The riveting interplay of industrialization, informalization, and exponential technological growth of recent years has shifted the attention from classical machine learning techniques to more sophisticated deep learning approaches; yet its intrinsic black-box nature has been impeding its widespread adoption in transparency-critical operations. In this rapidly evolving landscape, where the symbiotic relationship between research and practical applications has never been more interwoven, the contribution of this paper is twofold: advancing gradient faithfulness of CAM methods and exploring new frontiers beyond it. In the first part, we theorize three novel gradient-based CAM formulations, aimed at replacing and superseding traditional Grad-CAM-based methods by tackling and addressing the intricately and persistent vanishing and saturating gradient problems. As a consequence, our work introduces novel enhancements to Grad-CAM that reshape the conventional gradient computation by incorporating a customized and adapted technique inspired by the well-established and provably Expected Gradients’ difference-from-reference approach. Our proposed techniques– Expected Grad-CAM, Expected Grad-CAM++and Guided Expected Grad-CAM– as they operate directly on the gradient computation, rather than the recombination of the weighing factors, are designed as a direct and seamless replacement for Grad-CAM and any posterior work built upon it. In the second part, we build on our prior proposition and devise a novel CAM method that produces both high-resolution and class-discriminative explanation without fusing other methods, while addressing the issues of both gradient and CAM methods altogether. Our last and most advanced proposition, Hyper Expected Grad-CAM, challenges the current state and formulation of visual explanation and faithfulness and produces a new type of hybrid saliencies that satisfy the notion of natural encoding and perceived resolution. By rethinking faithfulness and resolution is possible to generate saliencies which are more detailed, localized, and less noisy, but most importantly that are composed of only concepts that are encoded by the layerwise models’ understanding. Both contributions have been quantitatively and qualitatively compared and assessed in a 5 to 10 times larger evaluation study on the ILSVRC2012 dataset against nine of the most recent and performing CAM techniques across six metrics. Expected Grad-CAM outperformed not only the original formulation but also more advanced methods, resulting in the second-best explainer with an Ins-Del score of 0.56. Hyper Expected Grad-CAM provided remarkable results across each quantitative metric, yielding a 0.15 increase in insertion when compared to the highest-scoring explainer PolyCAM, totaling to an Ins-Del score of 0.72.

[pt] A indústria mineral vem presenciando, ao longo das últimas décadas, uma redução da qualidade de minério de ferro extraído e o surgimento de novas demandas ambientais. Esta conjuntura fortalece a busca por produtos provenientes do minério de ferro que atendam aos requisitos da indústria siderúrgica, como é o caso de novos aglomerados de minério de ferro. A Microtomografia de Raios-X (microCT) permite a caracterização da estrutura tridimensional de uma amostra, com resolução micrométrica, de forma não-destrutiva. Entretanto, tal técnica apresenta diversas limitações. Quanto melhor a resolução, maior o tempo de análise e menor o volume de amostra adquirido. Modelos de Super Resolução (SR), baseados em Deep Learning, são uma poderosa ferramenta para aprimorar digitalmente a resolução de imagens tomográficas adquiridas em pior resolução. Este trabalho propõe o desenvolvimento de uma metodologia para treinar três modelos de SR, baseados na arquitetura EDSR, a partir de imagens tomográficas de briquetes de redução direta: Um modelo para aumento de resolução de 16 um para 6 um, outro para aumento de 6 um para 2 um, e o terceiro para aumento de 4 um para 2 um. Esta proposta tem como objetivo mitigar as limitações do microCT, auxiliando o desenvolvimento de novas metodologias de Processamento Digital de Imagens para os aglomerados. A metodologia inclui diferentes propostas para avaliação do desempenho da SR, como comparação de PSNR e segmentação de poros. Os resultados apontam que a SR foi capaz de aprimorar a resolução das imagens tomográficas e mitigar ruídos habituais da tomografia. / [en] The mining industry has been witnessing a reduction of extracted iron ore s quality and the advent of new environmental demands. This situation reinforces a search for iron ore products that meet the requirements of the steel industry, such as new iron ore agglomerates. X-ray microtomography (microCT) allows the characterization of a sample s three-dimensional structure, with micrometer resolution, in a non-destructive analysis. However, this technique presents several limitations. Better resolutions greatly increase analysis time and decrease the acquired sample’s volume. Super-Resolution (SR) models, based on Deep Learning, are a powerful tool to digitally enhance the resolution of tomographic images acquired at lower resolutions. This work proposes the development of a methodology to train three SR models, based on EDSR architecture, using tomographic images of direct reduction briquettes: A model for enhancing the resolution from 16 um to 6 um, another for enhancing from 6 um to 2 um, and the third for enhancing 4 um to 2 um. This proposal aims to mitigate the limitations of microCT, assisting the development and implementation of new Digital Image Processing methodologies for agglomerates. The methodology includes different proposals for SR s performance evaluation, such as PSNR comparison and pore segmentation. The results indicate that SR can improve the resolution of tomographic images and reduce common tomography noise.

