Les systèmes de vision artificielle sont de plus en plus présents en contexte routier. Ils sont installés sur l'infrastructure, pour la gestion du trafic, ou placés à l'intérieur du véhicule, pour proposer des aides à la conduite. Dans les deux cas, les systèmes de vision artificielle visent à augmenter la sécurité et à optimiser les déplacements. Une revue bibliographique retrace les origines et le développement des algorithmes de vision artificielle en contexte routier. Elle permet de démontrer l'importance des descripteurs d'images dans la chaîne de traitement des algorithmes. Elle se poursuit par une revue des descripteurs d'images avec une nouvelle approche source de nombreuses analyses, en les considérant en parallèle des applications finales. En conclusion, la revue bibliographique permet de déterminer quels sont les descripteurs d'images les plus représentatifs en contexte routier. Plusieurs bases de données contenant des images et les données météorologiques associées (ex : pluie, brouillard) sont ensuite présentées. Ces bases de données sont innovantes car l'acquisition des images et la mesure des conditions météorologiques sont effectuées en même temps et au même endroit. De plus, des capteurs météorologiques calibrés sont utilisés. Chaque base de données contient différentes scènes (ex: cible noir et blanc, piéton) et divers types de conditions météorologiques (ex: pluie, brouillard, jour, nuit). Les bases de données contiennent des conditions météorologiques naturelles, reproduites artificiellement et simulées numériquement. Sept descripteurs d'images parmi les plus représentatifs du contexte routier ont ensuite été sélectionnés et leur robustesse en conditions de pluie évaluée. Les descripteurs d'images basés sur l'intensité des pixels ou les contours verticaux sont sensibles à la pluie. A l'inverse, le descripteur de Harris et les descripteurs qui combinent différentes orientations sont robustes pour des intensités de pluie de 0 à 30 mm/h. La robustesse des descripteurs d'images en conditions de pluie diminue lorsque l'intensité de pluie augmente. Finalement, les descripteurs les plus sensibles à la pluie peuvent potentiellement être utilisés pour des applications de détection de la pluie par caméra.Le comportement d'un descripteur d'images en conditions météorologiques dégradées n'est pas forcément relié à celui de la fonction finale associée. Pour cela, deux détecteurs de piéton ont été évalués en conditions météorologiques dégradées (pluie, brouillard, jour, nuit). La nuit et le brouillard sont les conditions qui ont l'impact le plus important sur la détection des piétons. La méthodologie développée et la base de données associée peuvent être utilisées à nouveau pour évaluer d'autres fonctions finales (ex: détection de véhicule, détection de signalisation verticale).En contexte routier, connaitre les conditions météorologiques locales en temps réel est essentiel pour répondre aux deux enjeux que sont l'amélioration de la sécurité et l'optimisation des déplacements. Actuellement, le seul moyen de mesurer ces conditions le long des réseaux est l'installation de stations météorologiques. Ces stations sont coûteuses et nécessitent une maintenance particulière. Cependant, de nombreuses caméras sont déjà présentes sur le bord des routes. Une nouvelle méthode de détection des conditions météorologiques utilisant les caméras de surveillance du trafic est donc proposée. Cette méthode utilise des descripteurs d'images et un réseau de neurones. Elle répond à un ensemble de contraintes clairement établies afin de pouvoir détecter l'ensemble des conditions météorologiques en temps réel, mais aussi de pourvoir proposer plusieurs niveaux d'intensité. La méthode proposée permet de détecter les conditions normales de jour, de nuit, la pluie et le brouillard. Après plusieurs phases d'optimisation, la méthode proposée obtient de meilleurs résultats que ceux obtenus dans la littérature, pour des algorithmes comparables. / Computer vision systems are increasingly being used on roads. They can be installed along infrastructure for traffic monitoring purposes. When mounted in vehicles, they perform driver assistance functions. In both cases, computer vision systems enhance road safety and streamline travel.A literature review starts by retracing the introduction and rollout of computer vision algorithms in road environments, and goes on to demonstrate the importance of image descriptors in the processing chains implemented in such algorithms. It continues with a review of image descriptors from a novel approach, considering them in parallel with final applications, which opens up numerous analytical angles. Finally the literature review makes it possible to assess which descriptors are