The task of object detection and pose estimation has widely been done using template matching techniques. However, these algorithms are sensitive to outliers and occlusions, and have high latency due to their iterative nature. Recent research in computer vision and deep learning has shown great improvements in the robustness of these algorithms. However, one of the major drawbacks of these algorithms is that they are specific to the objects. Moreover, the estimation of pose depends significantly on their RGB image features. As these algorithms are trained on meticulously labeled large datasets for object's ground truth pose, it is difficult to re-train these for real-world applications. To overcome this problem, we propose a two-stage pipeline of convolutional neural networks which uses RGB images to localize objects in 2D space and depth images to estimate a 6DoF pose. Thus the pose estimation network learns only the geometric features of the object and is not biased by its color features. We evaluate the performance of this framework on LINEMOD dataset, which is widely used to benchmark object pose estimation frameworks. We found the results to be comparable with the state of the art algorithms using RGB-D images. Secondly, to show the transferability of the proposed pipeline, we implement this on ATLAS robot for a pick and place experiment. As the distribution of images in LINEMOD dataset and the images captured by the MultiSense sensor on ATLAS are different, we generate a synthetic dataset out of very few real-world images captured from the MultiSense sensor. We use this dataset to train just the object detection networks used in the ATLAS Robot experiment.

At most of the US signal, pedestrian walk timings run in concurrence with relevant vehicular traffic signal phases which means that major-street coordinated operations can be interrupted by a pedestrian call. Such interruption may increase delays and stops for major traffic flows. An alternative to this design is to increase the cycle length and embed pedestrian timings within the ring-barrier structure of the prevailing coordination plan. Both approaches have advantages and disadvantages. This study attempts a novel approach to address this situation by a comprehensive experimental evaluation of traffic performance under various pedestrian signal timing strategies. Findings show that either abovementioned approach works well for very low traffic demands. When the traffic demand increases findings cannot be generalized as they differ for major coordinated movements versus overall network performance. While coordinated movements prefer no interruption of the coordinated operations, the overall network performance is better in the other case. / Includes bibliography. / Thesis (M.S.)--Florida Atlantic University, 2018. / FAU Electronic Theses and Dissertations Collection

Diffusion tensor imaging (DTI) is a relatively new technology of magnetic resonance imaging, which enables us to observe the insight structure of the human body in vivo and non-invasively. It displays water molecule movement by a 3×3 diffusion tensor at each voxel. Tensor field processing, visualisation and tractography are all based on the diffusion tensors. The accuracy of estimating diffusion tensor is essential in DTI. This research focuses on exploring the potential improvements at the tensor estimation of DTI. We analyse the noise arising in the measurement of diffusion signals. We present robust methods, least median squares (LMS) and least trimmed squares (LTS) regressions, with forward search algorithm that reduce or eliminate outliers to the desired level. An investigation of the criterion to detect outliers is provided in theory and practice. We compare the results with the generalised non-robust models in simulation studies and applicants and also validated various regressions in terms of FA, MD and orientations. We show that the robust methods can handle the data with up to 50% corruption. The robust regressions have better estimations than generalised models in the presence of outliers. We also consider the multiple tensors problems. We review the recent techniques of multiple tensor problems. Then we provide a new model considering neighbours' information, the Bayesian single and double tensor models using neighbouring tensors as priors, which can identify the double tensors effectively. We design a framework to estimate the diffusion tensor field with detecting whether it is a single tensor model or multiple tensor model. An output of this framework is the Bayesian neighbour (BN) algorithm that improves the accuracy at the intersection of multiple fibres. We examine the dependence of the estimators on the FA and MD and angle between two principal diffusion orientations and the goodness of fit. The Bayesian models are applied to the real data with validation. We show that the double tensors model is more accurate on distinct fibre orientations, more anisotropic or similar mean diffusivity tensors. The final contribution of this research is in covariance tensor estimation. We define the median covariance matrix in terms of Euclidean and various non-Euclidean metrics taking its symmetric semi-positive definiteness into account. We compare with estimation methods, Euclidean, power Euclidean, square root Euclidean, log-Euclidean, Riemannian Euclidean and Procrustes median tensors. We provide an analysis of the different metric between different median covariance tensors. We also provide the weighting functions and define the weighted non-Euclidean covariance tensors. We finish with manifold-valued data applications that improve the illustration of DTI images in tensor field processing with defined non-weighted and weighted median tensors. The validation of non-Euclidean methods is studied in the tensor field processing. We show that the root square median estimator is preferable in general, which can effectively exclude outliers and clearly shows the important structures of the brain. The power Euclidean median estimator is recommended when producing FA map.