Der zunehmende Einsatz regenerativer Energiequellen erfordert die Nutzung von Energiezwischenspeichern, die umweltfreundlich, günstig und skalierbar sein sollten. Die Zink-Luft-Flussbatterie (ZLFB) kann perspektivisch diese Anforderungen erfüllen, wobei zur Bereitstellung der gespeicherten Energie eine Suspension aus Zinkpartikeln in einem gelierten Elektrolyt durch eine elektrochemische Zelle gepumpt wird. Um die Strömungsstruktur der ZLFB auszulegen und Fehlfunktionen zu vermeiden, ist ein grundlegendes Verständnis der Rheologie der Zinksuspension notwendig. Außerdem kann über die Einstellung einer geeigneten Strömung die bei der Entladung erreichte elektrische Leistungsdichte gesteigert werden. Bereits die Flüssigphase der Zinksuspension weist eine komplexe nicht-Newtonsche Rheologie auf, welche durch die Zugabe der Partikel komplexer wird. Für das grundlegende Verständnis der Rheologie werden daher Modellexperimente durchgeführt, wobei in dieser Arbeit ein L-förmiger Kanal mit Strömungsaufweitung untersucht wurde, um die komplexen strömungsmechanischen Eigenschaften der Zinksuspension abzubilden. Zur Erfassung des Strömungsfeldes ist eine Ortsauflösung von 1 … 2 mm in einem Messbereich von 20 × 15 mm2 erforderlich. Ultraschall ist prinzipiell geeignet, um das Strömungsfeld in der opaken Suspension zu erfassen, wobei die wesentliche Herausforderung in den starken Wellenfrontverzerrungen besteht, welche durch die Zinkpartikel eingebracht werden. Es konnte gezeigt werden, dass die Ultrasound Imaging Velocimetry (UIV) robuster gegenüber diesen Störungen ist, als die Ultraschall-Doppler-Velozimetrie (UDV). Die UIV wurde daher mittels Geschwindigkeitsnormal an die messtechnischen Randbedingungen der Zinksuspension angepasst und charakterisiert. Bei einer Ortsauflösung von 1,6 mm wurde eine Gesamtmessunsicherheit von 2,5 % axial und 4,1 % lateral zur Schallausbreitungsrichtung erreicht. Das im Modellexperiment gemessene Strömungsfeld weist eine Totzone an der Strömungsumlenkung auf, deren Auftreten durch eine von der Scherhistorie abhängige Viskosität erklärt werden kann. Dieser Effekt wird als Thixotropie bezeichnet. Durch die In-situ-Messung der Strömung in einer aktiven ZLFB kann eine Korrelation von Strömung und elektrischer Leistung erfolgen und die erzielte Leistungsdichte perspektivisch durch eine Anpassung der Strömung gesteigert werden. Bei der Messung im 2,6 mm hohen Anodenspalt muss aufgrund der komplexen Rheologie der Suspension und der daraus resultierenden hohen Geschwindigkeitsgradienten eine Ortsauflösung von unter 100 µm bei gleichzeitig kleiner Ultraschallfrequenz realisiert werden, da der Ultraschall für hohe Frequenzen nicht in die Zinksuspension eindringt. Um dieses Ziel zu erreichen, wurde die Super Resolution Ultrasound Particle Tracking Velocimetry (SRPTV) genutzt, welche Ortsauflösungen unterhalb des Beugungslimits ermöglicht. Einzelne nichtlineare Streupartikel werden mittels Harmonic Imaging isoliert abgebildet und verfolgt, wobei die durch die Zinkpartikel eingebrachten Wellenfrontstörungen durch einen kohärenzgewichteten Strahlformer kompensiert werden. Es wurde eine Ortsauflösung von 67 µm axial und 30 µm lateral zur Schallausbreitungsrichtung bei einer Anregungswellenlänge von 330 µm erreicht. Trotz der stark streuenden Zinksuspension, konnte so eine Messung in der aktiven ZLFB mit einer maximalen Messunsicherheit von 12,5 % durchgeführt werden. Dabei wurde eine Wandgleitgeschwindigkeit von 3 mm s−1 bei einer maximalen Geschwindigkeit von etwa 8 mm s−1 festgestellt. Die SRPTV kann darüber hinaus in anderen technischen Prozessen eingesetzt werden, in denen Suspensionsströmungen in kleinen Geometrien auftreten.:Symbolverzeichnis xiii Abkürzungsverzeichnis xv 1 Einleitung 1 1.1 Motivation und Zielstellung 1 1.2 Stand der Technik 4 1.3 Lösungsansatz und Struktur der Arbeit 8 2 Theoretische Grundlagen 11 2.1 Grundlagen der Schallausbreitung 11 2.1.1 Schallausbreitung in homogenen Medien 11 2.1.2 Schallausbreitung in inhomogenen Medien 13 2.2 Ultraschall Bildgebung 14 2.2.1 Phased-Array-Prinzip 