Until recently, methodologies to extract sub-pixel information from remotely sensed data have focused on linear un-mixing models and so called fuzzy classifiers. Recent research has suggested that neural networks have the potential for providing sub- pixel information. Neural networks offer an attractive alternative as they are non- parametric, they are not restricted to any number of classes, they do not assume that the spectral signatures of pixel components mix linearly and they do not necessarily have to be trained with pure pixels. The thesis tests the validity of neural networks for extracting sub-pixel information using a combination of qualitative and quantitative analysis tools. Previously published experiments use data sets that are often limited in terms of numbers of pixels and numbers of classes. The data sets used in the thesis reflect the complexity of the landscape. Preparation for the experiments is canied out by analysing the data sets and establishing that the network is not sensitive to particular choices of parameters. Classification results using a conventional type of target with which to train the network show that the response of the network to mixed pixels is different from the response of the network to pure pixels. Different target types are then tested. Although targets which provide detailed compositional information produce higher accuracies of classification for subsidiary classes, there is a trade off between the added information and added complexity which can decrease classification accuracy. Overall, the results show that the network seems to be able to identify the classes that are present within pixels but not their proportions. Experiments with a very accurate data set show that the network behaves like a pattern matching algorithm and requires examples of mixed pixels in the training data set in order to estimate pixel compositions for unseen pixels. The network does not function like an unmixing model and cannot interpolate between pure classes.

No description available.

The high resolution silicon Pixel detector is critical in event vertex reconstruction and in particle track reconstruction in the ATLAS detector. During the pixel data taking operation, some modules (Silicon Pixel sensor +Front End Chip+ Module Control Chip (MCC)) go to an auto-disable state, where the Modules don't send the data for storage. Modules become operational again after reconfiguration. The source of the problem is not fully understood. One possible source of the problem is traced to the occurrence of single event upset (SEU) in the MCC. Such a module goes to either a Timeout or Busy state. This report is the study of different types and rates of errors occurring in the Pixel data taking operation. Also, the study includes the error rate dependency on Pixel detector geometry.

No description available.

Artificial Intelligence is a machine learning technique that exists everywhere in our day-to-day life. From a simple Google search that provides answers to any questions, to autocorrect suggestions provided while writing emails, we encounter AI in every next phase of our life. Humans have developed an invisible trust in AI that remains unrecognized. Artificial intelligence (AI) development in architecture has been a protracted and intriguing process. Recent advances in text-to-image generating software powered by AI have proven to be an efficient tool for architects to visualize their designs with a different perspective and enhance the thinking process. However, the lack of the tool's ability to capture the designer's integrity has shown the requirement for human involvement. This thesis claims that human decision-making skills remain crucial despite AI-augmented design's benefits. By conducting a comparative analysis between human-developed architecture and AI-augmented designs through the process of AI text-to-image generating tool Stable Diffusion, the thesis argues that human brain involvement is necessary due to the lack of Stable Diffusion's ability to understand architectural drawings and elements, the ability to representing architectural depth through spaces and emotions, and its inadequate learning from the past design experiences. / Master of Architecture / Human communication has mainly based on gestures and visuals before the advent of writing and widespread literacy. Images have been one of the successful means of transiting design ideas. Past and present works of art have influenced the process of design thinking for architects. The human mind has always been able to capture past experiences and memories in the form of a collective database to convey new ideas. An average human brain can store up to 2.5 million gigabytes of memory. Artificial Intelligence is a computer language system that operates similarly to the human thinking process. The machine can learn from infinite gathered past data and provide exceptional results every time. It refers to developing intelligent computer systems that can mimic human problem-solving ability to an extent. With the active emergence of Artificial intelligence in the 21st century, there has been a rise in interest in generating realistic images by translating written descriptions. Through collaboration with human thinking processes and AI-generated images, designers can discover an additional tool to communicate their ideas. This thesis aims to summarize the evolution of AI in architecture and explore the potential use of text-to-image and image-to-image generating tools to transform the architectural design process.

No description available.

