ORCID

0009-0003-6037-2834

Keywords

infrared imaging, system design, target detection, CNN object detection

Abstract

The infrared spectrum contains various sub-bands (from near infrared (NIR) to long-wave infrared (LWIR), which are specialized at different detection tasks. Objects that are well camouflaged to human vision can be highly noticeable in NIR imagery. For vision navigation, the varying illumination condition and the cluttered urban background poses challenges in the visible band but can easily be resolved by using a thermal imager operating in LWIR. Despite this potential, practical deployment remains limited by the lack of suitable platforms, task-specific datasets, and long-duration field validation. This dissertation addresses these gaps through the design and evaluation of two infrared sensing systems mounted on moving platforms, each supported by a custom machine-vision object detector. The first is an NIR-based system developed for detecting invasive Burmese pythons in marshy South Florida environments, with active illumination enabling nighttime operation. The second is an LWIR-based system for detecting towers as persistent landmarks to establish vehicle geolocation in GNSS-denied conditions. Tower-based absolute geolocation measurements correct drifts that accumulate in inertial navigation and visual-odometry-based estimates over time. Because both applications involve specialized sensing conditions and target characteristics, custom object detectors are required rather than off-the-shelf models. In both systems, the trained detectors achieve low false positive rates, supporting reliable long-duration deployment. Together, these system designs establish a practical foundation for infrared-enabled target detection and navigation in cluttered environments and restrictive conditions.

Completion Date

2026

Semester

Spring

Committee Chair

Renshaw, C. Kyle

Degree

Doctor of Philosophy (Ph.D.)

College

College of Optics and Photonics

Format

PDF

Document Type

Dissertation

Identifier

DP0053198

Download

Share

COinS
 

Accessibility Statement

This item was created or digitized prior to April 24, 2027, or is a reproduction of legacy media created before that date. It is preserved in its original, unmodified state specifically for research, reference, or historical recordkeeping. In accordance with the ADA Title II Final Rule, the University Libraries provides accessible versions of archival materials upon request. To request an accommodation for this item, please submit an accessibility request form.