters. Images of moving cars, acquired by diverse analog systems of traffic law enforcement were used to test the algorithms. These images were digitalized by a scanner and finally formatted as JPG or JPEG (Joint Photographic Experts Group). The proposed system is classified as a vehicle license plate recognition system(VLPRS), whose purpose is the traffic law enforcement. The system is able to identify license plate numbers of drivers who jump red lights and exceed the speed limit. Several algorithms which constitute the overall system were evaluated and compared with the performance of similar ones from the literature. The main contributions of this work are: the localization plate algorithm, based on a neural network; the use of the Min/Max algorithm to post processing binary images, and the use of the method of rows and columns in the characters isolation step. To assess the system performance some tests were provided using the two proposed methods for plate localization which are based on the maximum correlation and a feedforward neural network. We concluded that the latter performed better than the former according to the simulation results obtained for the set of images used.

SYRACO est un SYstème de Reconnaissance Automatisée des COccolithes, développé à son origine par Luc Beaufort et Denis Dollfus à partir de 1995 et plus récemment avec Yves Gally. L'utilité d'un tel système est de permettre aux spécialistes un gain de temps majeur dans l'acquisition et le traitement des données. Dans ce travail, le système a été amélioré techniquement et sa reconnaissance a été étendue aux nannofossiles calcaires du Cénozoïque. Ce système fait le tri entre les nannofossiles et les non-nannofossiles avec une efficacité respectivement estimée à 75% et 90 %. Il s'appuie sur une nouvelle base d'images de référence d'espèces datant de l'Eocène Supérieur aux espèces vivantes, ce qui représente des centaines d'espèces avec une forte variabilité morphologique. Il permet de réaliser une classification en 39 morphogroupes par la combinaison de réseaux de neurones artificiels avec des modèles statistiques. Les résultats sont présentés sous forme de comptages automatisés, de données morphométriques (taille, masse...) et de mosaïques d'images. Il peut ainsi être utilisé pour des analyses biostratigraphiques et paléocéanographiques. / SYRACO is an automated recognition system of coccoliths, originally developed since 1995 by Luc Beaufort and Denis Dollfus, and more recently with the help of Yves Gally. The main purpose of this system is for specialists to save time in the acquisition and treatment of data. By this recent work, the system has been technically improved and its ability of recognition has been extended to calcareous nannofossils of the Cenozoic Era. It sorts nannofossils and non-nannofossils with a reliability respectively estimated to 75% and 90%. It is based on a new reference images database of species from the Upper Eocene up to living species. This represents hundreds of species with a high morphological variability. It leads to the establishment of a classification arranged in 39 morphogroups, combining artificial neural networks to statistical models. The results are presented as automated counting, morphometrical data (size, mass...) and mosaics of images. Those results can be valuable in biostratigraphical and paleoceanographical analyses.

