Search

The utilisation of 3D digitisation and visualisation has grown considerably since 2008 and is becoming an increasingly useful tool for the digital documentation and metric analysis of archaeological artefacts and skeletal remains. It provides public access to rare and fragile specimens of palaeontological and palaeopathological importance whilst reducing the physical impact on these remains. Research in engineering and computer vision provides some insight into the impact of surface properties such as colour, specularity, reflectance and shape on the quality of the recorded 3D image, but within the archaeological and palaeontological disciplines comparable work has not yet been developed. If archaeology and anthropology are to provide long term reliable data from archaeological and palaeontological specimens in a way that doesn’t require repeated re-digitisation, we need to understand the impacts that the taphonomic histories of such samples have on our ability to 3D record them. Understanding the relationship of these taphonomic histories and the surface and optical properties will promote informed choices about the suitability of recording techniques. This thesis considers the taphonomic processes that affect the preservation of bone over archaeological, forensic and palaeontological timescales and the effect this has on the quality of 3D digital models. The digital refit of fragmentary bone samples is considered in relation to the effect of taphonomic alterations to bone. Conclusions regarding the key taphonomic factors and 3D digital model quality are drawn and areas of further work are identified.

Bayesian methods offer a powerful framework for predicting and quantifying uncertainty in various practical systems. This study explores the application of Bayesian approaches in two distinct domains: structured light (SL) systems and pipeline corrosion analysis. In structured light systems, Bayesian method is utilized to build a complete coordinate uncertainty model from the root cause. Details on sources of aleatoric and epistemic uncertainty and its propagation to final reconstructed coordinates is discussed. Certain practical implementation issues and possible solutions are also discussed. Conversely, in pipeline corrosion analysis, Bayesian methodologies are employed to model uncertainty derived from acquired data. This application involves leveraging Bayesian inference to understand, model, and mitigate uncertainties arising from environmental factors and material behaviors. Through these two applications, Bayesian methods prove instrumental in navigating and managing uncertainties in practical systems, offering insights, predictions, and solutions vital for decision-making and system optimization.

The growth of data capacity in optical communications links, which form the critical backbone of the modern internet, is facing a slowdown due to fundamental nonlinear limitations, leading to an impending "capacity crunch" on the horizon. Current technology has already exhausted degrees of freedom such as wavelength, amplitude, phase and polarization, leaving spatial multiplexing as the last available dimension to be efficiently exploited. To minimize the significant energy requirements associated with digital signal processing, it is critical to explore the upper limit of unmixed spatial channels in an optical fiber, which necessitates ideally packing spatial channels either in real space or in momentum space. The former strategy is realized by uncoupled multi-core fibers whose channel count has already saturated due to reliability constraint limiting fiber sizes. The later strategy is realized by the unmixed multimode fiber whose high spatial efficiency suggest the possibility of high channel-count scalability but the right subset of mode ought to be selected in order to mitigate mode coupling that is ever-present due to the plethora of perturbations a fiber normally experiences. The azimuthal modes in ring-core fibers turn out to be one of the most spatially efficient in this regard, by exploiting light’s orbital angular momentum (OAM). Unmixed mode counts have reached 12 in a ~1km fiber and 24 in a ~10m fiber. However, there is a fundamental bottleneck for scalability of conventionally bound modes and their relatively high crosstalks restricts their utility to device length applications. In this thesis, we provide a fundamental solution to further fuel the unmixed-channel count in an MMF. We utilize the phenomenon of topological confinement, which is a regime of light guidance beyond conventional cutoff that has, to the best of our knowledge, never been demonstrated till publications based on the subject matter of this thesis. In this regime, light is guided by the centrifugal barrier created by light’s OAM itself rather than conventional total internal reflection arising from the index inhomogeneity of the fiber. The loss of these topologically confined modes (TCMs) decreases down to negligible levels by increasing the OAM of fiber modes, because the centrifugal barrier that keeps photons confined to a fiber core increases with the OAM value of the mode. This leads to low-loss transmission in a km-scale fiber of these cutoff modes. Crucially, the mode-dependent confinement loss of TCMs further lifts the degeneracy of wavevectors in the complex space, leading to frustration of phase-matched coupling. This thus allows further scaling the mode count that was previously hindered by degenerate mode coupling in conventionally bound fiber modes. The frustrated coupling of TCMs thus enables a record amount of unmixed OAM modes in any type of fiber that features a high index contrast, whether specially structured as a ring-core, or simply constructed as a step-index fiber. Using all these favorable attributes, we achieve up to 50 low-loss modes with record low crosstalk (approaching -45 dB/km) over a 130-nm bandwidth in a ~1km-long ring-core fiber. The TCM effect promises to be inherently scalable, suggesting that even higher modes counts can be obtained in the future using this design methodology. Hence, the use of TCMs promises breaking the record spectral efficiency, potentially making it the choice for transmission links in future Space-Division-Multiplexing systems. Apart from their chief attribute of significantly increasing the information content per photon for quantum or classical networks, we expect that this new light guidance may find other applications such as in nonlinear signal processing and light-matter interactions.