15 2.2.2 Plane wave imaging 16 2.2.3 Grenzen der Schallfeldfokussierung 16 2.3 Messung von Strömungsfeldern 17 2.3.1 Ultraschall-Doppler-Velozimetrie 18 2.3.2 Ultrasound Imaging Velocimetry 19 2.3.3 Ultrasound Particle Tracking Velocimetry 19 2.4 Nichtlineare akustische Effekte 20 2.4.1 Beschreibung von Linearität 20 2.4.2 Interaktion von nichtlinearen Streupartikeln und Schallwelle 20 2.4.3 Harmonic Imaging 21 3 Experimentelle Grundlagen 25 3.1 Charakterisierung der Zinksuspension 25 3.1.1 Zusammensetzung der Zinksuspension 25 3.1.2 Bestimmung von Dämpfung und Schallgeschwindigkeit 26 3.1.3 Bestimmung der spezifischen akustischen Impedanz 29 3.2 Messtechnik 30 3.2.1 Ultraschallforschungsplattform: Phased Array Ultrasound Dopp- ler Velocimeter 30 3.2.2 Ultraschallwandler 32 4 Verfahren zur Strömungsmessung im Modellexperiment 37 4.1 Experimenteller Aufbau 37 4.2 Untersuchung geeigneter Verfahren zur Messung von Strömungsfel- dern in der Zinksuspension 38 4.3 Optimierung der Signalverarbeitung und Charakterisierung der Messei- genschaften 44 4.3.1 Geschwindigkeitsnormal 44 4.3.2 Optimierung der Messsystemparameter 45 4.3.3 Charakterisierung der Messeigenschaften 48 4.3.4 Validierung 49 4.4 Messung der Suspensionsströmung im Modellexperiment 55 4.4.1 Messergebnisse 55 4.4.2 Vergleich von Simulation und Messung 58 4.5 Fazit 61 5 Verfahren zur In-situ-Strömungsmessung in einer Zink-Luft-Flussbatterie 63 5.1 Experimenteller Aufbau 63 5.2 Strömungsmessung unterhalb des Beugungslimits - Super Resolution Ultrasound Particle Tracking Velocimetry (SRPTV) 65 5.2.1 Nutzung nichtlinearer Streupartikel 68 5.2.2 Trennung von linearem und nichtlinearem Signalanteil 72 5.2.3 Strahlformung mit Kompensation der Streuung 76 5.2.4 Particle Tracking 80 5.3 Charakterisierung der Messeigenschaften 81 5.3.1 Vorgehen zur Charakterisierung der Messeigenschaften 82 5.3.2 Untersuchung der Positionsunsicherheit 83 5.3.3 Untersuchung der Geschwindigkeitsunsicherheit 92 5.4 Messung an einer aktiven Zink-Luft-Flussbatterie 95 5.4.1 Aufbau und Durchführung 95 5.4.2 Messergebnisse 97 5.4.3 Vergleich von Simulation und Messung 97 5.5 Fazit 102 6 Zusammenfassung und Ausblick 103 6.1 Erkenntnisse und Fortschritt 103 6.1.1 Ultrasound Imaging Velocimetry 103 6.1.2 Super Resolution Ultrasound Particle Tracking Velocimetry 104 6.1.3 Fazit 106 6.2 Ausblick und weiterführende Arbeiten 106 6.2.1 Messtechnik 106 6.2.2 Anwendung 107 Literaturverzeichnis 109 Publikationsverzeichnis 117 Artikel in Zeitschriften mit peer-review 117 Tagungsbeiträge 117 Patente 119 / For the efficient use of renewable energies, energy storage systems are required that are environmentally friendly, low priced and scalable. The zinc-air flow battery (ZAB), which is operated by pumping an opaque suspension of zinc particles in an gelled electrolyte through an electrochemical cell, is a promising candidate as energy storage system for these requirements. To design the fluidic structures and avoid malfunction, a fundamental understanding of the rheology of the zinc suspension is required. Additionally, the electrical performance of the cell can be imporved by optimizing the flow in the electrochemical cell. The liquid phase of the suspension itself has complex non-Newtonian properties, which are even more complex when the particles are considered. For the fundamental understanding of the suspension rheology, model experiments are conducted. In this work an L-shaped channel with a widening is used to represent relevant effects from the complex rheology of the suspension. To measure the flow field, a spatial resolution of 1 … 2 mm and a measurement area of 20 × 15 mm2 are required. Ultrasound can be used to measure the flow in opaque liquids, but wavefront distortions are introduced by the zinc particles. Established measurement methods for homogeneous opaque fluids, the Ultrasound Imaging Velocimetry (UIV) and the Ultrasound Doppler Velocimetry (UDV), were compared for the application at the suspension. The UIV has a 50 % lower random deviation, which makes it more suitable for