Search

This thesis presents a series of neural network classifiers for object detection in both optical and infrared images. The focus of this work is on efficient and accurate solutions. The thesis discusses the evolution of the highly efficient and tiny network Binary Classification Vision Transformer (BC-ViT) and how through thoughtful modifications and improvements, the BC-ViT can be utilized for tasks of increasing complexity. Chapter 2 discusses the creation of BC-ViT and its initial use case for underwater image classification of optical images. The BC-ViT is able to complete its task with an accuracy of 99.29\% while being comprised of a mere 15,981 total trainable parameters. Chapter 3, Waste Multi-Class Vision Transformer (WMC-ViT), introduces the usefulness of mindful algorithm design for the realm of multi-class classification on a mutually exclusive dataset. WMC-ViT shows that the task oriented design strategy allowed for a network to achieve an accuracy score of 94.27\% on a five class problem while still maintaining a tiny parameter count of 35,492. The final chapter demonstrates that by utilizing functional blocks of BC-ViT, a simple and effective target detection algorithm for infrared images can be created. The Edge Infrared Vision Transformer (EIR-ViT) showed admirable results with a high IoU of 79\% coupled with a massive reduction in parameters compared to current state of the art solutions at roughly 120,000. This thesis ends with summary of Chapters 2, 3, and 4 that further discusses the implications this design strategy offers.

Fluviokarst landscapes are dominated by both fluvial and karst features. Interpreting hydrologic pathways of fluviokarst can be confounded by the unknown connectivity of the various flow regimes. A combined discrete-continuum (CDC) hybrid numeric model for simulating the surface and subsurface hydrology and hydraulics in fluviokarst basins was formulated to investigate fluviokarst pathways. This model was applied to the Cane Run Royal Springs basin in Kentucky USA. A priori constraints on parameterization were avoided via multi-stage optimization utilizing Sobol sequencing and high performance computing. Modelling results provide evidence of hydrologic pathways dominated by fracture flow, epikarst transfer and runoff. Fractures in karst basins with high fracture-matrix permeability ratios may influence both springflow and streamflow. Swallet features can be as important as spring features as they are sink features in streamflow during hydrologic events. Inflections in spring hydrographs represent shifts in the surface-subsurface connectivity via the fractures, as opposed to shifts in dominant storage zones. Existing methods of dual- and triunal hydrograph separation of karst springflow may not be directly transferrable to fluviokarst springs. The numerical model herein has advantages of suggesting dominant pathways in complex terrane and highlighting unforeseen surface-subsurface connectivity. However, disadvantages include computational expense and previous site studies.