the most representative in road environments.Several databases containing images and associated meteorological data (e.g. rain, fog) are then presented. These databases are completely original because image acquisition and weather condition measurement are at the same location and the same time. Moreover, calibrated meteorological sensors are used. Each database contains different scenes (e.g. black and white target, pedestrian) and different kind of weather (i.e. rain, fog, daytime, night-time). Databases contain digitally simulated, artificial and natural weather conditions.Seven of the most representative image descriptors in road context are then selected and their robustness in rainy conditions is evaluated. Image descriptors based on pixel intensity and those that use vertical edges are sensitive to rainy conditions. Conversely, the Harris feature and features that combine different edge orientations remain robust for rainfall rates ranging in 0 – 30 mm/h. The robustness of image features in rainy conditions decreases as the rainfall rate increases. Finally, the image descriptors most sensitive to rain have potential for use in a camera-based rain classification application.The image descriptor behaviour in adverse weather conditions is not necessarily related to the associated final function one. Thus, two pedestrian detectors were assessed in degraded weather conditions (rain, fog, daytime, night-time). Night-time and fog are the conditions that have the greatest impact on pedestrian detection. The methodology developed and associated database could be reused to assess others final functions (e.g. vehicle detection, traffic sign detection).In road environments, real-time knowledge of local weather conditions is an essential prerequisite for addressing the twin challenges of enhancing road safety and streamlining travel. Currently, the only mean of quantifying weather conditions along a road network requires the installation of meteorological stations. Such stations are costly and must be maintained; however, large numbers of cameras are already installed on the roadside. A new method that uses road traffic cameras to detect weather conditions has therefore been proposed. This method uses a combination of a neural network and image descriptors applied to image patches. It addresses a clearly defined set of constraints relating to the ability to operate in real-time and to classify the full spectrum of meteorological conditions and grades them according to their intensity. The method differentiates between normal daytime, rain, fog and normal night-time weather conditions. After several optimisation steps, the proposed method obtains better results than the ones reported in the literature for comparable algorithms.

[pt] As partículas finas (PM2.5) são um dos principais poluentes atmosféricos associados a problemas de saúde. Estas partículas penetram no sistema respiratório, carreando desde metais traços a substâncias orgânicas. Apesar disso, a legislação ambiental brasileira ainda não tem estabelecido padrões para este poluente. Entretanto, Agencia Ambiental dos Estados Unidos (US.EPA) já tem adotado limites para exposições de curto (25 (micro)g m-3/diário) e longo (15 (micro)g m-3/anual) prazo. Esta tese teve quatro principais objetivos: (1) investigar a relação das condições meteorológicas, sazonalidade e bacias aéreas sobre as concentrações de PM2.5 na atmosfera; (2) avaliar modelos de previsão de qualidade do ar inovadores para estimar concentração de PM2.5 em locais com diferentes fontes de emissão; (3) validar método de extração e determinação pseudototal de metais traços presentes no material particulado, com espectrômetro de emissão ótica por plasma indutivamente acoplado (ICP-OES) de acordo com critérios estabelecidos pelo INMETRO; (4) quantificar carbono orgânico e metais traços presentes no material particulado fino para entender melhor como a atmosfera do estado do Rio de Janeiro tem sido afetada, devido aos vários tipos de emissão e condições meteorológicas. Amostradores de grandes volumes coletaram todas as amostras de PM2.5. Estes amostradores foram operados por 24 h, a cada seis dias, em locais com diferentes fontes de emissão (industrial, veicular, poeira do solo, etc.), no estado do Rio de Janeiro. As amostras foram coletadas pelo Instituto Estadual do Ambiente (INEA), no período de janeiro/11 até dezembro/13. Variáveis meteorológicas próximas (d(menor que)2 km) aos pontos de monitoramento de PM2.5 também foram obtidas na mesma frequência e período de amostragem. Em relação a este estudo, quatro resultados podem ser destacados. O primeiro, as concentrações médias diárias de PM2.5 variaram de 1-65 (micro)g m-3, ultrapassando em alguns pontos os limites