the flow measurement in the suspension and it was adapted to the measurement conditions in the suspension. At a spatial resolution of 1.66 mm, a velocity uncertainty of 2.5 % axial and 4.1 % lateral to the ultrasound propagation were achieved. The application of the UIV to the suspension flow in the model experiment revealed a thixotropic behavior of the fluid, which resulted in a dead flow zone opposite to the inlet of the channel. The in situ measurement of the flow in an active ZAB, allows to correlate electrical performance and flow and thereby an improvement of the cell performance by adapting the flow. For the measurement in the anodic channel with a width of 2.6 mm, a spatial resolution of 100 µm is required because of the high velocity gradients due to the non-Newtonian rheology of the suspension. The high spatial resolution has to be achieved at low ultrasound frequencies, since the ultrasound does not penetrate into the suspension for high frequencies. To achieve this, the Super Resolution Ultrasound Particle Tracking Velocimetry (SRPTV) was used, which allows a spatial resolution beyond the diffraction limit. Harmonic Imaging is used to image isolated non-linear tracer particles, which are tracked for velocity estimation. The speckle and image distortion due to the induced wavefront distortions are compensated with a coherence weighting beamformer. A spatial resolution of 67 µm axial and 30 µm lateral to the ultrasound propagation were achieved. Despite the strong scattering of the ultrasound at the zinc particles, a maximum velocity uncertainty of 12.5 % referred to the maximum velocity was achieved for the measurement in the active ZAB. A slip velocity of 3 mm at a maximum velocity of 8 mm was observed. The SRPTV can be applied to other technical processes, where suspension flows in small geometries play an important role.:Symbolverzeichnis xiii Abkürzungsverzeichnis xv 1 Einleitung 1 1.1 Motivation und Zielstellung 1 1.2 Stand der Technik 4 1.3 Lösungsansatz und Struktur der Arbeit 8 2 Theoretische Grundlagen 11 2.1 Grundlagen der Schallausbreitung 11 2.1.1 Schallausbreitung in homogenen Medien 11 2.1.2 Schallausbreitung in inhomogenen Medien 13 2.2 Ultraschall Bildgebung 14 2.2.1 Phased-Array-Prinzip 15 2.2.2 Plane wave imaging 16 2.2.3 Grenzen der Schallfeldfokussierung 16 2.3 Messung von Strömungsfeldern 17 2.3.1 Ultraschall-Doppler-Velozimetrie 18 2.3.2 Ultrasound Imaging Velocimetry 19 2.3.3 Ultrasound Particle Tracking Velocimetry 19 2.4 Nichtlineare akustische Effekte 20 2.4.1 Beschreibung von Linearität 20 2.4.2 Interaktion von nichtlinearen Streupartikeln und Schallwelle 20 2.4.3 Harmonic Imaging 21 3 Experimentelle Grundlagen 25 3.1 Charakterisierung der Zinksuspension 25 3.1.1 Zusammensetzung der Zinksuspension 25 3.1.2 Bestimmung von Dämpfung und Schallgeschwindigkeit 26 3.1.3 Bestimmung der spezifischen akustischen Impedanz 29 3.2 Messtechnik 30 3.2.1 Ultraschallforschungsplattform: Phased Array Ultrasound Dopp- ler Velocimeter 30 3.2.2 Ultraschallwandler 32 4 Verfahren zur Strömungsmessung im Modellexperiment 37 4.1 Experimenteller Aufbau 37 4.2 Untersuchung geeigneter Verfahren zur Messung von Strömungsfel- dern in der Zinksuspension 38 4.3 Optimierung der Signalverarbeitung und Charakterisierung der Messei- genschaften 44 4.3.1 Geschwindigkeitsnormal 44 4.3.2 Optimierung der Messsystemparameter 45 4.3.3 Charakterisierung der Messeigenschaften 48 4.3.4 Validierung 49 4.4 Messung der Suspensionsströmung im Modellexperiment 55 4.4.1 Messergebnisse 55 4.4.2 Vergleich von Simulation und Messung 58 4.5 Fazit 61 5 Verfahren zur In-situ-Strömungsmessung in einer Zink-Luft-Flussbatterie 63 5.1 Experimenteller Aufbau 63 5.2 Strömungsmessung unterhalb des Beugungslimits - Super Resolution Ultrasound Particle Tracking Velocimetry (SRPTV) 65 5.2.1 Nutzung nichtlinearer Streupartikel 68 5.2.2 Trennung von linearem und nichtlinearem Signalanteil 72 5.2.3 Strahlformung mit Kompensation der Streuung 76 5.2.4 Particle Tracking 80 5.3 Charakterisierung der Messeigenschaften 81 5.3.1 Vorgehen zur