Most neural networks use a normal convolutional layer that assumes that all input pixels are valid pixels. However, pixels added to the input through padding result in adding extra information that was not initially present. This extra information can be considered invalid. Invalid pixels can also be inside the image where they are referred to as holes in completion tasks like image inpainting. In this work, we look for a method that can handle both types of invalid pixels. We compare on the same test bench two methods previously used to handle invalid pixels outside the image (Partial and Edge convolutions) and one method that was designed for invalid pixels inside the image (Gated convolution). We show that Partial convolution performs the best in image classification while Gated convolution has the advantage on semantic segmentation. As for hotel recognition with masked regions, none of the methods seem appropriate to generate embeddings that leverage the masked regions. / Master of Science / A module at the heart of deep neural networks built for Artificial Intelligence is the convolutional layer. When multiple convolutional layers are used together with other modules, a Convolutional Neural Network (CNN) is obtained. These CNNs can be used for tasks such as image classification where they tell if the object in an image is a chair or a car, for example. Most CNNs use a normal convolutional layer that assumes that all parts of the image fed to the network are valid. However, most models zero pad the image at the beginning to maintain a certain output shape. Zero padding is equivalent to adding a black frame around the image. These added pixels result in adding information that was not initially present. Therefore, this extra information can be considered invalid. Invalid pixels can also be inside the image where they are referred to as holes in completion tasks like image inpainting where the network is asked to fill these holes and give a realistic image. In this work, we look for a method that can handle both types of invalid pixels. We compare on the same test bench two methods previously used to handle invalid pixels outside the image (Partial and Edge convolutions) and one method that was designed for invalid pixels inside the image (Gated convolution). We show that Partial convolution performs the best in image classification while Gated convolution has the advantage on semantic segmentation. As for hotel recognition with masked regions, none of the methods seem appropriate to generate embeddings that leverage the masked regions.

Segmentação de imagens é ainda uma etapa desafiadora do processo de reconhecimento de padrões. Entre as abordagens de segmentação, muitas são baseadas em particionamento em grafos, as quais apresentam alguns inconvenientes, sendo um deles o tempo de processamento muito elevado. Com as recentes pesquisas na teoria de redes complexas, as técnicas de reconhecimento de padrões baseadas em grafos melhoraram consideravelmente. A identificação de grupos de vértices pode ser considerada um processo de detecção de comunidades de acordo com a teoria de redes complexas. Como o agrupamento de dados está relacionado com a segmentação de imagens, esta também pode ser abordada através de redes complexas. No entanto, a segmentação de imagens baseado em redes complexas apresenta uma limitação fundamental, que é o número excessivo de nós na rede. Neste trabalho é proposta uma abordagem de redes complexas para segmentação de imagens de grandes dimensões que é ao mesmo tempo precisa e rápida. Para alcançar este objetivo, é incorporado o conceito de Super Pixels, visando reduzir o número de nós da rede. Os experimentos mostraram que a abordagem proposta produz segmentações de boa qualidade em baixo tempo de processamento. Além disso uma das principais contribuições deste trabalho é a determinação dos melhores parâmetros, uma vez que torna o método bastante independente dos parâmetros, o que não fora alcançado antes em nenhuma pesquisa da área / Image segmentation is still a challenging stage of the pattern recognition process. Amongst the various segmentation approaches, some are based on graph partitioning, many of which show some drawbacks, such as the high processing times. Recent trends on complex network theory have contributed considerably to the development of graph-based pattern recognition techniques. The identification of group of vertices can be considered a community detection process according to complex network theory. Since data clustering is closely related to image segmentation, image segmentation tasks can also be tackled by complex networks. However, complex network-based image segmentation poses a very important limitation: the excessive number of nodes of the underlying network. In this work we propose a approach based on complex networks suitable for the segmentation of image with large dimensions that is accurate and yet fast. To accomplish that, we have incorporated the concept of Super Pixels aiming at reducing the number of the nodes in the network. The results have shown that the proposed approach delivered accurate image segmentation within low computational times. Another contribution worth mentioning is the determination of the best values for the parameters needed by the underlying graphbased segmentation and community detection algorithms, which enabled the proposed approach to become less dependent on the parameters. To the best of our knowledge, this is a new contribution to the field