I dag är det vanligt att EU-medborgare företar en gränsöverskridning med sin familj inom EU. Samkönade par har möjlighet att ingå äktenskap i fem medlemsstater, men eftersom äktenskapet inte erkänns i övriga 22 medlemsstater riskerar det samkönade paret att mottagarstaten ser äktenskapet som ogiltigt i rättslig mening vid en gränsöverskridning inom EU. Samkönade par som redan är gifta kan dock erhålla visst skydd för de fall en medlemsstat föreskriver samkönat registrerat partnerskap. EU-rätten föreskriver inga minimikrav för hur registrerat partnerskap bör utformas i medlemsstaters nationella rättssystem. Det har i sin tur vilket lett till att rättskyddet för samkönat registrerat partnerskap ser olika ut bland medlemsstaterna. Samkönade civilstånd är därför rättsosäkert vid gränsöverskridningar sett ur ett helhetsperspektiv. Familjerätten är ett rättsområde där medlemsstaterna har suveränitet. Medlemsstaterna har därför själva rätten att besluta hur familjerätten utformas i det nationella rättssystemet, men det finns vissa restriktioner. Medlemsstaternas utövande av sin suveränitet måste ske i konformitet med etablerade EU-principer och annan relevant EU-rätt. Kommissionen har kommit med ett förslag om ett automatiskt erkännande av civilståndshandlingar och uppsatsens syfte är därför att utreda om ett automatiskt erkännande är kompatibelt med gällande EU-rätt. För att besvara syftet i uppsatsen utreds gällande EU-rätt med direkt eller indirekt inverkan på samkönade relationer. Vidare utreds förslaget om ett automatiskt erkännande av civilståndshandlingar i kommissionens grönbok om minskad byråkrati för medborgarna tillsammans med de yttranden som grönboken mottagit från olika organisationer. Genom att ställa ett automatiskt erkännande av civilståndshandlingar mot gällande EU-rätt leder slutsatsen fram till vilka hinder och möjligheter som föreligger vid genomförandet av förslaget. De hinder och möjligheter som föreligger talar för att ett automatiskt erkännande av civilståndshandlingar är genomförbart. EU-rätten ställer vissa krav på medlemsstaterna även om medlemsstaterna har suveränitet inom familjerätten. Följden blir att EU måste samarbeta med medlemsstaterna för att säkerställa rättssäkerheten av civilståndshandlingar när EU-medborgare vidtar gränsöverskridningar inom EU. / Today it’s common for EU-citizens to move across borders with their families within the EU. Same-sex couples have access to marriage in five Member States, but because their marriage is not recognized in the other 22 Member States the same-sex marriage might be void in a legal sense in another Member State. Same-sex couples who are already married may benefit from some protection in the event that a Member State provides same-sex registered partnership. EU law does not provide a minimum requirement when Member States develop legal protection for same-sex couples through registered partnership in their legal system. This in turn has led to the differences among the Member States when it comes to same-sex registered partnership. Therefore, a same-sex civil status doesn’t provide a legal certainty in cross-border situations when looked upon from a broader perspective. Family law is an area of law where Member States have sovereignty. Therefore, the Member States have the right to shape the family law themselves in their national legal system, although there are some restrictions. When Member States exercise their sovereignty they must still exercise it in conformity with established EU principals and other relevant EU law. The Commission has now proposed an automatic recognition of civil status records and the purpose for this thesis is therefore to analyze whether an automatic recognition of civil status record are compatible with existing EU law. In order to answer the purpose in the thesis, EU law with direct or indirect impact on same-sex relationships is analyzed. Furthermore, the proposal of an automatic recognition of civil status records in the Commissions Green Paper on less bureaucracy for citizens is analyzed together with the opinions the Green Paper received from different organizations. When looking at the compatibility of an automatic recognition of civil status records and existing EU law, the conclusion highlights the obstacles and opportunities that exist with the implementation of the proposal. The obstacles and opportunities that exist suggest that an automatic recognition of civil status records is achievable. EU law imposes certain requirements on Member States, although Member States have sovereignty over their national family law. The previous suggests that the EU must work together with the Member States in order to ensure the legal certainty of civil status records when EU-citizens cross the borders within the EU.

Submitted by Johnny Rodrigues (johnnyrodrigues@ufcg.edu.br) on 2018-10-01T15:06:04Z No. of bitstreams: 1 MATHEUS BEZERRA ESTRELA RODRIGUES - DISSERTAÇÃO PPGCC 2012..pdf: 7068761 bytes, checksum: 4b1283a1da5ca466fcf0357c33091a30 (MD5) / Made available in DSpace on 2018-10-01T15:06:04Z (GMT). No. of bitstreams: 1 MATHEUS BEZERRA ESTRELA RODRIGUES - DISSERTAÇÃO PPGCC 2012..pdf: 7068761 bytes, checksum: 4b1283a1da5ca466fcf0357c33091a30 (MD5) Previous issue date: 2012-02-29 / Impulsionado pelo crescimento no uso de vigilância eletrônica, essa pesquisa introduz o uso de uma técnica que demonstra eficiência no reconhecimento de faces em imagens, alterando o objeto de busca para pneus de veículos, visando o reconhecimento da presença do veículo na cena. A técnica aplicada para o reconhecimento é o algoritmo Viola-Jones. Essa técnica é dividida em dois momentos: o treinamento e a detecção. Na primeira etapa, vários treinamentos são executados, usando aproximadamente 7000 imagens diferentes. Para a etapa final, um detector de faces foi adaptado para reconhecer pneus, utilizando o treinamento da etapa anterior, e sua eficiência em reconhecer os pneus foi comparável à eficiência do detector de faces que usa treinamento de referência da biblioteca em software que é referência nesta área, OpenCV. O detector desenvolvido apresentou taxa de reconhecimento de 77%, quando o reconhecimento de faces obteve 80%. A taxa de falsos negativos também foi próxima, apresentando o detector de pneus 2% e o de faces 1%. / Motivated by the growing use of electronic surveillance, this research introduces the use of the Viola-Jones algorithm, which is known to be efficient in recognition of human faces in images, changing the object to be recognized to vehicle tires, aiming to detect vehicles in a scene. This approach divides the process in two steps: training and detection. Training was done using around 7000 different images of vehicles. For the detection step, work was done to adapt a face detector to detect vehicles tires. The tire detector was compared to a face detector that used a reference training for faces from OpenCV library. The tire detector showed 77% efficiency, whereas the face detector showed 80%. False negative numbers also showed similar closeness, as 2% for the tire detector and 1% for the reference face detector.