Charakterisierung der Messeigenschaften 82 5.3.2 Untersuchung der Positionsunsicherheit 83 5.3.3 Untersuchung der Geschwindigkeitsunsicherheit 92 5.4 Messung an einer aktiven Zink-Luft-Flussbatterie 95 5.4.1 Aufbau und Durchführung 95 5.4.2 Messergebnisse 97 5.4.3 Vergleich von Simulation und Messung 97 5.5 Fazit 102 6 Zusammenfassung und Ausblick 103 6.1 Erkenntnisse und Fortschritt 103 6.1.1 Ultrasound Imaging Velocimetry 103 6.1.2 Super Resolution Ultrasound Particle Tracking Velocimetry 104 6.1.3 Fazit 106 6.2 Ausblick und weiterführende Arbeiten 106 6.2.1 Messtechnik 106 6.2.2 Anwendung 107 Literaturverzeichnis 109 Publikationsverzeichnis 117 Artikel in Zeitschriften mit peer-review 117 Tagungsbeiträge 117 Patente 119

Il existe au niveau d’interfaces séparant des milieux de constantes diélectriques de signes opposés des ondes électromagnétiques confinées à proximité de ces interfaces. On parle d’ondes de surface. C’est notamment le cas des métaux et des cristaux polaires : on parle alors de plasmons-polaritons de surface et de phonons-polaritons de surface respectivement. L’objectif de cette thèse est de revisiter certains aspects théoriques associés à ces ondes de surface.Dans un premier temps, en nous basant sur le formalisme de Green, nous donnons un moyen d’obtenir une expression du champ des ondes de surface sous forme de somme de modes. En présence de pertes, ces ondes ont nécessairement un vecteur d’onde ou une pulsation complexe. Nous donnons ainsi deux expressions de leur champ, correspondant à chacun de ces deux cas, et discutons de l’opportunité d’utiliser l’une ou l’autre de ces expressions.Nous posons par la suite les bases d’une optique de Fourier et d’une optique géométrique des ondes de surface. Nous montrons comment obtenir une équation de Helmholtz à deux dimensions pour les ondes de surface, un principe d’Huygens-Fresnel pour les ondes de surface, ainsi qu’une équation eikonale pour les ondes de surface, qui s’applique sous certaines hypothèses. Nous nous intéressons également à la superlentille proposée par Pendry, qui s’appuie sur les ondes de surface. Nous étudions notamment le fonctionnement de cette superlentille en régime impulsionnel, et montrons qu’en présence de pertes, il est possible d’obtenir une meilleure résolution avec certaines formes d’impulsion par rapport au régime harmonique, au prix d’une importante baisse de signal toutefois.Nous développons ensuite un traitement quantique des ondes de surface. Nous calculons au préalable une expression de leur énergie, et nous donnons une expression de leur hamiltonien et de leurs opérateurs champ. Sans pertes, nous montrons que le facteur de Purcell prédit par notre théorie quantique est rigoureusement égal au facteur de Purcell calculé avec des outils classiques. Nous comparons ensuite ce facteur de Purcell à celui calculé classiquement avec pertes, et montrons sur un exemple que les pertes peuvent être négligées dans de nombreux cas. Nous donnons enfin une expression des coefficients d’Einstein associés aux ondes de surface permettant d’étudier la dynamique de l’inversion de population d’un milieu fournissant un gain aux ondes de surface. Nous appliquons par la suite ce formalisme quantique à l’interaction électrons-phonons-polaritons de surface dans les puits quantiques, notamment leur interaction avec un mode de phonon du puits particulièrement confiné grâce à un effet de constante diélectrique proche de zéro (epsilon near zero, ENZ). / Interfaces between materials having opposite dielectric constants support electromagnetic waves confined close to these interfaces called surface waves. For metals and polar crystals, they are respectively called surface plasmon-polaritons and surface phonon-polaritons. The goal of this thesis is to revisit some theoretical aspects associated to these surface waves.Using the Green formalism, we derive an expression of the surface wave field as a sum of modes. With losses, these waves must have a complex wave vector or frequency. Thus we give two expressions of their field, for each of these cases, and