Additive Manufacturing (AM), also known as industrial 3-D printing, is a manufacturing technique witnessing accelerating adoption across aerospace, medical, and tooling industries. Historically, AM has supported rapid prototyping, due to the ability to quickly build one-off components with large design freedom and relatively small investment cost for small scale batches. However, advancements in build speed and reductions in system cost are motivating the establishment of AM as a method for full-scale production. Opposed to traditional, subtractive techniques, the AM process creates material and geometry simultaneously, often with highly chaotic modes of material solidification. Specifically, powder bed fusion (PBF), the focus of this work, builds parts layer-by-layer from metallic powder, using high power-density lasers to melt or sinter each subsequent layer. Despite being the current de-facto standard for aerospace part production, PBF often embeds numerous AM-specific defects into components, such as lack of fusion defects, keyhole porosity, and spatter. Combined with the general AM build failure modes such as support structure failure and thermal deformation, part quality and reliability are among the primary factors inhibiting further adoption of PBF. The main objective of this work is to determine a method to improve part quality and reliability through real-time in-situ monitoring. A digital fringe projection (DFP) monitoring system is designed, constructed, and tested using structural health methodology. The DFP monitoring system will retrofit industrial 3-D printers to obtain full-field height-maps of each powder coating layer, a feature which is indicative of several AM-specific defect modes such as lack of fusion, support structure failure, and thermal deformation. A rigorous uncertainty modeling approach is developed to analyze impacts on height noise structures due to arbitrarily correlated pixel noise. A 3-D surrogate model is developed to analyze specific powder bed defect geometries, spatial filtering, and DFP monitoring system placement. Mounting fixtures and a calibration routine are demonstrated within an operational environment (EOS M 290 3-D printer). Finally, the effects of operational parameters of the DFP monitoring system are explored via a parametric detection study on lab-created powder defects. The initial development of this system will support the AM industry by creating a hardware and software framework which provides rich, real-time data sets indicative of part health to operators of industrial PBF printers.

Spatial augmented reality is a particular kind of augmented reality technique that uses projector to blend the real objects with virtual contents. Coincidentally, as a means of 3D shape measurement, structured light illumination makes use of projector as part of its system as well. It uses the projector to generate important clues to establish the correspondence between the 2D image coordinate system and the 3D world coordinate system. So it is appealing to build a system that can carry out the functionalities of both spatial augmented reality and structured light illumination. In this dissertation, we present all the hardware platforms we developed and their related applications in spatial augmented reality and structured light illumination. Firstly, it is a dual-projector structured light 3D scanning system that has two synchronized projectors operate simultaneously, consequently it outperforms the traditional structured light 3D scanning system which only include one projector in terms of the quality of 3D reconstructions. Secondly, we introduce a modified dual-projector structured light 3D scanning system aiming at detecting and solving the multi-path interference. Thirdly, we propose an augmented reality face paint system which detects human face in a scene and paints the face with any favorite colors by projection. Additionally, the system incorporates a second camera to realize the 3D space position tracking by exploiting the principle of structured light illumination. At last, a structured light 3D scanning system with its own built-in machine vision camera is presented as the future work. So far the standalone camera has been completed from the a bare CMOS sensor. With this customized camera, we can achieve high dynamic range imaging and better synchronization between the camera and projector. But the full-blown system that includes HDMI transmitter, structured light pattern generator and synchronization logic has yet to be done due to the lack of a well designed high speed PCB.