by Shiu-kwok Lau. / Bibliography: leaves 41-43 / Thesis (M.Ph.)--Chinese University of Hong Kong, 1985

Siu-man Ng. / Thesis (M.Ph.)--Chinese University of Hong Kong, 1987. / Bibliography: leaves 44-47.

by Ming-long Lam. / Thesis (M.Ph.)--Chinese University of Hong Kong, 1987. / Bibliography: leaves 41-43.

by Chiu Yiu Ming. / Thesis (M.Ph.)--Chinese University of Hong Kong, 1988. / Bibliography: leaves 127-129.

by Linda Hoi-ying Yau. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1991. / Includes bibliographical references. / Chapter Chapter 1 --- Preliminary / Chapter § 1.1 --- Introduction page --- p.1 / Chapter § 1.2 --- Notations page --- p.3 / Chapter Chapter 2 --- Multilevel Analysis of Structural Equation Models with Multivariate Normal Distribution / Chapter § 2.1 --- The Multilevel Structural Equation Model page --- p.4 / Chapter § 2.2 --- "First Stage Estimation of and Σkmkm-1---ki+1wo for i=1,...,m-1 page" --- p.5 / Chapter § 2:3 --- Second Stage Estimation of Structural Parameters page --- p.10 / Chapter Chapter 3 --- Generalization to Arbitrary and Elliptical Distributions / Chapter § 3.1 --- Asymptotically Distribution-Free Estimation page --- p.25 / Chapter § 3.2 --- Elliptical Distribution Estimation page --- p.30 / Chapter Chapter 4 --- Artificial Examples / Chapter § 4.1 --- Examples on Multivariate Normal Distribution Estimation Page --- p.34 / Chapter § 4.2 --- Examples on Elliptical Distribution Estimation page --- p.40 / Chapter §4.3 --- Findings and Summary Page --- p.42 / Chapter Chapter 5 --- Conclusion and Discussion page --- p.44 / References page --- p.47 / Figure 1 page --- p.49 / Appendices page --- p.50 / Tables Page --- p.59