discuss when each of these expressions should be used.We then give the basis of a surface wave Fourier optics and geometrical optics. We derive a 2D Helmholtz equation for surface waves, a Huygens-Fresnel principle for surface waves, and an eikonal equation for surface waves. We then take a look at Pendry’s superlens, in which surface waves play a major role. We study the behavior of the superlens in pulsed mode taking losses into account, and show that its resolution can be increased for some pulse shapes compared to the steady state, at the expense of a signal decay.We then develop a quantum treatment of surface waves. We first calculate their energy, and then give an expression of their hamiltonian and field operators. Without losses, we show that the Purcell factor given by our quantum theory is perfectly equal to the Purcell factor given by the classical theory. We then compare this Purcell factor to the lossy case on an example, and show that losses can often be neglected. We then derive the Einstein coefficients associated to surface wave emission and absorption, which allow studying the population inversion dynamics of a gain medium. We then use this quantum formalism to study the interaction between electrons and surface phonon-polaritons in quantum wells, particularly their interaction with a phonon mode which features high confinement thanks to an epsilon near zero (ENZ) effect.

No description available.

The goal of this project is to study novel nanoscale excitation volumes, sensitive enoughto study individual chromophores and go on to study new and exciting self assemblyapproaches to this problem. Small excitation volumes may be engineered using light con-finement inside apertures in metal films. These apertures enhance fluorescence emissionrates, quantum yields, decrease fluorescence quenching, enable higher signal-to-noiseratios and allow higher concentration single chromophore fluorescence, to be studied byrestricting this excitation volume. Excitation volumes are reported on using the chro-mophore's fluorescence by utilising fluorescence correlation spectroscopy, which monitorsfluctuations in fluorescence intensity. From the correlation in time, we can find the res-idence time, the number of chromophores, the volume in which they are diffusing andtherefore the fluorescence emission efficiency. Fluorescence properties are a probe ofthe local environment, a particularly powerful tool due to the high brightness (quantumyield) fluorescent dyes and sensitive photo-detection equipment both of which are readilyavailable, (such as avalanche photodiodes and photomultiplier tubes). Novel materialscombining the properties of conducting and non-conducting materials at scales muchsmaller than the incident wavelength are known as meta-materials. These allow combi-nations of properties not usually possible in natural materials at optical frequencies. Theproperties reported so far include; negative refraction, negative phase velocity, fluorescenceemission enhancement, lensing and therefore light confinement has also been proposed tobe possible. Instead of expensive and slow lithography methods many of these materialsmay be fabricated with self assembly techniques, which are truly nanoscopic and otherwiseinaccessible with even the most sophisticated equipment. It was found that nanoscaled volumes from ZMW and HMMs based on NW arrays wereall inefficient at enhancing fluorescence. The primary cause was the reduced fluorescencelifetime reducing the fluorescence efficiency, which runs contrary to some commentatorsin the literature. NW based lensing was found to possible in the blue region of the opticalspectrum in a HMM, without the background fluorescence normally associated with a PAAtemplate. This was achieved using a pseudo-ordered array of relatively large nanowireswith a period just smaller than lambda / 2 which minimised losses. Nanowires in the traditionalregime lambda / 10 produced significant scattering and lead to diffraction, such that they werewholly unsuitable for an optical lensing application.