Detection, characterization and separation of chiral structures as one of the fundamental constituents of biomolecules are crucial in chemical and biological applications. Conventionally, chiroptical techniques such as circular dichroism (CD) are exploited to identify the material chirality. These methods require substantial amount of sample material with a high concentration. since chirality effect (as an indirect electric to magnetic coupling between incident light and sample material) is typically very weak. Therefore, the observation of chirality might be masked under the electric properties in chiroptical techniques if a substantial amount of material is not provided. Consequently, these techniques are not suitable for detection of chirality for nanoscale samples and can’t provide information about their primary or secondary structure. On the other hand, separation of enantiomers is of great interest in pharmaceutical applications. Classical separation techniques which are used to separate molecular compounds and complexes in achiral environment, such as liquid-liquid extraction, chromatography and distillation are not universal for separation of all types of chiral molecules/compounds since not all compounds and host media are chemically compatible. This dissertation focuses on introducing new techniques for detection, characterization and manipulation of chiral nanoparticles and nanostructures using structured light. Specifically, a new technique involving the measurement of photo-induced optical force (near-field characteristics) rather than the measurement of conventional scattered power (far-field characteristics) is introduced which substantially enhances the resolution in the detection of enantiomer type of chiral samples down to sub-100 nm. Furthermore, by utilizing structured illumination (circularly polarized wave for detection of transverse chirality and combination of azimuthally and radially polarized waves in the so-called azimuthally-radially polarized beam for detection of longitudinal chirality) the prediction of spatial features of the structure of a chiral sample is facilitated. It’s been demonstrated that circularly polarized illumination can’t differentiate between chirality and anisotropy. To overcome this deficiency and differentiate between anisotropy and chirality, an engineered combination of vortex beams (ARPB) is utilized. Lastly, a novel scheme to manipulate and control chiral nanoparticles and to sort enantiomers is proposed.

Imaging at the nanoscale and/or at remote locations holds great promise for studies in fields as disparate as the life sciences and materials sciences. One such microscopy technique, stimulated emission depletion (STED) microscopy, is one of several fluorescence based imaging techniques that offers resolution beyond the diffraction-limit. All current implementations of STED microscopy, however, involve the use of free-space beam shaping devices to achieve the Gaussian- and donut-shaped Orbital Angular Momentum (OAM) carrying beams at the desired colors –-- a challenging prospect from the standpoint of device assembly and mechanical stability during operation. A fiber-based solution could address these engineering challenges, and perhaps more interestingly, it may facilitate endoscopic implementation of in vivo STED imaging, a prospect that has thus far not been realized because optical fibers were previously considered to be incapable of transmitting the OAM beams that are necessary for STED. In this thesis, we investigate fiber-based STED systems to enable endoscopic nanoscale imaging. We discuss the design and characteristics of a novel class of fibers supporting and stably propagating Gaussian and OAM modes. Optimization of the design parameters leads to stable excitation and depletion beams propagating in the same fiber in the visible spectral range, for the first time, with high efficiency (>99%) and mode purity (>98%). Using the fabricated vortex fiber, we demonstrate an all-fiber STED system with modes that are tolerant to perturbations, and we obtain naturally self-aligned PSFs for the excitation and depletion beams. Initial experiments of STED imaging using our device yields a 4-fold improvement in lateral resolution compared to confocal imaging. In an experiment in parallel, we show the means of using q-plates as free-space mode converters that yield alignment tolerant STED microscopy systems at wavelengths covering the entire visible spectrum, and hence dyes of interest in such imaging schematics. Our study indicates that the vortex fiber is capable of providing an all-fiber platform for STED systems, and for other imaging systems where the exploitation of spatio-spectral beam shaping is required.

Multipolar electromagnetic phenomena in sub-wavelength resonators are at the heart of metamaterial science and technology. Typically, researchers engineer multipolar light-matter interactions by modifying the size, shape, and composition of the resonators. Here, we instead engineer multipolar interactions by modifying properties of the incident radiation. In this dissertation, we propose a new framework for determining the scattering response of resonators based on properties of the local excitation field.First, we derive an analytical theory to determine the scattering response of spherical nanoparticles under any type of illumination. Using this theory, we demonstrate the ability to drastically manipulate the scattering properties of a spherical nanoparticle by varying the illumination and demonstrate excitation of a longitudinal quadrupole mode that cannot be accessed with conventional illumination. Next, we investigate the response of dielectric dimer structures illuminated by cylindrical vector beams. Using finite-difference time-domain simulations, we demonstrate significant modification of the scattering spectra of dimer antennas and reveal how the illumination condition gives rise to these spectra through manipulation of electric and magnetic mode hybridization. Finally, we present a simple and efficient numerical simulation based on local field principles for extracting the multipolar response of any resonator under illumination by structured light. This dissertation enhances the understanding of fundamental light-matter interactions in metamaterials and lays the foundation for researchers to identify, quantify, and manipulate multipolar light-matter interactions through optical beam engineering.

