Please use this identifier to cite or link to this item:
Design and implementation of adaptable onboard digital image detection algorithms for aurora observations
|Lim_Chester_r.pdf||Version for non-UH users. Copying/Printing is not permitted||4.13 MB||Adobe PDF||View/Open|
|Lim_Chester_uh.pdf||Version for UH users||4.17 MB||Adobe PDF||View/Open|
|Title:||Design and implementation of adaptable onboard digital image detection algorithms for aurora observations|
|Authors:||Lim, Chester Nian Vee|
|Issue Date:||Aug 2011|
|Publisher:||[Honolulu] : [University of Hawaii at Manoa], [August 2011]|
|Abstract:||Aurora observations are moving towards utilizing the capabilities of near-infrared (NIR) cameras onboard high-altitude, long-duration balloons (LDB). Preliminary aurora observations using NIR cameras were initially conducted on the ground where scientists were able to continuously record data and perform processing . Continuously recording data requires a large storage capacity. Currently, only ground-based processing algorithms exist for NIR aurora observations. Moving observations from ground-based to balloon-based require real-time, on-board processing algorithms to optimize the use of the limited onboard storage space by reducing the massive amount of collected data to be stored.|
The purpose of this thesis is to address the problem of reducing the on-board storage requirement while achieving high aurora detection rate, thus optimizing the use of the limited onboard storage space. The thesis applies ISAAC technology  that was developed at the NASA Jet Propulsion Laboratory (JPL) to the entire life cycle of building an on-board aurora image processing system in the context of balloon-borne payload. It includes an exploration of high-level on-board processing digital system architecture for a balloon-based aurora observations, development of real-time, adaptable, on-board, aurora image processing algorithms, prototyping a hardware/software codesign on ISAAC's FPGA-based platform for these algorithms, and performance evaluation of these algorithms through a series of metrics.
The outcomes of this thesis are a modular and reusable library of adaptable and fully functional aurora detection and data reduction algorithms that can be configured on a perapplication basis in response to different requirements of aurora observation.
|Description:||M.S. University of Hawaii at Manoa 2011.|
Includes bibliographical references.
|Appears in Collections:||M.S. - Electrical Engineering|
Items in ScholarSpace are protected by copyright, with all rights reserved, unless otherwise indicated.