Ce manuscrit de thèse présente l’étude des techniques de super-résolution latérale en nanoscopie optique, qui est une des nouvelles techniques d'imagerie haute résolution, aujourd'hui largement utilisée en biophysique et en imagerie médicale, pour imager et caractériser des nanostructures, tout en conservant les avantages de l'imagerie optique en champ lointain comme un vaste champ, la visualisation et l’analyse en temps réel…Un des défis futurs de la microscopie 3D super-résolue est d’éviter l’utilisation des marqueurs fluorescents. La microscopie interférométrique fait partie des techniques d’imagerie 3D sans marquage permettant la détection de nanostructures. Pour améliorer le pouvoir de détection de ce système optique, un premier protocole de traitement d’images a été développé et implémenté, permettant ainsi de révéler des structures initialement non mesurables. Puis, pour améliorer la résolution latérale du système, une nouvelle technique combinant l’interférométrie et le principe du nano-jet photonique a été développée permettant l’observation d’objets de taille inférieure à la limite de diffraction de l’instrument optique. / This manuscript presents the study of the lateral super-resolution techniques in optical nanoscopy, which is a new high-resolution imaging method now widely used in biophysics and medical imaging, to observe and measure nanostructures, with the advantages of far field optical imaging, such as a large field of view, visualization and analysis in real time…One of the future challenges of 3D super resolution microscopy is to avoid the use of fluorescent markers. Interferometric microscopy is a 3D label-free imaging technique enabling the detection of nanostructures. To improve the detection capability of this optical system, a first version of a protocol composed of image processing methods was developed and implemented, revealing structures initially unmeasurable. Then, to improve the lateral resolution of the system, a new technique combining interferometry and the principle of the photonic nano-jet has been developed, thus allowing the observation of objects of a size smaller than the diffraction limit of the optical instrument.

Made available in DSpace on 2016-06-02T19:03:57Z (GMT). No. of bitstreams: 1 5415.pdf: 8638421 bytes, checksum: 0e5c5abf95c786434202fdae3e69dc1e (MD5) Previous issue date: 2013-06-24 / Financiadora de Estudos e Projetos / The resolution enhancement of an image is always desirable, independently of its objective, but mainly if the image has the purpose of visual analysis. The hardware development for increasing the image resolution still has a higher cost than the algorithmic solutions for super-resolution. Like image restoration, super-resolution is also an ill-conditioned inverse problem, and has an infinite number of solutions. This work analyzes the iterative restoration methods (Van Cittert, Tikhonov-Miller and Conjugate Gradiente) which propose solutions for the ill-conditioning problem and compares them with the IBP method (Iterative Back Projection). The analysis of the found similarities is the basis of a generalization, such that other iterative restoration methods can have their properties adapted, as regularization of the ill-conditioning, noise reduction and other degradations and the increase of the convergence rate can be incorporated to the techniques of super-resolution. Two new methods were created as case studies of the proposed generalization: the first one is a super-resolution method for dynamic magnetic resonance imaging (MRI) of the swallowing process, that uses an adaptiveWiener filtering as regularization and a non-rigid registration; and the second one is a pan sharpening method of SPOT satellite bands, that uses sampling based on sensor s characteristics and non-adaptive Wiener filtering. / A melhora da resolução de uma imagem é sempre desejada, independentemente de seu objetivo, mas principalmente se destinada a análise visual. O desenvolvimento de hardware para o aumento de resolução de uma imagem em sua captura ainda possui o custo mais elevado do que as soluções algorítmicas de super resolução (SR). Assim como a restauração de imagens, a super resolução também é um problema inverso mal-condicionado e possui infinitas soluções. Este trabalho analisa métodos de restauração iterativos (Van Cittert, Tikhonov-Miller e Gradiente Conjugado) que proponham soluções para o problema do malcondicionamento e os compara com o método IBP (Iterative Back-Projection). A análise das semelhanças encontradas é base para uma generalização de modo que outros métodos iterativos de restauração possam ter suas propriedades adaptadas, tais como regularização do mal-condicionamento, redução do ruído e outras degradações e aumento na taxa de convergência, para que possam ser incorporadas à técnicas de super resolução. Dois novos métodos foram criados como estudo de caso da generalização proposta: o primeiro é um método de super-resolução para imageamento por ressonância magnética (MRI) dinâmico do processo de deglutição, que utiliza uma filtragem de Wiener adaptativa como regularização e registro não-rígido; o segundo é um método de pansharpening das bandas do satélite SPOT, que utiliza amostragem baseada nas características do sensor e filtragem de Wiener não-adaptativa.