Abstract: Extracting 3D geometry of an object from 2D images has been a popular topic in computer vision for decades. Many methods have been proposed to solve this problem with high accuracy and structured light methods are one of the most commonly used. Despite their high accuracy, there are limitations for existing methods. First, most existing methods can either be applied to obtain dense correspondences but limited to static scenes, or be applied to dynamic scenes but limited to sparse correspondences. Second, existing methods cannot be applied to scenes with global illuminations such as inter-reflection and projector defocus. Last but not least, existing structured light methods are rarely applied to underwater applications due to the difficulties of underwater camera calibration. Motivated by the above limitations, a new structured light method is presented to establish dense correspondences for dynamic scenes. Many simulated and real datasets are used to test the robustness of this method. Moreover, another novel structured light method is presented in this thesis to account for global illuminations such as inter-reflection, subsurface scattering and severe projector defocus. The experimental results demonstrate that the proposed method consistently outperforms two of the state-of-the-art methods. To address the limitation in underwater applications, two new underwater camera calibration methods are presented by using the physically correct refraction model. All the experimental results of applying these two methods are evaluated against the ground truth, and the simulated experiments are compared to the methods that produces the best results. The comparison demonstrates that our results are promising. The most significant advantage of these two methods is that no calibration object is required, which can be very difficult to access when the cameras are deployed undersea. Finally, a multi-camera multi-projector system is designed and developed to monitor the undersea habitat. Since 2014, the system has been deployed along the coast of B.C., collecting data that can be shared with researchers from around the world.

Human Motion capture has been an active area of research for many years and has applications in many fields such as gaming, entertainment, physical therapy, and ergonomics. Most commercially available motion capture systems use numerous markers to be placed on the body of the human subject requiring a significant setup time. In this dissertation, we develop the architecture and algorithms for markerless motion capture with a multi-view structured light system. In contrast to existing markerless approaches that use multiple camera streams, we reconstruct the scene by combining the views from three structured light stations using sinusoidal phase shift patterns, each equipped with one projector, a stereo pair of cameras for phase unwrapping, and a color camera. The three stations surround the subject and are time multiplexed to avoid interference. Phase-shifted sinusoidal patterns offer low decoding complexity, require as few as three projection frames per reconstruction, and are well suited for capturing dynamic scenes. In these systems, depth is reconstructed by determining the phase projected onto each pixel in the camera and establishing correspondences between camera and projector pixels. Typically, multiple periods are projected within the set of sinusoidal patterns, thus requiring phase unwrapping on the phase image before correspondences can be established. There are three novel contributions to this dissertation; first, we present a novel phase unwrapping algorithm across space and time in order to generate a temporally consistent point cloud. Specifically, we combine a quality guided phase unwrapping approach with absolute phase estimates from the stereo cameras to solve for the absolute phase of connected regions. Second, we develop a calibration method for multi-camera-projector systems in which sensors face each other as well as share a common viewpoint. We use a translucent planar sheet framed in PVC piping as a calibration target which is placed at multiple positions and orientations within a scene. In each position, the target is captured by the cameras while it is being illuminated by a set of patterns from various projectors. The translucent sheet allows the projected patterns to be visible from both sides, allowing correspondences between devices that face each other. The set of correspondences generated between the devices using this target are input into a bundle adjustment framework to estimate calibration parameters. Third, we develop algorithms to reconstruct dynamic geometry of a human subject using a template generated by the system itself. Specifically, we deform the template to each frame of the captured geometry by iteratively aligning each bone of the skeleton. This is done by searching for correspondences between the source template and the captured geometry, solving for rotation of bones, and enforcing constraints on each rotation to prevent the template from taking on anatomically unnatural poses. Once the geometry of the dynamic mesh is reconstructed, the template is textured using the color cameras from the multi-view structured-light system. We demonstrate the effectiveness of our approach both qualitatively and quantitatively for an actual sequence of a moving human subject by synthesizing arbitrary views of the dynamic scene.