La taille des pixels des capteurs d’image CMOS approche aujourd’hui lemicron. Dans ce contexte, le courant d’obscurité reste un paramètrecritique. Il se superpose au courant photogénéré en affectant la qualité del’image par l’apparition de pixels blancs. La contamination métalliqueintroduite au cours du procédé de fabrication joue un rôle prépondérant dansla création des défauts à l’origine de ce courant d’obscurité. Cette étude apermis d’établir les seuils de dangerosité de différents élémentsmétalliques sur la technologie imageur. L’origine de contaminationsaccidentelles a été identifiée lors de crises de rendement. Pour cela, untravail sur les techniques de détection a été mené par µPCD, DLTS, pompagede charge, SIMS, TEM et photoluminescence. La spectroscopie de courantd’obscurité (DCS), particulièrement efficace dans ce contexte, a étédéveloppée pour l’identification de contaminations en or, tungstène etmolybdène, avec des limites de détection qui atteignent 108 à 1010 at/cm3.Nous observons la quantification du courant d’obscurité et étudionsl’amplification du champ électrique sur le taux de génération afin demodéliser les pics de courant d’obscurité obtenus. Le comportement decertains métaux dans le silicium est précisé par ces expériences, et nousévaluons l’efficacité de piégeage de plusieurs substrats imageur. Ce travailconduit à la mise en place de protocoles de contrôle de la contaminationmétallique en salle blanche. / Pixels size of CMOS image sensors is now decreasing towards one micron. Inthat context, dark current is a critical parameter. It superimposes with thecurrent generated by photons and affects the image quality with whitepixels. The metallic contamination introduced during the fabrication processplays an important role in the generation of defects that induce this darkcurrent. This study has allowed to determine dangerousness thresholds ofseveral metals on the imager technology. The origin of some accidentalcontaminations has been identified during yield crisis. Some work withdetection techniques has been performed with µPCD, DLTS, charge pumping,SIMS, TEM and photoluminescence. Dark current spectroscopy (DCS),particularly adapted to this situation, has been developped for theidentification of gold, tunsgten and molybdenum contaminations, withdetection limits that reach 108 to 1010 at/cm3. We have observed the darkcurrent quantization and studied the electric field enhancement ofgeneration rate to model the dark current peaks obtained. The behavior ofsome metals in silicon is confirmed by these experiments and we haveevaluated the getter efficiency of different substrates for image sensors.This work has lead to the application of protocols for the metalliccontamination control in clean room.

Segmentação de imagens é ainda uma etapa desafiadora do processo de reconhecimento de padrões. Entre as abordagens de segmentação, muitas são baseadas em particionamento em grafos, as quais apresentam alguns inconvenientes, sendo um deles o tempo de processamento muito elevado. Com as recentes pesquisas na teoria de redes complexas, as técnicas de reconhecimento de padrões baseadas em grafos melhoraram consideravelmente. A identificação de grupos de vértices pode ser considerada um processo de detecção de comunidades de acordo com a teoria de redes complexas. Como o agrupamento de dados está relacionado com a segmentação de imagens, esta também pode ser abordada através de redes complexas. No entanto, a segmentação de imagens baseado em redes complexas apresenta uma limitação fundamental, que é o número excessivo de nós na rede. Neste trabalho é proposta uma abordagem de redes complexas para segmentação de imagens de grandes dimensões que é ao mesmo tempo precisa e rápida. Para alcançar este objetivo, é incorporado o conceito de Super Pixels, visando reduzir o número de nós da rede. Os experimentos mostraram que a abordagem proposta produz segmentações de boa qualidade em baixo tempo de processamento. Além disso uma das principais contribuições deste trabalho é a determinação dos melhores parâmetros, uma vez que torna o método bastante independente dos parâmetros, o que não fora alcançado antes em nenhuma pesquisa da área / Image segmentation is still a challenging stage of the pattern recognition process. Amongst the various segmentation approaches, some are based on graph partitioning, many of which show some drawbacks, such as the high processing times. Recent trends on complex network theory have contributed considerably to the development of graph-based pattern recognition techniques. The identification of group of vertices can be considered a community detection process according to complex network theory. Since data clustering is closely related to image segmentation, image segmentation tasks can also be tackled by complex networks. However, complex network-based image segmentation poses a very important limitation: the excessive number of nodes of the underlying network. In this work we propose a approach based on complex networks suitable for the segmentation of image with large dimensions that is accurate and yet fast. To accomplish that, we have incorporated the concept of Super Pixels aiming at reducing the number of the nodes in the network. The results have shown that the proposed approach delivered accurate image segmentation within low computational times. Another contribution worth mentioning is the determination of the best values for the parameters needed by the underlying graphbased segmentation and community detection algorithms, which enabled the proposed approach to become less dependent on the parameters. To the best of our knowledge, this is a new contribution to the field