Zugl.: Aachen, Techn. Hochsch., Diss., 2009.

Universiẗat, Diss., 1999--Hamburg.

This paper provides a fully automated method of finding high-quality seed points in 3D space from a stack of images of neuron cells. These seed points may then be used as initial starting points for automated local tracing algorithms, removing a time consuming required user interaction in current methodologies. Methods to collapse the search space and provide rudimentary topology estimates are also presented. / Master of Science

No description available.

This thesis presents the design and development of an accelerated signal-level radar simulator with an emphasis on space debris monitoring in the Low Earth Orbit. Space surveillance represents a major topic of concern to astronomers as the threat of space debris and orbital overpopulation looms – particularly due to the lack of effective mitigation techniques and the limitations of modern space-monitoring sensors. This work thus aimed to investigate and design possible tools that could be used for training, testing and research purposes, and thereby aid further study in the field. At present, there exist no three-dimensional, ray-traced, signal-level radar simulators available for public use. As such, this thesis proposes an open-source, ray-traced radar simulator that models the interactions between spaceborne targets and terrestrial radar systems. This utilises a ray-tracing algorithm to simulate the effects of debris size, shape, orientation, and material properties when computing radar signals in a typical simulation. The generated received signals, produced at the output of the simulator, were also verified against systems theory, and validated with an existing, well-established simulator. The developed software was designed to aid astronomers and researchers in space situational awareness applications through the simulation of radar designs for orbital surveillance experiments. Due to its open-source nature, it is also expected to be used in training and research environments involving the testing of space-monitoring systems under various simulation conditions. The software offers native support for measured Two-Line Element datasets and the Simplified General Perturbations #4 orbit propagation model, enabling the accurate modelling of targets and the dynamic orbital forces acting upon them. As a result, the software has aptly been named the Space Object Astrodynamics and Radar Simulator – or SOARS. SOARS was built upon the foundations of a general-purpose radar simulator known as the Flexible Extensible Radar Simulator – or FERS – which provided integrated radar models for propagation loss, antenna shapes, Doppler and phase shifts, Radar Cross Section modelling, pulse waveforms, high-accuracy clock mechanisms, and interpolation algorithms. While FERS lacked various features required for space-monitoring applications, many of its implementations were used in SOARS to minimise simulation limits and maximise signal rendering accuracy by supporting an arbitrary number of transmitters, receivers, and targets. The goal was thus to have the simulator limited only by the end-user's system, and to specialise the operation of the software towards space surveillance by integrating additional features – such as built-in models for environmental and system noise, multiscatter effects, and target modelling using meshes comprised of triangular primitives. After completing the software's development, the ray-traced simulator was compared against a more streamlined version of SOARS that made use of point-model approximations for quick-look simulations, and the trade-offs between both simulators (including software runtime, memory utilisation and simulation accuracy) were investigated and evaluated. This assessed the value of implementing ray tracing in a radar simulator operating primarily within space contexts and evaluated the results of both simulators using detection processing as a demonstrated application of the system. And while the use of ray tracing resulted in significant costs in speed and memory, the investigation found that the ray-traced simulator generated more reliable results relative to the point-model version – providing various advantages in test scenarios involving shadowing and multiscatter. The design of the SOARS software, as well as its point-model “baseline” alternative and the investigation into each simulator's advantages and disadvantages, are thus presented in this thesis. The developed programs were released as open-source tools under the GNU General Public Licence and are freely available for public use, modification, and distribution.

In mammals, the basic computations required for azimuthal sound localization are performed by a group of auditory brainstem nuclei known as the superior olivary complex (SOC). The lateral superior olive (LSO), in the SOC, aids in sound localization by computing intensity differences between sounds arriving at the two ears. It does this by comparing excitatory input from the ipsilateral anteroventral cochlear nucleus (AVCN) with inhibitory input from the ipsilateral medial nucleus of the trapezoid body (MNTB), which is driven by the contralateral AVCN. In order for sounds to be accurately localized, the AVCN-LSO and MNTB-LSO projections must be aligned with each other in a frequency-dependent manner. Rough alignment occurs over the course of development, but a significant amount of circuit refinement is required to achieve adult-like precision. Two types of refinement occur in these pathways: 1) physiological, or functional refinement; and 2) anatomical refinement. Little is known about the latter type of refinement in the AVCN-LSO pathway. In order to study this, I conducted a variety of experiments all aimed at anterogradely labeling a small number of cells projecting from the AVCN to the LSO in juvenile rats. I experimented with several approaches in order to develop the technique of ex vivo, sparse axon labeling in this area of the brain. I show the optimal technique developed after testing various tracers, application methods, and incubation times, among others. This optimized technique can now be used in a future experiment that will uncover and describe anatomical refinement in the AVCN-LSO pathway of the auditory brainstem. / Thesis / Master of Science (MSc)