adotados pela US.EPA. Estes resultados mostraram que concentrações de PM2.5 no RJ não é influenciada, em expressão, pela sazonalidade. Além disso, foi observado que as bacias aéreas definidas no Rio de Janeiro não têm sido confirmadas, e os locais mostraram uma semelhança de comportamento em função da sua fonte de emissão. O segundo, a aplicação do modelo Holt-Winters para previsão de PM2.5 simulou melhor a zona industrial, com RMSE (raiz do erro quadrático médio) entre 5,8-14,9 (micro)g m-3. Em contrapartida, a rede neural artificial associada a variáveis meteorológicas estimou melhor os resultados das zonas urbanas e rurais, com RMSE entre 4,2-9,3 (micro)g m-3. O terceiro, o método de extração e determinação pseudototal de metais por ICP-OES atendeu aos critérios de validação estabelecidos pelo INMETRO. Além disso, mostrou-se ser equivalente ao método US.EPA IO-3.1. Finalmente, as concentrações de carbono orgânico solúvel em água variaram de 0,8-4,9 (micro)g m-3. Os principais metais determinados foram: Na (5,8-13,6 (micro)g m-3), Al (1,6-6,7 (micro)g m-3) e Zn (1,9-6,6 (micro)g m-3). Foi verificado também que os fenômenos meteorológicos de superfície aumentam em 30 por cento a explicação da variância do modelo receptor (PCA), quando adicionados aos dados das substâncias químicas analisadas do PM2.5. Contudo, é crucial a aplicação de ferramentas quimiométricas para ajudar na caracterização e estimava das concentrações de poluentes atmosféricos. / [en] Fine particulate matters (PM2.5) are one of the primary air pollutants associated with health problems. These particles penetrate in the respiratory system, loading from trace metals to organic compounds. Neverthelere4ss, the Brazilian environmental legislation has not yet established standards for this pollutant. However, the US Environmental Agency (US.EPA) has already adopted limits for short-term (25 (micro)g m-3/daily) and long-term (15 (micro)g m-3/annual) exposures. This thesis had four main objectives: (1) to investigate the relation of weather conditions, seasonality and air basins on PM2.5 concentrations in the atmosphere; (2) to evaluate innovative air quality forecast models to estimate PM2.5 concentration in sites with different emission sources; (3) to validate method to extract and pseudo total determinate trace metals present in the particulate matter by inductively coupled plasma optical emission spectrometer (ICP-OES) according to criteria established by INMETRO; (4) to quantify organic carbon and trace metals present in fine particulate matter to better understand how the Rio de Janeiro State (RJ) atmosphere has been affected due to the various types of emission and weather conditions. High volumes samplers PM2.5 collected all PM2.5 samples. These samplers were operated for 24 h, every six days, in places with different emission sources (industrial, vehicular, soil dust, et caetera), in the Rio de Janeiro State. The samples were collected by the State Environmental Institute (INEA) during the period from January/2011 still December/2013. Meteorological variables nearby (d(less than)2 km) to PM2.5 monitoring points were also obtained at the same frequency and sampling period. Regarding this study, four results can be highlighted. The first one, the PM2.5 dailly concentrations average ranged from 1-65 (micro)g m-3, exceeding in some sites the limits adopted by US.EPA. These results showed that PM2.5 concentrations in RJ is not influenced, in expression, by the seasonality. In addition, it was observed that the defined RJ air basins have not been confirmed, and the local showed a similar performance according to their emission sources. The second one, the application of the Holt-Winters model for PM2.5 forecast simulated best industrial zone, with RMSE (root mean square error) between 5.8 to 14.9 (micro)g m-3. On the others hand, the artificial neural network associated with meteorological variables estimated best results from urban and rural areas, with RMSE between 4.2 to 9.3 (micro)g m-3. The third one, the method to extract and determine pseudo total metals by ICP-OES followed the validation criteria established by INMETRO. Furthermore, it was shown to be equivalent to US.EPA IO-3.1 method. Finally, the water-soluble organic carbon concentrations ranged from 0.8 to 4.9 (micro)g m-3. The principal metals determined were: Na (5.8-13.6 (micro)g m-3), Al (1.6-6.7 (micro)g m-3) and Zn (1.9-6.6 (micro)g m-3). It was also found that the surface meteorological phenomena increase at 30 percent the explicated variance of the receiver model (PCA) when added to PM2.5 chemical analysis data. Therefore, it is crucial the application of chemometric tools to help in the characterization and estimated air pollutant concentrations.