Dissertation defense: Christian Pattyn

March 5, 2025

Student Name: Christian Pattyn
Program: PhD NSME
Date: Thursday 3/13/25
Time: 1:30 pm
Place: CHTM 101
Committee Chair: Dr. Nathan Jackson

Targetless Characterization of Optical Blurring from Microphysical Scattering

by Christian Alexander Pattyn

B.Sc. Chemical Engineering, University of New Mexico, 2020
M.Sc. Biomedical Engineering, University of New Mexico, 2020
Ph.D. Nanoscience & Microsystems Engineering, University of New Mexico, 2025

Abstract

Atmospheric aerosols, such as fog, degrade optical signals via the scattering and absorption of light, making it difficult to recover information from our surroundings. The degraded visual environments (DVEs) created by these aerosols affect both the intensity and resolution in imaging systems. There are many modern solutions which can assist in signal recovery in these environments, but they all require information about the environment as an input. Current DVE characterization techniques typically rely on either high SWaP equipment which characterizes the aerosol or prior knowledge of the observed environment for comparison to degraded images. Because of these requirements, neither of these approaches are well-suited to widespread deployment on consumer vehicles, such as aircraft or automobiles. There is a clear need for a versatile, low SWaP technique which can characterize DVEs in a manner that is useful for current deblurring algorithms.

To address this need, I have developed a comparatively low SWaP technique which measures the angular distribution of light scattered by atmospheric aerosols. I demonstrate improved recovery of high spatial frequency optical information through atmospheric aerosols (fog) by characterizing the point spread function (PSF) of the aerosol environment. To do so I first pass a collimated beam through an experimentally generated fog-analogue. An f-theta lens then collects the scattered light and transforms its angular scattering profile into the linear domain, mapping the beam onto a focal plane. This angular map can then be used to calculate the anticipated PSF of the beam at an arbitrary stand-off distance. I then use the modelled PSF to deblur images of a resolution target taken simultaneously with the angular scattering measurement using deconvolution. My findings demonstrate that this approach can increase the resolution of the blurred images by much as 26% (from 3.17 𝑙𝑝𝑚𝑚⁄ to 4.00 𝑙𝑝𝑚𝑚⁄). I present this work as a low SWaP, low complexity method to enhance the resolution limit of images blurred by aerosol DVEs.