by Mow Wai Ho. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1991. / Bibliographies: leaves 98-104. / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Channel Model and Other Basic Assumptions --- p.5 / Chapter 1.2 --- Complexity Measure --- p.8 / Chapter 1.3 --- Maximum Likelihood Sequence Estimator --- p.9 / Chapter 1.4 --- The Viterbi Algorithm ´ؤ An Implementation of MLSE --- p.11 / Chapter 1.5 --- Error Performance of the Viterbi Algorithm --- p.14 / Chapter 1.6 --- Suboptimal Viterbi-like Algorithms --- p.17 / Chapter 1.7 --- Trends of Digital Transmission and MLSE --- p.19 / Chapter 2 --- New Formulation of MLSE --- p.21 / Chapter 2.1 --- The Truncated Viterbi Algorithm --- p.21 / Chapter 2.2 --- Choice of Truncation Depth --- p.23 / Chapter 2.3 --- Decomposition of MLSE --- p.26 / Chapter 2.4 --- Lattice Interpretation of MLSE --- p.29 / Chapter 3 --- The Closest Vector Problem --- p.34 / Chapter 3.1 --- Basic Definitions and Facts About Lattices --- p.37 / Chapter 3.2 --- Lattice Basis Reduction --- p.40 / Chapter 3.2.1 --- Weakly Reduced Bases --- p.41 / Chapter 3.2.2 --- Derivation of the LLL-reduction Algorithm --- p.43 / Chapter 3.2.3 --- Improved Algorithm for LLL-reduced Bases --- p.52 / Chapter 3.3 --- Enumeration Algorithm --- p.57 / Chapter 3.3.1 --- Lattice and Isometric Mapping --- p.58 / Chapter 3.3.2 --- Enumerating Points in a Parallelepiped --- p.59 / Chapter 3.3.3 --- Enumerating Points in a Cube --- p.63 / Chapter 3.3.4 --- Enumerating Points in a Sphere --- p.64 / Chapter 3.3.5 --- Comparisons of Three Enumeration Algorithms --- p.66 / Chapter 3.3.6 --- Improved Enumeration Algorithm for the CVP and the SVP --- p.67 / Chapter 3.4 --- CVP Algorithm Using the Reduce-and-Enumerate Approach --- p.71 / Chapter 3.5 --- CVP Algorithm with Improved Average-Case Complexity --- p.72 / Chapter 3.5.1 --- CVP Algorithm for Norms Induced by Orthogonalization --- p.73 / Chapter 3.5.2 --- Improved CVP Algorithm using Norm Approximation --- p.76 / Chapter 4 --- MLSE Algorithm --- p.79 / Chapter 4.1 --- MLSE Algorithm for PAM Systems --- p.79 / Chapter 4.2 --- MLSE Algorithm for Unimodular Channel --- p.82 / Chapter 4.3 --- Reducing the Boundary Effect for PAM Systems --- p.83 / Chapter 4.4 --- Simulation Results and Performance Investigation for Example Channels --- p.86 / Chapter 4.5 --- MLSE Algorithm for Other Lattice-Type Modulation Systems --- p.91 / Chapter 4.6 --- Some Potential Applications --- p.92 / Chapter 4.7 --- Further Research Directions --- p.94 / Chapter 5 --- Conclusion --- p.96 / Bibliography --- p.104

by Kwan-Moon Leung. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1990. / Bibliography: leaves 77-79. / Chapter Chapter 1 --- Introduction --- p.1 / Chapter Chapter 2 --- Estimation of the Model with Some Polytomous Entries Missed --- p.5 / Chapter §2.1 --- The Model --- p.5 / Chapter §2.2 --- Full Maximum Likelihood (FML) Estimation --- p.7 / Chapter Chapter 3 --- Estimation of the Model with Some Continuous and Polytomous Entries Missed --- p.13 / Chapter §3.1 --- The Model --- p.13 / Chapter §3.2 --- Pseudo Maximum Likelihood (PsML) Estimation --- p.15 / Chapter Chapter 4 --- Indirect Methods --- p.19 / Chapter §4.1 --- Listwise Deletion Method --- p.19 / Chapter §4.2 --- Mean Imputation Method --- p.19 / Chapter §4.3 --- Regression Imputation Method --- p.20 / Chapter Chapter 5 --- Computation of the Estimates --- p.23 / Chapter §5.1 --- Optimization Procedure --- p.23 / Chapter §5.2 --- Starting Value and Gradient Vector of the Model with Some Polytomous Entries Missed --- p.25 / Chapter §5.3 --- Starting Value and Gradient Vector of the Model with Some Continuous and Polytomous Entries Missed --- p.29 / Chapter Chapter 6 --- Partition Maximum Likelihood (PML) Estimation --- p.35 / Chapter §6.1 --- Motivation --- p.35 / Chapter §6.2 --- PML Procedure of the Model with Some Polytomous Entries Missed --- p.35 / Chapter §6.3 --- PML Procedure of the Model with Some Continuous and Polytomous Entries Missed --- p.37 / Chapter Chapter 7 --- Simulation Studies and Comparison --- p.39 / Chapter §7.1 --- Simulation Study I --- p.39 / Chapter §7.2 --- Simulation Study II --- p.44 / Chapter Chapter 8 --- Summary and Discussion --- p.43 / Tables / Appendix / References