A three-by-three polarization ray tracing matrix method is developed to calculate the polarization transformations associated with ray paths through optical systems. The relationship between the three-by-three polarization ray tracing matrix P method and the Jones calculus is shown in Chapter 2. The diattenuation, polarization dependent transmittance, is calculated via a singular value decomposition of the P matrix and presented in Chapter 3. In Chapter 4 the concept of retardance is critically analyzed for ray paths through optical systems. Algorithms are presented to separate the effects of retardance from geometric transformations. The parallel transport of vectors is associated with non-polarizing propagation through an optical system. A parallel transport matrix Q establishes a proper relationship between sets of local coordinates along the ray path, a sequence of ray segments. The proper retardance is calculated by removing this geometric transformation from the three-by-three polarization ray trace matrix. Polarization aberration is wavelength and spatial dependent polarization change that occurs as wavefrontspropagate through an optical system. Diattenuation and retardance of interfaces and anisotropic elements are common sources of polarizationaberrations. Two representations of polarization aberrationusing the Jones pupil and a polarization ray tracing matrix pupil, are presentedin Chapter 5. In Chapter 6 a new class of aberration, skew aberration is defined, as a component of polarization aberration. Skew aberration is an intrinsic rotation of polarization states due to the geometric transformation of local coordinates; skew aberration occurs independent of coatings and interface polarization. Skew aberration in a radially symmetric system primarily has the form of a tilt plus circular retardance coma aberration. Skew aberration causes an undesired polarization distribution in the exit pupil. A principal retardance is often defined within (-π, + π] range. In Chapter 7 an algorithm which calculates the principal retardance, horizontal retardance component, 45° retardance component, and circular retardance component for given retarder Jones matrices is presented. A concept of retarder space is introduced to understand apparent discontinuities in phase unwrapped retardance. Dispersion properties of retarders for polychromatic light is used to phase unwrap the principal retardance. Homogeneous and inhomogeneous compound retarder systems are analyzed and examples of multi-order retardance are calculated for thick birefringent plates. Mathematical description of the polarization properties of light and incoherent addition of light is presented in Chapter 8, using a coherence matrix. A three-by-three-by-three-by-three polarization ray tracing tensor method is defined in order to ray trace incoherent light through optical systems with depolarizing surfaces. The polarization ray tracing tensor relates the incident light’s three-by-three coherence matrix to the exiting light’s three-by-three coherence matrix. This tensor method is applicable to illumination systems and polarized stray light calculations where rays at an imaging surface pixel have optical path lengths which vary over many wavelengths. In Chapter 9 3D Stokes parameters are defined by expanding the coherence matrix with Gell-Mann matrices as a basis. The definition of nine-by-nine 3D Mueller matrix is presented. The 3D Mueller matrix relates the incident 3D Stokes parameters to the exiting 3D Stokes parameters. Both the polarization ray tracing tensor and 3D Mueller matrix are defined in global coordinates. In Chapter 10 a summary of my work and future work are presented followed by a conclusion.

La génération d'images de synthèse pour la production cinématographique n'a cessé d'évoluer durant ces dernières décennies. Pour le non-expert, il semble que les effets spéciaux aient atteint un niveau de réalisme ne pouvant être dépassé. Cependant, les logiciels mis à la disposition des artistes ont encore du progrès à accomplir. En effet, encore trop de temps est passé à attendre le résultat de longs calculs, notamment lors de la prévisualisation d'effets spéciaux. La lenteur ou la mauvaise qualité des logiciels de prévisualisation pose un réel problème aux artistes. Cependant, l'évolution des cartes graphiques ces dernières années laisse espérer une potentielle amélioration des performances de ces outils, notamment par la mise en place d'algorithmes hybrides rasterisation/ lancer de rayons, tirant profit de la puissance de calcul de ces processeurs, et ce, grâce à leur architecture massivement parallèle. Cette thèse explore les différentes briques logicielles nécessaires à la mise en place d'un pipeline de rendu complexe sur carte graphique, permettant une meilleure prévisualisation des effets spéciaux. Différentes contributions ont été apportées à l'entreprise durant cette thèse. Tout d'abord, un pipeline de rendu hybride a été développé (cf. Chapitre 2). Par la suite, différentes méthodes d'implémentation de l'algorithme de Path Tracing ont été testées (cf. Chapitre 3), de façon à accroître les performances du pipeline de rendu sur GPU. Une structure d'accélération spatiale a été implémentée (cf. Chapitre 4), et une amélioration de l'algorithme de traversée de cette structure sur GPU a été proposée (cf. Section 4.3.2). Ensuite, une nouvelle méthode de décorrélation d'échantillons, dans le cadre de la génération de nombres aléatoires a été proposée (cf. Section 5.4) et a donné lieu à une publication [Dufay et al., 2016]. Pour finir, nous avons tenté de combiner l'algorithme de Path Tracing et les solutions Many Lights, toujours dans le but d'améliorer la prévisualisation de l'éclairage global. Cette thèse a aussi donné lieu à la soumission de trois mémoires d'invention et a permis le développement de deux outils logiciels présentés en Annexe A. / Image synthesis for movie production never stopped evolving over the last decades. It seems it has reached a level of realism that cannot be outperformed. However, the software tools available for visual effects (VFX) artists still need to progress. Indeed, too much time is still wasted waiting for results of long computations, especially when previewing VFX. The delays or poor quality of previsualization software poses a real problem for artists. However, the evolution of graphics processing units (GPUs) in recent years suggests a potential improvement of these tools. In particular, by implementing hybrid rasterization/ray tracing algorithms, taking advantage of the computing power of these processors and their massively parallel architecture. This thesis explores the different software bricks needed to set up a complex rendering pipeline on the GPU, that enables a better previsualization of VFX. Several contributions have been brought during this thesis. First, a hybrid rendering pipeline was developed (cf. Chapter 2). Subsequently, various implementation schemes of the Path Tracing algorithm have been tested (cf. Chapter 3), in order to increase the performance of the rendering pipeline on the GPU. A spatial acceleration structure has been implemented (cf. Chapter 4), and an improvement of the traversal algorithm of this structure on GPU has been proposed (cf. Section 4.3.2). Then, a new sample decorrelation method, in the context of random number generation was proposed (cf. Section 5.4) and resulted in a publication [Dufay et al., 2016]. Finally, we combined the Path Tracing algorithm with the Many Lights solution, always with the aim of improving the preview of global illumination. This thesis also led to the submission of three patents and allowed the development of two software tools presented in Appendix A.

No description available.

No description available.