The use of structured light illumination techniques for three-dimensional data acquisition is, in many cases, limited to stationary subjects due to the multiple pattern projections needed for depth analysis. Traditional Composite Pattern (CP) multiplexing utilizes sinusoidal modulation of individual projection patterns to allow numerous patterns to be combined into a single image. However, due to demodulation artifacts, it is often difficult to accurately recover the subject surface contour information. On the other hand, if one were to project an image consisting of many thin, identical stripes onto the surface, one could, by isolating each stripe center, recreate a very accurate representation of surface contour. But in this case, recovery of depth information via triangulation would be quite difficult. The method described herein, Modified Composite Pattern (MCP), is a conjunction of these two concepts. Combining a traditional Composite Pattern multiplexed projection image with a pattern of thin stripes allows for accurate surface representation combined with non-ambiguous identification of projection pattern elements. In this way, it is possible to recover surface depth characteristics using only a single structured light projection. The technique described utilizes a binary structured light projection sequence (consisting of four unique images) modulated according to Composite Pattern methodology. A stripe pattern overlay is then applied to the pattern. Upon projection and imaging of the subject surface, the stripe pattern is isolated, and the composite pattern information demodulated and recovered, allowing for 3D surface representation. In this research, the MCP technique is considered specifically in the context of a Hidden Markov Process Model. Updated processing methodologies explained herein make use of the Viterbi algorithm for the purpose of optimal analysis of MCP encoded images. Additionally, we techniques are introduced which, when implemented, allow fully automated processing of the Modified Composite Pattern image.

Phase measuring Profilometry (PMP) is an important technique used in 3D data acquisition. Many variations of the PMP technique exist in the research world. The technique involves projecting phase shifted versions of sinusoidal patterns with known frequency. The 3D information is obtained from the amount of phase deviation that the target object introduces in the captured patterns. Using patterns based on single frequency result in projecting a large number of patterns necessary to achieve minimal reconstruction errors. By using more than one frequency, that is multi-frequency, the error is reduced with the same number of total patterns projected as in the single frequency case. The first major goal of our research work is to minimize the error in 3D reconstruction for a given scan time using multiple frequency sine wave patterns. A mathematical model to estimate the optimal frequency values and the number of phase shift patterns based on stochastic analysis is given. Experiments are conducted by implementing the mathematical model to estimate the optimal frequencies and the number of patterns projected for each frequency level used. The reduction in 3D reconstruction errors and the quality of the 3D data obtained shows the validity of the proposed mathematical model. The second major goal of our research work is the implementation of a post-processing algorithm based on stereo correspondence matching adapted to structured light illumination. Composite pattern is created by combining multiple phase shift patterns and using principles from communication theory. Composite pattern is a novel technique for obtaining real time 3D depth information. The depth obtained by the demodulation of captured composite patterns is generally noisy compared to the multi-pattern approach. In order to obtain realistic 3D depth information, we propose a post-processing algorithm based on dynamic programming. Two different communication theory principles namely, Amplitude Modulation (AM) and Double Side Band Suppressed Carrier (DSBSC) are used to create the composite patterns. As a result of this research work, we developed a series of low-cost structured light scanners based on the multi-frequency PMP technique and tested them for their accuracy in different 3D applications. Three such scanners with different camera systems have been delivered to Toyota for vehicle assembly line inspection. All the scanners use off the shelf components. Two more scanners namely, the single fingerprint and the palmprint scanner developed as part of the Department of Homeland Security grant are in prototype and testing stages.

A miniaturized spatial light modulator (SLM)-based structured-light illumination module with optical fiber input is designed to generate a coded 256 x 256 spots pattern for 3-D areal mapping applications. The projector uses the light from a He-Ne laser coupled to a polarization-maintaining (PM) fiber to illuminate a specially made hologram so that four virtual point sources are regenerated. The interference pattern of the four sources are filtered and modulated by an SLM. The output intensity can thus be encoded to form any arbitrary pattern through the electronic input applied to the SLM with a high speed. In this thesis, a complete optical diffraction analysis of the system is presented to provide guidelines for the optimal design of the system parameters. Through the theoretical analysis for square beam array generation, the important parameters for fabricating a hologram are given. The final system optical design and arrangement based on optical analysis are described. The detailed opto-mechanical construction of the LIM and the associated alignment, the computer simulation and the preliminary test results of the developed LIM are also provided.