Le cerveau est un organe complexe composé de neurones et des cellules non-neuronales. La communication entre les neurones a lieu via les synapses, dont le remodelage morphologique est considéré essentiel pour le traitement et le stockage des informations dans le cerveau des mammifères. Récemment, ce point de vue neuro-centré de la fonction synaptique a évolué, en prenant également en compte les processus gliaux à proximité immédiate de la synapse. Cependant, comme leur structure est bien en deçà de la résolution spatiale de la microscopie optique conventionnelle, les progrès dans les enquêtes dans leur environnement physiologique, le cerveau intact, ont été entravés. En effet, on sait peu sur les variations nanométriques de la morphologie des épines dendritiques et l'interaction avec les processus gliaux, et, finalement, comment elles affectent la transmission synaptique in vivo. Dans cette thèse, nous cherchons à visualiser la dynamique de la nano-morphologie des épines dendritiques et les processus gliaux dans le cortex à tonneaux de souris in vivo. Nous avons donc mis en place l’imagerie super-résolution 2P-STED en temps réel, ce qui permet une haute résolution spatiale et la pénétration profonde des tissus, chez la souris anesthésiée in vivo. Nous montrons que la nano-morphologie des épines est diversifiée, variable, mais globalement stable, et que les différences dans la morphologie des épines peut avoir un effet sur leur compartimentation in vivo. En outre, la mise en œuvre de l’imagerie super-résolution en double couleur in vivo et le développement d'une approche de marquage astrocytaire, nous ont permis de fournir la caractérisation à l'échelle nanométrique des interactions neurone-glie. Ces résultats apportent un aperçu sans précédent dans la dynamique de la synapse à l'échelle nanométrique in vivo, et ouvrent la voie à une meilleure compréhension de la façon dont les réarrangements morphologiques des synapses contribuent à la physiologie du cerveau. / The brain is a complex organ consisting of neurons and non-neuronal cells. Communication between neurons takes place via synapses, whose morphological remodeling is thought to be crucial for information processing and storage in the mammalian brain. Recently, this neuro-centric view of synaptic function has evolved, also taking into account the glial processes in close vicinity of the synapse. However, as their structure is well below the spatial resolution of conventional light microscopy, progress in investigating them in a physiological environment, the intact brain, has been impeded. Indeed, little is known on the nanoscale morphological variations of dendritic spines, the interaction with glial processes, and how these affect synaptic transmission in vivo. Here, we aim to visualize the dynamic nano-morphology of dendritic spines in mouse somatosensory cortex in vivo. We implemented super-resolution 2P-STED time-lapse imaging, which allows for high spatial resolution and deep tissue penetration, in anesthetized mice, and show that the nano-morphology of spines is diverse, variable, but on average stable, and that differences in spine morphology can have an effect on spine biochemical compartmentalization in vivo. Moreover, implementation of dual color in vivo super-resolution imaging and a novel astrocytic labeling approach provided the first steps towards nanoscale characterization of neuron-glia interactions in vivo. These findings bring new insights in synapse dynamics at the nanoscale in vivo, and our methodological endeavors help pave the way for a better understanding of how nanoscale aspects of spine morphology and their dynamics might contribute to brain physiology and animal behavior.

Úkolem této diplomové práce je navrhnout vlastní metodu pro zvýšení rozlišení v obraze scény, pokud je k dispozici více snímků dané scény. V teoretické části diplomové práce jsou jako nejlepší metody pro zvýšení rozlišení v obraze vybrány ty, které jsou založeny na principech zpracování signálu. Dále jsou popsány základní požadavky metod pro zvýšení rozlišení v obraze při přítomnosti více snímků stejné scény a jejich typická struktura. Následuje stručný přehled těchto metod a jejich vzájemné porovnání podle optimálních kritérií. Praktická část diplomové práce se zabývá samotným návrhem metody pro zvýšení rozlišení v obraze, pokud je k dispozici více snímků této scény. První navržená metoda je naimplementována a otestována. Při testování této metody je však  zjištěna její špatná funkčnost pro snímky scény s nízkým rozlišením, které vznikly vzájemnou rotací. Z toho důvodu je navržena vylepšená metoda pro zvýšení rozlišení v obraze. Tato metoda využívá při svém výpočtu robustních technik. Díky tomu je již vylepšená metoda nezávislá na rotaci mezi snímky scény s nízkým rozlišením. I tato metoda je řádně otestována a její výsledky jsou porovnány s výsledky první navržené metody pro zvýšení rozlišení v obraze. V porovnání výpočetních časů je lepší první navrhovaná metoda, avšak její výsledky pro obrazy obsahující rotace nejsou kvalitní. Oproti tomu pro obrazy, které vznikly pouze posunem při snímání scény, jsou tyto výsledky velice dobré. Vylepšená metoda je tedy využitelná zejména pro obrazy obsahující rotace. V závěru této práce je ještě navrženo jedno vylepšení, které by mohlo zlepšit výsledky druhé navrhnuté metody pro zvýšení rozlišení v obraze scény.